Polarity conversion method of vehicle-mounted extra-high voltage direct current withstand voltage test device

Abstract

The invention discloses a polarity conversion method of a vehicle-mounted extra-high voltage direct current withstand voltage test device. Initially measuring and collecting data through a preset low-voltage scheme, and predicting the voltage, current and leakage current of the equipment by using an LSTM model. And then, constructing a digital twinborn model, collecting data in real time in a high-voltage test stage, and importing the model to predict a power state so as to realize early warning analysis. Based on the early warning result, the test scheme is dynamically adjusted, the test scheme comprises voltage, time and a monitoring strategy, and finally an optimized target test scheme is generated, so that the accuracy and the safety of the power equipment test are improved.

Description

Technical Field

[0001] The present invention relates to the field of direct current testing and analysis, and more specifically to a polarity conversion method for a vehicle-mounted ultra-high voltage direct current withstand voltage test device. Background Art

[0002] With the large-scale deployment of DC transmission projects, the application of DC high-voltage generators is becoming more and more widespread. However, in the process of DC withstand voltage test, polarity conversion and early warning analysis are indispensable links, and their efficiency and accuracy directly affect the test results and equipment safety.

[0003] Traditional voltage withstand tests mainly rely on manual analysis, and often fail to take into account the impact of power transmission characteristics of different power equipment and the change characteristics of power test data on equipment voltage withstand warning. They lack the ability to perform simulation predictions in combination with real-time monitored power data, and the test schemes and parameters used are often relatively fixed, resulting in poor dynamic warning effects for different power equipment. At the same time, voltage withstand test warning analysis often relies on manual experience, and the practicability of warning analysis is poor, making it difficult to meet the testing needs of modern power systems. Summary of the invention

[0004] The invention overcomes the defects of the prior art and proposes a polarity conversion method for a vehicle-mounted ultra-high voltage direct current withstand voltage test device.

[0005] A first aspect of the present invention provides a polarity conversion method for a vehicle-mounted ultra-high voltage direct current withstand voltage test device, comprising:

[0006] Based on the vehicle-mounted UHV DC withstand voltage test device, the target power equipment is tested. In the first cycle, the test is performed based on the preset low-voltage solution, and different polarity conditions are tested through the automated polarity conversion method, and the first power test data of the target power equipment is collected;

[0007] Based on the time information of the first cycle, the first power test data is serialized to obtain serialized data, three LSTM-based prediction models are set based on the three dimensions of device voltage, device current, and leakage current, and the three prediction models are trained through the serialized data;

[0008] The three prediction models are combined to form a digital twin model. In the second cycle, the target power equipment is tested based on the preset high-voltage scheme. In multiple preset time periods in the second cycle, the polarity conversion is performed automatically, and the power data is collected in real time to obtain the second power test data.

[0009] In a plurality of preset time periods within the second cycle, based on a preset time period, data corresponding to the second power test data is imported into the digital twin model to perform power data prediction, obtain power prediction data, perform early warning analysis on the prediction data, and record the early warning time node and the early warning prediction data;

[0010] Through the early warning time nodes and early warning prediction data, the test plan of the target power equipment is regulated and analyzed in terms of test voltage, test time, and test monitoring, and a target test plan is generated.

[0011] In this scheme, the vehicle-mounted UHV DC withstand voltage test device tests the target power equipment. In the first cycle, the test is performed based on a preset low-voltage scheme, and different polarity conditions are tested through an automated polarity conversion method, and the first power test data of the target power equipment is collected, specifically:

[0012] Start the vehicle-mounted UHV DC withstand voltage test device to connect and test the target power equipment;

[0013] In the first cycle, the test is performed based on the preset low-voltage solution, different polarity conditions are tested through an automated polarity conversion method, and the first power test data of the target power equipment is collected in real time;

[0014] The first power test data includes equipment voltage, equipment current, and leakage current parameters during the test;

[0015] The preset low voltage scheme includes test voltage and current, polarity switching time, number of switching times and continuous test time before and after switching.

[0016] In this solution, based on the time information of the first cycle, the first power test data is serialized to obtain serialized data, and three LSTM-based prediction models are set based on the three dimensions of device voltage, device current, and leakage current. The three prediction models are trained through the serialized data, specifically:

[0017] Based on the time information of the first cycle, serializing the first power test data to obtain serialized data;

[0018] The serialized data includes device voltage, device current, and leakage current data;

[0019] Build three LSTM-based prediction models;

[0020] The serialized data is split based on the three dimensions of device voltage, device current, and leakage current, and imported into three prediction models for training respectively. The back propagation algorithm is used to perform feedback analysis and parameter tuning of the model prediction results.

