An intelligent power distribution automation terminal automatic testing method and system

Abstract

The application provides a kind of intelligent power distribution automation terminal automatic test method and system, it is related to automatic test technical field, method includes: the model information and environmental information of terminal to be measured are obtained;Test data of terminal to be measured are generated in combination with model information and environmental information, in combination with Lyapunov index;Test data are input to terminal to be measured, and output operation data;Dynamic threshold based on potential energy field is used to analyze operation data, and output abnormal data;Based on pre-training model and coupling weighted distance, abnormal data are fused and classified to determine fault type;Early warning is carried out according to fault type.Based on the analysis method of Lyapunov index and dynamic threshold, test can comprehensively cover all kinds of functions and limit state of equipment, so that potential abnormality and fault can be effectively found.

Description

Technical Field

[0001] This invention relates to the field of automatic testing technology, and in particular to an automatic testing method and system for intelligent power distribution automation terminals. Background Technology

[0002] Intelligent distribution automation terminals are intelligent management devices used in power system distribution networks. Through real-time monitoring, control, and protection of the distribution network, they can perform multiple functions such as data acquisition, fault detection, and load management, improving the automation, informatization, and intelligence levels of the distribution system and ensuring the stability and security of power supply.

[0003] The necessity of automated testing for intelligent distribution automation terminals lies in its ability to ensure the stability and reliability of all functions of the terminal equipment before and during operation. Automated testing can simulate various operating environments, verify the equipment's response speed, accuracy, and fault tolerance, and promptly identify potential faults. This avoids the incompleteness and errors inherent in manual testing, ensuring the efficient operation of the distribution network and the safe and stable operation of the power system.

[0004] However, existing power distribution terminal testing processes rely heavily on manual operation, resulting in low testing efficiency, long cycles, difficulty in timely detection of power distribution terminal anomalies, and susceptibility to human factors, leading to inaccurate test results and an inability to fully cover the complex functions and extreme states of the equipment. Summary of the Invention

[0005] In view of the shortcomings of the prior art, the purpose of this invention is to provide an automatic testing method for intelligent power distribution automation terminals, which can solve the technical problems of the prior art, such as the power distribution terminal testing process relying on manual operation, low testing efficiency, long cycle, difficulty in timely detection of power distribution terminal anomalies, and susceptibility to human factors, resulting in inaccurate test results and difficulty in fully covering the complex functions and extreme states of the equipment.

[0006] A first aspect of this invention provides an automatic testing method for intelligent power distribution automation terminals, comprising:

[0007] S1: Obtain the model information and environmental information of the terminal under test;

[0008] Environmental information includes temperature, humidity, and electromagnetic interference intensity.

[0009] S2: Combine model information and environmental information with Lyapunov index to generate test data for the terminal under test;

[0010] S3: Input test data into the terminal under test and output running data;

[0011] S4: Analyze the running data using a dynamic threshold based on the potential energy field and output abnormal data;

[0012] S5: Based on the pre-trained model and the coupled weighted distance, abnormal data are fused and classified to determine the fault type;

[0013] S6: Provide early warnings based on fault type.

[0014] A second aspect of this invention provides an automatic testing system for intelligent power distribution automation terminals, comprising: a processor and a memory;

[0015] The memory stores programs or instructions that can run on the processor, which, when executed by the processor, implement the steps of the automatic testing method for intelligent power distribution automation terminals as described in the first aspect.

[0016] A third aspect of the present invention provides a readable storage medium storing a program or instructions that, when executed by a processor, implement the steps of the automatic testing method for intelligent power distribution automation terminals as described in the first aspect.

[0017] The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following:

[0018] In this embodiment of the invention, compared to traditional manual operation, this automated testing process acquires the model and environmental information of the terminal under test and generates test data based on this information, reducing errors caused by human intervention. Simultaneously, the analysis method based on Lyapunov exponent and dynamic thresholds enables the test to comprehensively cover various functions and extreme states of the equipment, thereby effectively identifying potential anomalies and faults. Furthermore, data fusion combining a pre-trained model and coupled weighted distance allows for accurate identification of fault types and timely early warning, ensuring the stability and safety of the power distribution system and improving the overall reliability of the power distribution terminal. Attached Figure Description

[0019] The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Throughout the drawings, the same reference numerals denote the same parts. Obviously, the drawings described below are merely some embodiments of the present invention, and those skilled in the art can obtain other drawings based on these drawings without any creative effort.

