Health assessment method and system of power supply system based on artificial intelligence

Abstract

The invention relates to the technical field of power supply system health assessment, in particular to a health assessment method and system of a power supply system based on artificial intelligence. The method comprises the following steps: acquiring data of a power supply system, estimating initial performance parameters of the power supply system, evaluating an initial running state of the power supply system, and acquiring initial state data of the power supply system; performing power supply equipment load operation simulation based on the initial state data, and predicting a fault evolution trajectory; the transmission quality attenuation degree of the power supply system is estimated by analyzing the fault evolution trajectory of the power supply system, the loss increase trend of a transmission line is detected, and the performance decrease trend is evaluated in combination with the attenuation degree; evaluating health attenuation data of the power supply system according to the performance reduction trend; transmitting the data to an artificial intelligence model for deep learning training to obtain accurate power supply system health state evaluation data; according to the invention, health analysis is carried out on the power supply system, so that the power supply system is safer and more stable.

Description

Technical Field

[0001] The present invention relates to the technical field of power system health assessment, and particularly to a health assessment method and system for a power system based on artificial intelligence. Background Art

[0002] Power systems are playing an increasingly important role in various types of equipment. Whether in household appliances, industrial equipment, or in high-precision medical equipment, communication base stations and other places, the stability and reliability of the power system are crucial for the normal operation of the entire system. Especially in the fields of industry, automobiles, and power, faults in the power system often bring serious safety hazards. Conducting a health assessment of the power system and timely warning of its fault trend are key tasks to ensure the stability and reliability of the power system. During the long-term operation of the power system, it will be affected by various factors, such as load fluctuations, external environmental changes, unstable power quality, etc. These factors not only lead to a decline in the performance of the power supply, but also accelerate the aging process of each component in the power system. In particular, components such as transformers, rectifiers, and filter capacitors in the power system are prone to being affected by multiple loads such as thermal effects, overloads, and current ripples during long-term operation, and gradually show performance degradation, aging, and even failures. However, traditional power system health assessments have problems of inaccurate analysis of the aging trend of the power system and inaccurate analysis of the performance attenuation of the power system. Summary of the Invention

[0003] Based on this, it is necessary to provide a health assessment method and system for a power system based on artificial intelligence to solve at least one of the above technical problems.

[0004] To achieve the above object, a health assessment method and system for a power system based on artificial intelligence includes the following steps:

[0005] Step S1: Obtain power system data; estimate the initial performance parameters of the power system according to the power system data; conduct an assessment of the operating state of the power system based on the initial performance parameters of the power system to obtain the initial state data of the power system;

[0006] Step S2: Conduct a simulation of the load operation of the power equipment based on the initial state data of the power system to obtain the simulated load operation data of the power system; predict the fault evolution trajectory data of the power system according to the simulated load operation data of the power system;

[0007] Step S3: Estimate the degree of attenuation of the transmission quality of the power system according to the fault evolution trajectory data of the power system; detect the growth trend of the transmission line loss according to the degree of attenuation of the transmission quality of the power system; evaluate the downward trend of the performance of the power system according to the growth trend of the transmission line loss and the degree of attenuation of the transmission quality of the power system;

[0008] Step S4: Evaluate the power system health attenuation data for the downward trend of the power system performance; transmit the power system health attenuation data to a preset artificial intelligence model for model deep learning training to obtain a power system artificial intelligence model; perform a power system health status assessment based on the power system artificial intelligence model to obtain power system health status assessment data.

[0009] The present invention comprehensively monitors and health-assesses the power system through a series of precise steps, thereby realizing the accurate identification and prediction of the power system state, effectively improving the operation reliability and safety of the power system. By obtaining the real-time data of the power system and estimating the initial performance parameters, it provides basic data support for subsequent operation state assessment. The acquisition of the initial state data of the power system helps to clearly understand the initial working condition of the system, compare the subsequent operation changes, and timely discover potential abnormal situations. By establishing the simulated data of the power system load operation, it can dynamically predict the performance of the system under different load conditions, thereby providing a basis for predicting the fault evolution trajectory and being able to identify the weaknesses and vulnerable components of the system in advance. Further, the system can accurately judge the performance decline speed and fault development direction of the power system according to the transmission quality attenuation degree and line loss growth trend predicted by the power system fault evolution trajectory data. Through these assessments, it can master the changes in the power system performance in real time, scientifically predict the change trend of the system health state, perform maintenance and adjustment in a timely manner, and avoid the occurrence of sudden failures. The detection of the line loss growth trend helps to accurately evaluate the attenuation of the power system control accuracy. By tracking these attenuation trends, necessary repair measures can be taken in the early stage, effectively extending the service life of the system and preventing large-scale failures. By evaluating the downward trend of the power system performance and combining the health attenuation data, further in-depth analysis of the system health status can reveal the key factors affecting the system health and form power system health attenuation data. These data are trained through deep learning, endowing the artificial intelligence model with powerful prediction ability, enabling the model to accurately evaluate the future health state of the power system. The use of this artificial intelligence model can provide more intelligent and automated health assessment results, significantly improving the accuracy and timeliness of power system fault prediction. Combining the evaluation data of the intelligent model, specific maintenance suggestions and repair measures can be given according to the state of the system to ensure the continuous and stable operation of the power system, prevent equipment downtime caused by unclear system states, and improve the intelligent management level of the power system. The present invention is an optimized treatment of the traditional power system health assessment, solving the problems of inaccurate analysis of the power system aging trend and inaccurate analysis of the power system performance attenuation existing in the traditional power system health assessment. It improves the accuracy of the power system aging trend analysis and the accuracy of the power system performance attenuation analysis.

[0010] The present invention also provides a health assessment system for a power supply system based on artificial intelligence, which is used to execute the health assessment method for the power supply system based on artificial intelligence as described above. The health assessment system for the power supply system based on artificial intelligence includes:

[0011] An operating state assessment module, configured to obtain power supply system data; estimate initial performance parameters of the power supply system according to the power supply system data; and perform an operating state assessment of the power supply system based on the initial performance parameters of the power supply system to obtain initial state data of the power supply system.

[0012] A fault evolution trajectory prediction module, configured to perform a load operation simulation of the power supply equipment based on the initial state data of the power supply system to obtain load operation simulation data of the power supply system; and predict fault evolution trajectory data of the power supply system according to the load operation simulation data of the power supply system.

[0013] A fault evolution trajectory prediction module, configured to perform a load operation simulation of the power supply equipment based on the initial state data of the power supply system to obtain load operation simulation data of the power supply system; and predict fault evolution trajectory data of the power supply system according to the load operation simulation data of the power supply system.

[0014] A health state assessment module, configured to evaluate health decay data of the power supply system based on the performance degradation trend of the power supply system; transmit the health decay data of the power supply system to a preset artificial intelligence model for model deep learning training to obtain an artificial intelligence model of the power supply system; and perform a health state assessment of the power supply system according to the artificial intelligence model of the power supply system to obtain health state assessment data of the power supply system. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is a schematic diagram of the step flow of a health assessment method and system for a power supply system based on artificial intelligence;

[0016] Figure 2 is Figure 1 a detailed implementation step flow schematic diagram of step S2 in

[0017] Figure 3 is Figure 1 a detailed implementation step flow schematic diagram of step S3 in

[0018] The realization, functional features and advantages of the object of the present invention will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0019] The technical method of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative work fall within the scope of protection of the present invention.

[0020] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the drawings represent the same or similar parts, and thus repeated descriptions thereof will be omitted. Some of the block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0021] It should be understood that although the terms "first", "second", etc. are used here to describe each unit, these units should not be limited by these terms. These terms are only used to distinguish one unit from another. For example, without departing from the scope of the exemplary embodiments, the first unit can be called the second unit, and similarly the second unit can be called the first unit. The term "and / or" used here includes any and all combinations of one or more of the listed associated items.

[0022] To achieve the above object, please refer to Figures 1 to 3 , a health assessment method and system for a power supply system based on artificial intelligence, comprising the following steps:

[0023] Step S1: Obtain power supply system data; estimate the initial performance parameters of the power supply system according to the power supply system data; perform an assessment of the operating state of the power supply system based on the initial performance parameters of the power supply system to obtain the initial state data of the power supply system;

[0024] In the embodiments of the present invention, the operation data of the power supply system is obtained through appropriate sensors and monitoring devices. These sensors and devices can monitor the working state, load condition, voltage, current, temperature and other parameters of the power supply system in real time. Specifically, an embedded sensor system is used to install sensors on various important components of the power supply device (such as transformers, inverters, battery packs, etc.) to collect relevant current, voltage, power and thermal effect data. At the same time, the real-time data is transmitted to the central processing unit through network protocols (such as Modbus, CAN bus, etc.) for processing and analysis. Based on the collected power supply system data, the initial performance parameters of the power supply system are estimated through specific algorithms. This process can adopt various numerical methods, such as the least squares method, Kalman filtering, or the method of regression analysis based on historical data, to calculate the initial performance parameters of the power supply system, such as output power, efficiency, heat generation data, etc. Through these initial performance parameters, the operation state of the power supply system is further evaluated. When evaluating the operation state of the power supply system, model recognition technology is used in combination with the internal structure data of the power supply system to analyze whether it is in a normal operation state. For example, decision tree or support vector machine (SVM) algorithms in machine learning are used for state classification to calibrate whether the power supply system is in a normal, warning or fault state, and the evaluation result is output as the initial state data of the power supply system, which serves as the basis for subsequent analysis.

