Metering equipment intelligent evaluation method and system based on extreme scene simulation

Abstract

The invention belongs to the technical field of electric power metering equipment fault detection, and particularly relates to a metering equipment intelligent evaluation method and system based on extreme scene simulation. Aiming at the defect that the operation condition of equipment under extreme conditions is difficult to simulate in the electric power metering equipment detection in the prior art, the invention adopts the following technical scheme: the metering equipment intelligent evaluation method based on extreme scene simulation comprises the following steps: constructing a test parameter generation model based on a bright electric power large model, learning feature distribution of a historical fault mode through a self-attention mechanism in the test parameter generation model, and obtaining parameters exceeding the design specification of the to-be-tested metering equipment in the simulation extreme scene; deploying a high-precision sensor array; collecting data of the measured metering equipment through a high-precision sensor array in a simulated extreme scene; and performing fault analysis according to the acquired data of the measured metering equipment. According to the method, the single-dimension limitation of a traditional testing method is broken through, and the limit durability of the equipment is evaluated more comprehensively.

Description

Technical Field

[0001] The present invention belongs to the technical field of power metering equipment fault detection, and in particular relates to a metering equipment intelligent evaluation method and system based on extreme scenario simulation. Background Art

[0002] With the rapid development of smart grid technology, new types of metering and data collection equipment, such as IoT meters and energy controllers, are constantly emerging. While these devices enhance the intelligence of the grid, they also place higher demands on their stability and reliability. However, in actual use, some devices may reveal hidden issues under specific conditions, which are often difficult to fully uncover through conventional full-performance testing and field trials. Therefore, in-depth research on the extreme durability and potential defects of new metering and data collection equipment is an urgent need.

[0003] Specifically, existing methods and systems can detect equipment failures to a certain extent, but they still have many shortcomings. First, traditional methods and systems often perform fault analysis only after the equipment fails, which is a passive detection method and cannot prevent potential failures in advance. Second, traditional methods and systems rely on manual analysis, which is not only time-consuming and labor-intensive, but also easily affected by human factors, resulting in inaccurate analysis results. In addition, traditional methods and systems have difficulty simulating the operation of equipment under extreme conditions, and therefore cannot fully assess the ultimate durability of the equipment. Summary of the Invention

[0004] To address the shortcomings of existing power metering equipment testing, which struggles to simulate equipment operating under extreme conditions, this present invention provides an intelligent metering equipment evaluation method based on extreme scenario simulation. By simulating extreme operating conditions and abnormal operations, it conducts stress and destructive testing on metering and acquisition terminals to assess their ultimate durability and fault warning capabilities. This invention also provides an intelligent metering equipment evaluation system based on multi-dimensional fault analysis.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: an intelligent evaluation method for metering equipment based on extreme scenario simulation, the intelligent evaluation method for metering equipment based on extreme scenario simulation comprising:

[0006] A test parameter generation model based on the Guangming Electric Power large-scale model was constructed. The self-attention mechanism was used in the test parameter generation model to learn the characteristic distribution of historical failure modes and obtain parameters in simulated extreme scenarios that exceeded the design specifications of the metering equipment under test.

[0007] deploying high-precision sensor arrays;

[0008] Collect data from the measured measuring equipment through a high-precision sensor array in simulated extreme scenarios;

[0009] Perform fault analysis based on the collected data of the measured measuring equipment.

[0010] The present invention provides an intelligent evaluation method for metering equipment based on extreme scenario simulation. The test parameter generation model is based on the Guangming Electric Power large model. The test parameter generation model learns the characteristic distribution of historical failure modes through a self-attention mechanism to obtain parameters in simulated extreme scenarios that exceed the design specifications of the metering equipment to be tested. In the simulated extreme scenarios, the data of the metering equipment to be tested is collected through a high-precision sensor array, and fault analysis is performed, breaking through the single-dimensional limitations of traditional testing methods and more comprehensively evaluating the extreme durability of the equipment.

[0011] As an improvement, the high-precision sensor array is used to collect at least two of the following data: mechanical vibration, electromagnetic radiation, thermodynamic distribution, microscopic deformation, and chemical corrosion;

[0012] The high-precision sensor array integrates industrial high-speed cameras and acoustic emission equipment to capture transient anomalies.

[0013] As an improvement, the test dimensions of extreme scenarios include at least two of the following:

[0014] Voltage-current test: Continuously detect the heating condition of the equipment and the working condition of key components in overvoltage or overcurrent environment;

[0015] Electromagnetic-mechanical testing: applying high-frequency mechanical vibration in a strong magnetic field environment to detect the magnetostrictive effect of ferromagnetic materials;

[0016] Thermal-chemical testing: gradient temperature increase and simultaneous injection of corrosive aerosol to analyze the oxidation kinetics of the material;

[0017] Optical-electrical-thermal triple-field testing: Using high-energy laser local heating in conjunction with high-voltage pulses to test the breakdown characteristics of insulating materials;

[0018] Bio-electronic testing: Cultivating microbial colonies in a hot and humid environment to analyze the impact of biofilm growth on PCB impedance characteristics.

