Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Priority
Receipt is acknowledged of certified copies of papers required by 37 CFR 1.55.
Specification
The lengthy specification has not been checked to the extent necessary to determine the presence of all possible minor errors. Applicant’s cooperation is requested in correcting any errors of which applicant may become aware in the specification.
Claim Objections
Claim 17 is objected to because of the following informalities:
Claim 17 recites, “the trained machine learning model taking as input at least meat loss output measurements,” in ll. 12-13, recommend rephrasing to read, “ the machine learning model taking as input at least metal loss output measurements.”
Appropriate correction is required.
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 (i.e., changing from AIA to pre-AIA ) 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.
Claims 1-7, 9, & 11-15 are rejected under 35 U.S.C. 103 as being unpatentable over May et al. (US 2005/0068026 A1, Pub. Date Mar. 31, 2005, hereinafter, May), in view of Hardy et al. (US 2017/0168016 A1, Pub. Date Jun. 15, 2017, hereinafter, Hardy), and further in view of Atherton (US 2004/0189289 A1, Pub. Date Sep. 30, 2004, hereinafter, Atherton).
Regarding independent claim 1, May, teaches:
A method of pulsed eddy current (PEC) testing, the method comprising (Fig. 4; [Abstract] & [0004]):
receiving metal loss output measurements, the metal loss output measurements calculated based on an eddy current response (Fig. 1; [0003] & [0014]-[0019]: teaches calculating thickness/metal loss based on the eddy current response) captured by a probe at a location on a pipe or tank (Disclosed in combination: May: [0003] & [0014]-[0019]; Hardy: Fig. 8A; [0079]),
outputting the compensated metal loss output measurements as PEC test results (Disclosed in combination: May: Fig. 1; [0019]-[0020]; Hardy: [0034]).
PNG
media_image1.png
764
940
media_image1.png
Greyscale
PNG
media_image2.png
778
556
media_image2.png
Greyscale
May, in combination with Hardy, are silent in regard to:
wherein one or more energized high voltage cables are located within the pipe or tank;
generating compensated metal loss output measurements to compensate for errors in the metal loss output measurements caused by the energized high voltage cables; and
However, Atherton, in combination with Hardy, further teach:
wherein one or more energized high voltage cables are located within the pipe or tank (Disclosed in combination: Atherton: Fig. 1; [0036]: places an energized power cable directly inside the pipe alongside the inspection tool; Hardy: [0063]: discloses the probe and the presence of a parasitic external magnetic field 530);
generating compensated metal loss output measurements to compensate for errors in the metal loss output measurements caused by the energized high voltage cables (Disclosed in combination: Atherton: [0027]-[0029]: teaches the necessity of compensating for power cable interference; Hardy: [0063]: discloses the probe and the presence of a parasitic external magnetic field 530); and
PNG
media_image3.png
695
1071
media_image3.png
Greyscale
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 PEC thickness measurement system of May with the antiparallel noise-canceling coils of Hardy, and to apply this compensated measurement technique in an environment containing power cables as taught by Atherton, according to known methods. The motivation to do so stems from Atherton’s warning in paragraph [0027] that power cables act as a source of interference that can skew eddy current signal responses, making it of “great importance” to calculate corrections. A POSITA would implement hardware solutions like Hardy’s noise immunity configuration (canceling parasitic magnetic fields) to achieve the accurate corrections taught by Atherton when operating a tool powered by internal cables ([0036]), yielding predictable results (KSR).
Regarding dependent claim 2, May, teaches:
The method of claim 1 (Fig. 4; [Abstract] & [0004]),
May, is silent in regard to:
further comprising controlling the movement of the probe to another location on the pipe or tank.
However, Hardy, in combination with Atherton, further teach:
further comprising controlling the movement of the probe to another location on the pipe or tank (Disclosed in combination: Hardy: Figs. 8A & 9A; [0037], [0079] & [0082]: teaches moving the probe body to different locations, both circumferentially and longitudinally, to scan the pipe; Atherton: [Abstract]: reinforces the teaching of controlling the movement/traversing of an eddy current probe through a pipe to test different locations).
