DETAILED ACTION
Status of Claims
The following is a Final Office Action in response to applicant’s Request for Continued Examination filed on 06/07/2026.
Claims 1, 3, 4, 6, 9, 9, 11, 13, 14, and 17-20 are amended. Claims 2, 5, 8, 12, and 16 are cancelled. Claims 1, 3, 4, 6, 7, 9- 11, 13-15, and 17-20 are considered in this Office Action. Claims 1, 3, 4, 6, 7, 9-11, 13-15, and 17-20 are currently pending.
Response to Amendments
Applicant’s amendments necessitated the new ground(s) of rejection set forth in this Office Action.
Applicant’s amendments with respect to the 35 U.S.C. §112(a) rejection to claims have been considered, applicant’s amendments overcome the 35 USC 112(a) rejection. The rejection is therefore withdrawn.
Applicant’s arguments and amendments with respect to the 35 U.S.C. §103 rejection to claims have been considered, however they are primarily raised in light of applicant’s amendments and therefore is moot. An updated the 35 U.S.C. §103 rejection will address applicant’s amendment.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The text of those sections of Title 35, U.S. Code not included in this action can be found in a prior Office action.
The factual inquiries set forth in Graham v. John Deere Co., 383 U.S. 1, 148 USPQ 459 (1966), that are applied for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claims 1, 11, and 19 are rejected under 35 U.S.C. 103 as being unpatentable over in view of Gupta (US 20140207298 A1, hereinafter “Gupta”) in view of Pavlovski (WO 2016069330 A1, hereinafter “Pavlovski”) in view of Marhoefer (US 20120130556 A1, hereinafter “Marhoefer”).
Claim 1
Gupta teaches:
A method of […]the method comprising, by a computer:([0093] The energy disaggregation on energy use data collected by utilities for all customers, while [0026] disaggregation models used to derive solar capacity for a specific home by reviewing and analyzing a historical net power signature of the home. [0069] a series of a utility 610 providing electricity or other services to a home 670. As the energy enters the home 670, a ZigBee or other HAN 620 may measure the energy provided. [0042] disaggregating low frequency energy consumption data that includes solar panel generation; net power signature represents the net usage of the home, including any contribution from solar panels. [0028] Data may be obtained and/or accessed in various manners. For example, a current clamp (CT clamp) may be utilized. The use of two (2) CT clamps may generally be required, with one CT claim positioned at or proximate to the net meter (which may indicate net power draw for the house), and a second CT clamp positioned at or proximate to the solar system (which may indicate power captured and contributed by the solar system). [0041] When using low frequency whole-house energy consumption data, the energy contribution of solar panels must be determined and disaggregated. Such disaggregation may be based upon meteorological data. [0048] Next, a regression model may be trained with weather data and the number of hours from sunrise to sunset as one or more independent variables, and solar intensity as the dependent variable. [0027] energy disaggregation may be determined and/or impacted by the type of data and/or how the data is obtained or accessed. For example, data types may include power signals, or meteorological data or conditions. Power signals may be obtained in low frequency or high frequency samples. Low frequency data may be sampled--for example--hourly, while high or higher frequency data may be sampled--for example--each minute. Meteorological data or conditions may include information such as, but not limited to, (i) skycover or cloud cover (which may be set forth as a percentage or ratio of cover to clear sky); (ii) temperature; (iii) wind-speed; (iv) dew point; and (v) sunrise/sunset times),
wherein the two or more dual consumers are identified by joint processing timestamped data informative of individual grid power consumption GPCs of the consumers from the plurality of consumers and timestamped data informative of the one or more weather conditions([0027] Techniques for energy disaggregation may be determined and/or impacted by the type of data and/or how the data is obtained or accessed. For example, data types may include power signals, or meteorological data or conditions. Power signals may be obtained in low frequency or high frequency samples. Low frequency data may be sampled--for example--hourly, while high or higher frequency data may be sampled--for example--each minute. Meteorological data or conditions may include information such as, but not limited to, (i) skycover or cloud cover (which may be set forth as a percentage or ratio of cover to clear sky); (ii) temperature; (iii) wind-speed; (iv) dew point; and (v) sunrise/sunset times. [0041] Such disaggregation may be based upon meteorological data, [0093] The energy disaggregation on energy use data collected by utilities for all customers);
wherein a GPC of a given consumer from the plurality of consumers is obtained by measuring by a power consumption meter associated therewith([0028] energy usage data may be obtained from Green Button (an industry effort to provide transparent energy usage data, which is generally provided in hourly intervals); from Smart Meters--for example using a Smart Meter Home Area Network channel);
and wherein the two or more dual consumers are identified with no measurements and without using prior knowledge of their individual alternative power production([0025] predicting solar output on unseen homes using training data from different locations around the world. This predictive model may be applicable to locations other than where it was trained. For example, a model may be trained on the west coast of the United States (e.g., California), but may be used to predict solar output on the east coast of the United States (e.g., Connecticut). [0026] disaggregation models may be used to derive solar capacity for a specific home by reviewing and analyzing a historical net power signature of the home. Such models may not require any special hardware to be installed at or on the home to predict such solar capacity. [0106] Moreover, the energy use data for a target set of participating homes (where each home needed to sign up and is known) or a participating region (where each house did not need to sign up and could have still participated in the program) may be compared before and after the program);
separately for each of the identified two or more dual consumers, using a trained Forecasting Machine Learning (FML) Model to provide individual forecast of alternative power production by connected alternative power source([0042] disaggregating low frequency energy consumption data that includes solar panel generation, where net power signature represents the net usage of the home, including any contribution from solar panels. FIG. 3 indicates an exemplary solar power signal over the same three (3) day period. It can be seen from the net power curve in FIG. 2 that the net power becomes negative between sunrise and sunset. Similarly, in FIG. 3 it can be seen that the solar power signal is generally represented by a single major curve per day, between sunrise and sunset. [0048]a regression model may be trained with weather data and the number of hours from sunrise to sunset as one or more independent variables, and solar intensity as the dependent variable. [0049] Solar intensity may be seen as the normalized version of solar generation, and may be stated in the range from 0 to 1. Normalization of the dependent variable may be desirable when using a regression model, because it generally permits or allows the model to be easily trained. In accordance with some embodiments of the present invention, a radial basis function (RBF) support vector machine combined with RBF neural networks may be used. [0051] models learned based upon data collected over a year for one hundred (100) homes, and were tested upon approximately twenty-five (25) homes to confirm results. [0053] an exemplary flow for training and predicting solar generation based upon low frequency consumption data. Based upon both the solar intensity and the solar capacity, at 460 the solar generation may be predicted. [0054] FIG. 5, an example of a predicted solar panel generation and ground truth generation for a specific home),
wherein the FML model is trained on historical individual GPCs and weather conditions data to forecast the alternative power production ([0026] disaggregation models may be used to derive solar capacity for a specific home by reviewing and analyzing a historical net power signature of the home. [0048]a regression model may be trained with weather data and the number of hours from sunrise to sunset as one or more independent variables, and solar intensity as the dependent variable. [0053] an exemplary flow for training and predicting solar generation based upon low frequency consumption data. Based upon both the solar intensity and the solar capacity, at 460 the solar generation may be predicted. [0054] FIG. 5, an example of a predicted solar panel generation and ground truth generation for a specific home);
and controlling thermostat set-point change in a set of points connected to the grid([0071] With regard to utility demand response concerns, a utility may send a peak demand reduction signal on a peak usage day either using ZigBee or WiFi. The PCT may then cut back the energy usage by reducing the cooling/heating cycles or relax the thermostat set point by a few degrees).