[0020] Figure 1 This is a flowchart illustrating an automatic testing method for an intelligent power distribution automation terminal provided in an embodiment of the present invention;

[0021] Figure 2This is a schematic diagram of the structure of an automatic testing system for intelligent power distribution automation terminals provided in an embodiment of the present invention. Detailed Implementation

[0022] To enable those skilled in the art to better understand the technical solutions in the embodiments of the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. It should be understood that these descriptions are merely exemplary and are not intended to limit the scope of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0023] The automatic testing method for intelligent power distribution automation terminals provided by the present invention will be described in detail below with reference to the accompanying drawings, through specific embodiments and application scenarios.

[0024] Reference manual attached Figure 1 The diagram shows a flowchart of an automatic testing method for an intelligent power distribution automation terminal provided by an embodiment of the present invention.

[0025] This invention provides an automatic testing method for intelligent power distribution automation terminals, which may include the following steps:

[0026] S1: Obtain the model information and environmental information of the terminal under test.

[0027] Environmental information includes temperature, humidity, and electromagnetic interference intensity. By acquiring the model information of the terminal under test and environmental information—factors that affect the device's operating performance—the accuracy and relevance of the test data are ensured. Conducting comprehensive testing under different environmental conditions helps improve the reliability and accuracy of the test results.

[0028] S2: Combine model information and environmental information with Lyapunov indices to generate test data for the terminal under test.

[0029] The Lyapunov index is a method for measuring system stability and sensitivity. It describes how a small disturbance under initial conditions changes over time, reflecting the degree of disorder in the system. In the testing of intelligent power distribution automation terminals, the Lyapunov index can be used to predict and analyze the dynamic behavior of terminal equipment under different operating environments, helping to generate realistic test data and simulate various extreme conditions of the equipment.

[0030] In one possible implementation, the test data includes electromagnetic voltage and current data, electronic voltage and current data, digital voltage and current data, and remote control data.

[0031] Understandably, the test data covers different types of measurement methods. This data comprehensively reflects the voltage and current changes of the power distribution terminal under different technical means, as well as the remote control status, ensuring multi-dimensional testing and evaluation of the equipment's performance under various operating conditions, and improving the accuracy and reliability of the tests.

[0032] In one possible implementation, S2 specifically includes:

[0033] S201: Establish the environmental information tensor.

[0034] The specific formula for the environmental information tensor is as follows:

[0035] ;

[0036] in, Represents an environmental information tensor. Indicates the first i Normalized intensity of environmental factors i =1,2,..., n , n This indicates the total number of environmental factors, and the subscript T indicates transpose.

[0037] Among them, the i The normalization intensity of environmental factors refers to the normalization process of the actual measured values ​​of environmental factors so that their values ​​fall within a uniform standard range (such as 0 to 1). This processing method eliminates the differences in the dimensions and units of different environmental factors, enabling various environmental factors to be compared and integrated at the same scale. This allows for a more effective description of the impact of the environment on equipment, improving the comparability of data and the accuracy of analysis.

[0038] S202: Combining environmental information tensors, the coupling effect of environmental information is described in differential form.

[0039] The specific formula is as follows:

[0040] ;

[0041] in, The gradient operator represents the environmental information tensor. This indicates the partial derivative. This represents the matrix of material coefficients for the terminal under test related to the model information. This represents the environmental mutation operator. Indicates the time.

[0042] Among them, the material coefficient matrix of the terminal under test related to the model information is a parameter matrix reflecting the material properties of the equipment and its model. It takes into account the response and behavior of different materials of the equipment under specific environmental conditions and can adjust the impact of environmental factors on equipment performance. By combining with environmental information, this matrix helps to more accurately simulate the dynamic changes of different models of equipment under different environmental conditions, improving the accuracy of test data and the evaluation of equipment performance.

[0043] It should be noted that the coupling effect of environmental information is described in differential form. The gradient operator of the environmental information tensor, combined with partial derivatives and the environmental catastrophe operator, reflects the changes in environmental factors over time and their interactions. The material coefficient matrix considers the characteristics of the equipment model, ensuring accurate simulation of the equipment's response under different environmental conditions and enhancing the real-world relevance and accuracy of the test data.