PNG
media_image4.png
612
942
media_image4.png
Greyscale
PNG
media_image5.png
577
697
media_image5.png
Greyscale
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to control the movement of the probe to another location on the pipe or tank, as taught by Hardy and Atherton, according to known methods. The motivation to combine these teachings is provided by Hardy in paragraphs [0079] & [0082], which state that displacing or scanning the probe along the pipe’s circumference or longitudinal axis serves the purpose of covering a larger area of the pipe and reducing the time of inspection. A POSITA would implement the capability to move or traverse the probe (as in Hardy/Atherton) along with May’s thickness evaluation algorithms to perform comprehensive, full-pipe structural integrity inspection(s) rather than limited single static point(s), yielding predictable results (KSR).
Regarding dependent claim 3, May, teaches:
The method of claim 1 (Fig. 4; [Abstract] & [0004]), further comprising creating eddy currents by causing an electrical current to be supplied to the probe (Disclosed in combination: May: [0018]; Hardy: [0042]; both references teach supplying an electrical current to the probe/coil to initiate the PEC process)
May, is silent in regard to:
and causing the electrical current to be cut-off from the probe.
However, Hardy, further teaches:
and causing the electrical current to be cut-off from the probe ([0042]-[0044]: teaches cutting off the electrical current from the tester coil to trigger the magnetic variation that creates the eddy currents in the object).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to create eddy currents by supplying an electrical current to the probe and subsequently cutting off, as taught by Hardy, in the PEC method of May, according to known methods. The motivation to combine these teachings is that the “cut-off” phase is the required physical mechanism to produce the sharp magnetic field variation necessary to induce the transient eddy currents into the test object. A POSITA would operate May’s pulse generator in this manner, as it defines the standard and necessary operation of a PEC testing device, combined with Hardy’s “cut-off” phase, and yield predictable results (KSR).
Regarding dependent claim 4, May, teaches:
The method of claim 3 (Fig. 4; [Abstract] & [0004]), further comprising calculating the metal loss output measurements based at least in part on the eddy currents (Disclosed in combination: May: [0014] & [0023]: teaches capturing the eddy current response and mathematically calculating the thickness/loss of the metal based on that response; Hardy: [0001]-[0002] & [0034]-[0035]: teaches analyzing the eddy current signal to calculate “wall loss”).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to calculate metal loss output measurements based on the eddy currents, as taught by May and Hardy, according to known methods, because determining wall/metal loss is the fundamental, well-known purpose of utilizing Pulsed Eddy Current testing on pipes and tanks. A POSITA would configure the signal analyzer to output the thickness difference as a “metal loss” measurement to inform the operator of corrosion or structural degradation, thus yielding predictable results (KSR).
Regarding dependent claim 5, May, teaches:
The method of claim 1 (Fig. 4; [Abstract] & [0004]), wherein the pipe or tank comprises a ferrous material (Disclosed in combination: May: [0003]: teaches that the container (tank) or pipe being tested by the pulsed eddy current system can be made of steel. Steel is an alloy of iron, making it a ferrous material; Hardy: [0086]: provides an example where the electrically conductive object being inspected via the PEC method is carbon steel, a ferrous material; Atherton: [0003] & [0037]: further corroborates the well-known industry standard of testing steel and ferromagnetic/ferrous pipes).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention that the pipe or tank of the combined May, Hardy, and Atherton system comprises a ferrous material, because Hardy provides a working example of inspecting a “carbon steel” object, and Atherton discusses testing “ferromagnetic” and “steel pipes”, according to known methods. The motivation to utilize a ferrous material for the pink or tank is simply the intended real-world application of assessing structural degradation, wall loss, and corrosion in standard industrial steel pipelines and containers, yielding predictable results (KSR).