[0019] As an improvement, the extreme scenario includes at least two of the following:

[0020] Extreme load test: simulates the current to increase to several times the normal value in a short period of time and / or suddenly increases to several times the normal value, while monitoring the internal temperature of the equipment and whether the components are burned or the performance is degraded;

[0021] Multi-factor superposition test: Test by combining different environmental conditions, including high temperature, vibration, humidity, and chemical corrosion;

[0022] Long-term endurance testing: By simulating the continuous operation of the equipment for months or even years, the slow changes in its performance are observed;

[0023] Intelligent scenario simulation test: Based on historical failure data, automatically generate the test scenarios that are most likely to cause problems:

[0024] For data collection terminals, scenarios where they simultaneously receive massive amounts of commands are simulated to test the limits of their data processing capabilities. For rural users, metering equipment and data collection terminals are installed outdoors to simulate the instantaneous high-voltage pulses caused by lightning strikes and test the effectiveness of lightning protection devices.

[0025] Destructive critical point testing: By gradually increasing the test intensity, the critical value of device failure is accurately found.

[0026] As an improvement, extreme scenarios include extreme load testing. The extreme load testing process includes:

[0027] Deep decoupling: In-depth analysis of multi-source data, stripping away noise, and decoupling coupled composite faults into independent basic failure feature units;

[0028] Intelligent reconstruction and deduction: Based on decoupling characteristics, causal reasoning is used to proactively combine and deduce more likely or extreme unexpected compound failure modes that have never occurred in history but are bound to exist in physical logic;

[0029] Generate dynamic load curves: Based on the reconstructed unexpected composite failure mode and combined with the specific model and design characteristics of the device under test, a highly customized load curve is dynamically synthesized. This load curve includes non-uniform amplitudes and complex timing combinations, rather than simple multiple overcurrents.

[0030] As an improvement, the extreme scenario includes multi-factor superposition testing. The multi-factor superposition testing process includes:

[0031] Construct a three-dimensional environment matrix;

[0032] Weight distribution;

[0033] Start the load loading system;

[0034] Detect whether there are data anomalies;

[0035] If no data anomaly is detected, the loading continues and an environmental correlation map is generated; if data anomaly is detected, the parameters of the load loading system are dynamically adjusted.

[0036] As an improvement, the extreme scenario includes a long-term endurance test. The long-term endurance test process includes:

[0037] Set up accelerated aging model;

[0038] Long-short-term memory model is used for lifespan prediction;

[0039] Load typical working conditions;

[0040] Continue for a preset duration;

[0041] Detect whether performance degradation exceeds a threshold;

[0042] If the performance degradation does not exceed the threshold within the preset time, the operating parameters are adjusted; if the performance degradation exceeds the threshold within the preset time, the aging point is located and a life curve is generated.

[0043] As an improvement, extreme scenarios include intelligent scenario simulation tests. The intelligent scenario simulation test process includes:

[0044] Input device topology;

[0045] Knowledge graph call;

[0046] Generate abnormal scenarios using generative adversarial networks;

[0047] Digital twin preview: obtain and screen several scenarios;

[0048] Loading of the tested equipment;

[0049] Data discrepancy analysis;

[0050] If the data difference analysis is greater than the threshold, the parameters of the generative adversarial network are modified; if the data difference is less than the threshold, optimization suggestions are output.

[0051] As an improvement, the extreme scenario includes destructive critical point testing, and the destructive critical point process includes:

[0052] Set safety thresholds;

[0053] Apply load in a certain step size;

[0054] 3D deformation monitoring;

[0055] When the critical value is reached, emergency braking protection;

[0056] Perform reverse engineering analysis using micro-computed tomography to generate material improvement recommendations.

[0057] The intelligent evaluation system for measuring equipment based on multi-dimensional fault analysis is used to implement the aforementioned intelligent evaluation system for measuring equipment based on multi-dimensional fault analysis. The intelligent evaluation system for measuring equipment based on multi-dimensional fault analysis includes:

[0058] The model building module builds a test parameter generation model based on the Guangming Electric Power large model. In this test parameter generation model, the characteristic distribution of historical failure modes is learned through the self-attention mechanism to obtain parameters in simulated extreme scenarios that exceed the design specifications of the metering equipment under test.

[0059] Extreme scenario simulation device, used to simulate extreme scenarios that exceed the design specifications of the metering equipment to be tested;

[0060] High-precision sensor arrays are used to collect data from measured measuring equipment in extreme scenarios;

[0061] The fault analysis module is used to perform fault analysis on the collected data of the measured measuring equipment.

[0062] The beneficial effects of the intelligent evaluation method for metering equipment based on extreme scenario simulation of the present invention are: the test parameter generation model is based on the Guangming Electric Power large model, and the characteristic distribution of historical failure modes is learned in the test parameter generation model through the self-attention mechanism to obtain parameters in simulated extreme scenarios that exceed the design specifications of the metering equipment to be tested. In the simulated extreme scenarios, the data of the metering equipment to be tested is collected through a high-precision sensor array, and fault analysis is performed, breaking through the single-dimensional limitations of traditional testing methods and more comprehensively evaluating the extreme durability of the equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 This is a functional block diagram of a metering equipment intelligent evaluation system based on multi-dimensional fault analysis according to an embodiment of the present invention.