[0021] In this solution, the three prediction models are combined to form a digital twin model. In the second cycle, the target power equipment is tested based on the preset high-voltage solution. In multiple preset time periods in the second cycle, automatic polarity conversion is performed, and power data is collected in real time to obtain the second power test data, specifically:

[0022] The data interfaces in the three prediction models are standardized by unified data format settings and interfaces, and unified communication protocols are set for data interaction to achieve model coupling and form a digital twin model with unified input and output;

[0023] The second cycle is divided into a plurality of time periods, and the target power equipment is tested based on a preset high voltage scheme;

[0024] During a plurality of preset time periods in the second cycle, automatic polarity conversion is performed, and power data is collected in real time to obtain second power test data.

[0025] In this solution, among the multiple preset time periods in the second cycle, based on a preset time period, the corresponding data of the second power test data is imported into the digital twin model for power data prediction, the power prediction data is obtained, the prediction data is analyzed for early warning, and the early warning time node and the early warning prediction data are recorded, specifically:

[0026] Selecting a current preset time period from among a plurality of preset time periods in the second cycle;

[0027] From the second power test data, obtain the power data corresponding to the current preset time period and import it into the digital twin model for data prediction and analysis, the prediction time is the next time period, and obtain the power data prediction;

[0028] Conduct equipment operation early warning analysis on the forecast data, and record the early warning time nodes and early warning forecast data;

[0029] Early warning analysis is performed for each time period, and all early warning time nodes and early warning forecast data are integrated and counted to obtain the early warning data set in the second cycle.

[0030] In this scheme, the test scheme of the target power equipment is regulated and analyzed in terms of test voltage, test time, and test monitoring through the warning time node and the warning prediction data, and a target test scheme is generated, specifically:

[0031] Obtain the warning time node and warning prediction data of a preset time period through the warning data set, perform warning range analysis on the equipment voltage, equipment current and leakage current based on the warning prediction data, and reset the test voltage and test current within the preset time period. Set the test duration of the test voltage and test current based on the warning level of the warning prediction data, and generate the withstand voltage test process of the time period based on the test voltage, test current and test duration;

[0032] Based on the early warning data set, early warning analysis is performed on multiple preset time periods of the entire second cycle, and multiple voltage withstand test processes are integrated to form a target test plan.

[0033] The second aspect of the present invention further provides a polarity conversion system for a vehicle-mounted ultra-high voltage DC withstand voltage test device, the system comprising: a memory, a processor, the memory comprising a polarity conversion program for the vehicle-mounted ultra-high voltage DC withstand voltage test device, and the polarity conversion program for the vehicle-mounted ultra-high voltage DC withstand voltage test device is executed by the processor to implement the following steps:

[0034] Based on the vehicle-mounted UHV DC withstand voltage test device, the target power equipment is tested. In the first cycle, the test is performed based on the preset low-voltage solution, and different polarity conditions are tested through the automated polarity conversion method, and the first power test data of the target power equipment is collected;

[0035] Based on the time information of the first cycle, the first power test data is serialized to obtain serialized data, three LSTM-based prediction models are set based on the three dimensions of device voltage, device current, and leakage current, and the three prediction models are trained through the serialized data;

[0036] The three prediction models are combined to form a digital twin model. In the second cycle, the target power equipment is tested based on the preset high-voltage scheme. In multiple preset time periods in the second cycle, the polarity conversion is performed automatically, and the power data is collected in real time to obtain the second power test data.

[0037] In a plurality of preset time periods within the second cycle, based on a preset time period, data corresponding to the second power test data is imported into the digital twin model to perform power data prediction, obtain power prediction data, perform early warning analysis on the prediction data, and record the early warning time node and the early warning prediction data;

[0038] Through the early warning time nodes and early warning prediction data, the test plan of the target power equipment is regulated and analyzed in terms of test voltage, test time, and test monitoring, and a target test plan is generated.

[0039] The third aspect of the present invention also provides a computer-readable storage medium, which includes a polarity conversion program for a vehicle-mounted ultra-high voltage DC withstand voltage test device. When the polarity conversion program for the vehicle-mounted ultra-high voltage DC withstand voltage test device is executed by a processor, the steps of the polarity conversion method for the vehicle-mounted ultra-high voltage DC withstand voltage test device as described in any one of the above items are implemented.