Regarding dependent claim 6, May, teaches:
The method of claim 1 (Fig. 4; [Abstract] & [0004]), wherein the PEC test results indicate an estimated average thickness of the pipe or tank at the location (Disclosed in combination: May: Fig. 1; [0014], [0019] & [0023]: Fig. 1 illustrates the test results monitor displaying “thickness” as an output; Hardy: [0001]-[0002]: confirms that inferring/estimating the thickness at the inspected location is the standard result of the PEC test).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention that the combined Pulsed Eddy Current testing as taught by May and Hardy, according to known methods, evaluates a volumetric area of the metal beneath the sensor footprint, therefore the estimated thickness calculated from the eddy current decay curve is inherently an average thickness of the location beneath the probe. May discloses the method wherein the PEC test results indicated an estimated thickness of the pipe or tank at the location, estimating “physical parameters”, “wall thickness, i.e., thickness of the metal object beneath the eddy current sensor” as the output results. Hardy further corroborates that the fundamental purpose of the PEC system is to infer “material thickness”. Therefore, estimating and calculating the thickness of the metal object beneath the eddy current sensor as taught by May and Hardy, yields an estimated average thickness (KSR) of the material over the footprint (inspection area) of the probe. PEC works via magnetic field volumes rather than single-point electronic lasers, resulting in thickness estimation that is an average of the area measured.
Regarding dependent claim 7, May, teaches:
The method of claim 1 (Fig. 4; [Abstract] & [0004]), wherein the outputting is to one or both of a storage device and a user interface (Disclosed in combination: May: Fig. 1; [0017] & [0019]: memory 25 corresponds to the “storage device” and display 26/monitor 26 corresponds to the user interface; Atherton: [0031]: provides secondary support listing various types of storage devices used to record the eddy current data).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to output the data to a user interface and/or storage device as taught by May and Atherton, according to known methods. The motivation to do so is because an operator would either immediately view the test results to identify pipeline degradation or store the data for subsequent analytical review. May discloses outputting the calculated properties (results) to display 26 (user interface) and/or saving them to memory 25 (storage device). Atherton teaches the standard industry practice of outputting and storing the data on hard drives, disks, or solid-state memories. The combination of prior art teachings is a combination of known elements and/or methods to improve the processing/outputting of results to a user, for ease of interpretation, yielding predictable results (KSR).
Regarding independent claim 9, May, teaches:
A system for pulsed eddy current (PEC) testing, the system comprising (Fig. 1; [0016]):
a PEC tester configured to collect a plurality of metal loss output measurements at various locations of a pipe or tank made up of a ferrous material (Disclosed in combination: May: [0003]; Hardy: [0034], [0079], [0082] & [0086]: teaches scanning the pipe circumferentially and longitudinally, collecting at various locations, and specifies the object inspected is made of “carbon steel” ferrous material)),
a processing system comprising:
a memory comprising computer readable instructions; and
a processing device for executing the computer readable instructions, the computer readable instructions controlling the processing device to perform operations comprising (Fig. 1; [0017] & [0019]: teaches a processing device, executing computer-readable instructions/stored programs from the memory):
receiving the metal loss output measurements ([0019]);
outputting the compensated metal loss output measurements as PEC test results (Disclosed in combination: May: Fig. 1; [0019]-[0020]; Hardy: [0034]).
May, in combination with Hardy, are silent in regard to:
wherein one or more energized high voltage cables are located within the pipe or tank; and
generating compensated metal loss output measurements to compensate for errors in the metal loss output measurements caused by the energized high voltage cables; and
However, Atherton, in combination with Hardy, further teach:
wherein one or more energized high voltage cables are located within the pipe or tank (Disclosed in combination: Atherton: Fig. 1; [0036]: places the energized power cable inside the pipe; Hardy: [0063]); and
generating compensated metal loss output measurements to compensate for errors in the metal loss output measurements caused by the energized high voltage cables (Disclosed in combination: Atherton: [0027]-[0029]: teaches the necessity of compensating for power cable interference; Hardy: [0063]: discloses the probe and the presence of a parasitic external magnetic field 530); and
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 computerized PEC system of May to include the noise-canceling signal compensation of Hardy, and to apply this system in a pipe containing an energized high-voltage cable as taught by Atherton, according to known methods. The motivation to do so stems from Atherton’s warning in paragraph [0027] that power cables can skew the electromagnetic responses (errors), making it of “great importance” to calculate corrections. A POSITA would implement hardware solutions like Hardy’s noise-cancellation (compensation) algorithms into May’s computer memory, and further include Atherton’s accurate corrections when operating a tool powered by internal cables, to output accurate, compensated metal loss test results, yielding predictable results (KSR).