[0064] Figure 2 This is a data flow diagram of a metering equipment intelligent evaluation system based on multi-dimensional fault analysis according to an embodiment of the present invention.

[0065] Figure 3 This is a flow chart of an intelligent evaluation method for metering equipment based on extreme scenario simulation according to an embodiment of the present invention.

[0066] Figure 4 This is a flow chart of an embodiment of the present invention, in which an intelligent evaluation method for metering equipment based on extreme scenario simulation is used to perform an extreme load test.

[0067] Figure 5 This is a flow chart of a multi-factor superposition test performed by an intelligent evaluation method for metering equipment based on extreme scenario simulation according to an embodiment of the present invention.

[0068] Figure 6 This is a flow chart of a long-term endurance test of a measuring equipment intelligent evaluation method based on extreme scenario simulation according to an embodiment of the present invention.

[0069] Figure 7 This is a flow chart of an embodiment of the present invention when performing an intelligent scenario simulation test using an intelligent evaluation method for metering equipment based on extreme scenario simulation.

[0070] Figure 8 This is a flow chart of a destructive critical point test performed by an intelligent evaluation method for measuring equipment based on extreme scenario simulation according to an embodiment of the present invention. DETAILED DESCRIPTION

[0071] The technical solutions of the embodiments of the present invention are explained and described below, but the following embodiments are only preferred embodiments of the present invention and are not exhaustive. Based on the embodiments in the embodiments, other embodiments obtained by those skilled in the art without creative work are all within the scope of protection of the present invention.

[0072] See also Figures 1 to 8 The embodiment of the present invention provides an intelligent evaluation method for metering equipment based on extreme scenario simulation, and the intelligent evaluation method for metering equipment based on extreme scenario simulation includes:

[0073] A test parameter generation model based on the Guangming Electric Power large-scale model was constructed. The self-attention mechanism was used in the test parameter generation model to learn the characteristic distribution of historical failure modes and obtain parameters in simulated extreme scenarios that exceeded the design specifications of the metering equipment under test.

[0074] deploying high-precision sensor arrays;

[0075] Collect data from the measured measuring equipment through a high-precision sensor array in simulated extreme scenarios;

[0076] Perform fault analysis based on the collected data of the measured measuring equipment.

[0077] The present invention provides an intelligent evaluation method for metering equipment based on extreme scenario simulation. The test parameter generation model is based on the Guangming Electric Power large model. The test parameter generation model learns the characteristic distribution of historical failure modes through a self-attention mechanism to obtain parameters in simulated extreme scenarios that exceed the design specifications of the metering equipment to be tested. In the simulated extreme scenarios, the data of the metering equipment to be tested is collected through a high-precision sensor array, and fault analysis is performed, breaking through the single-dimensional limitations of traditional testing methods and more comprehensively evaluating the extreme durability of the equipment.

[0078] Example 1

[0079] See also Figure 1 The first embodiment of the present invention, an intelligent metering equipment evaluation system based on multi-dimensional fault analysis, is integrated into a smart metering industrial control platform. The smart metering industrial control platform includes three functional modules: full performance testing, on-site test installation testing, and fault analysis testing. The fault analysis testing module, a multi-dimensional fault analysis-based intelligent metering equipment evaluation system, simulates extreme operating conditions and abnormal operations to perform stress and destructive testing on metering acquisition terminals to assess their ultimate durability and fault warning capabilities.

[0080] The intelligent evaluation system for metrological equipment based on multi-dimensional fault analysis is designed to implement the fault analysis and inspection process and enhance support for the simulation system. It has functions such as fault analysis and inspection plan template management, fault analysis and inspection plan customization management, fault analysis and inspection plan execution, research and inspection data result display, and multi-dimensional auxiliary analysis of research and inspection results. It realizes the customization of research and inspection plans and helps on-site metrology personnel explore more research testing possibilities for equipment and systems.

[0081] The system architecture of the intelligent evaluation system for metering equipment based on multi-dimensional fault analysis includes:

[0082] Hardware layer: includes metering and collection equipment (such as electricity meters, collection terminals, etc.), simulation area (including dual-mode communication test unit), high-supply high-metering, high-supply low-metering users and other test equipment, as well as sensors and communication modules for data collection and transmission;

[0083] Data layer: responsible for storing and managing various types of data generated during the fault analysis process, including equipment status data, fault data, test condition data, etc.

[0084] Application layer: Contains core functional modules such as fault analysis and inspection data result display, multi-dimensional auxiliary analysis, result archiving, inspection condition management, and inspection plan management;

[0085] User interface layer: Provides a graphical user interface (GUI) to facilitate user interaction with the system, such as setting test conditions and viewing analysis results.