[0044] S203: Combine the environmental information tensor and its gradient operator to construct a test parameter generator.

[0045] The specific formula for calculating the test parameter generator is as follows:

[0046] ;

[0047] in, The gradient operator represents the environmental information tensor. This represents the gradient operator drift term. Representation of time t The relevant Wiener process, This represents the feature vector of the terminal under test corresponding to the model information. Represents the mutation threshold constant. This indicates a test parameter generator.

[0048] The gradient operator drift term describes the average change trend of the gradient of the environmental information tensor over time. It reflects the long-term trend of the impact of environmental changes on device performance and is used to adjust the dynamic parameters in the model to more accurately predict the device's performance under different environmental conditions. By incorporating the drift term, the test parameter generator can better adapt to environmental changes, improving the accuracy and reliability of test data.

[0049] It should be noted that a test parameter generator was constructed by combining the environmental information tensor and its gradient operator. This generator uses differential equations to describe the impact of environmental changes on test parameters. In this way, test parameters related to environmental factors and equipment characteristics can be dynamically generated, ensuring that the testing process takes into account various complex external and internal factors.

[0050] S204: Use the test parameter generator to generate alternative test data.

[0051] S205: Use the Lyapunov index to filter candidate test data and output test data.

[0052] It should be noted that this process, by establishing an environmental information tensor and describing the coupling effects of environmental factors, combined with differential equations and gradient operators, generates highly accurate and scientific test data. Through the differential form of coupling effects, the impact of different environmental conditions on the equipment can be simulated, making the generated test data more closely resemble actual working scenarios. The test parameter generator, combining environmental information, equipment model, and time factors, can generate diverse candidate test data and uses the Lyapunov index to select the data that best meets the testing requirements. This method ensures the comprehensiveness, accuracy, and reliability of the test data, effectively improving the scientific rigor of the testing process and the ability to evaluate the equipment's performance under extreme conditions.

[0053] In one possible implementation, S205 specifically includes:

[0054] S2051: Obtain the maximum divergence rate threshold of the test data of the terminal under test.

[0055] The maximum divergence rate threshold for test data refers to the maximum permissible rate of change of data over a long period during testing. Specifically, it sets an upper limit representing the maximum degree of change in test data within a certain time interval. If the data change exceeds this threshold, it indicates potential instability or abnormal behavior in the system, requiring parameter adjustments and data regeneration. The maximum divergence rate threshold ensures the stability and reliability of test data, preventing excessively fluctuating abnormal data from affecting test results, and helps guarantee the accuracy of the testing process and the evaluation of equipment performance. It is understood that those skilled in the art can set the size of the maximum divergence rate threshold according to actual needs; this invention does not impose any limitations on this.

[0056] S2052: Calculate the Lyapunov index of the test data.

[0057] S2053: Determine whether the Lyapunov exponent is less than the maximum divergence threshold of the test data. If so, output the alternative test data as the test data. Otherwise, adjust the coefficient matrix of the terminal material to be tested and execute step S204.

[0058] The specific formula for determining this is:

[0059] ;

[0060] in, This represents the maximum divergence threshold. Represents the logarithmic function. T Indicates time interval, and They represent t Time and t + T The test data vector at time step, This represents the Lyapunov index.

[0061] It should be noted that the stability of the test data is ensured by obtaining the maximum divergence threshold of the test data from the terminal under test and calculating the Lyapunov exponent. The Lyapunov exponent measures the sensitivity of test data to changes over time. If the exponent is less than the maximum divergence threshold, it indicates that the data is relatively stable and can be used as valid test data output. Otherwise, the material coefficient matrix of the terminal under test is adjusted, and alternative test data is regenerated. The advantage of this process is that it uses the Lyapunov exponent to ensure the long-term stability and reliability of the test data, avoids interference from unstable or abnormal data, and improves the accuracy and comprehensiveness of the test results.

[0062] S3: Input test data into the terminal under test and output running data.

[0063] It should be noted that by inputting test data into the terminal under test and outputting operational data, the performance of the equipment under various working environments can be comprehensively and realistically captured. This process ensures that all functions of the equipment are verified under real-world operating conditions, helps identify potential problems, ensures the accuracy and comprehensiveness of test results, and further improves the reliability and stability of the power distribution terminal.