[0040] The present invention discloses a polarity conversion method for a vehicle-mounted ultra-high voltage direct current withstand voltage test device. The low-voltage scheme is preliminarily measured and data is collected, and the LSTM model is used to predict the voltage, current and leakage current of the equipment. Then, a digital twin model is constructed to collect data in real time during the high-voltage test phase, and the model is imported to predict the power state to achieve early warning analysis. Based on the early warning results, the test scheme is dynamically adjusted, including voltage, time and monitoring strategy, and finally an optimized target test scheme is generated to improve the accuracy and safety of power equipment testing. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 A flow chart showing a polarity conversion method of a vehicle-mounted ultra-high voltage direct current withstand voltage test device according to the present invention is shown;

[0042] Figure 2 A block diagram of a polarity conversion system of a vehicle-mounted ultra-high voltage direct current withstand voltage test device of the present invention is shown. DETAILED DESCRIPTION

[0043] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0044] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited to the specific embodiments disclosed below.

[0045] Figure 1 A flow chart of a polarity conversion method of a vehicle-mounted ultra-high voltage direct current withstand voltage test device of the present invention is shown.

[0046] like Figure 1 As shown, the first aspect of the present invention provides a polarity conversion method for a vehicle-mounted ultra-high voltage direct current withstand voltage test device, including automatic polarity conversion and corresponding test data analysis, prediction and equipment warning, specifically:

[0047] S102, testing the target power equipment based on the vehicle-mounted ultra-high voltage DC withstand voltage test device, testing based on a preset low-voltage solution in a first cycle, testing different polarity conditions through an automated polarity conversion method, and collecting first power test data of the target power equipment;

[0048] S104, based on the time information of the first cycle, serialize the first power test data to obtain serialized data, set three LSTM-based prediction models based on the three dimensions of device voltage, device current, and leakage current, and train the three prediction models through the serialized data;

[0049] S106, combining the three prediction models to form a digital twin model, testing the target power equipment based on a preset high-voltage scheme in a second cycle, performing automatic polarity conversion in multiple preset time periods in the second cycle, and collecting power data in real time to obtain second power test data;

[0050] S108, in a plurality of preset time periods within the second cycle, based on a preset time period, importing the data corresponding to the second power test data into the digital twin model to perform power data prediction, obtain power prediction data, perform early warning analysis on the prediction data, and record the early warning time node and the early warning prediction data;

[0051] S110, through the early warning time node and the early warning prediction data, the test voltage, test time, and test monitoring of the test plan of the target power equipment are regulated and analyzed, and a target test plan is generated.

[0052] It should be noted that the target power equipment is generally a cable equipment, and its insulation, stability and other power performance are tested by using an ultra-high voltage DC withstand voltage test device.

[0053] According to an embodiment of the present invention, the vehicle-mounted UHV DC withstand voltage test device tests the target power equipment. In the first cycle, the test is performed based on a preset low-voltage solution, and different polarity conditions are tested through an automated polarity conversion method, and the first power test data of the target power equipment is collected, specifically:

[0054] Start the vehicle-mounted UHV DC withstand voltage test device to connect and test the target power equipment;

[0055] In the first cycle, the test is performed based on the preset low-voltage solution, different polarity conditions are tested through an automated polarity conversion method, and the first power test data of the target power equipment is collected in real time;

[0056] The first power test data includes equipment voltage, equipment current, and leakage current parameters during the test;

[0057] The preset low voltage scheme includes test voltage and current, polarity switching time, number of switching times and continuous test time before and after switching.

[0058] It should be noted that the DC withstand voltage test device of the present invention is a vehicle-mounted ultra-high voltage test device, which can be moved on a vehicle and can be used for ultra-high voltage DC testing.

[0059] The DC withstand voltage test device includes:

[0060] DC high voltage generator: used to generate high voltage DC power supply;

[0061] Transformer: Increases the input AC voltage to a higher voltage level, and then rectifies it to obtain DC high voltage;

[0062] Measuring instruments: including voltmeter, ammeter, microammeter;

[0063] Protection resistor: limits the short-circuit current when the test product breaks down, used to protect the test device;

[0064] Automatic polarity conversion device: automatic polarity adjustment by adjusting the diode.