[0025] Step S2: Perform a load operation simulation of the power supply device based on the initial state data of the power supply system to obtain the load operation simulation data of the power supply system; predict the fault evolution trajectory data of the power supply system according to the load operation simulation data of the power supply system;

[0026] In the embodiment of the present invention, when the load operation model of the power supply equipment is established by using the initial state data of the power supply system, the working performance characteristics of each power supply equipment under different loads are considered. Specifically, the operating performance of these devices includes the load adaptability of the transformer, the output power stability of the inverter, etc. The performance of the transformer under different load conditions mainly involves its temperature rise, magnetic field strength and electrical insulation performance, while the output power stability of the inverter is related to its efficiency change under fluctuating load and long-term operation. To more comprehensively describe the load operation of these devices, numerical simulation methods such as finite element method (FEM) or multi-body dynamics model (MBD) are used. Through these numerical simulation methods, the working environment, temperature change, current fluctuation, power output and other parameters of the power supply equipment under different load states can be calculated in detail, so as to accurately obtain the load operation simulation data of the power supply system. These simulation data can not only reflect the performance of the power supply equipment under normal working conditions, but also reveal the degradation of equipment performance and potential fault hazards under high load or abnormal load conditions. By simulating the load operation of the power supply system, key parameters such as temperature rise, current fluctuation, power output, etc. can be extracted, and these data play an important role in the subsequent fault prediction process. For example, the temperature rise data obtained through simulation reveals the overheating problem of the equipment under long-term high load, while the current fluctuation can reflect the reaction of the equipment under sudden load changes or unstable load conditions. The fluctuation of power output can help predict the efficiency reduction or performance degradation of the equipment under load changes. Based on the load operation simulation data, the next step is to predict the fault evolution trajectory of the power system through time series prediction or neural network algorithms (such as long short-term memory network LSTM). In this process, the aging, wear and electrical failure of the power supply equipment under long-term load changes are taken into account. Long-term load changes will cause the performance of the power supply equipment to gradually decline, especially when the mechanical and electrical components of the equipment age or become damaged under frequent load fluctuations, temperature rise or high-load working conditions. Deep learning methods such as LSTM can effectively process time series data, learn the potential patterns in historical data, and accurately predict the future fault evolution trajectory of the power supply equipment. LSTM can capture long-term dependencies when processing time series data, which helps to model and predict the fault trends that occur in the long-term operation of the power supply equipment. By training on historical data, the LSTM model can identify fault signs in the power supply system, such as sudden increase in equipment load, abnormal temperature rise, and drastic current fluctuations, which are often important signals that the equipment is about to fail. In the process of predicting the fault evolution trajectory, the aging rate of different equipment, the type of fault, and the impact of environmental factors on the equipment are taken into account.The aging rates and failure types of different power supply devices vary under the same working environment. For example, due to being in a high-load state for a long time, the aging speed of the insulation material of a transformer will be faster; while an inverter will experience increased losses of internal electronic components due to working under fluctuating loads for a long time. The model can conduct targeted modeling for different devices, reflecting different working characteristics and fault evolution laws. In addition, environmental factors such as temperature, humidity, and air pollution will also affect the fault evolution process of the devices. These factors are appropriately integrated through dynamic modeling methods and adjusted in real time according to the actual situation. In practical applications, the fault evolution trajectory prediction model is continuously updated to ensure accurate prediction of the fault evolution of the power supply system.

[0027] Step S3: Estimate the attenuation degree of the power supply system transmission quality based on the power supply system fault evolution trajectory data; detect the growth trend of the transmission line loss according to the attenuation degree of the power supply system transmission quality; evaluate the decline trend of the power supply system performance based on the growth trend of the transmission line loss and the attenuation degree of the power supply system transmission quality;

[0028] In the embodiments of the present invention, based on the fault evolution trajectory data obtained in step S2, the attenuation of the transmission quality of the power supply system is further analyzed. The core of this analysis is to identify the attenuation trend of the transmission quality of the power supply system by long-term monitoring of the current and voltage fluctuations in the power supply system, combined with the operating environment and load changes of the power supply system, and using advanced time-frequency domain analysis techniques such as Fourier transform or wavelet analysis. The Fourier transform can convert the time-domain signal into a frequency-domain signal, helping to identify changes in frequency components, so as to more accurately capture high-frequency noise or low-frequency fluctuations in the power supply system, while wavelet analysis can provide better time-frequency local characteristics and sensitively capture the instantaneous changes of the signal. Through these analysis techniques, the attenuation of the transmission quality of the power supply system at different time scales can be identified, such as voltage fluctuations, current instability, frequency offset, etc., to help determine whether there is a risk of a decline in the transmission quality of the power supply system. In the process of analyzing the attenuation of the transmission quality of the power supply system, the change data of current and voltage play a crucial role in detecting the loss of the transmission line. By real-time monitoring of the current and voltage data, combined with the load conditions of the power supply system, the change trends of the line resistance, power loss, and losses caused by environmental factors (such as temperature, humidity, etc.) are calculated. For example, an increase in temperature leads to an increase in the resistance of the cable, thereby increasing the power loss of the line, and changes in humidity lead to the aging of the insulating material, thus affecting the transmission quality of the power supply system. By collecting these data, the loss level of the power supply system can be accurately quantified, and the loss growth trend of the transmission line can be predicted based on these data. Through the analysis of the transmission line loss data, the risks faced by the power supply system in future operation are further evaluated. When the loss increases to a certain extent, it indicates that some components of the power supply system have begun to age or malfunction, such as the deterioration of the insulating material of the transformer, poor wire contact, or equipment overload. Combining these loss data, potential transmission line faults or performance degradation problems are identified, so as to take necessary maintenance measures in advance to avoid system failures or serious performance degradation. Combining the loss growth trend of the transmission line with the attenuation degree of the transmission quality of the power supply system, the overall performance degradation trend of the power supply system is comprehensively evaluated. This evaluation uses regression analysis or weighted average method to process multi-dimensional data. Regression analysis can fit the relationship between the attenuation of transmission quality and loss growth based on historical data, and predict the performance changes of the power supply system within a certain period in the future. The weighted average method assigns different weights according to the importance of different data sources, and comprehensively evaluates the health status and performance change trend of the power supply system based on multiple indicators. Based on these comprehensive data, the remaining service life of the power supply system is more accurately estimated. Through the comprehensive analysis of various parameters, it is judged whether the power supply system needs maintenance, repair, or component replacement.

[0029] Step S4: Evaluate the power system health decay data for the power system performance degradation trend; transmit the power system health decay data to a preset artificial intelligence model for model deep learning training to obtain a power system artificial intelligence model; perform power system health status evaluation based on the power system artificial intelligence model to obtain power system health status evaluation data.

[0030] In the embodiments of the present invention, these data are obtained through real-time monitoring of the power supply system and compared with historical data to obtain the trend of the health attenuation of the power supply system. The health attenuation data of the power supply system not only includes the changes in the current performance indicators of the system, but also details factors such as system aging and failure risks. These data are transmitted to a preset artificial intelligence model through a data acquisition system for further processing and analysis. The selection and training process of the artificial intelligence model is the key to the health assessment of the power supply system. Generally, convolutional neural networks (CNNs) and recurrent neural networks (RNNs) are widely used to process such time series data. CNNs are suitable for processing data with spatial structures and can extract local features of the performance indicators of the power supply system; while RNNs are particularly suitable for processing sequence data and can capture the temporal dependencies of the performance indicators of the power supply system during long-term operation. In this embodiment, a hybrid model combining CNN and RNN is selected to give full play to their respective advantages and accurately learn the health status and attenuation characteristics from the historical operation data of the power supply system. The process of model training mainly relies on a large amount of historical data, which includes the performance of the power supply system under different working conditions, fault records, and maintenance records, etc. The training process takes the health attenuation data as input and propagates it through the layers in the network structure to gradually extract important features from the data. These features are non-linearly transformed through activation functions (such as ReLU or Sigmoid) to help the model identify complex health attenuation patterns. During the training process, the model calculates the error between the predicted value and the actual value, and uses the backpropagation algorithm to reverse the error to each parameter layer of the model to update the weights and biases in the network. To optimize the training effect of the model, an optimization algorithm (such as the Adam optimizer) is used for gradient descent. In each iteration, the Adam optimizer dynamically adjusts the learning rate according to the historical gradients to ensure efficient learning while avoiding overfitting or gradient explosion. During the training process, regularization methods, such as Dropout or L2 regularization, are introduced to further improve the generalization ability of the model and avoid overfitting on the training data. After the training is completed, the artificial intelligence model can predict the current health status of the power supply system by analyzing new real-time data. This evaluation result includes information such as the operating health of the power supply system, the risk of failures occurring, and maintenance suggestions. Through these evaluation results, the operation and maintenance personnel can take corresponding maintenance or repair measures in a timely manner to avoid major failures of the power supply system and ensure its long-term stable operation. The accuracy and reliability of the model directly depend on the quality and quantity of the training data. Continuously accumulating the operation data of the power supply system and retraining the model regularly are important measures to improve the evaluation accuracy.

[0031] Preferably, step S1 includes the following steps:

[0032] Step S11: Obtain power supply system data;

[0033] In the embodiment of the present invention, real-time acquisition is performed through sensors and a monitoring system. The sensors are deployed at key parts of the power supply equipment, such as transformers, battery packs, inverters, etc., to collect data such as current, voltage, temperature, power, frequency, etc. These data are transmitted to the central data processing platform in real time through industrial communication protocols (such as Modbus, Profibus, etc.). The data acquisition frequency is set according to the operation cycle of the power supply system to ensure dynamic monitoring of various parameters of the power supply system. For example, current and voltage data are collected once per second, and temperature data are collected once per minute.