While Gupta teaches [0042] disaggregating low frequency energy consumption data that includes solar panel generation, where net power signature represents the net usage of the home, including any contribution from solar panels. FIG. 3 indicates an exemplary solar power signal over the same three (3) day period. It can be seen from the net power curve in FIG. 2 that the net power becomes negative between sunrise and sunset. Similarly, in FIG. 3 it can be seen that the solar power signal is generally represented by a single major curve per day, between sunrise and sunset. [0048]a regression model may be trained with weather data and the number of hours from sunrise to sunset as one or more independent variables, and solar intensity as the dependent variable. [0049] Solar intensity may be seen as the normalized version of solar generation, and may be stated in the range from 0 to 1. Normalization of the dependent variable may be desirable when using a regression model, because it generally permits or allows the model to be easily trained. In accordance with some embodiments of the present invention, a radial basis function (RBF) support vector machine combined with RBF neural networks may be used. [0051] models learned based upon data collected over a year for one hundred (100) homes and were tested upon approximately twenty-five (25) homes to confirm results. [0053] an exemplary flow for training and predicting solar generation based upon low frequency consumption data. Based upon both the solar intensity and the solar capacity, at 460 the solar generation may be predicted. [0054] FIG. 5, an example of a predicted solar panel generation and ground truth generation for a specific home. [0071] With regard to utility demand response concerns, a utility may send a peak demand reduction signal on a peak usage day either using ZigBee or WiFi. The PCT may then cut back the energy usage by reducing the cooling/heating cycles or relax the thermostat set point by a few degrees. Gupta does not explicitly teach the following. However analogues reference in the field of energy predication model, Pavlovski teaches:
managing an electrical power grid operatively connected to a plurality of consumers in a geographical area, the plurality of consumers including dual consumers and non-dual consumers([0003] Load forecasting in a utility grid is an integral part of energy management. Growing penetration of distributed energy resources such as solar photovoltaic ("PV") power plants, wind power plants, and power storage plants has changed conventional utility practices for generating, transmitting, and distributing electric power. [0004] A group of power plants, electrical energy consumption devices, and associated infrastructure spread over a geographical area may be referred to as a utility grid. [0011] defining two or more load forecast zones, each load forecast zone being associated with a load profile type and a climate zone type; assigning each of the one or more loads to one of the two or more load forecast zones based on the load profile type and the climate zone type associated with the load),
train a model in accordance with a forecast of the one or more weather conditions in the geographical area(Pavlovski teaches[0051]The net load forecasting system 500 is configured to forecast net load at least partially based on the at least one forecasting variable provided by the environment forecasting component 750. Typical environmental forecasting variables predicted by the environmental forecasting component 750 include weather forecasts, storm warnings, wind speed, air density, air turbidity, irradiance, atmospheric turbulence, rain conditions, snow conditions, air temperature, and humidity. the intermittent distributed energy resource forecasting component 530 may anticipate the future weather conditions at the site of an intermittent distributed energy resource 200 within a selected forecast horizon.);
using the individual forecasts for the two or more dual consumers to provide a total forecast of total alternative power production for a group ofthe identified dual consumers(Pavlovski teaches [0043] forecasts electric power generation by intermittent distributed energy resources 200 within the load forecast zones 900 individually, or in clusters. [0041] The net load forecasting component 550 is configured to produce net load forecasts for the utility grid 100, or any portion thereof, based on the data generated by the intermittent distributed energy resource forecasting component 530 and the load forecasting component 540, while [0046]providing independent forecasts for the distributed energy resources and for the load groups, and combining these forecasts into a single net load forecast for the load forecast zones 900);
and using the provided total forecast of total alternative power production to trigger one or more management actions with regard to power production in the electrical power grid(Pavlovski teaches [0043] forecasts electric power generation by intermittent distributed energy resources 200 within the load forecast zones 900 individually, or in clusters. [0041] The net load forecasting component 550 is configured to produce net load forecasts for the utility grid 100, or any portion thereof, based on the data generated by the intermittent distributed energy resource forecasting component 530 and the load forecasting component 540, while [0046]providing independent forecasts for the distributed energy resources and for the load groups, and combining these forecasts into a single net load forecast for the load forecast zones 900, while [0063] the energy management system 120 is connected to the net load forecasting system 500 and is configured to request net load forecasts from the net load forecasting system 500 to maintain operating conditions in the utility grid 100 within a desired range).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the system for power distribution grid management of Gupta with Pavlovski to include managing an electrical power grid operatively connected to a plurality of consumers in a geographical area, the plurality of consumers including dual consumers and non-dual consumers, train a model in accordance with a forecast of the one or more weather conditions in the geographical area, using the individual forecasts for the two or more dual consumers to provide a total forecast of total alternative power production for a group of
While Gupta teaches [0042] disaggregating low frequency energy consumption data that includes solar panel generation, where net power signature represents the net usage of the home, including any contribution from solar panels. FIG. 3 indicates an exemplary solar power signal over the same three (3) day period. It can be seen from the net power curve in FIG. 2 that the net power becomes negative between sunrise and sunset. Similarly, in FIG. 3 it can be seen that the solar power signal is generally represented by a single major curve per day, between sunrise and sunset. [0048]a regression model may be trained with weather data and the number of hours from sunrise to sunset as one or more independent variables, and solar intensity as the dependent variable. [0049] Solar intensity may be seen as the normalized version of solar generation, and may be stated in the range from 0 to 1. Normalization of the dependent variable may be desirable when using a regression model, because it generally permits or allows the model to be easily trained. In accordance with some embodiments of the present invention, a radial basis function (RBF) support vector machine combined with RBF neural networks may be used. [0051] models learned based upon data collected over a year for one hundred (100) homes and were tested upon approximately twenty-five (25) homes to confirm results. [0053] an exemplary flow for training and predicting solar generation based upon low frequency consumption data. Based upon both the solar intensity and the solar capacity, at 460 the solar generation may be predicted. [0054] FIG. 5, an example of a predicted solar panel generation and ground truth generation for a specific home. [0071] With regard to utility demand response concerns, a utility may send a peak demand reduction signal on a peak usage day either using ZigBee or WiFi. The PCT may then cut back the energy usage by reducing the cooling/heating cycles or relax the thermostat set point by a few degrees. Gupta does not explicitly teach the following. However analogues reference in the field of energy predication model, Marhoefer teaches:
wherein the one or more management actions comprise providing commands for automatically power distributing between the power grid and a power storage facility(Marhoefer: [0059] The results of these calculations are a series of instructions and override commands 122 transmitted to the thin-client device/gateway 106. These instructions and override commands create a set of digital signals 130 transmitted to the switch/actuator (generally contained within a battery-based inverter) 105 that control the flow of electricity among the grid, solar panels, storage device and building load, as well as charging, idling and discharging the storage device.),
wherein the one or more management actions comprise issuing by the computer at least one command related to at least one of: charging/discharging one or more batteries connected to the grid ([0059]The results of these calculations are a series of instructions and override commands 122 transmitted to the thin-client device/gateway 106. These instructions and override commands create a set of digital signals 130 transmitted to the switch/actuator (generally contained within a battery-based inverter) 105 that control the flow of electricity among the grid, solar panels, storage device and building load, as well as charging, idling and discharging the storage device).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the system for power distribution grid management of Gupta and Pavlovski with Marhoefer to include the one or more management actions comprise providing commands for automatically power distributing between the power grid and a power storage facility and the one or more management actions comprise issuing by the computer at least one command related to at least one of: charging/discharging one or more batteries connected to the grid. Doing so would aid in managing the implementation, inclusion and aggregation of distributed renewable energy sources and energy storage in the electric grid in a way that maximizes and optimizes energy distribution, and hence value for asset owners, aggregators, utilities and regional transmission organizations. [0002].