[0064] S4: Analyze the running data using a dynamic threshold based on the potential energy field and output abnormal data.

[0065] Among them, dynamic thresholding based on potential energy fields is a method for setting thresholds using potential energy field theory. By considering the energy changes and dynamic behavior of the equipment under different operating states, it sets a highly sensitive threshold to identify abnormal data. The potential energy field can simulate the situation of external disturbances during equipment operation, dynamically adjusting the threshold to more accurately identify potential abnormal or fault signals. By analyzing operating data using dynamic thresholding based on potential energy fields, abnormal behavior during equipment operation can be effectively identified. This method considers the changes in the equipment under different operating states, making anomaly detection more flexible and accurate, avoiding false positives and false negatives caused by static thresholds, thereby improving the accuracy of fault location and the comprehensiveness of testing.

[0066] In one possible implementation, S4 specifically includes:

[0067] S401: Maps the running data within a preset sliding window size to a potential energy field.

[0068] It should be noted that those skilled in the art can set the preset sliding window size according to actual needs, and this invention does not limit this.

[0069] The potential energy function of the potential energy field is as follows:

[0070] ;

[0071] ;

[0072] ;

[0073] in, Indicates the first i One running data point The potential energy value, This indicates the number of steps between historical running data points and the current running data point. Indicates and j Relevant historical operational data weighting coefficients, Indicates the preset sliding window size. Indicates the first in the sliding window j Historical operational data points, e Represents the natural constant. This represents the field strength attenuation factor. This represents the variance of the running data.

[0074] It should be noted that the potential energy value of each data point is calculated by mapping the running data within a preset sliding window to a potential energy field. The potential energy function reflects the similarity and changing trend between data points by considering the distance difference between historical data points and the current data point, and by using weighting coefficients and field strength attenuation factors. The size of the sliding window can be adjusted according to actual needs to ensure effective capture of the dynamic characteristics of device operation within a certain time range, providing an accurate representation of the potential energy field.

[0075] S402: Determine the dynamic threshold by combining the potential energy field and the potential energy field gradient.

[0076] The specific formula for calculating the dynamic threshold is as follows:

[0077] ;

[0078] ;

[0079] in, express t Real-time dynamic threshold This represents the mean gradient of the potential energy field. Represents the gradient of the potential energy field. Represents the potential energy curvature weight. Represents a symbolic function.

[0080] It should be noted that a dynamic threshold is calculated by combining the potential energy field and its gradient, and the threshold is adjusted to adapt to changes in the operating data. The dynamic threshold is jointly determined by the mean of the potential energy field gradient and the potential energy curvature weight. The potential energy curvature weight is adjusted according to the sign and magnitude of the potential energy field gradient, ensuring improved detection sensitivity when data changes drastically, while avoiding false alarms when data changes are small. This method can dynamically adjust the detection criteria according to actual conditions, improving the accuracy and adaptability of anomaly detection.

[0081] S403: Uses dynamic thresholds to parse running data and outputs abnormal data.

[0082] It's important to note that a potential energy field and dynamic threshold method are used to analyze operational data, ensuring accurate identification of anomalies. First, operational data within a specific time window is mapped to a potential energy field. The potential energy value for each data point is calculated using a potential energy function, reflecting the device's operating status. Then, by combining the gradient of the potential energy field, a dynamic threshold is calculated, adjusting the detection sensitivity in real time to adapt to changes in the device's operating environment. This process accurately identifies abnormal data during operation, avoiding misjudgments caused by static thresholds and improving the accuracy and stability of fault detection. Its advantage lies in dynamically adjusting the threshold, allowing the system to flexibly respond to changes in actual data, ensuring efficient and accurate anomaly detection.

[0083] In one possible implementation, S403 specifically includes:

[0084] If the potential energy field corresponding to the running data is greater than the dynamic threshold, and the gradient magnitude of the potential energy field gradient is greater than the maximum gradient magnitude within the preset sliding window size, the running data is determined to be abnormal data.

[0085] It should be noted that those skilled in the art can set the size of the preset sliding window according to actual needs, and this invention does not limit this.