Regarding dependent claim 11, May, teaches:
The system of claim 9 (Fig. 1; [0016]),
May, is silent in regard to:
wherein the operations further comprise creating eddy currents by causing an electrical current to be supplied to the PEC tester and causing the electrical current to be cut-off from the PEC tester.
However, Hardy, further teaches:
wherein the operations further comprise creating eddy currents by causing an electrical current to be supplied to the PEC tester ([0035] & [0042]: teaches supplying an electrical current to the PEC probe’s coil during a transmission phase to build up the magnetic field) and causing the electrical current to be cut-off from the PEC tester ([0042]-[0044]: teaches cutting off the electrical current from the tester coil to trigger the magnetic variation that creates the eddy currents in the object).
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 pulsed eddy current testing system of May to incorporate the specific electrical current supply and cut-off operations taught by Hardy, according to known methods. The motivation for this modification is to implement a reliable and optimized method for generated transient magnetic fields within the PEC inspection system. A POSITA would recognize that to successfully execute the PEC testing taught by May, the system requires a specific electrical modulation scheme to induce the eddy currents. Incorporating Hardy’s teaching of supplying and then abruptly cutting off the current provides the electromagnetic transition necessary to induce and subsequently measure the decaying eddy currents. Combining these references represents the application of a known technique (Hardy’s current modulation phases for PEC) to a known system (May’s PEC inspection device) to yield the predictable result of measuring wall thickness and corrosion in industrial structures (KSR).
Regarding dependent claim 12, May, teaches:
The system of claim 11 (Fig. 1; [0016]), wherein the operations further comprise calculating the metal loss output measurements based at least in part on the eddy currents (Disclosed in combination: May: [0003] & [0023]: teaches capturing the eddy current response and mathematically calculating the thickness/loss of the metal based on that response; Hardy: [0001]-[0002] & [0034]-[0035]: describes an analyzer that calculates “wall loss” (metal loss) by applying algorithms to the received eddy current signal).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to calculate metal loss output measurements based on the eddy currents, as taught by May and Hardy, according to known methods, because determining wall/metal loss is the fundamental, well-known purpose of utilizing Pulsed Eddy Current testing on pipes and tanks. A POSITA would configure the signal analyzer to output the thickness difference as a “metal loss” measurement to inform the operator of corrosion or structural degradation, thus yielding predictable results (KSR).
Regarding dependent claim 13, May, teaches:
The system of claim 9 (Fig. 1; [0016]), wherein the pipe or tank comprises a ferrous material (Disclosed in combination: May: [0003]: teaches that the container (tank) or pipe being tested by the pulsed eddy current system can be made of steel. Steel is an alloy of iron, making it a ferrous material; Hardy: [0086]: provides an example where the electrically conductive object being inspected via the PEC method is carbon steel, a ferrous material; Atherton: [0003] & [0037]: further corroborates the well-known industry standard of testing steel and ferromagnetic/ferrous pipes).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention that the pipe or tank of the combined May, Hardy, and Atherton system comprises a ferrous material, because Hardy provides a working example of inspecting a “carbon steel” object, and Atherton discusses testing “ferromagnetic” and “steel pipes”, according to known methods. The motivation to utilize a ferrous material for the pink or tank is simply the intended real-world application of assessing structural degradation, wall loss, and corrosion in standard industrial steel pipelines and containers, yielding predictable results (KSR).