[0086] See also Figure 2 ,The data flow process of the complete detection process includes:

[0087] (1) The collection microservice sends commands to the on-site equipment through the communication service. The communication management service provides communication functions such as device access, collection service, scheduling service, and message parsing.

[0088] (2) After the field equipment collects data, it is sent to the collection communication service, and after passing through the communication management layer, it is stored in the distributed relationship database.

[0089] (3) The detection platform monitors the message data, collects it and sends it to the laboratory service. Through the interface interaction service, it interacts with the relevant microservices for data. The on-site test microservice provides services such as data analysis and test result output.

[0090] (4) Data is mainly stored in a distributed relational database, while specific business application data can be stored in the form of files.

[0091] (5) Business applications query or update data stored in the distributed relational database.

[0092] In this embodiment, the intelligent evaluation method for metering equipment based on extreme scenario simulation includes:

[0093] A test parameter generation model based on the Guangming Electric Power large-scale model was constructed. The self-attention mechanism was used in the test parameter generation model to learn the characteristic distribution of historical failure modes and obtain parameters in simulated extreme scenarios that exceeded the design specifications of the metering equipment under test.

[0094] deploying high-precision sensor arrays;

[0095] Collect data from the measured measuring equipment through a high-precision sensor array in simulated extreme scenarios;

[0096] Perform fault analysis based on the collected data of the measured measuring equipment.

[0097] In this embodiment, the high-precision sensor array is used to collect at least two of the following data: 12 types of physical field data, including mechanical vibration (0-20kHz), electromagnetic radiation (DC-40GHz), thermodynamic distribution (-50℃~300℃), microscopic deformation (nanoscale), and chemical corrosion (gas / liquid concentration).

[0098] In this embodiment, a high-precision sensor array integrates an industrial high-speed camera (1000fps) and an acoustic emission device to capture transient abnormal phenomena.

[0099] In this embodiment, the test dimensions for extreme scenarios include the following:

[0100] Voltage-current test: Continuously detect the heating condition of the equipment and the working condition of key components in overvoltage or overcurrent environment;

[0101] Electromagnetic-mechanical testing: applying high-frequency mechanical vibration in a strong magnetic field environment to detect the magnetostrictive effect of ferromagnetic materials;

[0102] Thermal-chemical testing: gradient temperature increase and simultaneous injection of corrosive aerosol to analyze the oxidation kinetics of the material;

[0103] Optical-electrical-thermal triple-field testing: Using high-energy laser local heating in conjunction with high-voltage pulses to test the breakdown characteristics of insulating materials;

[0104] Bio-electronic testing: Cultivating microbial colonies in a hot and humid environment to analyze the impact of biofilm growth on PCB impedance characteristics.

[0105] Specifically, during voltage-current testing, the Guangming model uses a self-attention mechanism to analyze the multi-dimensional correlation of overvoltage / overcurrent events in the historical fault library (such as the nonlinear relationship between the current mutation rate and the thermal breakdown threshold of a specific chip). For example, the model dynamically generates a non-uniform step load curve: based on a preset safety boundary (such as 150% of the nominal value), it intelligently superimposes high-frequency microsecond current spikes (peak value accurate to 102%-105% of the design margin) and simultaneously injects specific harmonic components (such as the 13th harmonic, with weights generated by training from historical insulation failure cases), thereby accurately triggering the chain reaction of "local overheating-material lattice instability", exposing the transient thermal breakdown critical point that cannot be captured in traditional step-up pressure testing (positioning accuracy reaches ±1%), and simultaneously generating a correlation model between the component degradation rate and the current spectrum characteristics, ultimately discovering the tolerance of the equipment under high-intensity working conditions.

[0106] During electromagnetic-mechanical testing, the model uses self-attention weights to analyze massive amounts of ferromagnetic material failure data, identify the hidden variable relationship between the magnetostrictive effect and the vibration frequency (such as the acceleration effect of 5kHz vibration on the domain wall motion of silicon steel sheets under a 1T magnetic field), and dynamically generate a spatiotemporal coupled stress field: embedding an asymmetric mechanical vibration waveform in a strong magnetic field environment (the amplitude is adaptively adjusted according to the material fatigue cumulative damage model), and adjusting the magnetic field gradient direction in real time to simulate the eddy current distribution under real working conditions, thereby revealing the energy threshold (unit: J / mm) for microcrack initiation and propagation. 2 ), established a failure prediction equation for the coupling of magnetic, vibration and thermal fields, increased the crack detection sensitivity to the millimeter level (five times higher than the traditional single stress test), and finally discovered the law of microcrack generation in the equipment under extreme electromagnetic vibration.