[0086] The specific formula for judging abnormal data is as follows:

[0087] ;

[0088] in, This represents the preset field strength abrupt change coefficient. express t Running data at all times The potential energy value, This indicates taking the maximum value.

[0089] It should be noted that those skilled in the art can set the magnitude of the preset field strength mutation coefficient according to actual needs, and this invention does not limit this.

[0090] It should be noted that the system determines whether data is abnormal by checking if the potential energy field corresponding to the running data is greater than a dynamic threshold and if the gradient magnitude of the potential energy field exceeds the maximum value within a preset sliding window. Specifically, if the potential energy value is greater than the threshold and the gradient magnitude exceeds a preset proportion of the maximum gradient magnitude, the data is considered abnormal. This method can accurately identify abnormal situations with large fluctuations or sudden changes in data, thereby effectively improving the sensitivity and accuracy of fault detection.

[0091] S5: Based on the pre-trained model and the coupled weighted distance, abnormal data are fused and classified to determine the fault type.

[0092] It should be noted that fusing and classifying anomalous data through a pre-trained model and coupled weighted distance can integrate multiple data sources and consider their interrelationships, thereby more accurately locating the fault type. This method overcomes the misjudgment that may be caused by a single factor, ensuring more comprehensive and accurate fault type identification, and helps improve the system's fault diagnosis capabilities and response speed.

[0093] In one possible implementation, the pre-trained models include a BERT pre-trained model for extracting text-based anomalies, a ViT pre-trained model for extracting image-based anomalies, and a TST pre-trained model for extracting time-series anomalies.

[0094] In one possible implementation, S5 specifically includes:

[0095] S501: Extract feature vectors from historical anomaly data corresponding to different fault types using a pre-trained model, and construct a coupling relationship matrix.

[0096] Characterizes the nonlinear dependencies between anomalous data.

[0097] The coupling matrix is ​​as follows:

[0098] ;

[0099] in, This represents the Sigmoid function. Represents a learnable symmetric matrix. and They represent the first i The first historical anomaly data and the first j One historical anomaly data point, Indicates calculation and The coupling strength between them.

[0100] It should be noted that extracting feature vectors from historical anomaly data using a pre-trained model and constructing a coupling matrix can effectively characterize the nonlinear dependencies between anomaly data. Using the sigmoid function and a learnable symmetric matrix W allows the coupling strength between data points to be dynamically adjusted according to the characteristics of the actual data, thereby more accurately capturing the complex correlations between different fault types. This method avoids the limitations of traditional linear models and improves the accuracy and robustness of fault identification.

[0101] S502: Define the set of fault type prototypes.

[0102] The fault type prototype set refers to a set of feature vectors representing different fault modes. Each prototype vector represents a specific fault type and contains the characteristics of typical abnormal data under that fault type. By defining these prototype sets, they can be used as reference templates in subsequent abnormal data processing to determine whether the data to be detected conforms to a specific fault type. The definition of the prototype set helps the system more effectively match abnormal data with known fault modes, thereby improving the accuracy and reliability of fault diagnosis.

[0103] Optionally, the fault type prototype set can be generated by collecting historical data on different fault types, extracting typical features of each fault, and using clustering or classification algorithms to generate representative feature vectors for each type of fault. These prototype vectors are used to describe the core features of various fault modes, ensuring accurate identification of different types of faults during the detection process, thereby improving the accuracy and efficiency of fault diagnosis.

[0104] S503: Adjust the prototype vector in the fault type prototype set by using the coupling relationship matrix.

[0105] The specific formula adjustment is as follows:

[0106] ;

[0107] in, Indicates the first v The prototype vector of the fault type, Indicates the assignment operator. This represents the total number of historical outlier data samples. Indicates the first i The first historical anomaly data and the first v The coupling strength of historical anomaly data corresponding to the fault type. This represents the ReLU activation function.

[0108] It should be noted that adjusting the fault type prototype vector through the coupling relationship matrix can effectively fuse the correlation strength of similar abnormal data, making the prototype vector more accurately reflect the distribution of real fault modes. This adjustment overcomes the shortcomings of traditional clustering methods that ignore the interrelationships between data, improving the accuracy and robustness of fault type identification. Furthermore, by using the ReLU activation function, the model's adaptability to nonlinear features can be enhanced, further improving fault detection performance.