[0065] The automatic polarity conversion device can use an automatic system to adjust the polarity, and the polarity adjustment is achieved by adjusting the direction of the diode. Through the system of the present invention, based on the user setting, click the positive polarity output, the device automatically detects whether it is positive polarity, if it is, it directly outputs, if not, the system automatically adjusts the diode direction to become positive polarity, and then outputs.

[0066] In addition, the automatic polarity conversion device can control the time node of the test conversion and the duration of the test before and after the conversion through the system to ensure that the test process complies with the user's predetermined plan.

[0067] According to an embodiment of the present invention, based on the time information of the first cycle, the first power test data is serialized to obtain serialized data, and three LSTM-based prediction models are set based on the three dimensions of device voltage, device current, and leakage current. The three prediction models are trained through the serialized data, specifically:

[0068] Based on the time information of the first cycle, serializing the first power test data to obtain serialized data;

[0069] The serialized data includes device voltage, device current, and leakage current data;

[0070] Build three LSTM-based prediction models;

[0071] The serialized data is split based on the three dimensions of device voltage, device current, and leakage current, and imported into three prediction models for training respectively. The back propagation algorithm is used to perform feedback analysis and parameter tuning of the model prediction results.

[0072] It should be noted that the three LSTM-based prediction models correspond to prediction models of three equipment test parameters, namely, equipment voltage, equipment current, and leakage current. In the subsequent data coupling, the three models are combined to form a digital twin model.

[0073] According to an embodiment of the present invention, the three prediction models are combined to form a digital twin model. In the second cycle, the target power equipment is tested based on a preset high-voltage scheme. In multiple preset time periods in the second cycle, automatic polarity conversion is performed, and power data is collected in real time to obtain second power test data, specifically:

[0074] The data interfaces in the three prediction models are standardized by unified data format settings and interfaces, and unified communication protocols are set for data interaction to achieve model coupling and form a digital twin model with unified input and output;

[0075] The second cycle is divided into a plurality of time periods, and the target power equipment is tested based on a preset high voltage scheme;

[0076] During a plurality of preset time periods in the second cycle, automatic polarity conversion is performed, and power data is collected in real time to obtain second power test data.

[0077] It should be noted that the preset high-voltage scheme is a user-set scheme. The digital twin model is built based on three prediction models. Through the digital twin model, it is possible to analyze, display, and predict the corresponding data of the power equipment in real time based on the input data, and realize information monitoring and data analysis of the power equipment test. Multiple preset time periods are divided based on a certain time interval and are used for periodic analysis of power data. Each time period includes an automated polarity conversion process, which is used to analyze the withstand voltage of the power equipment before and after the polarity conversion. The second power test data is real-time acquisition data, including data collected and integrated in multiple time periods. A digital twin model with unified input and output, that is, only includes one data input interface and one data output interface.

[0078] In the first cycle, a preset low-voltage scheme is set to collect power operation data of power equipment for training prediction models and building data twin models. In the second cycle, a preset high-voltage scheme is set to conduct actual UHV withstand voltage tests, and data prediction and early warning analysis are performed based on the tests.

[0079] According to an embodiment of the present invention, among the multiple preset time periods within the second cycle, based on a preset time period, the data corresponding to the second power test data is imported into the digital twin model for power data prediction, power prediction data is obtained, and early warning analysis is performed on the prediction data, and the early warning time node and the early warning prediction data are recorded, specifically:

[0080] Selecting a current preset time period from among a plurality of preset time periods in the second cycle;

[0081] From the second power test data, obtain the power data corresponding to the current preset time period and import it into the digital twin model for data prediction and analysis, the prediction time is the next time period, and obtain the power data prediction;

[0082] Conduct equipment operation early warning analysis on the forecast data, and record the early warning time nodes and early warning forecast data;

[0083] Early warning analysis is performed for each time period, and all early warning time nodes and early warning forecast data are integrated and counted to obtain the early warning data set in the second cycle.

[0084] According to an embodiment of the present invention, the test scheme of the target power equipment is subjected to regulation and analysis of the test voltage, test time, and test monitoring through the warning time node and the warning prediction data, and a target test scheme is generated, specifically:

[0085] Obtain the warning time node and warning prediction data of a preset time period through the warning data set, perform warning range analysis on the equipment voltage, equipment current and leakage current based on the warning prediction data, and reset the test voltage and test current within the preset time period. Set the test duration of the test voltage and test current based on the warning level of the warning prediction data, and generate the withstand voltage test process of the time period based on the test voltage, test current and test duration;

[0086] Based on the early warning data set, early warning analysis is performed on multiple preset time periods of the entire second cycle, and multiple voltage withstand test processes are integrated to form a target test plan.