Claim 11/19
Gupta teaches:
One or more computers comprising processors ([0008] processor; network; and figure 6) a system capable of […] the operations comprising::([0093] The energy disaggregation on energy use data collected by utilities for all customers, while [0026] disaggregation models used to derive solar capacity for a specific home by reviewing and analyzing a historical net power signature of the home. [0069] a series of a utility 610 providing electricity or other services to a home 670. As the energy enters the home 670, a ZigBee or other HAN 620 may measure the energy provided. [0042] disaggregating low frequency energy consumption data that includes solar panel generation; net power signature represents the net usage of the home, including any contribution from solar panels. [0028] Data may be obtained and/or accessed in various manners. For example, a current clamp (CT clamp) may be utilized. The use of two (2) CT clamps may generally be required, with one CT claim positioned at or proximate to the net meter (which may indicate net power draw for the house), and a second CT clamp positioned at or proximate to the solar system (which may indicate power captured and contributed by the solar system). [0041] When using low frequency whole-house energy consumption data, the energy contribution of solar panels must be determined and disaggregated. Such disaggregation may be based upon meteorological data. [0048] Next, a regression model may be trained with weather data and the number of hours from sunrise to sunset as one or more independent variables, and solar intensity as the dependent variable. [0027] energy disaggregation may be determined and/or impacted by the type of data and/or how the data is obtained or accessed. For example, data types may include power signals, or meteorological data or conditions. Power signals may be obtained in low frequency or high frequency samples. Low frequency data may be sampled--for example--hourly, while high or higher frequency data may be sampled--for example--each minute. Meteorological data or conditions may include information such as, but not limited to, (i) skycover or cloud cover (which may be set forth as a percentage or ratio of cover to clear sky); (ii) temperature; (iii) wind-speed; (iv) dew point; and (v) sunrise/sunset times),
wherein the two or more dual consumers are identified by joint processing timestamped data informative of individual grid power consumption GPCs of the consumers from the plurality of consumers and timestamped data informative of the one or more weather conditions([0027] Techniques for energy disaggregation may be determined and/or impacted by the type of data and/or how the data is obtained or accessed. For example, data types may include power signals, or meteorological data or conditions. Power signals may be obtained in low frequency or high frequency samples. Low frequency data may be sampled--for example--hourly, while high or higher frequency data may be sampled--for example--each minute. Meteorological data or conditions may include information such as, but not limited to, (i) skycover or cloud cover (which may be set forth as a percentage or ratio of cover to clear sky); (ii) temperature; (iii) wind-speed; (iv) dew point; and (v) sunrise/sunset times. [0041] Such disaggregation may be based upon meteorological data, [0093] The energy disaggregation on energy use data collected by utilities for all customers);
wherein a GPC of a given consumer from the plurality of consumers is obtained by measuring by a power consumption meter associated therewith([0028] energy usage data may be obtained from Green Button (an industry effort to provide transparent energy usage data, which is generally provided in hourly intervals); from Smart Meters--for example using a Smart Meter Home Area Network channel);
and wherein each given dual consumer is identified in accordance with a relationship between the consumer's individual GPC and the one or more weather conditions during a certain time period with no measurements and without using prior knowledge of individual alternative power production([0025] predicting solar output on unseen homes using training data from different locations around the world. This predictive model may be applicable to locations other than where it was trained. For example, a model may be trained on the west coast of the United States (e.g., California), but may be used to predict solar output on the east coast of the United States (e.g., Connecticut). [0026] disaggregation models may be used to derive solar capacity for a specific home by reviewing and analyzing a historical net power signature of the home. Such models may not require any special hardware to be installed at or on the home to predict such solar capacity. [0106] Moreover, the energy use data for a target set of participating homes (where each home needed to sign up and is known) or a participating region (where each house did not need to sign up and could have still participated in the program) may be compared before and after the program, while [0059] techniques may be desirable that may differentiate spikes from solar signals caused by weather or by appliance usage. In accordance with some embodiments of the present invention, techniques of differentiating such spikes may comprise: (i) identifying correlations between weather and spikes in the data; (ii) establishing spikes caused by weather; (iii) determining features used in appliance usage by using waveform characteristics and transitions; (iv) training a classification model with two (2) classes: weather caused spikes and appliance usage spikes; and (v) performing disaggregation only on spikes that are not determined to be caused by weather);
separately for each of the identified two or more dual consumers, using a trained Forecasting Machine Learning (FML) Model to provide individual forecast of alternative power production by connected alternative power source([0042] disaggregating low frequency energy consumption data that includes solar panel generation, where net power signature represents the net usage of the home, including any contribution from solar panels. FIG. 3 indicates an exemplary solar power signal over the same three (3) day period. It can be seen from the net power curve in FIG. 2 that the net power becomes negative between sunrise and sunset. Similarly, in FIG. 3 it can be seen that the solar power signal is generally represented by a single major curve per day, between sunrise and sunset. [0048]a regression model may be trained with weather data and the number of hours from sunrise to sunset as one or more independent variables, and solar intensity as the dependent variable. [0049] Solar intensity may be seen as the normalized version of solar generation, and may be stated in the range from 0 to 1. Normalization of the dependent variable may be desirable when using a regression model, because it generally permits or allows the model to be easily trained. In accordance with some embodiments of the present invention, a radial basis function (RBF) support vector machine combined with RBF neural networks may be used. [0051] models learned based upon data collected over a year for one hundred (100) homes, and were tested upon approximately twenty-five (25) homes to confirm results. [0053] an exemplary flow for training and predicting solar generation based upon low frequency consumption data. Based upon both the solar intensity and the solar capacity, at 460 the solar generation may be predicted. [0054] FIG. 5, an example of a predicted solar panel generation and ground truth generation for a specific home),
wherein the FML model is trained on historical individual GPCs and weather conditions data to forecast the alternative power production ([0026] disaggregation models may be used to derive solar capacity for a specific home by reviewing and analyzing a historical net power signature of the home. [0048]a regression model may be trained with weather data and the number of hours from sunrise to sunset as one or more independent variables, and solar intensity as the dependent variable. [0053] an exemplary flow for training and predicting solar generation based upon low frequency consumption data. Based upon both the solar intensity and the solar capacity, at 460 the solar generation may be predicted. [0054] FIG. 5, an example of a predicted solar panel generation and ground truth generation for a specific home);
and controlling thermostat set-point change in a set of points connected to the grid([0071] With regard to utility demand response concerns, a utility may send a peak demand reduction signal on a peak usage day either using ZigBee or WiFi. The PCT may then cut back the energy usage by reducing the cooling/heating cycles or relax the thermostat set point by a few degrees).
While Gupta teaches [0042] disaggregating low frequency energy consumption data that includes solar panel generation, where net power signature represents the net usage of the home, including any contribution from solar panels. FIG. 3 indicates an exemplary solar power signal over the same three (3) day period. It can be seen from the net power curve in FIG. 2 that the net power becomes negative between sunrise and sunset. Similarly, in FIG. 3 it can be seen that the solar power signal is generally represented by a single major curve per day, between sunrise and sunset. [0048]a regression model may be trained with weather data and the number of hours from sunrise to sunset as one or more independent variables, and solar intensity as the dependent variable. [0049] Solar intensity may be seen as the normalized version of solar generation, and may be stated in the range from 0 to 1. Normalization of the dependent variable may be desirable when using a regression model, because it generally permits or allows the model to be easily trained. In accordance with some embodiments of the present invention, a radial basis function (RBF) support vector machine combined with RBF neural networks may be used. [0051] models learned based upon data collected over a year for one hundred (100) homes and were tested upon approximately twenty-five (25) homes to confirm results. [0053] an exemplary flow for training and predicting solar generation based upon low frequency consumption data. Based upon both the solar intensity and the solar capacity, at 460 the solar generation may be predicted. [0054] FIG. 5, an example of a predicted solar panel generation and ground truth generation for a specific home. [0071] With regard to utility demand response concerns, a utility may send a peak demand reduction signal on a peak usage day either using ZigBee or WiFi. The PCT may then cut back the energy usage by reducing the cooling/heating cycles or relax the thermostat set point by a few degrees. Gupta does not explicitly teach the following. However analogues reference in the field of energy predication model, Pavlovski teaches:
memory, the one or more computers configured, via computer-executable instructions, to perform operations for operating, in a cloud computing environment([0031] the data processing system 300 may also be a virtual machine. The data processing system 300 includes an input device 310, at least one central processing unit ("CPU") 320, memory 330, a display 340, and an interface device 350. The memory 330 may include a variety of storage devices including internal memory and external mass storage typically arranged in a hierarchy of storage as understood by those skilled in the art. For example, the memory 330 may include databases, random access memory ("RAM"), read-only memory ("ROM"), Hash memory, and/or disk devices. The interface device 350 may include one or more network connections. The data processing system 300 may be adapted for communicating with other data processing systems (e.g., similar to the data processing system 300) over a network 351 via the interface device 350. ), for [….] managing an electrical power grid operatively connected to a plurality of consumers in a geographical area, the plurality of consumers including dual consumers and non-dual consumers([0003] Load forecasting in a utility grid is an integral part of energy management. Growing penetration of distributed energy resources such as solar photovoltaic ("PV") power plants, wind power plants, and power storage plants has changed conventional utility practices for generating, transmitting, and distributing electric power. [0004] A group of power plants, electrical energy consumption devices, and associated infrastructure spread over a geographical area may be referred to as a utility grid. [0011] defining two or more load forecast zones, each load forecast zone being associated with a load profile type and a climate zone type; assigning each of the one or more loads to one of the two or more load forecast zones based on the load profile type and the climate zone type associated with the load),
train a model in accordance with a forecast of the one or more weather conditions in the geographical area(Pavlovski teaches[0051]The net load forecasting system 500 is configured to forecast net load at least partially based on the at least one forecasting variable provided by the environment forecasting component 750. Typical environmental forecasting variables predicted by the environmental forecasting component 750 include weather forecasts, storm warnings, wind speed, air density, air turbidity, irradiance, atmospheric turbulence, rain conditions, snow conditions, air temperature, and humidity. the intermittent distributed energy resource forecasting component 530 may anticipate the future weather conditions at the site of an intermittent distributed energy resource 200 within a selected forecast horizon.);
using the individual forecasts for the two or more dual consumers to provide a total forecast of total alternative power production for a group ofthe identified dual consumers(Pavlovski teaches [0043] forecasts electric power generation by intermittent distributed energy resources 200 within the load forecast zones 900 individually, or in clusters. [0041] The net load forecasting component 550 is configured to produce net load forecasts for the utility grid 100, or any portion thereof, based on the data generated by the intermittent distributed energy resource forecasting component 530 and the load forecasting component 540, while [0046]providing independent forecasts for the distributed energy resources and for the load groups, and combining these forecasts into a single net load forecast for the load forecast zones 900);
and using the provided total forecast of total alternative power production to trigger one or more management actions with regard to power production in the electrical power grid(Pavlovski teaches [0043] forecasts electric power generation by intermittent distributed energy resources 200 within the load forecast zones 900 individually, or in clusters. [0041] The net load forecasting component 550 is configured to produce net load forecasts for the utility grid 100, or any portion thereof, based on the data generated by the intermittent distributed energy resource forecasting component 530 and the load forecasting component 540, while [0046]providing independent forecasts for the distributed energy resources and for the load groups, and combining these forecasts into a single net load forecast for the load forecast zones 900, while [0063] the energy management system 120 is connected to the net load forecasting system 500 and is configured to request net load forecasts from the net load forecasting system 500 to maintain operating conditions in the utility grid 100 within a desired range).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the system for power distribution grid management of Gupta with Pavlovski to include managing an electrical power grid operatively connected to a plurality of consumers in a geographical area, the plurality of consumers including dual consumers and non-dual consumers, train a model in accordance with a forecast of the one or more weather conditions in the geographical area, using the individual forecasts for the two or more dual consumers to provide a total forecast of total alternative power production for a group ofproduction in the electrical power grid. Doing so would use the forecasted power production/generation by the energy management system to reduce the imbalance between the available energy resources and intended uses thereby maintaining operating conditions in the utility grid within their standard or desired ranges. [0005].