[0034] Step S12: Extract the internal structure data of the power supply system according to the power supply system data;

[0035] In the embodiment of the present invention, after obtaining the power supply system data, the internal structure data of the power supply system are then extracted according to these data. The internal structure of the power supply system includes information such as the connection relationship between devices, the type, specifications, and performance parameters of the devices. By parsing the device list and schematic diagram of the power supply system, and combining the installation location, type, function of the devices and their roles in the system, the physical structure data of the power supply system are extracted. At this time, the extraction of the internal structure data of the system can be automatically processed through image recognition technology and knowledge graphs, and the physical connection relationship and electrical connection relationship of the devices are abstracted into a network model of the system. This network model not only includes the structure of the devices themselves, but also includes the electrical and thermodynamic interaction relationships between the devices, further supporting subsequent fault prediction and performance evaluation.

[0036] Step S13: Estimate the initial performance parameters of the power supply system according to the power supply system data and the internal structure data of the power supply system;

[0037] In the embodiment of the present invention, based on the power supply system data obtained in step S11 and the internal structure data of the power supply system extracted in step S12, the initial performance parameters of the power supply system are estimated. To estimate the initial performance parameters of the system, according to the technical specifications, design parameters of the devices and the actual operating environment, the theoretical performance values of the power supply system are calculated through formulas and models. For example, through current, voltage, and power data, preliminary performance indicators such as the efficiency, power factor, and load rate of the devices are calculated. In addition, considering the operating environment of the power supply system, such as environmental factors such as temperature, humidity, and air pressure, models are established for these influencing factors, and corrections and adjustments are made in combination with the data provided by environmental sensors. When estimating the initial performance parameters, statistical methods such as regression analysis and the least squares method are used to model and optimize historical data and real-time data to obtain accurate initial performance parameters.

[0038] Step S14: Based on the internal structure data of the power supply system and the initial performance parameters of the power supply system, conduct an initial state assessment of the power supply system to obtain the initial state data of the power supply system.

[0039] In the embodiment of the present invention, the initial state of the power supply system is evaluated by combining the internal structure data of the power supply system in step S12 and the initial performance parameters in step S13. The goal of the initial state evaluation is to evaluate the overall health level of the system based on the current working state of the power supply system. This process involves the calculation and comprehensive evaluation of multiple parameters, including the operating efficiency of power supply equipment, load adaptability, temperature change trend, electrical stability, etc. By analyzing the internal structure data of the power supply system, it is determined whether the operating state of each device meets the preset working standards. Then, according to the initial performance parameters, it is judged whether each performance index is within the normal range, such as the load rate of the device, the voltage fluctuation range, etc. If some parameters deviate from the normal range, it indicates that there are potential problems in the system. Finally, based on the operating state data of the power supply system, a comprehensive evaluation is carried out to calculate the initial health score of the power supply system. This score can be used as a benchmark for subsequent evaluation and fault prediction, providing a reference for subsequent operations. In actual operation, the evaluation method uses a multi-factor comprehensive evaluation method or an evaluation model based on fuzzy mathematics to perform weighted summation on each performance index to obtain the initial health state data of the power supply system.

[0040] Preferably, step S13 includes the following steps:

[0041] Step S131: Identify the connection relationship characteristics of the power supply system components according to the internal structure data of the power supply system, and extract the internal topological structure of the power supply system based on the connection relationship characteristics of the power supply system components;

[0042] In the embodiment of the present invention, by obtaining the internal structure data of the power supply system, including information such as device type, device number, connection port, connection type, power capacity, etc., the component connection relationship of the power supply system is constructed. The internal structure data of the power supply system usually includes the specific configuration of the power supply equipment and the connection methods between various devices, such as current transmission paths, signal links, power flow paths, etc. These connection relationships are converted into digital formats through electrical drawings or CAD drawings in the industrial automation system, and are analyzed and extracted using image recognition technology or graphics processing algorithms. By processing these data, an electrical topology diagram of the power supply system is constructed, and each component and its connection method are clearly marked in the topology diagram, providing necessary structured data for subsequent analysis. The topological structure data of the power supply system usually includes device nodes (such as transformers, inverters, circuit breakers, etc.) and connections between devices (such as cables, connectors, etc.).

[0043] Step S132: Calculate the output efficiency of the power supply system according to the power supply system data;

[0044] In an embodiment of the present invention, after obtaining the component connection relationship and topology structure of the power supply system, the output efficiency of the power supply system is calculated through real-time monitoring and historical data input. The calculation of the power supply system output efficiency is based on the input and output power data of the power supply system. The input power data is usually provided by the voltage and current sensors of the power supply, and the output power is provided by the power monitor of the load device. The basic formula for calculating the output efficiency is output efficiency = output power / input power. During this process, the working state of the system is considered, for example, the influence of fluctuations in the power supply load, the temperature of the device, the humidity of the environment, etc. on the efficiency. By monitoring and modeling these variables, combining real-time data, using statistical analysis methods (such as regression analysis) to dynamically estimate the efficiency, and regularly calibrating the model to ensure that the calculation results are consistent with the actual performance of the power supply system. Real-time monitoring and optimizing the working parameters of the power supply system helps to more accurately evaluate the output efficiency of the power supply device.

[0045] Step S133: Calculate the heat generation data of the power supply system based on the output efficiency of the power supply system;

[0046] In an embodiment of the present invention, by calculating the output efficiency of the power supply system, the heat generation situation of the power supply system is further estimated. During the operation of the power supply system, part of the electrical energy is converted into heat energy, usually generated when current flows through resistors, transformers, and power electronic devices (such as inverters, rectifiers). According to the working efficiency and power output of the power supply device, a thermodynamic model is used to estimate the heat generated inside the system. Specifically, the heat loss formula Q = P×(1 - η) is used, where Q is the heat generation data, P is the input power, and η is the efficiency value. Combining the power and efficiency of each device, the heat generation data of each component is calculated, and the total heat generation data of the entire power supply system is obtained by summing up. This data is crucial for judging the heat dissipation capacity and temperature change trend of the system. During the calculation process, the influence of environmental factors (such as temperature, humidity) on heat generation is considered to more accurately evaluate the heat dissipation requirements of the power supply system and the problem of heat accumulation.

[0047] Step S134: Estimate the temperature rise stability parameters of the power supply system based on the heat generation data of the power supply system and the internal topology structure of the power supply system;

[0048] In the embodiments of the present invention, data and internal topology of the power supply system are utilized to further estimate the temperature rise stability parameters of the power supply system. The temperature rise stability mainly refers to whether the temperature rise trend of the power supply equipment during long-term operation will affect the stability and long-term performance of the system. According to the heat generation data, the temperature field simulation is carried out using the heat conduction equation or the finite element method (FEM) to calculate the temperature distribution of each device in the system. The heat generation data provides the basic data for the temperature field simulation, while the topology helps to define how heat is transferred between various devices. Through simulation, the temperature rise data of each device is obtained, and further the temperature stability of the system under specific load conditions is calculated. If a certain component in the system has too high a temperature due to poor heat dissipation or design defects, it will affect the stability and long-term operation ability of the entire power supply system. During this process, a computational fluid dynamics (CFD) simulation tool is used to perform more refined modeling and analysis of the thermal effects of the system.

[0049] Step S135: Evaluate the regulated voltage performance data of the power supply system based on the internal topology of the power supply system and the output efficiency of the power supply system;

[0050] In the embodiments of the present invention, based on the internal topology and output efficiency data of the power supply system, the regulated voltage performance of the power supply system is evaluated. The regulated voltage performance is the ability of the power supply system to maintain a stable output voltage under conditions such as load fluctuations and external interference. By using the topology of the power supply system, the relationship between the input voltage and output voltage of each device is determined, and combined with the power supply output efficiency, the voltage fluctuations under different loads and environmental conditions are calculated. Through long-term monitoring of the voltage data, statistical methods (such as standard deviation or coefficient of variation) are used to evaluate the regulated voltage performance of the power supply system. In addition, by establishing a dynamic voltage stability model and considering the effects of load changes, current fluctuations and environmental factors, the regulated voltage ability of the system is predicted. The evaluation of the regulated voltage performance can also be combined with the heat generation situation of the power supply system to ensure that the system temperature rise will not cause deterioration of voltage fluctuations. If the regulated voltage performance data exceeds the preset threshold, it means that the power supply system faces stability risks.

[0051] Step S136: Estimate the initial performance parameters of the power supply system according to the regulated voltage performance data of the power supply system and the temperature rise stability parameters of the power supply system.

[0052] In the embodiments of the present invention, the initial performance parameters of the power supply system are estimated based on the voltage regulation performance data and temperature rise stability parameters of the power supply system. The initial performance parameters of the power supply system mainly include multiple indicators such as the output power, efficiency, temperature stability, and load adaptability of the system. These parameters directly determine the overall performance and operating stability of the power supply system. The voltage regulation performance data reflects the ability of the power supply system to maintain the stability of the output voltage under different loads and environmental conditions. If the output voltage of the power supply system fluctuates less under load changes or external environmental disturbances, it indicates that the system has strong voltage regulation ability, which means that the power supply system can maintain a stable voltage output within a wide load range. When estimating the initial performance parameters, the voltage regulation performance data is the key basis for evaluating the operating reliability of the power supply system. The temperature rise stability parameter reflects the temperature change of the power supply system during long-term operation. Since power supply equipment generates heat during operation, excessive temperature will affect the efficiency, stability, and service life of the system. The evaluation of the temperature rise stability parameter can help judge the heat dissipation ability of the system under different operating loads. If the power supply system can maintain a low temperature under high load, it indicates that the heat dissipation design and thermal management ability of the system are strong, and the temperature stability in the initial performance parameters is relatively ideal. By combining the voltage regulation performance data and the temperature rise stability parameter, the output power and efficiency of the power supply system are comprehensively analyzed. When the voltage regulation performance is good and the temperature rise is stable, the output power and efficiency of the system can usually be maintained at a high level, reflecting that the system has strong load adaptability and can operate efficiently under various load conditions. Through further analysis of the voltage regulation performance and temperature rise stability data, and by using multi-dimensional data fusion methods, such as the weighted average method, principal component analysis (PCA), etc., the initial performance parameters of the power supply system are calculated.