[0087] It is worth mentioning that in the traditional voltage withstand test process, the power transmission characteristics of different power equipment and the impact of the change characteristics of power test data on the equipment voltage withstand warning are often not taken into account. There is a lack of the ability to conduct simulation predictions in combination with real-time monitored power data. The test schemes and parameters used are often relatively fixed, resulting in poor dynamic warning effects for different power equipment. At the same time, voltage withstand test warning analysis often relies on manual experience, and the practicability of warning analysis is poor.

[0088] Based on this, the present invention builds LSTM-based prediction models for the three dimensions of equipment voltage, equipment current, and leakage current in the test process of the power equipment. During the test, the corresponding power data of the equipment is converted into serialized data. The conversion process is based on the time information dimension, and the serialized data of the first week is used to learn and train the power data trend and change law of the power equipment during the test process. At the same time, the three models are data coupled to form a data twin model, which effectively realizes the information monitoring and prediction of the power equipment, improves the information analysis capability of the withstand voltage test, and reduces the situation of manpower participation in the test analysis.

[0089] Further, in an embodiment of the present invention, a high-voltage withstand voltage test is performed in the second cycle through a digital twin model. In the high-voltage withstand voltage test, data of power equipment in different time periods are collected for prediction, and early warning analysis is performed based on the real-time prediction results of the data in the current time period. The early warning analysis includes comparing the predicted data with the power data in the normal state. If it exceeds the warning range, the early warning information is recorded to obtain the early warning time node and the early warning prediction data. Through the early warning time node and the early warning prediction data, the test plan for the target power equipment can be reset to improve the test efficiency and the accuracy of the test. For example, in a certain time period in the second cycle, the predicted data of the next time period of the target power equipment under the preset voltage and current conditions is predicted, and an early warning analysis is performed based on the predicted data to determine whether there is a situation exceeding the warning range. If so, the early warning information is recorded. Through the early warning information, the test plan in the current time period can be further adjusted, such as increasing the voltage or adjusting the duration, adjusting the number of polarity conversions and the time node, etc. The adjustment of the test plan is aimed at the predicted data in the current time period, and the target test plan based on the complete cycle is formed through the analysis of each predicted time period. Through the predictive analysis and target test scheme of the present invention, it is possible to effectively discover the fault warnings and equipment safety hazards hidden in the power test data, and to achieve accurate fault warnings and accurate test evaluation of leakage current for power equipment.

[0090] This target test plan can be applied to the test evaluation of the third cycle, and conduct targeted early warning analysis on the target power equipment, reduce the involvement of human experience, and improve information analysis capabilities.

[0091] It should be noted that the target test plan includes the test set voltage and current for the target power equipment, the test duration corresponding to the set voltage and current, etc.

[0092] According to an embodiment of the present invention, it also includes:

[0093] After the second cycle of the experiment is completed, the digital twin model is used to obtain the forecast data for each preset time period;

[0094] Compare the predicted data of each preset time period with the corresponding data segment in the second power test data, and calculate the data difference based on the three dimensions of equipment voltage, equipment current, and leakage current;

[0095] The data difference calculation is specifically to calculate the difference between the predicted value and the actual value based on the data of the three dimensions to obtain three difference values, and to perform weighted average of the three difference values ​​based on the preset weights to obtain the deviation value;

[0096] Compare and analyze the deviation value with the threshold, and select the preset time period exceeding the threshold and mark it as the second warning time node;

[0097] Based on the preset high-voltage scheme and combined with the second warning time node, fault diagnosis, pressure resistance safety and pressure resistance stability analysis are performed on the target.

[0098] It should be noted that in the process of predictive analysis, there is often a certain deviation between the predicted value and the actual value. When there is a deviation, it means that the test process corresponding to the time node has certain safety hazards or faults. The embodiment of the present invention analyzes the deviation of the predicted data, records the time nodes where the prediction deviation exists, and conducts test feedback analysis in combination with the preset high-voltage scheme. Through predictive analysis, it is possible to detect faults and safety hazards.

[0099] Deviation calculation can be adjusted based on preset weights, and can dynamically adjust the emphasis of test analysis results based on different power test data to screen out early warning points.