While Gupta teaches [0042] disaggregating low frequency energy consumption data that includes solar panel generation, where net power signature represents the net usage of the home, including any contribution from solar panels. FIG. 3 indicates an exemplary solar power signal over the same three (3) day period. It can be seen from the net power curve in FIG. 2 that the net power becomes negative between sunrise and sunset. Similarly, in FIG. 3 it can be seen that the solar power signal is generally represented by a single major curve per day, between sunrise and sunset. [0048]a regression model may be trained with weather data and the number of hours from sunrise to sunset as one or more independent variables, and solar intensity as the dependent variable. [0049] Solar intensity may be seen as the normalized version of solar generation, and may be stated in the range from 0 to 1. Normalization of the dependent variable may be desirable when using a regression model, because it generally permits or allows the model to be easily trained. In accordance with some embodiments of the present invention, a radial basis function (RBF) support vector machine combined with RBF neural networks may be used. [0051] models learned based upon data collected over a year for one hundred (100) homes and were tested upon approximately twenty-five (25) homes to confirm results. [0053] an exemplary flow for training and predicting solar generation based upon low frequency consumption data. Based upon both the solar intensity and the solar capacity, at 460 the solar generation may be predicted. [0054] FIG. 5, an example of a predicted solar panel generation and ground truth generation for a specific home. [0071] With regard to utility demand response concerns, a utility may send a peak demand reduction signal on a peak usage day either using ZigBee or WiFi. The PCT may then cut back the energy usage by reducing the cooling/heating cycles or relax the thermostat set point by a few degrees. Gupta does not explicitly teach the following. However analogues reference in the field of energy predication model, Marhoefer teaches:
wherein the one or more management actions comprise providing commands for automatically power distributing between the power grid and a power storage facility(Marhoefer: [0059] The results of these calculations are a series of instructions and override commands 122 transmitted to the thin-client device/gateway 106. These instructions and override commands create a set of digital signals 130 transmitted to the switch/actuator (generally contained within a battery-based inverter) 105 that control the flow of electricity among the grid, solar panels, storage device and building load, as well as charging, idling and discharging the storage device.),
wherein the one or more management actions comprise issuing by the computer at least one command related to at least one of: charging/discharging one or more batteries connected to the grid ([0059]The results of these calculations are a series of instructions and override commands 122 transmitted to the thin-client device/gateway 106. These instructions and override commands create a set of digital signals 130 transmitted to the switch/actuator (generally contained within a battery-based inverter) 105 that control the flow of electricity among the grid, solar panels, storage device and building load, as well as charging, idling and discharging the storage device).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the system for power distribution grid management of Gupta and Pavlovski with Marhoefer to include the one or more management actions comprise providing commands for automatically power distributing between the power grid and a power storage facility and the one or more management actions comprise issuing by the computer at least one command related to at least one of: charging/discharging one or more batteries connected to the grid. Doing so would aid in managing the implementation, inclusion and aggregation of distributed renewable energy sources and energy storage in the electric grid in a way that maximizes and optimizes energy distribution, and hence value for asset owners, aggregators, utilities and regional transmission organizations. [0002].
Claims 3, 9, 13, 17, and 20 are rejected under 35 U.S.C. 103 as being unpatentable over in view of Gupta in view of Pavlovski in view of Marhoefer, as applied in claims 1, 11, and 19 and further in view of in view of Fabio Mantovani (US 2018/0216961 A1, hereinafter “Mantovani”).
Claim 3
While Gupta teaches [0042] disaggregating low frequency energy consumption data that includes solar panel generation, where net power signature represents the net usage of the home, including any contribution from solar panels. FIG. 3 indicates an exemplary solar power signal over the same three (3) day period. It can be seen from the net power curve in FIG. 2 that the net power becomes negative between sunrise and sunset. Similarly, in FIG. 3 it can be seen that the solar power signal is generally represented by a single major curve per day, between sunrise and sunset. [0048]a regression model may be trained with weather data and the number of hours from sunrise to sunset as one or more independent variables, and solar intensity as the dependent variable. [0049] Solar intensity may be seen as the normalized version of solar generation and may be stated in the range from 0 to 1. Normalization of the dependent variable may be desirable when using a regression model, because it generally permits or allows the model to be easily trained. In accordance with some embodiments of the present invention, a radial basis function (RBF) support vector machine combined with RBF neural networks may be used. [0051] models learned based upon data collected over a year for one hundred (100) homes and were tested upon approximately twenty-five (25) homes to confirm results. [0053] an exemplary flow for training and predicting solar generation based upon low frequency consumption data. Based upon both the solar intensity and the solar capacity, at 460 the solar generation may be predicted. [0054] FIG. 5, an example of a predicted solar panel generation and ground truth generation for a specific home. [0071] With regard to utility demand response concerns, a utility may send a peak demand reduction signal on a peak usage day either using ZigBee or WiFi. The PCT may then cut back the energy usage by reducing the cooling/heating cycles or relax the thermostat set point by a few degrees. Gupta does not explicitly teach the following. However, analogues reference in the field of energy predication model, Mantovani teaches:
The method of Claim 1, wherein a consumer is identified as having inverse relationship between the data informative of the consumer's individual GPC and the data informative of one or more weather conditions comparing to other consumers in a group of similar consumers ([0022] methods for identifying unconnected solar customers. [0071] customers can be required to have a negative correlation between hourly kW and irradiance (GHI). Table 7 and [0095] the unauthorized access detection system and method can construct a ‘neighborhood’ of fifty closest neighbors to a customer suspected of failing. Further, in some embodiments, an average of their normalized average total channel 1 output can be calculated and used to compare the 3-day moving average of the difference between the customer and their neighborhood. Examiner notes a negative correlation is an inverse relationship, where hourly KW is consumption (GPC type) and irradiance is weather/solar condition).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the system for power distribution grid management of Gupta, Pavlovski, and Marhoefer with Mantovani to include a dual consumer is identified as having inverse relationship between the data informative of the consumer's individual GPC and the data informative of one or more weather conditions comparing to other consumers in a group of similar consumers. Doing so would provide more up to date and accurate source of distributed solar photovoltaic system location data to capture locational value of distributed solar photovoltaic and plan for hosting capacity while understanding system impacts and minimizing costs [0007].