Regarding dependent claim 14, May, teaches:
The system of claim 9 (Fig. 1; [0016]), wherein the PEC test results indicate an estimated average thickness of the pipe or tank at the location (Disclosed in combination: May: Fig. 1; [0014], [0019] & [0023]: Fig. 1 illustrates the test results monitor displaying “thickness” as an output; Hardy: [0001]-[0002]: confirms that inferring/estimating the thickness at the inspected location is the standard result of the PEC test).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention that the combined Pulsed Eddy Current testing as taught by May and Hardy, according to known methods, evaluates a volumetric area of the metal beneath the sensor footprint, therefore the estimated thickness calculated from the eddy current decay curve is inherently an average thickness of the location beneath the probe. May discloses the method wherein the PEC test results indicated an estimated thickness of the pipe or tank at the location, estimating “physical parameters”, “wall thickness, i.e., thickness of the metal object beneath the eddy current sensor” as the output results. Hardy further corroborates that the fundamental purpose of the PEC system is to infer “material thickness”. Therefore, estimating and calculating the thickness of the metal object beneath the eddy current sensor as taught by May and Hardy, yields an estimated average thickness (KSR) of the material over the footprint (inspection area) of the probe. PEC works via magnetic field volumes rather than single-point electronic lasers, resulting in thickness estimation that is an average of the area measured.
Regarding dependent claim 15, May, teaches:
The system of claim 9 (Fig. 1; [0016]), wherein the outputting is to one or both of a storage device and a user interface (Disclosed in combination: May: Fig. 1; [0017] & [0019]: memory 25 corresponds to the “storage device” and display 26/monitor 26 corresponds to the user interface; Atherton: [0031]: provides secondary support listing various types of storage devices used to record the eddy current data).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to output the data to a user interface and/or storage device as taught by May and Atherton, according to known methods. The motivation to do so is because an operator would either immediately view the test results to identify pipeline degradation or store the data for subsequent analytical review. May discloses outputting the calculated properties (results) to display 26 (user interface) and/or saving them to memory 25 (storage device). Atherton teaches the standard industry practice of outputting and storing the data on hard drives, disks, or solid-state memories. The combination of prior art teachings is a combination of known elements and/or systems and/or methods to improve the processing/outputting of results to a user, for ease of interpretation, yielding predictable results (KSR).
Claims 8, 10 & 16 are rejected under 35 U.S.C. 103 as being unpatentable over May, in view of Hardy, in view of Atherton, and further in view of Vaganay et al. (US 2023/0003687 A1, Pub. Date Jan. 5, 2023, hereinafter, Vaganay).
Regarding dependent claim 8, May, teaches:
The method of claim 1 (Fig. 4; [Abstract] & [0004]),
May, in combination with Hardy, and Atherton, are silent in regard to:
wherein the PEC results are expressed as a percentage of thickness of a wall of the pipe or tank relative to a nominal thickness.
However, Vaganay, further teaches:
wherein the PEC results are expressed as a percentage of thickness of a wall of the pipe or tank relative to a nominal thickness (Fig. 13; [0257]-[0258]: teaches expressing the PEC measurements as a percentage relative to the nominal/original thickness of the wall. Fig. 13 further illustrates the exact percentage output on a user interface).
PNG
media_image6.png
818
893
media_image6.png
Greyscale
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 output of the PEC testing method of May, Hardy, and Atherton to express the results as a percentage relative to a nominal thickness, as taught by Vaganay, according to known methods. The motivation to combine is provided by Vaganay, formatting the data as a percentage of the nominal thickness allows an operator to quickly configure display settings (e.g., color-coded heat maps) to easily identify specific regions that have a high risk of leakage due to severe wall loss (Vaganay: [0257]-[0259]), yielding predictable results (KSR).
Regarding dependent claim 10, May, teaches:
The system of claim 9 (Fig. 1; [0016]),
May, in combination with Hardy, are silent in regard to:
wherein the operations further comprise controlling the movement of the PEC tester to another location on the pipe or tank.