[0100] The deviation value calculation formula is:

[0101] D S =V1×K1+A1×K2+A2×K3

[0102] Among them, K1, K2, K3 are preset weights, V1 is the device voltage difference, A1 is the device current difference, A2 is the leakage current difference, D S is the deviation value.

[0103] Figure 2 A block diagram of a polarity conversion system of a vehicle-mounted ultra-high voltage direct current withstand voltage test device of the present invention is shown.

[0104] The second aspect of the present invention further provides a polarity conversion system 2 for a vehicle-mounted ultra-high voltage DC withstand voltage test device, the system comprising: a memory 21, a processor 22, the memory comprising a polarity conversion program for the vehicle-mounted ultra-high voltage DC withstand voltage test device, and the polarity conversion program for the vehicle-mounted ultra-high voltage DC withstand voltage test device is executed by the processor to implement the following steps:

[0105] Based on the vehicle-mounted UHV DC withstand voltage test device, the target power equipment is tested. In the first cycle, the test is performed based on the preset low-voltage solution, and different polarity conditions are tested through the automated polarity conversion method, and the first power test data of the target power equipment is collected;

[0106] Based on the time information of the first cycle, the first power test data is serialized to obtain serialized data, three LSTM-based prediction models are set based on the three dimensions of device voltage, device current, and leakage current, and the three prediction models are trained through the serialized data;

[0107] The three prediction models are combined to form a digital twin model. In the second cycle, the target power equipment is tested based on the preset high-voltage scheme. In multiple preset time periods in the second cycle, the polarity conversion is performed automatically, and the power data is collected in real time to obtain the second power test data.

[0108] In a plurality of preset time periods within the second cycle, based on a preset time period, data corresponding to the second power test data is imported into the digital twin model to perform power data prediction, obtain power prediction data, perform early warning analysis on the prediction data, and record the early warning time node and the early warning prediction data;

[0109] Through the early warning time nodes and early warning prediction data, the test plan of the target power equipment is regulated and analyzed in terms of test voltage, test time, and test monitoring, and a target test plan is generated.

[0110] It should be noted that the target power equipment is generally a cable equipment, and its insulation, stability and other power performance are tested by using an ultra-high voltage DC withstand voltage test device.

[0111] According to an embodiment of the present invention, the vehicle-mounted UHV DC withstand voltage test device tests the target power equipment. In the first cycle, the test is performed based on a preset low-voltage solution, and different polarity conditions are tested through an automated polarity conversion method, and the first power test data of the target power equipment is collected, specifically:

[0112] Start the vehicle-mounted UHV DC withstand voltage test device to connect and test the target power equipment;

[0113] In the first cycle, the test is performed based on the preset low-voltage solution, different polarity conditions are tested through an automated polarity conversion method, and the first power test data of the target power equipment is collected in real time;

[0114] The first power test data includes equipment voltage, equipment current, and leakage current parameters during the test;

[0115] The preset low voltage scheme includes test voltage and current, polarity switching time, number of switching times and continuous test time before and after switching.

[0116] It should be noted that the DC withstand voltage test device of the present invention is a vehicle-mounted ultra-high voltage test device, which can be moved on a vehicle and can be used for ultra-high voltage DC testing.

[0117] The DC withstand voltage test device includes:

[0118] DC high voltage generator: used to generate high voltage DC power supply;

[0119] Transformer: Increases the input AC voltage to a higher voltage level, and then rectifies it to obtain DC high voltage;

[0120] Measuring instruments: including voltmeter, ammeter, microammeter;

[0121] Protection resistor: limits the short-circuit current when the test product breaks down, used to protect the test device;

[0122] Automatic polarity conversion device: automatic polarity adjustment by adjusting the diode.

[0123] The automatic polarity conversion device can use an automatic system to adjust the polarity, and the polarity adjustment is achieved by adjusting the direction of the diode. Through the system of the present invention, based on the user setting, click the positive polarity output, the device automatically detects whether it is positive polarity, if it is, it directly outputs, if not, the system automatically adjusts the diode direction to become positive polarity, and then outputs.

[0124] In addition, the automatic polarity conversion device can control the time node of the test conversion and the duration of the test before and after the conversion through the system to ensure that the test process complies with the user's predetermined plan.