Claim 9
While Gupta teaches [0042] disaggregating low frequency energy consumption data that includes solar panel generation, where net power signature represents the net usage of the home, including any contribution from solar panels. FIG. 3 indicates an exemplary solar power signal over the same three (3) day period. It can be seen from the net power curve in FIG. 2 that the net power becomes negative between sunrise and sunset. Similarly, in FIG. 3 it can be seen that the solar power signal is generally represented by a single major curve per day, between sunrise and sunset. [0048] a regression model may be trained with weather data and the number of hours from sunrise to sunset as one or more independent variables, and solar intensity as the dependent variable. [0049] Solar intensity may be seen as the normalized version of solar generation and may be stated in the range from 0 to 1. Normalization of the dependent variable may be desirable when using a regression model, because it generally permits or allows the model to be easily trained. In accordance with some embodiments of the present invention, a radial basis function (RBF) support vector machine combined with RBF neural networks may be used. [0051] models learned based upon data collected over a year for one hundred (100) homes and were tested upon approximately twenty-five (25) homes to confirm results. [0053] an exemplary flow for training and predicting solar generation based upon low frequency consumption data. Based upon both the solar intensity and the solar capacity, at 460 the solar generation may be predicted. [0054] FIG. 5, an example of a predicted solar panel generation and ground truth generation for a specific home. [0071] With regard to utility demand response concerns, a utility may send a peak demand reduction signal on a peak usage day either using ZigBee or WiFi. The PCT may then cut back the energy usage by reducing the cooling/heating cycles or relax the thermostat set point by a few degrees. Gupta does not explicitly teach the following. However, analogues reference in the field of energy predication model, Mantovani teaches:
The method of Claim 1, wherein the group of identified dual consumers is constituted by at least one of: all identified dual consumers, identified dual consumers having the same type of the alternative energy source, identified dual consumers having similar GPC patterns, identified dual consumers having similar GPC requirements([0016] determination of unauthorized interconnection of the customer includes pulling a record of list of customers identified to have solar but not net energy metering (NEM), and matching the record with estimated system size).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the system for power distribution grid management of Gupta, Pavlovski, and Marhoefer with Mantovani to include the group of identified dual consumers is constituted by at least one of: all identified dual consumers, identified dual consumers having the same type of the alternative energy source, identified dual consumers having similar GPC patterns, identified dual consumers having similar GPC requirements. Doing so would provide more up to date and accurate source of distributed solar photovoltaic system location data to capture locational value of distributed solar photovoltaic and plan for hosting capacity while understanding system impacts and minimizing costs [0007].
Claim 13/20
While While Gupta teaches [0042] disaggregating low frequency energy consumption data that includes solar panel generation, where net power signature represents the net usage of the home, including any contribution from solar panels. FIG. 3 indicates an exemplary solar power signal over the same three (3) day period. It can be seen from the net power curve in FIG. 2 that the net power becomes negative between sunrise and sunset. Similarly, in FIG. 3 it can be seen that the solar power signal is generally represented by a single major curve per day, between sunrise and sunset. [0048]a regression model may be trained with weather data and the number of hours from sunrise to sunset as one or more independent variables, and solar intensity as the dependent variable. [0049] Solar intensity may be seen as the normalized version of solar generation, and may be stated in the range from 0 to 1. Normalization of the dependent variable may be desirable when using a regression model, because it generally permits or allows the model to be easily trained. In accordance with some embodiments of the present invention, a radial basis function (RBF) support vector machine combined with RBF neural networks may be used. [0051] models learned based upon data collected over a year for one hundred (100) homes and were tested upon approximately twenty-five (25) homes to confirm results. [0053] an exemplary flow for training and predicting solar generation based upon low frequency consumption data. Based upon both the solar intensity and the solar capacity, at 460 the solar generation may be predicted. [0054] FIG. 5, an example of a predicted solar panel generation and ground truth generation for a specific home. [0071] With regard to utility demand response concerns, a utility may send a peak demand reduction signal on a peak usage day either using ZigBee or WiFi. The PCT may then cut back the energy usage by reducing the cooling/heating cycles or relax the thermostat set point by a few degrees. Gupta does not explicitly teach the following. However, analogues reference in the field of energy predication model, Mantovani teaches:
The system of Claim 11, wherein a dual consumer is identified as having inverse relationship between the data informative of the consumer's individual GPC and the data informative of one or more weather conditions by one of the following: i) comparing to other consumers in a group of similar consumers; ii) with the help of a machine learning model trained to identify patterns of GPC depending on weather conditions in the geographical area ([0022] methods for identifying unconnected solar customers. [0071] customers can be required to have a negative correlation between hourly kW and irradiance (GHI). Table 7 and [0095] the unauthorized access detection system and method can construct a ‘neighborhood’ of fifty closest neighbors to a customer suspected of failing. Further, in some embodiments, an average of their normalized average total channel 1 output can be calculated and used to compare the 3-day moving average of the difference between the customer and their neighborhood. Examiner notes a negative correlation is an inverse relationship, where hourly KW is consumption (GPC type) and irradiance is weather/solar condition).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the system for power distribution grid management of Gupta, Pavlovski, and Marhoefer with Mantovani to include a dual consumer is identified as having inverse relationship between the data informative of the consumer's individual GPC and the data informative of one or more weather conditions by one of the following: i) comparing to other consumers in a group of similar consumers; ii) with the help of a machine learning model trained to identify patterns of GPC depending on weather conditions in the geographical area. Doing so would provide more up to date and accurate source of distributed solar photovoltaic system location data to capture locational value of distributed solar photovoltaic and plan for hosting capacity while understanding system impacts and minimizing costs [0007].
Claim 17
While Gupta teaches [0042] disaggregating low frequency energy consumption data that includes solar panel generation, where net power signature represents the net usage of the home, including any contribution from solar panels. FIG. 3 indicates an exemplary solar power signal over the same three (3) day period. It can be seen from the net power curve in FIG. 2 that the net power becomes negative between sunrise and sunset. Similarly, in FIG. 3 it can be seen that the solar power signal is generally represented by a single major curve per day, between sunrise and sunset. [0048]a regression model may be trained with weather data and the number of hours from sunrise to sunset as one or more independent variables, and solar intensity as the dependent variable. [0049] Solar intensity may be seen as the normalized version of solar generation, and may be stated in the range from 0 to 1. Normalization of the dependent variable may be desirable when using a regression model, because it generally permits or allows the model to be easily trained. In accordance with some embodiments of the present invention, a radial basis function (RBF) support vector machine combined with RBF neural networks may be used. [0051] models learned based upon data collected over a year for one hundred (100) homes and were tested upon approximately twenty-five (25) homes to confirm results. [0053] an exemplary flow for training and predicting solar generation based upon low frequency consumption data. Based upon both the solar intensity and the solar capacity, at 460 the solar generation may be predicted. [0054] FIG. 5, an example of a predicted solar panel generation and ground truth generation for a specific home. [0071] With regard to utility demand response concerns, a utility may send a peak demand reduction signal on a peak usage day either using ZigBee or WiFi. The PCT may then cut back the energy usage by reducing the cooling/heating cycles or relax the thermostat set point by a few degrees. Gupta does not explicitly teach the following. However, analogues reference in the field of energy predication model, Mantovani teaches:
The system of Claim 14, wherein the group of identified dual consumers is constituted by at least one of: all identified dual consumers, identified dual consumers having the same type of the alternative energy source, identified dual consumers having similar GPC patterns, identified dual consumers having similar GPC requirements([0016] determination of unauthorized interconnection of the customer includes pulling a record of list of customers identified to have solar but not net energy metering (NEM), and matching the record with estimated system size).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the system for power distribution grid management of Gupta, Pavlovski, and Marhoefer with Mantovani to include the group of identified dual consumers is constituted by at least one of: all identified dual consumers, identified dual consumers having the same type of the alternative energy source, identified dual consumers having similar GPC patterns, identified dual consumers having similar GPC requirements. Doing so would provide more up to date and accurate source of distributed solar photovoltaic system location data to capture locational value of distributed solar photovoltaic and plan for hosting capacity while understanding system impacts and minimizing costs [0007].
Claims 4, 6, and 14 are rejected under 35 U.S.C. 103 as being unpatentable over Gupta in view of Pavlovski in view of Marhoefer, as applied in claims 1 and 11, and further in view of Charles McBrearty (US 2016/0306906 A1, hereinafter “McBrearty”).