[0107] During thermo-chemical testing, the temperature-concentration-time dependencies in historical corrosion data are decoupled using a self-attention mechanism. The model generates a dynamic corrosion protocol: During a temperature ramp of 50°C / min, the Cl- aerosol concentration is adjusted in real time (fluctuating between 50 and 200 ppm) according to the material oxidation kinetics equation. A pulsed high-concentration corrosive medium (500 ppm for 0.5 seconds) is injected into the critical temperature range (e.g., 80-120°C). This captures the transient chemical-thermodynamic synergistic effects of passive film rupture, quantifies the nonlinear attenuation coefficient (defined as the second-order derivative of the corrosion rate with respect to temperature), and precisely locates the critical temperature at which the material transitions from uniform corrosion to pitting corrosion (with an error of <±2°C), ultimately revealing the nonlinear attenuation characteristics of high-temperature accelerated corrosion.

[0108] During the optical-electrical-thermal three-field test, the model integrates the coupling relationship between laser energy distribution, electric field intensity gradient and heat conduction path through self-attention weights. First, a spatiotemporal synchronized composite stress field is generated: 10kW / m 2When the laser scans and heats in a spiral path, a high-voltage pulse sequence of seconds is applied synchronously (the voltage amplitude is dynamically adjusted by ±200V according to local temperature feedback) to construct a field strength-temperature gradient field in the non-uniform insulating medium, thereby achieving millisecond-level dynamic tracking of the breakdown path under composite stress and drawing a cloud map of the insulation failure probability (resolution 0.1mm). 2 ), the critical point positioning accuracy was improved to ±0.5% (the traditional single-field test accuracy was only ±5%), and the synergistic breakdown effect of dielectric space charge accumulation and laser hot spot was discovered, ultimately locating the critical point of insulation failure under composite stress with an accuracy of ±0.5%.

[0109] During the bio-electronic test, the model uses the self-attention mechanism to associate microbial metabolic activity (such as the secretion rate of extracellular polymers) with environmental parameters (temperature, humidity, and pH) to dynamically construct a biofilm growth incentive function: in an environment with a relative humidity of 95%, the dry-wet cycle (5-30 minutes) is periodically switched according to the characteristics of the PCB coating material, and a specific amino acid combination is injected to accelerate the generation of corrosive metabolites, thereby determining the biofilm thickness-impedance characteristic mutation threshold (for example, 0.1mm thickness causes a 40% decrease in impedance), and establishing a quantitative relationship model between the microbial community diversity index and the leakage current (R 2 >0.95), revealing the critical humidity point (RH92.7±0.3%) at which the hydrolase activity of the coating surges, and ultimately discovering the threshold of the insulation performance mutation of the device coating under biodegradation.

[0110] In this embodiment, extreme scenarios include the following:

[0111] Extreme load test: simulates the current to increase to several times the normal value in a short period of time and / or suddenly increases to several times the normal value, while monitoring the internal temperature of the equipment and whether the components are burned or the performance is degraded;

[0112] Multi-factor superposition test: Test by combining different environmental conditions, including high temperature, vibration, humidity, and chemical corrosion;

[0113] Long-term endurance testing: By simulating the continuous operation of the equipment for months or even years, the slow changes in its performance are observed;

[0114] Intelligent scenario simulation test: Based on historical failure data, automatically generate the test scenarios that are most likely to cause problems:

[0115] For data collection terminals, scenarios where they simultaneously receive massive amounts of commands are simulated to test the limits of their data processing capabilities. For rural users, metering equipment and data collection terminals are installed outdoors to simulate the instantaneous high-voltage pulses caused by lightning strikes and test the effectiveness of lightning protection devices.

[0116] Destructive critical point testing: By gradually increasing the test intensity, the critical value of device failure is accurately found.

[0117] In this embodiment, the extreme scenario includes an extreme load test, and the extreme load test process includes:

[0118] Deep Decoupling: This model thoroughly analyzes multi-source data, strips away noise, and decouples coupled compound faults into independent fundamental failure signature units. Rather than simply matching historical faults, the model deeply analyzes multi-source data (electrical parameters, environmental data, component status, text reports, etc.), strips away noise, and decouples coupled compound faults (such as "overcurrent + temperature rise leading to sampling distortion") into independent fundamental failure signature units (such as pure overcurrent stress characteristics and the temperature rise-performance degradation relationship of specific components).

[0119] Intelligent reconstruction and deduction: Based on decoupling characteristics and applying causal reasoning, we proactively combine and deduce unexpected, complex failure modes that have never occurred in history but are inherently logically present. For example, we intelligently combine the instantaneous high-voltage waveform characteristics of model A with the sustained high-temperature aging pattern of model B to generate a new scenario of "high-voltage shock combined with high-temperature aging."

[0120] Generate dynamic load curves: Based on the reconstructed unexpected composite failure modes, combined with the specific model and design features of the device under test (such as known weak points), a highly customized load curve is dynamically synthesized. This load curve includes non-uniform amplitude (precisely slightly exceeding the design margin) and complex timing combinations (multiple decoupling stresses such as overcurrent, harmonics, and power switching are applied in a specific order to simulate the worst real-world coupling conditions), rather than a simple multiple overcurrent.