[0109] S504: Determine the fault type corresponding to the abnormal data based on the adjusted prototype vector.

[0110] It should be noted that a pre-trained model extracts features from historical anomaly data and constructs a coupling matrix to characterize the nonlinear dependencies between data points. The coupling strength between different anomaly data points is calculated using the sigmoid function and a symmetric matrix, thus capturing the interrelationships between data more accurately. The adjusted fault type prototype set, by fusing the correlation information of similar anomaly data, makes the prototype vectors closer to the distribution center of real fault modes. This method overcomes the limitation of traditional clustering algorithms that ignore the relationships between data, improves the accuracy and robustness of fault type identification, and ensures accurate classification of complex anomaly data.

[0111] In one possible implementation, S504 specifically includes:

[0112] S5041: Extract the feature vectors corresponding to the abnormal data.

[0113] S5042: Calculate the coupling weighted distance between the feature vector and the adjusted prototype vector.

[0114] S5043: Output the fault type corresponding to the minimum coupling weighted distance.

[0115] The specific formula for determining the fault type is as follows:

[0116] ;

[0117] in, Indicates the fault type label, This indicates the selection of the fault type that minimizes the weighted coupling distance. This represents the feature vector corresponding to the abnormal data. This represents the coupling strength coefficient.

[0118] It should be noted that this step determines the fault type by extracting the feature vectors of the abnormal data and calculating the weighted coupling distance between them and the adjusted prototype vector. By minimizing the weighted coupling distance, the system selects the fault type label that best matches the abnormal data. This method considers the distance between the feature vector and the prototype, as well as the coupling strength between data, and can more accurately reflect the relationship between fault types. By introducing the weighted coupling distance and coupling strength coefficient, the correlation between data can be better captured, improving the accuracy and robustness of fault identification and avoiding the biases of traditional methods.

[0119] S6: Provide early warnings based on fault type.

[0120] In practical applications, efficient and accurate fault diagnosis is achieved through multiple steps. First, the equipment model and environmental information are acquired, and test data is generated using the Lyapunov exponent to ensure comprehensiveness and accuracy. Next, potential energy fields and dynamic threshold methods are used to identify abnormal data, avoiding misjudgments caused by static thresholds. Then, a pre-trained model and coupled weighted distance analysis are used to accurately determine the fault type, overcoming the limitation of traditional methods that ignore the relationships between data. Finally, early warnings are issued based on the fault type, ensuring timely problem detection and improving system stability. The advantage of this method lies in its improved accuracy and robustness through dynamic adjustment and multi-dimensional analysis, ensuring stable equipment operation and reliability.

[0121] The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following:

[0122] In this embodiment of the invention, compared to traditional manual operation, this automated testing process acquires the model and environmental information of the terminal under test and generates test data based on this information, reducing errors caused by human intervention. Simultaneously, the analysis method based on Lyapunov exponent and dynamic thresholds enables the test to comprehensively cover various functions and extreme states of the equipment, thereby effectively identifying potential anomalies and faults. Furthermore, data fusion combining a pre-trained model and coupled weighted distance allows for accurate identification of fault types and timely early warning, ensuring the stability and safety of the power distribution system and improving the overall reliability of the power distribution terminal.

[0123] Reference manual attached Figure 2 The diagram shows a structural schematic of an automatic testing system for an intelligent power distribution automation terminal provided in an embodiment of the present invention.

[0124] This invention provides an automatic testing system 20 for intelligent power distribution automation terminals, comprising: a processor 201 and a memory 202;

[0125] The memory 202 stores programs or instructions that can run on the processor 201. When the program or instructions are executed by the processor 201, they implement the steps of the above-described automatic testing method for intelligent power distribution automation terminals and achieve the same technical effect. To avoid repetition, the present invention will not elaborate further.

[0126] It should be understood that the processor 201 in this embodiment of the invention may be a central processing unit (CPU), or it may be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor.

[0127] It should also be understood that the memory 202 in the embodiments of the present invention can be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. The non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. The volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of random access memory are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM), and direct memory bus RAM (DR RAM).

[0128] The above embodiments can be implemented, in whole or in part, by software, hardware (such as circuits), firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of the present invention are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. A semiconductor medium can be a solid-state drive.