[0053] Preferably, step S2 includes the following steps:

[0054] Step S21: Simulate the load operation of the power supply equipment based on the initial state data of the power supply system to obtain the load operation simulation data of the power supply system;

[0055] In the embodiments of the present invention, the load operation of the power supply equipment is simulated based on the initial state data of the power supply system to obtain the load operation simulation data of the power supply system. The load operation simulation of the power supply equipment is carried out by simulating the working states of the power supply equipment under different load conditions. This simulation takes into account the actual operating environment of the power supply equipment, including parameters such as input voltage, load changes, and environmental temperature. In this step, numerical simulation methods, such as finite element analysis (FEA) or multi-body dynamics (MBD) models, are used to simulate the working states of each power supply equipment (such as transformers, inverters, etc.). By setting different load conditions and simulating the power supply equipment under these conditions, the load operation data of the power supply system is obtained. The load operation data includes parameters such as the temperature change, current fluctuation, power output, and efficiency of the power supply equipment.

[0056] Step S22: Detect the overload condition of the power supply system load according to the simulated data of the power supply system load operation;

[0057] In the embodiment of the present invention, the overload condition of the power supply system load is detected according to the simulated data of the power supply system load operation. By analyzing the simulated data of the power supply system load operation, it is detected whether there is an overload situation when the power supply device is in a high load state. The overload of the power supply device refers to the situation where the actual load of the device exceeds its designed rated load, which usually leads to a decline in device performance, too high temperature, and even failure. In this step, indicators such as current, voltage, and power are compared with the rated parameters of the power supply device to determine whether there is a load overload. If the load exceeds the rated range of the device, it is determined as a load overload. This process is dynamically monitored and warned by setting thresholds or using machine learning algorithms to detect the load condition of the system in real time to ensure timely discovery of potential overload risks.

[0058] Step S23: Estimate the accumulation of the thermal effect of the power supply system based on the overload condition of the power supply system load to obtain the accumulation data of the thermal effect of the power supply system;

[0059] In the embodiments of the present invention, under the overload state of the load, the power supply device will generate more heat than normal. When current passes through each component in the power supply system (such as transformers, rectifiers, inverters, battery packs, etc.), Joule heat caused by resistance will be generated, and this heat will increase the temperature of the device. As the load increases, the heat accumulation inside the device will also intensify, which will lead to accelerated material aging, especially the insulating materials, conductors, and semiconductor components in the power supply device. If the device bears an overload for a long time, it will cause premature aging of the winding insulation layer of the transformer, a decrease in the charge and discharge efficiency of the battery, and even serious faults such as short circuits or fires. To accurately estimate the accumulated data of the thermal effect of the power supply system, a thermal analysis model is used. The thermal analysis model can quantify the heat transfer, thermal effect accumulation, and thermal decay processes of each component in the power supply system. These models usually include heat conduction equations, heat convection equations, and radiative heat transfer equations, etc., and can describe the thermal behavior of the device under different loads and environmental temperatures. Specifically, the heat conduction equation is used to calculate the heat transfer process between the components inside the device. Different materials inside the device (such as copper, aluminum, steel, etc.) have different thermal conductivities, and the thermal conductivity characteristics of these materials affect the heat transfer speed and mode inside the device. The finite element analysis (FEA) technology is a commonly used calculation tool, which is widely used in the simulation and prediction of the thermal effects of the power supply system. By dividing the power supply system into a finite number of grid units, the finite element method is used to accurately simulate the temperature distribution and thermal effect accumulation process of each component under different loads and environmental conditions. In the finite element model, considering the geometric shape, material properties, working state of each component, and the influence of the external environment, high-precision heat conduction calculations can be achieved. Through this method, the heat accumulation data of the power supply system under a specific load can be obtained, and then the temperature change trend of the device can be predicted. The accumulated data of the thermal effect obtained through the above method can not only reflect the current thermal load condition of the power supply system, but also reveal the temperature change trend of the device under the overload condition. These data are of great significance for predicting the thermal decay of the power supply system equipment. For example, when the power supply device operates in a long-term overload state, the accumulated data of the thermal effect can reflect the performance degradation of the device caused by continuous high temperature, and timely identify existing heat loss problems, such as aging of the transformer coil and increase in the internal resistance of the battery.

[0060] Step S24: Predict the fault evolution trajectory data of the power supply system based on the overload condition of the power supply system and the accumulated data of the thermal effect of the power supply system.

[0061] In the embodiments of the present invention, the load overload condition is utilized to identify the fault area that appears in the power supply system. When the power supply device operates under an overload condition, it will experience excessive current and voltage fluctuations, which will further exacerbate its mechanical wear, electrical stress and thermal load, resulting in the key components of the device (such as transformers, fans, battery packs, etc.) exceeding the design load range. If these overload conditions are not handled in a timely manner, it will usually lead to a decline in the performance of the device during long-term operation. By long-term monitoring of the load overload condition and combining with the change trend of the load, a preliminary judgment is made on the fault location that appears in the device. Based on the accumulated data of the thermal effect, the impact of the thermal degradation effect on the health status of the device is further quantified. The accumulated data of the thermal effect reflects the heat accumulation process generated by the device under the condition of exceeding the load standard. As the heat accumulates continuously, the materials inside the power supply system (such as cables, electronic components, insulating materials, etc.) will age, degenerate or suffer from structural damage. In this process, the heat-sensitive parts of the device will fail in advance. For example, due to excessive temperature, the insulation layer of the transformer will break or burn out, and the battery pack will show phenomena such as overheating expansion and capacity attenuation. By analyzing these thermal effect data, the rate of decline in the performance of the device is predicted, and combined with the load overload information, the specific time and type of the fault occurrence are further inferred. The fault evolution trajectory of the power supply system is accurately predicted by means of historical fault data and advanced prediction models. The historical fault data includes the working state, load condition, environmental parameters, etc. of the device when a fault occurred in the past, and these data provide a valuable empirical basis for fault prediction. By reviewing and statistically analyzing the historical fault data, the rules of fault occurrence in the power supply system under different working states are identified, and a fault prediction model suitable for the current state of the power supply system is constructed accordingly. Common prediction models include methods such as time series analysis, regression analysis and neural networks. Among them, time series analysis helps to capture the correlation between load changes and device aging, regression analysis quantifies the influence degree of each factor on the occurrence of faults, and neural networks can automatically learn the non-linear relationship from large-scale data, thereby improving the accuracy of fault prediction. Specifically for the training of the model, the service life, aging rate, historical fault mode of the power supply device and the influence of the external environment (such as temperature, humidity, etc.) on the device need to be considered during the prediction process. For example, when using a neural network model, through deep learning technology, a model that can accurately reflect the health status of the power supply system is trained from the real-time operation data and historical fault data of the power supply device. During the training process, the input data includes the load history, temperature change, fault record, etc. of the device, and the model continuously adjusts its weights through the backpropagation algorithm to form a network that can predict the future fault evolution trajectory. Combining with the trained model, the fault evolution trajectory prediction is carried out. Specifically, the model will predict the fault probability, fault type and the specific time point of their occurrence at a certain future time point based on the working state, load condition and thermal effect data of the current power supply device.For example, the model predicts that at a certain point in time, the transformer in the power supply system will suffer insulation damage due to overheating, or the battery pack will experience capacity attenuation due to long-term high-temperature operation, thereby affecting the stability of the power supply system.

[0062] Preferably, step S23 includes the following steps:

[0063] Step S231: Identify the abnormal state of power conversion in the power supply system load overload condition recognition system;

[0064] In the embodiment of the present invention, the abnormal state of power conversion in the power supply system load overload condition recognition system is identified. The load overload of the power supply system usually leads to a decrease in the efficiency of the power conversion process, thereby affecting the overall performance of the power supply. In this step, the load of the power supply system is monitored in real time, and by collecting and analyzing parameters such as current and voltage, it is identified whether there is a load overload phenomenon. Sensors and data acquisition systems are used to continuously monitor the input and output of the power supply equipment. Specifically, according to the comparison between the actual load and the rated load of the power supply system, it is judged whether the system is in an overload state. Once it is identified that the power supply system is in a load overload state, the abnormalities occurring in the power conversion process are further analyzed, such as unstable output current, power factor decrease, etc. These abnormalities indicate that there is a deviation in the power conversion ability of the power supply system, which has an impact on the subsequent system stability. By setting the thresholds of current and voltage, once the preset range is exceeded, the identification of the abnormal state of power conversion can be triggered.

[0065] Step S232: Statistically analyze the growth trend of the power loss of the system according to the abnormal state of the system power conversion;

[0066] In the embodiment of the present invention, the growth trend of the power loss of the system is statistically analyzed according to the abnormal state of the system power conversion. After the power conversion abnormal state appears in the power supply system, the internal loss of the system will increase, manifested as a decrease in the power conversion efficiency. To evaluate the growth trend of the loss, the power output and input power of the power supply system are compared in real time. By calculating the difference between the actual output power and the output power in the ideal state, the power loss of the system is obtained. On this basis, through statistical analysis of the power loss data and using time series analysis methods, the growth trend of the loss is obtained. This process includes collecting and processing the operation data of the power supply system over a period of time. By comparing the power losses in different time periods, the increase amplitude and trend change of the loss are further calculated. Through multiple data collections and analyses, a loss growth trend curve is generated.