However, Atherton, in combination with Vaganay, further teach:
wherein the operations further comprise controlling the movement of the PEC tester to another location on the pipe or tank (Disclosed in combination: Atherton: [0020]: supports the mechanism of driving an eddy current tool toother locations within a pipe; Vaganay: [0005], [0120] & [0193]: teaches a control unit that directs a vehicle carrying the PEC tester to travel to secondary uninspected locations inside the tank via a propeller mechanism).
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 mobile pulsed eddy current inspection system of Vaganay to incorporate the pipe-traversing traction system of Atherton, according to known methods. The motivation for this modification is to adapt the inspection vehicle’s locomotion mechanism to the specific geometry and operational environment of the structure being inspected. A POSITA would recognize that navigating a fluid-filled tank (as in Vaganay) optimally utilizes a propeller. Whereas navigating a constrained, possibly dry tubular structure like a pipe necessitates a surface-contact drive system, such as the traction mechanism taught by Atherton. Combining these references represents a simple substitution of one known elements (a propeller-based drive system) for another known element (traction-based drive system) to yield predictable results (KSR). This modification is a design choice based on the inspection environment (pipe versus tank), allowed the pulsed eddy current tester to reposition itself to secondary locations to ensure comprehensive defect and wall-loss across various industrial infrastructures.
Regarding dependent claim 16, May, teaches:
The system of claim 9 (Fig. 1; [0016]),
May, in combination with Hardy, and Atherton, are silent in regard to:
wherein the PEC results are expressed as a percentage of thickness of a wall of the pipe or tank relative to a nominal thickness.
However, Vaganay, further teaches:
wherein the PEC results are expressed as a percentage of thickness of a wall of the pipe or tank relative to a nominal thickness (Fig. 13; [0257]-[0258]: teaches expressing the PEC measurements as a percentage relative to the nominal/original thickness of the wall. Fig. 13 further illustrates the exact percentage output on a user interface).
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 output of the PEC testing system of May, Hardy, and Atherton to express the results as a percentage relative to a nominal thickness, as taught by Vaganay, according to known methods. The motivation to combine is provided by Vaganay, formatting the data as a percentage of the nominal thickness allows an operator to quickly configure display settings (e.g., color-coded heat maps) to easily identify specific regions that have a high risk of leakage due to severe wall loss (Vaganay: [0257]-[0259]), yielding predictable results (KSR).
Claims 17 & 19-20 are rejected under 35 U.S.C. 103 as being unpatentable over Koenig et al. (US 2016/0290966 A1, Pub. Date Oct. 6, 2016, hereinafter, Koenig), in view of May.
Regarding independent claim 17, Koenig, teaches:
A machine learning system, comprising (Disclosed in combination: Koenig: [0073] & [Claim 46]; May: [0017]-[0019]: Koenig provides the “machine learning” operations, both references provide the hardware):
a memory comprising computer readable instructions (Disclosed in combination: Koenig: [0046]-[0047], [0051]-[0052] & [Claim 46]; May: Fig. 1; [0017]-[0019]); and
a processing device for executing the computer readable instructions, the computer readable instructions controlling the processing device to perform operations comprising (Disclosed in combination: Koenig: [0032], [0046]-[0047], [0051]-[0052], [0064] & [Claim 46]: provides “machine learning” operations; May: Fig. 1; [0017]-[0019]: discloses the processing device and memory executing instructions to run the predictive (machine learning) algorithms):
receiving training data as input, the training data comprising pulsed eddy current (PEC) measurements from field tests, and pipe profiles from field testing ([0046]-[0047], [0061], [0071]-[0074] & [Claim 46]: collects PEC signals from calibration samples to create “training data”);
preprocessing the training data by performing feature extraction on the PEC measurements, data normalization, and scaling ([0073], [Claim 46], [Claim 49] & [Claim 50]: discloses extracting features, normalizing, and scaling the data (dividing/subtracting)); and
generating a trained machine learning model using results of the preprocessing of the training data, the trained machine learning model taking as input at least meat loss output measurements from a PEC tester and generated compensated metal loss output measurements ([0073], [Claim 46], [Claim 47], [Claim 48], [Claim 49] & [Claim 50]: trains the ML algorithm with the data, applying it to lift-off compensated PEC features to measure remaining wall thickness (metal loss)).