Claim 4
While Gupta teaches [0042] disaggregating low frequency energy consumption data that includes solar panel generation, where net power signature represents the net usage of the home, including any contribution from solar panels. FIG. 3 indicates an exemplary solar power signal over the same three (3) day period. It can be seen from the net power curve in FIG. 2 that the net power becomes negative between sunrise and sunset. Similarly, in FIG. 3 it can be seen that the solar power signal is generally represented by a single major curve per day, between sunrise and sunset. [0048]a regression model may be trained with weather data and the number of hours from sunrise to sunset as one or more independent variables, and solar intensity as the dependent variable. [0049] Solar intensity may be seen as the normalized version of solar generation, and may be stated in the range from 0 to 1. Normalization of the dependent variable may be desirable when using a regression model, because it generally permits or allows the model to be easily trained. In accordance with some embodiments of the present invention, a radial basis function (RBF) support vector machine combined with RBF neural networks may be used. [0051] models learned based upon data collected over a year for one hundred (100) homes and were tested upon approximately twenty-five (25) homes to confirm results. [0053] an exemplary flow for training and predicting solar generation based upon low frequency consumption data. Based upon both the solar intensity and the solar capacity, at 460 the solar generation may be predicted. [0054] FIG. 5, an example of a predicted solar panel generation and ground truth generation for a specific home. [0071] With regard to utility demand response concerns, a utility may send a peak demand reduction signal on a peak usage day either using ZigBee or WiFi. The PCT may then cut back the energy usage by reducing the cooling/heating cycles or relax the thermostat set point by a few degrees. Gupta does not explicitly teach the following. However, analogues reference in the field of energy predication model, McBrearty teaches:
The method of Claim 1, wherein a dual consumer is identified, with the help of a machine learning model trained to identify patterns of GPC depending on weather conditions in the geographical area, as having inverse relationship between the data informative of the consumer's individual GPC and the data informative of one or more weather conditions([0036] describes he Geographic Average may be calculated by an algorithm which produces an average of the Normalized Performances for each area covered. [0054] construct a large data set of example patterns for certain types of systems, define certain features that tend to be characteristic of the systems in the data set, and use either statistical correlation techniques, or machine learning optimization (e.g. neural networks) to define classification thresholds in order to automatically identify a system type and use the developed thresholds feature sets and data history thresholds to automatically classify data streams according to different system types, wherein the characteristics of the monitored data are influenced by geographic-specific characteristics, like sunrise/sunset times or weather characteristics. As long as the monitored data stream has associated timestamps, it is possible to determine the location by finding a geographic location that would best match the observed energy consumption or generation characteristics of the monitored data stream. For example, if the type of generation source (or consumption) is known, one can model the expected behavior of this generation or consumption data stream under actual recorded weather around the world, and the best fitting match statistically is likely to be the actual physical location).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the system for power distribution grid management Gupta, Pavlovski, and Marhoefer with McBrearty to include a dual consumer is identified, with the help of a machine learning model trained to identify patterns of GPC depending on weather conditions in the geographical area, as having inverse relationship between the data informative of the consumer's individual GPC and the data informative of one or more weather conditions. Doing so would aid in understanding output of solar resource assessments are databases that catalog the regional intensity of the solar resource, on a given number of minutes increment or an hourly basis. [0002].
Claim 6
While While Gupta teaches [0042] disaggregating low frequency energy consumption data that includes solar panel generation, where net power signature represents the net usage of the home, including any contribution from solar panels. FIG. 3 indicates an exemplary solar power signal over the same three (3) day period. It can be seen from the net power curve in FIG. 2 that the net power becomes negative between sunrise and sunset. Similarly, in FIG. 3 it can be seen that the solar power signal is generally represented by a single major curve per day, between sunrise and sunset. [0048]a regression model may be trained with weather data and the number of hours from sunrise to sunset as one or more independent variables, and solar intensity as the dependent variable. [0049] Solar intensity may be seen as the normalized version of solar generation, and may be stated in the range from 0 to 1. Normalization of the dependent variable may be desirable when using a regression model, because it generally permits or allows the model to be easily trained. In accordance with some embodiments of the present invention, a radial basis function (RBF) support vector machine combined with RBF neural networks may be used. [0051] models learned based upon data collected over a year for one hundred (100) homes and were tested upon approximately twenty-five (25) homes to confirm results. [0053] an exemplary flow for training and predicting solar generation based upon low frequency consumption data. Based upon both the solar intensity and the solar capacity, at 460 the solar generation may be predicted. [0054] FIG. 5, an example of a predicted solar panel generation and ground truth generation for a specific home. [0071] With regard to utility demand response concerns, a utility may send a peak demand reduction signal on a peak usage day either using ZigBee or WiFi. The PCT may then cut back the energy usage by reducing the cooling/heating cycles or relax the thermostat set point by a few degrees. Gupta does not explicitly teach the following. However, analogues reference in the field of energy predication model, McBrearty teaches:
The method of Claim 1, further comprising processing the timestamped GPC and weather condition data to identify types of alternative power sources respectively connected to the identified two or more dual consumers, wherein, for a given dual consumer, the FML Model corresponds to an identified type of a connected alternative power source([0052] The method may next comprise the step of defining at least one characteristic feature for each at least one system type to provide at least one system type and correlated characteristic feature and saving the at least one system type and correlated characteristic feature in the at least one data server (516). A characteristic feature may be, for example, time of sunrise, time of sunset, associated timestamps, energy consumption, wind system output, weather, configuration, time of year, user habits, system size, tracker versus fixed, energy profile shape, east west orientation, north-south orientation, homeowner type, heating type, temperature sensitivity, consumption data, utilized energy, utilized generation, system derate factors and air conditioning status. [0054] construct a large data set of example patterns for certain types of systems, define certain features that tend to be characteristic of the systems in the data set, and use either statistical correlation techniques, or machine learning optimization (e.g. neural networks) to define classification thresholds in order to automatically identify a system type [0054] s long as the monitored data stream has associated timestamps, it is possible to determine the location by finding a geographic location that would best match the observed energy consumption or generation characteristics of the monitored data stream. For example, if the type of generation source (or consumption) is known, one can model the expected behavior of this generation or consumption data stream under actual recorded weather around the world (including sunrise sunset times), By maintaining a time history of weather conditions across a large area (e.g., North America and Europe), and modeling PV system outputs across all of these geographies, one can match the time history of a PV system's output to the geography of data it most closely matches. This matching process can be improved by narrowing the range of geographic areas (e.g. using the preceding longitudinal technique), or by improving the estimated model output of a PV system via leveraging known system characteristics (either known beforehand or identified as described in the section below about “Identification of system information for a solar PV system”). Output from a wind system is highly dependent on wind conditions. By maintaining a time history of weather conditions across a large area, and modeling the wind system outputs across all the locations, one can match the time history of a wind system's output to the geographic location it most closely matches).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Gupta, Pavlovski, and Marhoefer with McBrearty to include processing the timestamped GPC and weather condition data to identify types of alternative power sources respectively connected to the identified two or more dual consumers, wherein, for a given dual consumer, the FML Model corresponds to an identified type of a connected alternative power source. Doing so would aid in understanding output of solar resource assessments are databases that catalog the regional intensity of the solar resource, on a given number of minutes increment or an hourly basis. [0002].