[0067] Step S233: Determine the degree of attenuation of the working efficiency of the power supply system based on the load overload condition of the power supply system;

[0068] In the embodiments of the present invention, the degree of attenuation of the power supply system working efficiency is determined based on the overload condition of the power supply system load. When the power supply system is in an overload state, the working efficiency of the system usually decreases, specifically manifested as a decrease in the power output of the power supply and abnormal system temperature rise. Based on the load data and loss trend of the power supply system, the current working state of the power supply system is evaluated by calculating the degree of efficiency attenuation. The ratio of efficiency attenuation is determined by comparing the rated power and the actual output power of the power supply system. At the same time, according to the system load conditions, considering the thermal effect inside the power supply and the load conditions, the amplitude of efficiency attenuation is further estimated. At this time, the main method used is the power efficiency calculation based on the load model. By comparing the actual operating conditions of the power supply system under different loads, the degree of efficiency attenuation is comprehensively obtained. This degree of attenuation not only reflects the output performance of the power supply equipment but also reveals the stability problems faced by the power supply system.

[0069] Step S234: Estimate the growth of heat energy conversion according to the degree of attenuation of the power supply system working efficiency and the growth trend of system power loss;

[0070] In the embodiments of the present invention, the growth of heat energy conversion is estimated according to the degree of attenuation of the power supply system working efficiency and the growth trend of system power loss. After identifying the degree of attenuation of the power supply system working efficiency and the growth trend of power loss, the heat energy conversion inside the system is predicted. Since the attenuation of working efficiency is usually accompanied by an increase in energy loss, the heat generation of the power supply equipment will also increase. At this time, it is necessary to use the heat conduction and heat balance models to simulate and analyze the thermal effect of the power supply system. The specific operation is as follows: By collecting the temperature data, load data, and loss growth trend data of the power supply equipment, combined with the internal structure and thermal performance parameters of the power supply system (such as heat dissipation efficiency, heat conduction coefficient, etc.), the growth of heat energy conversion of the system under the overload condition is estimated. Through the heat energy conversion model, the heat accumulation and distribution of the system under different working states are simulated, and the change trend of temperature rise is predicted.

[0071] Step S235: Predict the thermal runaway state of the power supply system according to the growth of heat energy conversion;

[0072] In an embodiment of the present invention, the thermal runaway state of the power supply system is predicted based on the growth of thermal energy conversion. When the growth of thermal energy conversion in the power supply system reaches a certain level, it will cause the system to experience thermal runaway. Thermal runaway usually manifests as overheating of the power supply system, a sharp rise in temperature, etc., which may further lead to equipment failures and even safety problems such as fires. To predict the thermal runaway state of the power supply system, it is necessary to analyze based on the thermal energy conversion data predicted in the previous step in combination with the thermal runaway threshold of the system. Specifically, a maximum temperature rise threshold of the system is set, and it is determined whether it is approaching this threshold according to the growth of thermal energy conversion. Then, by predicting the temperature change trend of the equipment, it is evaluated whether there is a risk of excessive temperature. If the temperature of the system continues to rise and there is no effective heat dissipation mechanism or cooling measure, thermal runaway is triggered. This process relies on temperature monitoring and a thermal analysis model, and numerical calculations and simulations are used to early warn of the thermal runaway state of the power supply system.

[0073] Step S236: Based on the thermal runaway state of the power supply system and the growth of thermal energy conversion, estimate the accumulation of the thermal effect of the power supply system to obtain the thermal effect accumulation data of the power supply system.

[0074] In an embodiment of the present invention, based on the thermal runaway state of the power supply system and the growth of thermal energy conversion, estimate the accumulation of the thermal effect of the power supply system to obtain the thermal effect accumulation data of the power supply system. When there is a risk of thermal runaway in the power supply system, it is necessary to estimate the long-term accumulation of the thermal effect. This process combines the growth of thermal energy conversion with the thermal runaway state data. By establishing a thermal effect accumulation model, the development trend of the thermal effect of the equipment under different load conditions is predicted. According to the working time, load condition, temperature change, and loss data of the power supply system, the thermal conduction equation and the thermodynamic model are used for the cumulative calculation of the thermal effect. Based on these thermal effect data, predict the degree of thermal degradation of the equipment in the power supply system after long-term use and its impact on performance. Through the thermal effect accumulation data, obtain the thermal failure time and the estimated remaining life of the power supply system.

[0075] Preferably, step S24 includes the following steps:

[0076] Step S241: Based on the thermal effect accumulation data of the power supply system, analyze the deterioration trend of the transformer insulation material to obtain the deterioration trend data of the transformer insulation material;

[0077] In the embodiments of the present invention, based on the accumulated data of the thermal effect of the power supply system, the degradation trend of the transformer insulation material is analyzed to obtain the degradation trend data of the transformer insulation material. As a core device in the power supply system, the degradation of the insulation material of the transformer directly affects the stability and safety of the system. To analyze the degradation trend of the insulation material, the accumulated data of the thermal effect of the power supply system is collected and analyzed. This process is based on the temperature monitoring data inside the power supply system and the thermal effect accumulation model to obtain the temperature rise data at different time points during the operation of the transformer. Temperature is directly related to the degradation of the insulation material. Every time the temperature rises by a certain amplitude, it will accelerate the chemical aging process of the insulation material. By matching and analyzing the temperature rise data of the transformer under the load operation state with the thermal response of the transformer insulation material, the speed and trend of material degradation are deduced. The specific method is as follows: The surface and internal temperature of the transformer are monitored by devices such as temperature sensors and infrared thermal imagers, and the heat conduction equation and heat flow model are used to combine the temperature change with the material degradation characteristics, and then the degradation trend data of the transformer insulation material is calculated.

[0078] Step S242: Estimate the inter-turn short circuit situation of the transformer according to the degradation trend data of the transformer insulation material;

[0079] In the embodiments of the present invention, the inter-turn short circuit situation of the transformer is estimated according to the degradation trend data of the transformer insulation material. The inter-turn short circuit of the transformer is one of the common faults caused by the degradation of the insulation material. Especially under high-temperature conditions, the damage of the insulation layer causes a short circuit between the windings. According to the degradation trend data of the insulation material obtained in step S241, the probability and risk of the transformer occurring an inter-turn short circuit are further estimated. In this process, the degradation data of the transformer insulation material is associated with the heat accumulation generated by the current passing through the winding, and the change trend of the transformer winding temperature under different load conditions is calculated. Based on the relationship between the electrical characteristics of the material (such as dielectric strength and dielectric constant) and the temperature change, the decrease in its insulation ability is predicted. At this time, using the electrical insulation degradation model combined with the thermal effect data, a prediction model of the transformer inter-turn short circuit is established, and different time periods are analyzed to estimate the probability of the transformer having an inter-turn short circuit in a future period of time.

[0080] Step S243: Detect the abnormal growth degree of the transformer induced current according to the inter-turn short circuit situation of the transformer;

[0081] In the embodiments of the present invention, the abnormal growth degree of the induced current of the transformer is detected according to the inter-turn short circuit situation of the transformer. After an inter-turn short circuit occurs in the transformer, it will cause the current to abnormally increase in the transformer winding, thereby triggering electrical abnormal problems. To detect such abnormal growth, it is necessary to monitor the current change in the winding in real time through a current sensor installed in the transformer. According to the inter-turn short circuit prediction data obtained in step S242, the threshold of the current monitoring system is set to capture the increasing trend of abnormal current in real time. Abnormal current growth is usually accompanied by phenomena such as current fluctuation, frequency change, or sudden rise in the current value. When the current exceeds the set safety range, the system will record and analyze the amplitude and frequency characteristics of the current growth, and identify the root cause of the abnormal current through techniques such as frequency domain analysis. In this process, monitoring devices such as current transformers and voltage sensors are used, combined with a data acquisition system for real-time monitoring. Once an abnormality is found, an alarm is issued in a timely manner and the data is stored for subsequent analysis and evaluation. The goal of this step is to confirm the degree of abnormal current growth inside the transformer.

[0082] Step S244: Perform a thermal feedback increasing effect analysis on the thermal effect accumulation data of the power supply system according to the abnormal growth degree of the induced current of the transformer, and obtain thermal feedback increasing effect data;

[0083] In the embodiments of the present invention, a thermal feedback increasing effect analysis is performed on the thermal effect accumulation data of the power supply system according to the abnormal growth degree of the induced current of the transformer, and thermal feedback increasing effect data is obtained. After the abnormal growth of the induced current occurs in the transformer, additional thermal effects will be generated, further exacerbating the thermal load of the system. In this process, according to the abnormal current data detected in step S243, a feedback analysis of the thermal load of the power supply system is performed. The specific operation is as follows: By real-time monitoring the transformer temperature and load conditions, combined with the thermal feedback effect model, calculate the change in heat accumulation caused by the growth of the induced current. The thermal feedback effect refers to the additional heat caused by abnormal current, which further increases the overall temperature of the system, thereby accelerating the deterioration of the equipment. In the analysis, the abnormal increment of the current is combined with the thermal effect accumulation data of the power supply system, and thermal conduction simulation and heat flow calculation methods are used to deduce the influence degree of the thermal feedback. The purpose of this step is to quantify the increasing influence of the thermal feedback effect on the thermal effect accumulation of the power supply system, and provide accurate data for subsequent fault warning. Through this analysis, the thermal runaway risk of the system under different operating conditions can be accurately predicted.

[0084] Step S245: Predict the fault evolution trajectory data of the power supply system according to the load overload condition of the power supply system and the thermal feedback increasing effect data.