[0121] The extreme load test of this embodiment goes beyond traditional extreme load tests based on experience or standards. It can proactively design extreme working conditions that simulate unknown risks, accurately induce deep-seated design defects or boundary failures (such as protection logic vulnerabilities and chain failures of related components) of equipment under unexpected composite stresses, significantly improve test coverage and efficiency, discover hidden dangers that cannot be revealed by conventional rough overloads, and provide accurate failure physics basis for design optimization.

[0122] In this embodiment, the extreme scenario includes a multi-factor superposition test, and the multi-factor superposition test process includes:

[0123] Construct a three-dimensional environment matrix;

[0124] Weight distribution;

[0125] Start the load loading system;

[0126] Detect whether there are data anomalies;

[0127] If no data anomaly is detected, the loading continues and an environmental correlation map is generated; if data anomaly is detected, the parameters of the load loading system are dynamically adjusted.

[0128] Failures in real-world scenarios are often caused by a combination of factors. The system performs tests by combining different environmental conditions. For example, the high-temperature and vibration test exposes the device to 85°C temperatures while simultaneously applying high-frequency mechanical vibrations to observe whether connections loosen or break due to thermal expansion and vibration. The humidity and chemical corrosion test exposes the device to salt spray in a 95% humidity environment to measure the aging rate of the device casing under long-term corrosion.

[0129] In this embodiment, the extreme scenario includes a long-term endurance test, and the long-term endurance test process includes:

[0130] Set up accelerated aging model;

[0131] Long-short-term memory model is used for lifespan prediction;

[0132] Load typical working conditions;

[0133] Continue for a preset duration;

[0134] Detect whether performance degradation exceeds a threshold;

[0135] If the performance degradation does not exceed the threshold within the preset time, the operating parameters are adjusted; if the performance degradation exceeds the threshold within the preset time, the aging point is located and a life curve is generated.

[0136] Traditional testing typically focuses only on short-term performance. However, this example simulates continuous operation of the device for months or even years to observe slow changes in performance. For example, the device is operated continuously at rated power for 1,000 hours, and the consumption of key components is recorded.

[0137] In this embodiment, the extreme scenario includes an intelligent scenario simulation test, and the intelligent scenario simulation test process includes:

[0138] Input device topology;

[0139] Knowledge graph call;

[0140] Generate abnormal scenarios using generative adversarial networks;

[0141] Digital twin preview: obtain and screen several scenarios;

[0142] Loading of the tested equipment;

[0143] Data discrepancy analysis;

[0144] If the data difference analysis is greater than the threshold, the parameters of the generative adversarial network are modified; if the data difference is less than the threshold, optimization suggestions are output.

[0145] Based on historical fault data, the method in this embodiment automatically generates test scenarios that are "most likely to cause problems." For example, for data collection terminals, a scenario of simultaneously receiving massive instructions (e.g., tens of thousands of control commands per second) is simulated to test the limits of their data processing capabilities. For rural users, metering equipment and data collection terminals are installed outdoors to simulate the transient high-voltage pulses caused by lightning strikes to test the effectiveness of lightning protection devices.

[0146] In this embodiment, the extreme scenario includes a destructive critical point test, and the destructive critical point process includes:

[0147] Set safety thresholds;

[0148] Apply load in a certain step size;

[0149] 3D deformation monitoring;

[0150] When the critical value is reached, emergency braking protection;

[0151] Perform reverse engineering analysis using micro-computed tomography to generate material improvement recommendations.

[0152] By gradually increasing the test intensity, we can pinpoint the critical failure point for a device. For example, for insulating materials, we can continuously increase the voltage until it breaks down, and then record the maximum withstand value before breakdown. For mechanical structures, we can gradually increase the load until deformation occurs, analyzing the upper limit of their compressive resistance.

[0153] The intelligent evaluation system for measuring equipment based on multi-dimensional fault analysis is used to implement the aforementioned intelligent evaluation system for measuring equipment based on multi-dimensional fault analysis. The intelligent evaluation system for measuring equipment based on multi-dimensional fault analysis includes:

[0154] The model building module builds a test parameter generation model based on the Guangming Electric Power large model. In this test parameter generation model, the characteristic distribution of historical failure modes is learned through the self-attention mechanism to obtain parameters in simulated extreme scenarios that exceed the design specifications of the metering equipment under test.

[0155] Extreme scenario simulation device, used to simulate extreme scenarios that exceed the design specifications of the metering equipment to be tested;

[0156] High-precision sensor arrays are used to collect data from measured measuring equipment in extreme scenarios;

[0157] The fault analysis module is used to perform fault analysis on the collected data of the measured measuring equipment.