Claim 14
While While Gupta teaches [0042] disaggregating low frequency energy consumption data that includes solar panel generation, where net power signature represents the net usage of the home, including any contribution from solar panels. FIG. 3 indicates an exemplary solar power signal over the same three (3) day period. It can be seen from the net power curve in FIG. 2 that the net power becomes negative between sunrise and sunset. Similarly, in FIG. 3 it can be seen that the solar power signal is generally represented by a single major curve per day, between sunrise and sunset. [0048] a regression model may be trained with weather data and the number of hours from sunrise to sunset as one or more independent variables, and solar intensity as the dependent variable. [0049] Solar intensity may be seen as the normalized version of solar generation and may be stated in the range from 0 to 1. Normalization of the dependent variable may be desirable when using a regression model, because it generally permits or allows the model to be easily trained. In accordance with some embodiments of the present invention, a radial basis function (RBF) support vector machine combined with RBF neural networks may be used. [0051] models learned based upon data collected over a year for one hundred (100) homes and were tested upon approximately twenty-five (25) homes to confirm results. [0053] an exemplary flow for training and predicting solar generation based upon low frequency consumption data. Based upon both the solar intensity and the solar capacity, at 460 the solar generation may be predicted. [0054] FIG. 5, an example of a predicted solar panel generation and ground truth generation for a specific home. [0071] With regard to utility demand response concerns, a utility may send a peak demand reduction signal on a peak usage day either using ZigBee or WiFi. The PCT may then cut back the energy usage by reducing the cooling/heating cycles or relax the thermostat set point by a few degrees. Gupta does not explicitly teach the following. However, analogues reference in the field of energy predication model,, McBrearty teaches:
The system of Claim 11, wherein the computer is further configured to process the timestamped GPC and weather condition data to identify types of alternative power sources respectively connected to the identified two or more dual consumers, wherein, for a given dual consumer, the FML Model corresponds to an identified type of a connected alternative power source ([0052] The method may next comprise the step of defining at least one characteristic feature for each at least one system type to provide at least one system type and correlated characteristic feature and saving the at least one system type and correlated characteristic feature in the at least one data server (516). A characteristic feature may be, for example, time of sunrise, time of sunset, associated timestamps, energy consumption, wind system output, weather, configuration, time of year, user habits, system size, tracker versus fixed, energy profile shape, east west orientation, north-south orientation, homeowner type, heating type, temperature sensitivity, consumption data, utilized energy, utilized generation, system derate factors and air conditioning status. [0054] construct a large data set of example patterns for certain types of systems, define certain features that tend to be characteristic of the systems in the data set, and use either statistical correlation techniques, or machine learning optimization (e.g. neural networks) to define classification thresholds in order to automatically identify a system type [0054] s long as the monitored data stream has associated timestamps, it is possible to determine the location by finding a geographic location that would best match the observed energy consumption or generation characteristics of the monitored data stream. For example, if the type of generation source (or consumption) is known, one can model the expected behavior of this generation or consumption data stream under actual recorded weather around the world (including sunrise sunset times), By maintaining a time history of weather conditions across a large area (e.g., North America and Europe), and modeling PV system outputs across all of these geographies, one can match the time history of a PV system's output to the geography of data it most closely matches. This matching process can be improved by narrowing the range of geographic areas (e.g. using the preceding longitudinal technique), or by improving the estimated model output of a PV system via leveraging known system characteristics (either known beforehand or identified as described in the section below about “Identification of system information for a solar PV system”). Output from a wind system is highly dependent on wind conditions. By maintaining a time history of weather conditions across a large area, and modeling the wind system outputs across all the locations, one can match the time history of a wind system's output to the geographic location it most closely matches).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the system for power distribution grid management of Gupta, Pavlovski, and Marhoefer with McBrearty to include processing the timestamped GPC and weather condition data to identify types of alternative power sources respectively connected to the identified two or more dual consumers, wherein, for a given dual consumer, the FML Model corresponds to an identified type of a connected alternative power source. Doing so would aid in understanding output of solar resource assessments are databases that catalog the regional intensity of the solar resource, on a given number of minutes increment or an hourly basis. [0002].
Claims 7 and 15 are rejected under 35 U.S.C. 103 as being unpatentable over Gupta in view of Pavlovski in view of Marhoefer in view of McBrearty, as applied in claims 6 and 14, and further in view of Mantovani.
Claim 7
While While Gupta teaches [0042] disaggregating low frequency energy consumption data that includes solar panel generation, where net power signature represents the net usage of the home, including any contribution from solar panels. FIG. 3 indicates an exemplary solar power signal over the same three (3) day period. It can be seen from the net power curve in FIG. 2 that the net power becomes negative between sunrise and sunset. Similarly, in FIG. 3 it can be seen that the solar power signal is generally represented by a single major curve per day, between sunrise and sunset. [0048]a regression model may be trained with weather data and the number of hours from sunrise to sunset as one or more independent variables, and solar intensity as the dependent variable. [0049] Solar intensity may be seen as the normalized version of solar generation, and may be stated in the range from 0 to 1. Normalization of the dependent variable may be desirable when using a regression model, because it generally permits or allows the model to be easily trained. In accordance with some embodiments of the present invention, a radial basis function (RBF) support vector machine combined with RBF neural networks may be used. [0051] models learned based upon data collected over a year for one hundred (100) homes and were tested upon approximately twenty-five (25) homes to confirm results. [0053] an exemplary flow for training and predicting solar generation based upon low frequency consumption data. Based upon both the solar intensity and the solar capacity, at 460 the solar generation may be predicted. [0054] FIG. 5, an example of a predicted solar panel generation and ground truth generation for a specific home. [0071] With regard to utility demand response concerns, a utility may send a peak demand reduction signal on a peak usage day either using ZigBee or WiFi. The PCT may then cut back the energy usage by reducing the cooling/heating cycles or relax the thermostat set point by a few degrees. Gupta does not explicitly teach the following. However, analogues reference in the field of energy predication model, Mantovani teaches:
The method of Claim 6, wherein the type of the connected alternative energy source is identified with the help of a machine learning model trained to identify patterns of alternative power production being a function of the one or more weather conditions in one or more geographical areas([0049] and Fig.4 describe a load plot for a customer connected to an electric grid where the load goes negative, meaning the customer is using less power than they are generating; see also Table 2 items 15-16 and [0068]-[0069] describe obtaining the load prediction for a customer by correlating the weather with the power use of the building. [0007] By combining load data, weather patterns, customer locations, and interconnection data, it is possible to identify patterns which are consistent with interconnected solar photovoltaic systems).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the system for power distribution grid management of Gupta, Pavlovski, Marhoefer, and McBrearty with Mantovani to include the type of the connected alternative energy source is identified with the help of a machine learning model trained to identify patterns of alternative power production being a function of the one or more weather conditions in one or more geographical areas. Doing so would help efficiently capture locational value of distributed power grid and its consumer and plan for hosting capacity by adjusting and making changes while understanding system impacts and minimizing costs.
Claim 15
While While Gupta teaches [0042] disaggregating low frequency energy consumption data that includes solar panel generation, where net power signature represents the net usage of the home, including any contribution from solar panels. FIG. 3 indicates an exemplary solar power signal over the same three (3) day period. It can be seen from the net power curve in FIG. 2 that the net power becomes negative between sunrise and sunset. Similarly, in FIG. 3 it can be seen that the solar power signal is generally represented by a single major curve per day, between sunrise and sunset. [0048]a regression model may be trained with weather data and the number of hours from sunrise to sunset as one or more independent variables, and solar intensity as the dependent variable. [0049] Solar intensity may be seen as the normalized version of solar generation, and may be stated in the range from 0 to 1. Normalization of the dependent variable may be desirable when using a regression model, because it generally permits or allows the model to be easily trained. In accordance with some embodiments of the present invention, a radial basis function (RBF) support vector machine combined with RBF neural networks may be used. [0051] models learned based upon data collected over a year for one hundred (100) homes and were tested upon approximately twenty-five (25) homes to confirm results. [0053] an exemplary flow for training and predicting solar generation based upon low frequency consumption data. Based upon both the solar intensity and the solar capacity, at 460 the solar generation may be predicted. [0054] FIG. 5, an example of a predicted solar panel generation and ground truth generation for a specific home. [0071] With regard to utility demand response concerns, a utility may send a peak demand reduction signal on a peak usage day either using ZigBee or WiFi. The PCT may then cut back the energy usage by reducing the cooling/heating cycles or relax the thermostat set point by a few degrees. Gupta does not explicitly teach the following. However, analogues reference in the field of energy predication model, Mantovani teaches:
The system of Claim 14, wherein the type of the connected alternative energy source is identified with the help of a machine learning model trained to identify patterns of alternative power production being a function of the one or more weather conditions in one or more geographical areas ([0049] and Fig.4 describe a load plot for a customer connected to an electric grid where the load goes negative, meaning the customer is using less power than they are generating; see also Table 2 items 15-16 and [0068]-[0069] describe obtaining the load prediction for a customer by correlating the weather with the power use of the building. [0007] By combining load data, weather patterns, customer locations, and interconnection data, it is possible to identify patterns which are consistent with interconnected solar photovoltaic systems).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the system for power distribution grid management of Gupta, Pavlovski, Marhoefer, and McBrearty with Mantovani to include the type of the connected alternative energy source is identified with the help of a machine learning model trained to identify patterns of alternative power production being a function of the one or more weather conditions in one or more geographical areas. Doing so would help efficiently capture locational value of distributed power grid and its consumer and plan for hosting capacity by adjusting and making changes while understanding system impacts and minimizing costs.
Claims 10 and 18 are rejected under 35 U.S.C. 103 as being unpatentable over in view of Gupta in view of Pavlovski in view of Marhoefer, as applied in claim 1 and 11, and further Nagata Satoshi (WO 2008117392 A1, hereinafter “Satoshi”) and in view of Harish Bharti (US 2017/0102683 A1, hereinafter “Bharti”).