[0085] In the embodiments of the present invention, the fault evolution trajectory data of the power supply system is predicted based on the load overload condition of the power supply system and the data of the thermal feedback increasing effect. The load overload of the power supply system is closely related to the thermal feedback effect, and the combined action of the two triggers the fault evolution of the power supply system. In this step, the combined influence of load overload and thermal feedback is analyzed by combining the data obtained in steps S241 to S244. By collecting the load data, thermal effect data, turn-to-turn short circuit data, and abnormal current increase data of the power supply system, a comprehensive prediction is made using the fault evolution model. This model combines the performance degradation trend of the system under different loads, the accumulation of thermal effects, and the fault evolution path to simulate the evolution process of the power supply system under different fault states. Through in-depth analysis of these data, the type of fault that occurs in the power supply system and the time of fault occurrence are predicted, so as to provide early warning information for maintenance personnel. The specific operation is to calculate the future fault trajectory through the system fault diagnosis algorithm and historical fault data, so as to obtain the fault evolution trajectory data of the power supply system.

[0086] Preferably, step S245 includes the following steps:

[0087] Estimate the influence degree of the phase margin of the power supply system according to the load overload condition of the power supply system;

[0088] In the embodiments of the present invention, the load overload condition of the power supply system is closely related to the phase margin of the system. When there is a load overload, the load and working state of the power supply system exceed the predetermined safe working range, resulting in a decrease in the phase margin. During the implementation process, the load data of the power supply system is collected, and the specific values of the load changes are obtained by using the real-time load monitoring sensor and the load analysis model. According to the load data and the frequency response characteristics of the power supply system, combined with the phase margin calculation formula in control theory, the phase margin changes of the power supply system under different loads are calculated. The phase margin reflects the anti-interference ability of the system to interference. Under the overload condition, the phase margin of the system often decreases. By analyzing the relationship between the load data and the phase margin, the degree of change of the phase margin of the power supply system under the load overload condition can be estimated.

[0089] Analyze the power loop oscillation trend according to the influence degree of the phase margin of the power supply system;

[0090] In the embodiments of the present invention, the phase margin is a key parameter for the stability of the power supply system. When the phase margin is insufficient, the stability of the power supply system decreases, leading to loop oscillation. Analyze the oscillation trend, calculate the influence degree of the phase margin obtained, and combine with the frequency response curve of the power supply system. Use system stability analysis methods (such as Nyquist diagram or Bode diagram analysis) to evaluate the impact of the phase margin change on the system. Utilize the stability criterion of the control system to analyze how the change of the phase margin affects the oscillation behavior of the system under load overload conditions. By real-time monitoring the feedback signal of the system, detect the frequency response and phase shift of the loop, and calculate the probability of the oscillation trend based on these data. This analysis process uses the steady-state and transient response data of the control system, combines with mathematical modeling methods, simulates the oscillation trend of the system under different load conditions, and obtains the evolution direction and degree of the power supply loop oscillation.

[0091] Determine the exacerbation of the power supply voltage fluctuation based on the oscillation trend of the power supply loop and the influence degree of the phase margin of the power supply system;

[0092] In the embodiments of the present invention, the oscillation of the power supply loop will directly cause the power supply voltage fluctuation. Especially when the phase margin of the power supply system decreases, the oscillation will exacerbate the voltage fluctuation. In this step, based on the obtained oscillation trend of the power supply loop, combine with the voltage signal of the power supply system to analyze the exacerbation of the power supply voltage fluctuation. In the specific implementation process, real-time monitor the change of the power supply output voltage, use the voltage sensor to collect the system voltage data, and use digital signal processing methods to perform spectrum analysis on the voltage fluctuation. By analyzing the frequency components and amplitude changes of the voltage signal, determine the increasing trend of the power supply voltage fluctuation. Combine with the influence data of the phase margin of the power supply system to evaluate the exacerbation degree of the voltage fluctuation, and analyze the exacerbation phenomenon of the voltage fluctuation caused by the oscillation. Through modeling calculation, obtain the exacerbation value of the power supply voltage fluctuation under different load and phase margin conditions.

[0093] Predict the increasing characteristics of the power supply voltage ripple according to the power supply voltage fluctuation;

[0094] In the embodiments of the present invention, after the power supply voltage fluctuation exacerbates, the ripple of the power supply voltage also increases, which will affect the operation stability of the equipment. According to the obtained power supply voltage fluctuation data, further analyze the increasing characteristics of the voltage ripple, collect the high-frequency fluctuation data of the power supply output voltage, and use frequency domain analysis techniques such as fast Fourier transform (FFT) to extract the spectrum characteristics of the voltage ripple. By analyzing the amplitude and frequency characteristics of the voltage ripple, combine with the influence data of the load overload and phase margin of the power supply system to predict the increasing trend of the power supply voltage ripple. The increase of the ripple is usually accompanied by the decrease of the system stability. The voltage ripple affects the performance of electronic equipment. By predicting the ripple characteristics, the voltage quality problems occurring in the power supply system can be identified and measures can be taken in advance.

[0095] Detect abnormal electrolyte distribution inside the power supply based on the power supply voltage ripple increase characteristics and thermal feedback incremental effect data;

[0096] In an embodiment of the present invention, the abnormal distribution of electrolytes inside the power supply capacitor is usually related to the increase of the power supply voltage ripple, and the uneven distribution of electrolytes causes the capacitance to decay, thereby affecting the stability of the power supply system. According to the obtained voltage ripple increase characteristics, combined with the thermal feedback incremental effect data, the abnormal distribution of electrolytes inside the power supply is detected, and the distribution changes of electrolytes under different temperature and load conditions are calculated by using the internal temperature data of the capacitor and the internal structure model of the capacitor, using a thermal simulation analysis method. The uneven distribution of electrolytes can cause changes in the electrical characteristics of the capacitor, resulting in increased voltage fluctuations. By real-time monitoring of the temperature changes and electrolyte flow states of the power supply capacitor, sensors are used to obtain data and analyze the abnormal conditions of the electrolyte distribution. When the electrolyte distribution is abnormal, the performance of the capacitor will decline, which will lead to system instability.

[0097] Predict the power supply capacitor failure condition based on the abnormal distribution of electrolytes inside the power supply;

[0098] In an embodiment of the present invention, abnormal electrolyte distribution is one of the main reasons for the failure of the power supply capacitor. According to the obtained electrolyte distribution abnormality data, combined with the electrical characteristics of the capacitor, the failure condition of the power supply capacitor is predicted. In this process, the aging process of the capacitor due to uneven electrolyte distribution is analyzed by combining the electrolyte distribution data with the durability model of the capacitor. By establishing a capacitor failure prediction model based on factors such as temperature, voltage, and load, the time and probability of capacitor failure are further estimated. This process requires the collection of working parameters of the power supply capacitor, such as voltage, current, and temperature, and the failure time point of the capacitor is predicted by a data analysis method, and the failure condition of the power supply capacitor is obtained in combination with the operating status of the system.

[0099] The power supply system fault evolution trajectory data is predicted based on the power supply capacitor failure status and the abnormal distribution of electrolytes inside the power supply.

[0100] In an embodiment of the present invention, the failure of the power supply capacitor is usually accompanied by the occurrence of a system failure. According to the failure status of the power supply capacitor and the abnormal status of the electrolyte distribution, the fault evolution trajectory of the power supply system is predicted. Based on the obtained capacitor failure data, combined with the load data and voltage fluctuation data of the power supply system, the fault evolution model is used to model the fault trajectory of the system. This process uses the health status data of the power supply system to predict the failure mode and its evolution trend of the power supply equipment in the future. By integrating the status data of each component in the power supply system (such as power supply capacitors, electrolytes, loads, temperature, etc.), combined with historical fault data, the expansion path and time series of the fault are simulated, so as to obtain the time and type of the power supply system failure.

[0101] Preferably, step S3 includes the following steps:

[0102] Step S31: Estimate the attenuation degree of the power system transmission quality according to the power system fault evolution trajectory data;

[0103] In the embodiment of the present invention, through the power system fault evolution trajectory data, the attenuation degree of the power system transmission quality is evaluated, and the operation data of the power system within a period of time is collected, including basic electrical parameters such as voltage, current, and frequency. By deeply analyzing the historical records of the power system fault evolution trajectory data, a power system state evolution model is established. This model predicts the fault modes of each component of the power system through time series and how it affects the transmission quality of the system. Using these data, calculate the attenuation trend of each node of the system, and quantitatively evaluate the impact of abnormal fluctuations in the system operation on the transmission quality. By statistically analyzing the transmission data of different currents and voltages in the power system, potential fault signs are identified, especially those fluctuations caused by overload, aging, or abnormal conditions. By integrating these historical data and adopting techniques based on time series prediction (such as ARIMA model or prediction algorithm), effectively estimate how the transmission quality of the power system will attenuate in the future for a period of time. At this time, the attenuation degree of the system transmission quality is correlated with the fault evolution trajectory data, and the evaluation of the transmission quality is adjusted in real time through iterative update of the data.

[0104] Step S32: Detect the growth trend of the transmission line loss according to the attenuation degree of the power system transmission quality;

[0105] In the embodiments of the present invention, according to the degree of attenuation of the transmission quality of the power supply system obtained in step S31, the increasing trend of the line loss of the transmission line is further analyzed and detected. The attenuation of the transmission quality of the power supply system is usually accompanied by an increase in line loss, especially in the case of high load or aging. A real-time transmission monitoring device (such as current and voltage sensors and power meters) is used to collect the voltage and current data of the transmission line of the power supply system. After filtering and denoising these data, a method based on load analysis and power loss calculation is used to accurately determine the loss level of the transmission line. To accurately analyze the increasing trend of the transmission line loss, factors such as system load, length of the transmission line, and resistance need to be comprehensively considered. By real-time monitoring the voltage and current at each node on the line and combining with the theoretical models of losses in the power system (such as Ohm's law and power loss formula), the real-time data of the line loss in the power supply system is calculated. These data are compared with the transmission quality attenuation data of the power supply system to obtain the increasing trend of the transmission line loss. According to the above data, statistical analysis methods (such as regression analysis, trend analysis, etc.) are used to evaluate the changing trend of the line loss over time. If the attenuation of the transmission quality of the power supply system intensifies, the line loss often shows a certain linear or non-linear growth pattern. By continuously tracking and predicting these trends, the health status of the transmission line can be accurately reflected.