[0158] The beneficial effects of the intelligent evaluation method for metering equipment based on extreme scenario simulation according to the first embodiment of the present invention are as follows:

[0159] First, it breaks through the single-dimensional limitations of traditional testing methods and builds an active exploration system with full-scene and multi-factor linkage. Existing technologies usually rely on manually set fixed parameters (such as testing only voltage or temperature), and the data acquisition frequency is low (seconds) and the dimension is single (such as monitoring only current or vibration). In contrast, this embodiment deploys 12 types of high-precision sensors to synchronously collect multi-physical field data such as mechanical vibration, electromagnetic radiation, and microscopic deformation, and combines industrial high-speed cameras (1000 frames / second) to capture microsecond-level transient anomalies (such as material cracking). For example, in electromagnetic-mechanical testing, traditional methods can only test magnetic field strength or vibration frequency alone, while this embodiment can simultaneously apply a 1T strong magnetic field and 5kHz high-frequency vibration to monitor the microscopic deformation of ferromagnetic materials in real time and accurately detect magnetostrictive cracks that cannot be identified by traditional means. At the same time, the Guangming large model extracts complex features such as vibration spectrum entropy in real time, and the data screening efficiency is more than 5 times that of manual analysis, and the error rate is reduced from 8% to 0.3%.

[0160] Secondly, this embodiment solves the industry problem that multi-factor coupled failures are difficult to reproduce in real scenarios. Existing technologies mostly use step-by-step testing (such as measuring high temperature first and then measuring vibration), which cannot simulate complex scenarios where multiple factors act simultaneously in the actual environment. This embodiment generates superimposed test scenarios through intelligent algorithms. For example, it simultaneously applies high-frequency mechanical vibrations in a high-temperature environment of 85°C, and directly observes the fracture process of connecting components under the combined action of thermal expansion and vibration. For outdoor equipment lightning protection testing, traditional methods only simulate standard lightning waveforms, while this solution uses the Guangming Big Model to learn historical lightning data, which can generate instantaneous high-voltage pulses containing asymmetric harmonic distortion, and more realistically restore the complex electromagnetic environment at the moment of lightning strike. Experiments show that this method can detect 23% of potential defects missed by traditional tests, such as the failure of the communication module of a certain type of acquisition terminal under composite pulses.

[0161] In addition, this embodiment incorporates biological environmental factors into the electronic equipment reliability evaluation system for the first time, filling the gap in the industry. Traditional environmental testing only focuses on physical and chemical factors such as temperature and humidity, ignoring the impact of long-term microbial erosion. This embodiment quantitatively analyzes the damage law of PCB board biofilm growth on insulation performance by cultivating microbial colonies in a hot and humid environment. For example, a certain IoT surface coating has a colony density of 10^4 CFU / cm 2 When tested, the insulation resistance value plummeted by 60%, while conventional salt spray testing under the same conditions only detected a 20% performance degradation. This innovative testing method provides a direct basis for the design of anti-biological corrosion protection for outdoor equipment, helping manufacturers optimize coating material formulations and extend the equipment's field service life by more than three years.

[0162] The embodiment of the present invention also provides a metering equipment intelligent evaluation system based on multi-dimensional fault analysis, which is used to implement the aforementioned metering equipment intelligent evaluation system based on multi-dimensional fault analysis. The metering equipment intelligent evaluation system based on multi-dimensional fault analysis includes:

[0163] The model building module builds a test parameter generation model based on the Guangming Electric Power large model. In this test parameter generation model, the characteristic distribution of historical failure modes is learned through the self-attention mechanism to obtain parameters in simulated extreme scenarios that exceed the design specifications of the metering equipment under test.

[0164] Extreme scenario simulation device, used to simulate extreme scenarios that exceed the design specifications of the metering equipment to be tested;

[0165] High-precision sensor arrays are used to collect data from measured measuring equipment in extreme scenarios;

[0166] The fault analysis module is used to perform fault analysis on the collected data of the measured measuring equipment.

[0167] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Those skilled in the art will understand that the present invention includes, but is not limited to, the contents described in the above specific embodiments. Any modifications that do not deviate from the functional and structural principles of the present invention are intended to be included within the scope of the claims.

Claims

1. An intelligent evaluation method for metering equipment based on extreme scenario simulation, characterized by: The intelligent evaluation method for metering equipment based on extreme scenario simulation includes: A test parameter generation model based on the Guangming Electric Power large-scale model was constructed. The self-attention mechanism was used in the test parameter generation model to learn the characteristic distribution of historical failure modes and obtain parameters in simulated extreme scenarios that exceeded the design specifications of the metering equipment under test. deploying high-precision sensor arrays; Collect data from the measured measuring equipment through a high-precision sensor array in simulated extreme scenarios; Perform fault analysis based on the collected data of the measured measuring equipment.

2. The method for intelligent evaluation of metering equipment based on extreme scenario simulation according to claim 1 is characterized in that: The high-precision sensor array is used to collect at least two of the following data: mechanical vibration, electromagnetic radiation, thermodynamic distribution, micro deformation, and chemical corrosion; The high-precision sensor array integrates industrial high-speed cameras and acoustic emission equipment to capture transient anomalies.