Koenig, is silent in regard to:
simulated data from finite element modeling simulations,
However, May, further teaches:
simulated data from finite element modeling simulations ([0026]-[0029] & [Claim 12]: teaches supplementing calibration datasets with finite element simulations),
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 machine learning system of Koenig to incorporate the finite element modeling simulations taught by May, into the training data set, according to known methods. The motivation for this modification is to create a more robust, comprehensive, and cost-effective training dataset for the machine learning algorithm. A POSITA would recognize that relying on physical calibration blocks and field tests (as in Koenig) to train an algorithm is expensive, time-consuming, and limits the training data to physical mock-ups. Incorporating May’s teaching of generating training data via finite element computational simulations, a POSITA could simulate thousands of permutations of pipe profiles, defect size, and lift-off scenarios to feed into Koenig’s machine learning model. Combining these references represents the application of a known technique (using FEM simulated data to train a PEC estimation model) to a known method (Koenig’s ML feature extraction pipeline) to yield the predictable result of an accurate, trained predictive model capable of compensating for complex metal loss outputs without requiring exhaustive physical calibration samples (KSR).
Regarding dependent claim 19, Koenig, teaches:
The machine learning system of claim 17 (Disclosed in combination: Koenig: [0073] & [Claim 46]: established the ML system; May: [0015], [0017]-[0019] & [0026]-[0029]: provides hardware and algorithms for ML), wherein generating the trained machine learning model comprises ([0073] & [Claim 46]: establishes the fundamental process of generating/training the ML model using collected datasets) aligning features extracted (Disclosed in combination: Koenig: [0073] & [Claim 46]: teaches combining extracted features into a dataset; May: [0015] & [0026]-[0029]: combines physical and simulated parameters (features extracted) to derive a single function)
Koenig, is silent in regard to:
from simulation and field test data.
However, May, further teaches:
from simulation and field test data ([0026]-[0029]: lists the two data sources: physical calibration sample measurements (field test data) and finite element and computational simulations, mixing them creates the combined dataset).
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 machine learning dataset generation process of Koenig to incorporate and align features extracted from the finite element simulations taught by May, alongside the features extracted from the physical field tests, according to known methods. The motivation for this modification is to overcome the physical and economic limitations of relying solely on physical manufacturing calibration pipes or tanks with permutations of wall loss, defect geometry, and lift-off variation is expensive and time-consuming. Finite element simulations can generate massive datasets, covering edge-case defects. Incorporating May’s teaching to utilize both data sources, a POSITA would be motivated to combine, and therefore mathematically align, the simulated features with the physical field test features into a single training matrix. This ensures the machine learning algorithm developed in Koenig is trained on a dataset that is broad (via simulation) and empirically grounded (via field testing). This represents a combination of known methods (Koenig’s ML feature compilation and May’s multi-source data generation) to yield the predictable results of a robust, accurate predictive model capable of generalizing to unseen field conditions (KSR).
Regarding dependent claim 20, Koenig, teaches:
The machine learning system of claim 17 (Disclosed in combination: Koenig: [0073] & [Claim 46]: established the ML system; May: [0015], [0017]-[0019] & [0026]-[0029]: provides hardware and algorithms for ML), wherein generating the trained machine learning model comprises ([0073] & [Claim 46]: establishes the generation of the trained ML model) fine tuning the trained machine learning model (Disclosed in combination: Koenig: [0073] & [Claim 46]; May: [0015] & [0026]-[0029]: updates a baseline simulation model with new empirical calibration data to correct for field errors is the functional equivalent of fine-tuning the model) using additional field test data ([0073] & [Claim 46]: teaches acquiring new signals from physical calibration samples in the field to use as training data).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to generate the trained machine learning model of Koenig and May by fine-tuning a baseline model using additional field test data, according to known methods. A POSITA would be motivated to utilize the calibration steps taught by Koenig (acquiring a PEC signal on a physical calibration sample) not just to build a model, but to update, adjust, or “fine-tune” the weights and parameters of the existing baseline simulation model. Applying additional field test data (Koenig’s physical calibration samples) to adjust an already trained machine learning model (May’s simulated transfer function) is a standard machine learning practice designed to yield the predictable result of localized, accurate wall thickness measurements that compensate for specific field conditions (KSR).