[0106] Step S33: Analyze the attenuation of the control precision of the power supply system based on the increasing trend of the transmission line loss and the degree of attenuation of the transmission quality of the power supply system;

[0107] In the embodiments of the present invention, the control precision is one of the key factors of the performance of the power supply system. Especially in the case of attenuation of transmission quality and increase in line loss, the decrease in control precision will exacerbate the instability of the system. In this step, combining the increasing trend of the transmission line loss obtained in step S32 and the transmission quality attenuation data in step S31, the attenuation of the control precision of the power supply system is analyzed. By collecting the control signal and feedback signal of the power supply system (such as adjustment signal, control current and voltage data, etc.), the response data of the closed-loop control system is used to quantitatively analyze the control precision. Since the increase in the transmission line loss usually leads to a decrease in the response speed of the power supply system, the attenuation of the control precision is usually manifested in aspects such as an increase in the lag of the system control response and a decrease in stability. Using control system analysis methods (for example, system stability analysis, PID regulation analysis, etc.), the changing trend of the control precision over time is effectively analyzed, especially how the control precision is affected in the case of load overload or increase in line loss. Through the comprehensive analysis of the transmission quality, line loss, and control signal feedback of the power supply system, a mathematical model regarding the attenuation of the control precision of the power supply system can be established to further reveal the reasons for the decrease in control precision and its evolution trend.

[0108] Step S34: Evaluate the downward trend of the power supply system performance based on the growth trend of the transmission line loss and the attenuation of the power supply system control accuracy.

[0109] In the embodiment of the present invention, combining the growth trend of the transmission line loss obtained in step S32 with the attenuation of the control accuracy in step S33, comprehensively evaluate the overall downward trend of the power supply system performance. The performance degradation of the power supply system usually includes aspects such as reduced power transmission efficiency, insensitive control response, and reduced system stability. Based on the growth of the transmission line loss and the attenuation of the control accuracy, through statistical methods such as linear regression and non-linear regression, the change trends of each variable are integrated to obtain the downward trend of the power supply system performance. Further analyze these data, and by constructing a power supply system health index model, quantify the decline process of the power supply system performance. This process comprehensively judges whether the power supply system is in a healthy working state according to the evaluation of multiple dimensions such as the power transmission efficiency, response speed, and stability of the system. If the trends of increasing transmission loss and decreasing control accuracy of the system continue, the performance of the power supply system will gradually decline, resulting in failures or inability to meet load requirements. Through this comprehensive analysis, the performance decline of the power supply system in a certain period of time in the future can be accurately predicted.

[0110] Preferably, step S4 includes the following steps:

[0111] Step S41: Conduct a power supply system health attenuation evaluation on the downward trend of the power supply system performance to obtain power supply system health attenuation data;

[0112] In the embodiment of the present invention, through continuous monitoring and data collection, obtain various operating parameters of the power supply system within a period of time, including key indicators such as current, voltage, temperature, frequency, and power. These data are collected in real time by sensors and monitoring devices and stored in the central database. By analyzing these parameters, identify the decline trend of the system performance. For each parameter, use time series analysis methods (such as trend analysis and moving average method) to judge whether the system shows a decline trend, and conduct a comparative analysis of the current system state based on historical data. Calculate the main indicators of system health attenuation, such as current fluctuation, number of temperature overrun times, efficiency loss, etc., to further quantify the attenuation degree of the power supply system. Compare the attenuation trend with the normal operating parameters of the power supply system to identify the part that deviates from the normal value. By inputting these data into the health attenuation evaluation model, the model will conduct a multi-dimensional comprehensive analysis by combining the historical performance and real-time data of the power supply system, and output the health attenuation data of the power supply system. The attenuation data includes, but is not limited to, indicators such as the decline in the working efficiency of the power supply system, enhanced thermal effect, and weakened load bearing capacity. Through the evaluation of the attenuation data, the health status of the power supply system can be comprehensively reflected.

[0113] Step S42: Analyze the power system health attenuation factors based on the power system health attenuation data to obtain the power system health attenuation factor data;

[0114] In the embodiment of the present invention, based on the power system health attenuation data obtained in the foregoing steps, various factors leading to the decline of the power system are further analyzed. This process identifies factors highly correlated with health attenuation through correlation analysis among different parameters within the system. For example, unstable voltage, frequent current fluctuations, high temperature, etc. are all common factors for the decline of the power system. Through statistical methods such as multiple regression analysis and variance analysis, the influence degree of these factors is further quantified. A detailed analysis is carried out on each component of the power system, including transformers, load control, thermal management systems, protection systems, etc. Based on the operating data of different components, a factor model for the health decline of the power system is constructed. Based on the data collected by sensors, such as temperature, humidity, vibration, etc., combined with the system fault history data, the root causes of the decline and their change trends are identified. Through this analysis, the key factors for the decline of the power system can be clarified and further sorted into power system health decline factor data, covering multiple decline reasons such as equipment aging, abnormal load, and unstable power input.

[0115] Step S43: Classify the fault situations based on the power system health attenuation data to obtain the power system fault situation classification data;

[0116] In the embodiment of the present invention, based on the power system health attenuation data obtained in step S41, the classification process of the power system fault situations is started. By calibrating the historical data of different fault types of the power system, the fault situations are divided into different categories, such as electrical faults, thermal effect faults, load overload faults, etc. By using clustering analysis (such as K-means or DBSCAN algorithms) to classify the attenuation data, various fault patterns occurring in the system can be automatically identified. Each fault category corresponds to one or more influencing factors, such as temperature exceeding the standard, abnormal vibration, current fluctuation, etc. of the power system. Classification algorithms such as decision trees or support vector machines can also be used, combined with the health decline data, to accurately classify and calibrate the faults of the power system. Through deep learning of the historical fault data, the characteristic manifestations of the power system in different fault scenarios can be identified, so as to accurately classify the current fault data into the corresponding fault types. The classification data of each fault type will be further refined according to the severity, occurrence frequency, and influence degree of the fault to form the power system fault situation classification data.

[0117] Step S44: Input the power system fault situation classification data and the power system health attenuation factor data into a preset artificial intelligence model for model deep learning training to obtain a power system artificial intelligence model;

[0118] In the embodiment of the present invention, the power system fault situation classification data obtained in step S43 and the power system health attenuation factor data in step S42 are jointly input into a preset artificial intelligence model for deep learning training. The model uses deep learning methods such as convolutional neural network (CNN), recurrent neural network (RNN), or long short-term memory network (LSTM) to capture complex non-linear relationships in the data. The fault situation classification data and the health attenuation factor data are subjected to data cleaning and standardization processing to remove noise and outliers. Then, these data are trained through the artificial intelligence model to learn the correlation between various decays and faults within the system, and the model parameters are adjusted according to the training results to enable accurate prediction of the health status and fault evolution of the power system. Through the layer-by-layer training of the multi-layer neural network, the model can extract features from the input health attenuation data and establish a deep relationship with the fault type, attenuation trend, and system performance. After sufficient training, the artificial intelligence model can generate health assessment and fault prediction results for the power system. The model can not only identify the current health status of the system but also predict future faults and decay trends.

[0119] Step S45: Evaluate the health status of the power system according to the power system artificial intelligence model to obtain power system health status evaluation data.

[0120] In the embodiments of the present invention, when evaluating the health status of the power supply system based on the artificial intelligence model of the power supply system trained in step S44, it is necessary to input the real-time operation data of the power supply system (such as current, voltage, power, temperature, frequency, etc.) into the model. These data are monitored and collected in real time by sensors and undergo necessary preprocessing (such as data cleaning, standardization, and normalization) to ensure their quality and consistency. The preprocessed data is input into the trained artificial intelligence model, and the neural network or algorithm inside the model will deeply analyze these data and evaluate the health status of the power supply system based on the previous learning results. The model uses the features and rules extracted from historical data to identify the current health status of the system, perform fault diagnosis, attenuation analysis, and health risk assessment. During the evaluation process, the model diagnoses the current state of the system to check for any abnormalities or faults. Through the deep learning of the model, key fault features in the power supply system (such as excessive voltage fluctuations, too high temperature, or frequency fluctuations, etc.) can be identified, and based on these features, it can be determined whether there are faults or abnormalities in the power supply system. In addition, the model will also analyze the attenuation degree of the power supply system. By comparing with historical health data, it can be determined whether the decline of the system is within the normal range. If the decline speed exceeds the expected range, the model will identify potential health risks and give early warning signals, indicating that maintenance or repair is required. In addition to diagnosing the current state, the model can also predict the future health status of the power supply system. This prediction not only includes predicting the probability of a fault occurring but also covers aspects such as the evolution trend of the fault and the decline speed. The model will estimate the fault probability of the power supply system within a future period of time based on the input data and the rules learned during the training process, and predict the change trajectory of the system health status. The prediction results include the decline amplitude of the system performance, the type of faults that will occur, and their occurrence time within a certain future period, helping decision-makers take measures in advance to prevent the occurrence of faults or delay the decline process.