3. The intelligent evaluation method for metering equipment based on extreme scenario simulation according to claim 1 is characterized in that: The test dimensions for extreme scenarios include at least two of the following: Voltage-current test: Continuously detect the heating condition of the equipment and the working condition of key components in overvoltage or overcurrent environment; Electromagnetic-mechanical testing: applying high-frequency mechanical vibration in a strong magnetic field environment to detect the magnetostrictive effect of ferromagnetic materials; Thermal-chemical testing: gradient temperature increase and simultaneous injection of corrosive aerosol to analyze the oxidation kinetics of the material; Optical-electrical-thermal triple-field testing: Using high-energy laser local heating in conjunction with high-voltage pulses to test the breakdown characteristics of insulating materials; Bio-electronic testing: Cultivating microbial colonies in a hot and humid environment to analyze the impact of biofilm growth on PCB impedance characteristics.

4. The method for intelligent evaluation of metering equipment based on extreme scenario simulation according to claim 1 is characterized in that: Extreme scenarios include at least two of the following: Extreme load test: simulates the current to increase to several times the normal value in a short period of time and / or suddenly increases to several times the normal value, while monitoring the internal temperature of the equipment and whether the components are burned or the performance is degraded; Multi-factor superposition test: Test by combining different environmental conditions, including high temperature, vibration, humidity, and chemical corrosion; Long-term endurance testing: By simulating the continuous operation of the equipment for months or even years, the slow changes in its performance are observed; Intelligent scenario simulation test: Based on historical failure data, automatically generate the test scenarios that are "most likely to cause problems": For the acquisition terminal, simulate the scenario of receiving massive instructions at the same time to test the limit of its data processing capability; For rural users, metering equipment and data collection terminals are installed outdoors to simulate the instantaneous high-voltage pulse during a lightning strike and test the effectiveness of lightning protection devices; Destructive critical point testing: By gradually increasing the test intensity, the critical value of device failure is accurately found.

5. The intelligent evaluation method for metering equipment based on extreme scenario simulation according to claim 1 is characterized in that: Extreme scenarios include extreme load testing, and the extreme load testing process includes: Deep decoupling: In-depth analysis of multi-source data, stripping away noise, and decoupling coupled composite faults into independent basic failure feature units; Intelligent reconstruction and deduction: Based on decoupling characteristics, causal reasoning is used to proactively combine and deduce more likely or extreme unexpected compound failure modes that have never occurred in history but are bound to exist in physical logic; Generate dynamic load curves: Based on the reconstructed unexpected composite failure mode and combined with the specific model and design characteristics of the device under test, a highly customized load curve with non-uniform amplitude and complex timing combinations is dynamically synthesized.

6. The method for intelligent evaluation of metering equipment based on extreme scenario simulation according to claim 1 is characterized in that: Extreme scenarios include multi-factor superposition testing, and the multi-factor superposition testing process includes: Construct a three-dimensional environment matrix; Weight distribution; Start the load loading system; Detect whether there are data anomalies; If no data anomaly is detected, the loading continues and an environmental correlation map is generated; if data anomaly is detected, the parameters of the load loading system are dynamically adjusted.

7. The method for intelligent evaluation of metering equipment based on extreme scenario simulation according to claim 1 is characterized in that: Extreme scenarios include long-term endurance testing, which includes: Set up accelerated aging model; Long-short-term memory model is used for lifespan prediction; Load typical working conditions; Continue for a preset duration; Detect whether performance degradation exceeds a threshold; If the performance degradation does not exceed the threshold within the preset time, the operating parameters are adjusted; if the performance degradation exceeds the threshold within the preset time, the aging point is located and a life curve is generated.

8. The method for intelligent evaluation of metering equipment based on extreme scenario simulation according to claim 1 is characterized in that: Extreme scenarios include intelligent scenario simulation tests, and the intelligent scenario simulation test process includes: Input device topology; Knowledge graph call; Generate abnormal scenarios using generative adversarial networks; Digital twin preview: obtain and screen several scenarios; Loading of the tested equipment; Data discrepancy analysis; If the data difference analysis is greater than the threshold, the parameters of the generative adversarial network are modified; if the data difference is less than the threshold, optimization suggestions are output.

9. The method for intelligent evaluation of metering equipment based on extreme scenario simulation according to claim 1 is characterized in that: Extreme scenarios include destructive critical point testing, and the destructive critical point process includes: Set safety thresholds; Apply load in a certain step size; 3D deformation monitoring; When the critical value is reached, emergency braking protection; Perform reverse engineering analysis using micro-computed tomography to generate material improvement recommendations.

10. An intelligent evaluation system for measuring equipment based on multi-dimensional fault analysis, characterized by: The intelligent evaluation system for measuring equipment based on multi-dimensional fault analysis is used to implement any one of claims 1 to 9, and the intelligent evaluation system for measuring equipment based on multi-dimensional fault analysis comprises: The model building module builds a test parameter generation model based on the Guangming Electric Power large model. In this test parameter generation model, the characteristic distribution of historical failure modes is learned through the self-attention mechanism to obtain parameters in simulated extreme scenarios that exceed the design specifications of the metering equipment under test. Extreme scenario simulation device, used to simulate extreme scenarios that exceed the design specifications of the metering equipment to be tested; High-precision sensor arrays are used to collect data from measured measuring equipment in extreme scenarios; The fault analysis module is used to perform fault analysis on the collected data of the measured measuring equipment.

Images