Claim 18 is rejected under 35 U.S.C. 103 as being unpatentable over Koenig et al. (US 2016/0290966 A1, Pub. Date Oct. 6, 2016, hereinafter, Koenig), in view of May, and further in view of Hardy.
Regarding dependent claim 18, Koenig, teaches:
The machine learning system of claim 17 (Disclosed in combination: Koenig: [0073] & [Claim 46]: established the ML system; May: [0015], [0017]-[0019] & [0026]-[0029]: provides hardware and algorithms for ML), wherein generating the trained machine learning model comprises mapping (Disclosed in combination: Koenig: [0073] & [Claim 46]: established the ML system; May: [0015] & [0026]-[0029]: teaches building the model by mapping measurement inputs to simulated object properties) to features extracted from the PEC measurements (Disclosed in combination: Koenig: [0073] & [Claim 46]: uses the “feature extraction”; May: [0015] & [0026]-[0029]: teaches “fit coefficients” which are the mathematical features extracted from the raw PEC signal mapped to the properties).
Koenig, is silent in regard to:
an artificial intelligence-based surrogate model relative permeability
However, May, further teaches:
an artificial intelligence-based surrogate model ([0015] & [0026]-[0029]: creates surrogate model by training an empirical transfer function (multivariate regression, an AI/ML technique) to approximate finite element computational simulations) relative permeability ([0015]: lists permeability as one of the target properties being mapped by the model)
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 machine learning system of Koenig to incorporate the mapping of features to permeability using surrogate modeling techniques taught by May, according to known methods. The motivation for this modification is to improve the accuracy and robustness of the machine learning model’s thickness predictions. A POSITA would recognize that in real-world scenarios, variations in the magnetic permeability of the pipe or tank wall distort the induced eddy currents. By incorporating May’s teaching to map the extracted features relative to permeability, the machine learning system can compensate for the material variations. Combining the references represents the simple combination of known elements (Koenig’s ML pipeline and May’s permeability surrogate mapping) to yield the predictable result of an accurate predictive model that isolates true metal loss from permeability fluctuations (KSR).
Koenig, in combination with May, are silent in regard to:
and magnetic field strengths
However, Hardy, further teaches:
and magnetic field strengths ([0035]: confirms that the PEC sensors are measuring the magnetic fields)
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to configure the machine learning system of Koenig and May to map features extracted from the magnetic fields strengths, as detailed by Hardy, according to known methods. The motivation for this modification is to ground the machine learning feature extraction in the physical parameters being measured by the hardware. A POSITA understands that a PEC probe does not directly measure “thickness”; it measures the decay of the produced magnetic field over time. Incorporating Hardy’s teaching ensures the pipeline data is accurately configured to recognize the raw input data, from which the ML features are extracted, are precisely time-varying magnetic field strengths. This is the application of a known physical property of the sensor (Hardy) to a known data processing method (Koenig/May) to achieve the predictable result of linking the raw sensor hardware outputs to the software feature extraction algorithms (KSR).
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to HUGO NAVARRO whose telephone number is (571)272-6122. The examiner can normally be reached Monday-Friday 08:30-5:00 pm EST.
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, Eman Alkafawi can be reached at 571-272-4448. 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.
/HUGO NAVARRO/ Examiner, Art Unit 2858 April 7, 2026
/EMAN A ALKAFAWI/Supervisory Patent Examiner, Art Unit 2858 4/14/2026
Prosecution Insights