[0121] The present invention also provides a health assessment system for a power supply system based on artificial intelligence, which is used to execute the health assessment method for a power supply system based on artificial intelligence as described above. The health assessment system for a power supply system based on artificial intelligence includes:

[0122] An operating state assessment module, which is used to obtain power supply system data; estimate the initial performance parameters of the power supply system according to the power supply system data; and perform an operating state assessment of the power supply system based on the initial performance parameters of the power supply system to obtain the initial state data of the power supply system.

[0123] A fault evolution trajectory prediction module, which is used to perform a load operation simulation of the power supply equipment based on the initial state data of the power supply system to obtain the load operation simulation data of the power supply system; and predict the fault evolution trajectory data of the power supply system according to the load operation simulation data of the power supply system.

[0124] A fault evolution trajectory prediction module, which is used to perform load operation simulation of power supply equipment based on the initial state data of the power supply system to obtain load operation simulation data of the power supply system; and predict the fault evolution trajectory data of the power supply system according to the load operation simulation data of the power supply system.

[0125] A health status evaluation module, which is used to evaluate the health attenuation data of the power supply system based on the performance degradation trend of the power supply system; transmit the health attenuation data of the power supply system to a preset artificial intelligence model for model deep learning training to obtain an artificial intelligence model of the power supply system; and evaluate the health status of the power supply system according to the artificial intelligence model of the power supply system to obtain the health status evaluation data of the power supply system.

[0126] The above are only specific embodiments of the present invention, enabling those skilled in the art to understand or implement the present invention. Various modifications to these embodiments will be obvious to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to these embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A health assessment method for a power supply system based on artificial intelligence, characterized in that, It includes the following steps: Step S1: Obtain power system data; Estimate the initial performance parameters of the power system according to the power system data; Conduct an assessment of the operating state of the power system based on the initial performance parameters of the power system to obtain the initial state data of the power system; Step S2: Conduct a simulation of the load operation of the power equipment based on the initial state data of the power system to obtain the simulated load operation data of the power system; Predict the fault evolution trajectory data of the power system according to the simulated load operation data of the power system; Step S3: Estimate the degree of attenuation of the transmission quality of the power system according to the fault evolution trajectory data of the power system; Detect the growth trend of the transmission line loss according to the degree of attenuation of the transmission quality of the power system; Evaluate the downward trend of the performance of the power system according to the growth trend of the transmission line loss and the degree of attenuation of the transmission quality of the power system; Step S4: Evaluate the health attenuation data of the power system for the downward trend of the performance of the power system; Transmit the health attenuation data of the power system to a preset artificial intelligence model for model deep learning training to obtain an artificial intelligence model of the power system; Conduct a health state assessment of the power system according to the artificial intelligence model of the power system to obtain the health state assessment data of the power system.

2. The health assessment method of the power supply system based on artificial intelligence according to claim 1, characterized in that, Step S1 includes the following steps: Step S11: Obtain power system data; Step S12: Extract the internal structure data of the power system according to the power system data; Step S13: Estimate the initial performance parameters of the power system according to the power system data and the internal structure data of the power system; Step S14: Conduct an initial state assessment of the power system based on the internal structure data of the power system and the initial performance parameters of the power system to obtain the initial state data of the power system.

3. The health assessment method of the power supply system based on artificial intelligence according to claim 2, wherein Step S13 includes the following steps: Step S131: Identify the connection relationship characteristics of the power system components according to the internal structure data of the power system, and extract the internal topology structure of the power system based on the connection relationship characteristics of the power system components; Step S132: Calculate the output efficiency of the power system according to the power system data; Step S133: Calculate the heat generation data of the power system based on the output efficiency of the power system; Step S134: Estimate the temperature rise stability parameters of the power system according to the heat generation data of the power system and the internal topology structure of the power system; Step S135: Evaluate the voltage regulation performance data of the power system based on the internal topology structure of the power system and the output efficiency of the power system; Step S136: Estimate the initial performance parameters of the power system according to the voltage regulation performance data of the power system and the temperature rise stability parameters of the power system.

4. The health assessment method of the power supply system based on artificial intelligence according to claim 1, characterized in that Step S2 includes the following steps: Step S21: Conduct a simulation of the load operation of the power equipment based on the initial state data of the power system to obtain the simulated load operation data of the power system; Step S22: Detect the overload condition of the power system according to the simulated load operation data of the power system; Step S23: Conduct a prediction of the accumulation of the thermal effect of the power system based on the overload condition of the power system to obtain the accumulated thermal effect data of the power system; Step S24: Predict the fault evolution trajectory data of the power system based on the overload condition of the power system and the accumulated thermal effect data of the power system.

5. The health assessment method of the power supply system based on artificial intelligence according to claim 4, wherein, Step S23 includes the following steps: Step S231: Identify the abnormal state of system power conversion for the overload condition of the power system; Step S232: Statistically analyze the growth trend of the system power loss based on the abnormal state of the system power conversion; Step S233: Determine the attenuation degree of the power supply system working efficiency based on the overload condition of the power supply system load; Step S234: Estimate the growth of heat energy conversion according to the attenuation degree of the power supply system working efficiency and the growth trend of the system power loss; Step S235: Predict the thermal runaway state of the power supply system according to the growth of heat energy conversion; Step S236: Based on the thermal runaway state of the power supply system and the growth of heat energy conversion, estimate the accumulation of the thermal effect of the power supply system to obtain the thermal effect accumulation data of the power supply system.

6. The health assessment method of the power supply system based on artificial intelligence according to claim 4, wherein, Step S24 includes the following steps: Step S241: Analyze the deterioration trend of the transformer insulation material based on the thermal effect accumulation data of the power supply system to obtain the deterioration trend data of the transformer insulation material; Step S242: Estimate the inter-turn short circuit situation of the transformer according to the deterioration trend data of the transformer insulation material; Step S243: Detect the abnormal growth degree of the transformer induced current according to the inter-turn short circuit situation of the transformer; Step S244: Perform a thermal feedback increasing effect analysis on the thermal effect accumulation data of the power supply system according to the abnormal growth degree of the transformer induced current to obtain the thermal feedback increasing effect data; Step S245: Predict the power supply system fault evolution trajectory data according to the overload condition of the power supply system load and the thermal feedback increasing effect data.

7. The health assessment method of the power supply system based on artificial intelligence according to claim 6, wherein, Step S245 includes the following steps: Estimate the influence degree of the power supply system phase margin according to the overload condition of the power supply system load; Analyze the power loop oscillation trend according to the influence degree of the power supply system phase margin; Determine the aggravation of the power supply voltage fluctuation based on the power loop oscillation trend and the influence degree of the power supply system phase margin; Predict the characteristic of the increase in the power supply voltage ripple according to the power supply voltage fluctuation; Detect the abnormal condition of the internal electrolyte distribution of the power supply based on the characteristic of the increase in the power supply voltage ripple and the thermal feedback increasing effect data; Predict the power supply capacitor failure condition according to the abnormal condition of the internal electrolyte distribution of the power supply; Predict the power supply system fault evolution trajectory data based on the power supply capacitor failure condition and the abnormal condition of the internal electrolyte distribution of the power supply.

8. The health assessment method of the power supply system based on artificial intelligence according to claim 1, wherein Step S3 includes the following steps: Step S31: Estimate the attenuation degree of the power supply system transmission quality according to the power supply system fault evolution trajectory data; Step S32: Detect the growth trend of the transmission line loss according to the attenuation degree of the power supply system transmission quality; Step S33: Analyze the attenuation of the power supply system control accuracy based on the growth trend of the transmission line loss and the attenuation degree of the power supply system transmission quality; Step S34: Evaluate the decline trend of the power supply system performance according to the growth trend of the transmission line loss and the attenuation of the power supply system control accuracy.

9. The health assessment method of the power supply system based on artificial intelligence according to claim 1, characterized in that Step S4 includes the following steps: Step S41: Conduct a power supply system health attenuation assessment on the decline trend of the power supply system performance to obtain the power supply system health attenuation data; Step S42: Analyze the power supply system health attenuation factors based on the power supply system health attenuation data to obtain the power supply system health attenuation factor data; Step S43: Perform a classification process on the fault conditions based on the power supply system health attenuation data to obtain the power supply system fault condition classification data; Step S44: Classify the power system fault situation data and the power system health attenuation factor data to a preset artificial intelligence model for model deep learning training to obtain a power system artificial intelligence model; Step S45: Perform a health status assessment of the power system according to the power system artificial intelligence model to obtain power system health status assessment data.

10. A health assessment system for a power supply system based on artificial intelligence, characterized in that, For implementing the health assessment method of the power system based on artificial intelligence as described in claim 1, the health assessment system of the power system based on artificial intelligence includes: An operating status assessment module, configured to obtain power system data; estimate the initial performance parameters of the power system according to the power system data; perform an operating status assessment of the power system based on the initial performance parameters of the power system to obtain power system initial state data; A fault evolution trajectory prediction module, configured to perform a load operation simulation of the power equipment based on the power system initial state data to obtain power system load operation simulation data; predict power system fault evolution trajectory data according to the power system load operation simulation data; A performance degradation trend assessment module, configured to estimate the attenuation degree of the power system transmission quality according to the power system fault evolution trajectory data; detect the growth trend of the transmission line loss according to the attenuation degree of the power system transmission quality; evaluate the performance degradation trend of the power system according to the growth trend of the transmission line loss and the attenuation degree of the power system transmission quality; A health status assessment module, configured to evaluate the power system health attenuation data according to the power system performance degradation trend; transmit the power system health attenuation data to a preset artificial intelligence model for model deep learning training to obtain a power system artificial intelligence model; perform a health status assessment of the power system according to the power system artificial intelligence model to obtain power system health status assessment data.

Images