[0129] It should be understood that, in various embodiments of the present invention, the order of the above-mentioned process numbers does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0130] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0131] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the devices, apparatuses, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0132] In the several embodiments provided by this invention, it should be understood that the disclosed devices, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0133] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0134] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0135] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, essentially, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0136] This invention provides a readable storage medium comprising: storing a program or instructions on the readable storage medium, wherein when the program or instructions are executed by a processor, the program or instructions implement the steps of the above-described automatic testing method for intelligent power distribution automation terminals and achieve the same technical effect. To avoid repetition, this invention will not elaborate further.

[0137] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the embodiments of the present invention, and are not intended to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the protection scope of the present invention.

Claims

1. A method for automatically testing an intelligent power distribution automation terminal, characterized in that, The method comprises the following steps: S1: obtaining model information and environment information of a terminal to be tested; S2: combining the model information and the environment information, and combining a Lyapunov index to generate test data of the terminal to be tested; S3: inputting the test data into the terminal to be tested, and outputting running data; S4: analyzing the running data by using a dynamic threshold based on a potential energy field, and outputting abnormal data; S5: fusing and classifying the abnormal data based on a pre-trained model and a coupling weighted distance, and determining a fault type; S6: performing early warning according to the fault type; The S2 specifically comprises the following steps: S201: establishing an environment information tensor; S202: combining the environment information tensor, and describing coupling effects of the environment information in a differential form; S203: combining the environment information tensor and a gradient operator of the environment information tensor, and constructing a test parameter generator; S204: generating candidate test data by using the test parameter generator; S205: screening the candidate test data by using the Lyapunov index, and outputting the test data; The S4 specifically comprises the following steps: S401: mapping the running data in a preset sliding window size into a potential energy field; S402: combining the potential energy field and a potential energy field gradient to determine the dynamic threshold; S403: analyzing the running data by using the dynamic threshold, and outputting the abnormal data.

2. The intelligent power distribution automation terminal automatic test method according to claim 1, characterized in that, The test data comprises electromagnetic voltage and current data, electronic voltage and current data, digital voltage and current data, and remote control data.

3. The intelligent power distribution automation terminal automatic test method of claim 1, wherein, The S205 specifically comprises the following steps: S2051: obtaining a maximum divergence rate threshold of test data of the terminal to be tested; S2052: calculating a Lyapunov index of the test data; S2053: determining whether the Lyapunov index is less than the maximum divergence rate threshold of the test data, if yes, outputting the candidate test data as the test data, otherwise, adjusting a material coefficient matrix of the terminal to be tested, and executing step S204.

4. The intelligent power distribution automation terminal automatic testing method of claim 1, wherein, The S403 specifically comprises the following steps: In a case where the potential energy field corresponding to the running data is greater than the dynamic threshold, and a gradient module length of the potential energy field gradient is greater than a maximum value of gradient module lengths in the preset sliding window size, the running data is determined as abnormal data.

5. The intelligent power distribution automation terminal automatic testing method of claim 1, wherein, The pre-trained model comprises a BERT pre-trained model for extracting text type abnormal data, a ViT pre-trained model for extracting image type abnormal data, and a TST pre-trained model for extracting time sequence type abnormal data.

6. The intelligent power distribution automation terminal automatic testing method of claim 1, wherein, The S5 specifically comprises the following steps: S501: constructing a coupling relationship matrix by extracting feature vectors of historical abnormal data corresponding to different fault types by using the pre-trained model; S502: defining a fault type prototype set; S503: adjusting a prototype vector in the fault type prototype set by using the coupling relationship matrix; S504: determining a fault type corresponding to the abnormal data based on the adjusted prototype vector.

7. The intelligent power distribution automation terminal automatic test method according to claim 6, wherein, The S504 specifically comprises the following steps: S5041: extracting a feature vector corresponding to the abnormal data; S5042: calculating a coupling weighted distance between the feature vector and the adjusted prototype vector; S5043: output the fault type corresponding to the minimum coupling weighted distance.

8. An automatic test system for intelligent power distribution automation terminal, characterized in that, Comprise: a processor and a memory; the memory stores programs or instructions executable on the processor, and the programs or instructions are executed by the processor to realize the steps of the intelligent power distribution automation terminal automatic test method in any one of claims 1 to 7.

Images