Claim 10/18
While Gupta teaches [0042] disaggregating low frequency energy consumption data that includes solar panel generation, where net power signature represents the net usage of the home, including any contribution from solar panels. FIG. 3 indicates an exemplary solar power signal over the same three (3) day period. It can be seen from the net power curve in FIG. 2 that the net power becomes negative between sunrise and sunset. Similarly, in FIG. 3 it can be seen that the solar power signal is generally represented by a single major curve per day, between sunrise and sunset. [0048]a regression model may be trained with weather data and the number of hours from sunrise to sunset as one or more independent variables, and solar intensity as the dependent variable. [0049] Solar intensity may be seen as the normalized version of solar generation, and may be stated in the range from 0 to 1. Normalization of the dependent variable may be desirable when using a regression model, because it generally permits or allows the model to be easily trained. In accordance with some embodiments of the present invention, a radial basis function (RBF) support vector machine combined with RBF neural networks may be used. [0051] models learned based upon data collected over a year for one hundred (100) homes and were tested upon approximately twenty-five (25) homes to confirm results. [0053] an exemplary flow for training and predicting solar generation based upon low frequency consumption data. Based upon both the solar intensity and the solar capacity, at 460 the solar generation may be predicted. [0054] FIG. 5, an example of a predicted solar panel generation and ground truth generation for a specific home. [0071] With regard to utility demand response concerns, a utility may send a peak demand reduction signal on a peak usage day either using ZigBee or WiFi. The PCT may then cut back the energy usage by reducing the cooling/heating cycles or relax the thermostat set point by a few degrees. Gupta does not explicitly teach the following. However, analogues reference in the field of energy predication model, Satoshi teaches:
The method of Claim 1, wherein training the FML model comprises: using historical data informative of the one or more weather conditions and of individual GPC by different consumers to obtain a plurality of FML models having different parameters and initially trained to forecast GPC in accordance with a forecast of the one or more weather conditions; using the plurality of initially trained FML models to forecast GPC in accordance with the one or more weather conditions forecasted for a testing period and thereby obtaining a set of net load data time series forecasted by the plurality of FML models([0028]In the present invention, firstly, the "total photovoltaic power generation amount", "day maximum power demand amount" and "total power demand amount" of each electric power supplier and demander of the next day are estimated (predicted). For the estimation, weather forecasts and past weather information of the next day of each region of the supply and demander and the adjacent region, and "total solar cell power generation amount", "day maximum power demand amount", "day total power demand amount" Calendar information (day of the week, public holiday), theoretical solar radiation amount data is input to the hierarchical neural network. The neural network learns the data combination of the climate patterns of each electric power supplier and demander's area and surrounding areas and the actual results of the total electricity generation amount and the electricity demand of the area as a pattern and collates the weather forecast pattern of the next day with the past pattern And performs nonlinear interpolation estimation with. In this pattern learning, the model is updated using observation data everyday, so estimation accuracy also continues to improve day by day. In addition, it responds to environmental changes within each region of each electric power supplier and demander (solar battery total capacity, customer's change, long-term climate change, meteorological abnormal weather, etc.) by autonomous model update. It is unnecessary to construct each supplier / demander database within the area of each power customer. [0029]The above prediction is executed in the following procedure. (I) Prepare a neural network model that predicts the amount of electricity generation and electricity demand (if it does not exist, create a temporary model with dummy data). The past actual data of this model, the weather forecast of the next day, calendar information, and the amount of solar radiation (theoretical value) at the sunny day of the day are input. (At this time, it is desirable to add not only the area but also the weather information of the neighboring area in order to improve the prediction accuracy.) (Ii) Estimate total power generation amount, maximum power demand amount, total power demand amount. (Nonlinear interpolation estimation by pattern matching) (iii) Collect actual data for neural network relearning. Collect actual data and prepare for neural network re-learning. The actual data is various actual data (power generation amount, maximum power, total power, weather, calendar information, theoretical solar radiation amount) in a certain past period including the current day. (Iv) Re-learn to the neural network using back propagation (error back propagation method). (V) In the updated neural network, predict the total power generation amount, maximum power demand amount, total power demand amount of the next day. Hereinafter, the accuracy of the prediction data is increased by repeating (i) to (v)).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the system for power distribution grid management of Gupta, Pavlovski, and Marhoefer with Satoshi to include using historical data informative of the one or more weather conditions and of individual GPC by different consumers to obtain a plurality of FML models having different parameters and initially trained to forecast GPC in accordance with a forecast of the one or more weather conditions; using the plurality of initially trained FML models to forecast GPC in accordance with the one or more weather conditions forecasted for a testing period and thereby obtaining a set of net load data time series forecasted by the plurality of FML models. Doing so would help efficiently capture locational value of distributed power grid and its consumer and plan for hosting capacity while understanding system impacts and minimizing costs.
While While Gupta teaches [0042] disaggregating low frequency energy consumption data that includes solar panel generation, where net power signature represents the net usage of the home, including any contribution from solar panels. FIG. 3 indicates an exemplary solar power signal over the same three (3) day period. It can be seen from the net power curve in FIG. 2 that the net power becomes negative between sunrise and sunset. Similarly, in FIG. 3 it can be seen that the solar power signal is generally represented by a single major curve per day, between sunrise and sunset. [0048]a regression model may be trained with weather data and the number of hours from sunrise to sunset as one or more independent variables, and solar intensity as the dependent variable. [0049] Solar intensity may be seen as the normalized version of solar generation, and may be stated in the range from 0 to 1. Normalization of the dependent variable may be desirable when using a regression model, because it generally permits or allows the model to be easily trained. In accordance with some embodiments of the present invention, a radial basis function (RBF) support vector machine combined with RBF neural networks may be used. [0051] models learned based upon data collected over a year for one hundred (100) homes and were tested upon approximately twenty-five (25) homes to confirm results. [0053] an exemplary flow for training and predicting solar generation based upon low frequency consumption data. Based upon both the solar intensity and the solar capacity, at 460 the solar generation may be predicted. [0054] FIG. 5, an example of a predicted solar panel generation and ground truth generation for a specific home. [0071] With regard to utility demand response concerns, a utility may send a peak demand reduction signal on a peak usage day either using ZigBee or WiFi. The PCT may then cut back the energy usage by reducing the cooling/heating cycles or relax the thermostat set point by a few degrees. Gupta does not explicitly teach the following. However, analogues reference in the field of energy predication model, Bharti teaches:
and comparing the forecasted net load data time series in the set of net load data time series with net load data time series measured during the testing period and selecting a FML model providing the best net load forecast for the testing period, thereby giving rise to the trained FML ([0070] Referring again to FIG. 4, at 108 each of the models identified at 102 is run to generate respective sets of energy load forecasts for one (or more) of the combinations of the time scale periods and grid hierarchy elements as a function of its/their respective set(s) of currently-prioritized contextual influencing factors. [0071] At 110 the generated energy load forecasts are compared to actual, historic energy loads for the time scale period/grid hierarchy element combinations to determine differences between the historic forecast energy loads and the energy loads generated as a function of the current set of prioritized contextual influencing factors. Aspects perform prioritization and weight assignment of the contextual variables with respect to historical time period as the load forecast output is compared to the actual energy load for that time period).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the system for power distribution grid management of Gupta, Pavlovski, Marhoefer, and Satoshi with Bharti to include comparing the forecasted net load data time series in the set of net load data time series with net load data time series measured during the testing period and selecting a FML model providing the best net load forecast for the testing period, thereby giving rise to the trained FML, as taught by Bharti. Doing so would help efficiently capture locational value of distributed power grid and its consumer and plan for hosting capacity while understanding system impacts and minimizing costs.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure:
James P. Cipriani (US 20170286838 A1): method for predicting solar power generation receives historical power profile data and historical weather micro-forecast data at a given location for a set of days. Based on power output features for the days, clusters are generated. A classification model that assigns a day to a generated cluster according to weather features is created. For each cluster, a regression model that takes as input weather features and outputs predicted solar power is built. A system includes a sensor for collecting meteorological data at a solar farm, a meter for measuring photovoltaic power output of the solar farm, and a computer processor for executing instructions to predict solar power generation at the solar farm according to the method disclosed, based on data from the sensor and the meter, for a predefined time period. Further instructions predict solar power generation at the solar farm based on a micro-forecast for the solar farm.
Bing Dong (US 20180203160 A1): directed to multiple photovoltaic power inverters that convert sunlight to power. An amount of power for each of the inverters can be measured over a period of time. These measurements, along with other data, can be collected. The collected measurements can be used to generate artificial neural networks that predict the output of each inverter based on input parameters. Using these neural networks, the total solar power generation forecast for the photovoltaic system can be predicted.
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to REHAM K ABOUZAHRA whose telephone number is (571)272-0419. The examiner can normally be reached M-F 7:00 AM to 5:00 PM.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Brian Epstein can be reached at (571)-270-5389. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/REHAM K ABOUZAHRA/Examiner, Art Unit 3625
/BRIAN M EPSTEIN/Supervisory Patent Examiner, Art Unit 3625
Prosecution Insights