[0125] According to an embodiment of the present invention, based on the time information of the first cycle, the first power test data is serialized to obtain serialized data, and three LSTM-based prediction models are set based on the three dimensions of device voltage, device current, and leakage current. The three prediction models are trained through the serialized data, specifically:

[0126] Based on the time information of the first cycle, serializing the first power test data to obtain serialized data;

[0127] The serialized data includes device voltage, device current, and leakage current data;

[0128] Build three LSTM-based prediction models;

[0129] The serialized data is split based on the three dimensions of device voltage, device current, and leakage current, and imported into three prediction models for training respectively. The back propagation algorithm is used to perform feedback analysis and parameter tuning of the model prediction results.

[0130] It should be noted that the three LSTM-based prediction models correspond to prediction models of three equipment test parameters, namely, equipment voltage, equipment current, and leakage current. In the subsequent data coupling, the three models are combined to form a digital twin model.

[0131] According to an embodiment of the present invention, the three prediction models are combined to form a digital twin model. In the second cycle, the target power equipment is tested based on a preset high-voltage scheme. In multiple preset time periods in the second cycle, automatic polarity conversion is performed, and power data is collected in real time to obtain second power test data, specifically:

[0132] The data interfaces in the three prediction models are standardized by unified data format settings and interfaces, and unified communication protocols are set for data interaction to achieve model coupling and form a digital twin model with unified input and output;

[0133] The second cycle is divided into a plurality of time periods, and the target power equipment is tested based on a preset high voltage scheme;

[0134] During a plurality of preset time periods in the second cycle, automatic polarity conversion is performed, and power data is collected in real time to obtain second power test data.

[0135] It should be noted that the preset high-voltage scheme is a user-set scheme. The digital twin model is built based on three prediction models. Through the digital twin model, it is possible to analyze, display, and predict the corresponding data of the power equipment in real time based on the input data, and realize information monitoring and data analysis of the power equipment test. Multiple preset time periods are divided based on a certain time interval and are used for periodic analysis of power data. Each time period includes an automated polarity conversion process, which is used to analyze the withstand voltage of the power equipment before and after the polarity conversion. The second power test data is real-time acquisition data, including data collected and integrated in multiple time periods. A digital twin model with unified input and output, that is, only includes one data input interface and one data output interface.

[0136] In the first cycle, a preset low-voltage scheme is set to collect power operation data of power equipment for training prediction models and building data twin models. In the second cycle, a preset high-voltage scheme is set to conduct actual UHV withstand voltage tests, and data prediction and early warning analysis are performed based on the tests.

[0137] According to an embodiment of the present invention, among the multiple preset time periods within the second cycle, based on a preset time period, the data corresponding to the second power test data is imported into the digital twin model for power data prediction, power prediction data is obtained, and early warning analysis is performed on the prediction data, and the early warning time node and the early warning prediction data are recorded, specifically:

[0138] Selecting a current preset time period from among a plurality of preset time periods in the second cycle;

[0139] From the second power test data, obtain the power data corresponding to the current preset time period and import it into the digital twin model for data prediction and analysis, the prediction time is the next time period, and obtain the power data prediction;

[0140] Conduct equipment operation early warning analysis on the forecast data, and record the early warning time nodes and early warning forecast data;

[0141] Early warning analysis is performed for each time period, and all early warning time nodes and early warning forecast data are integrated and counted to obtain the early warning data set in the second cycle.

[0142] According to an embodiment of the present invention, the test scheme of the target power equipment is subjected to regulation and analysis of the test voltage, test time, and test monitoring through the warning time node and the warning prediction data, and a target test scheme is generated, specifically:

[0143] Obtain the warning time node and warning prediction data of a preset time period through the warning data set, perform warning range analysis on the equipment voltage, equipment current and leakage current based on the warning prediction data, and reset the test voltage and test current within the preset time period. Set the test duration of the test voltage and test current based on the warning level of the warning prediction data, and generate the withstand voltage test process of the time period based on the test voltage, test current and test duration;

[0144] Based on the early warning data set, early warning analysis is performed on multiple preset time periods of the entire second cycle, and multiple voltage withstand test processes are integrated to form a target test plan.

[0145] It is worth mentioning that in the traditional voltage withstand test process, the power transmission characteristics of different power equipment and the impact of the change characteristics of power test data on the equipment voltage withstand warning are often not taken into account. There is a lack of the ability to conduct simulation predictions in combination with real-time monitored power data. The test schemes and parameters used are often relatively fixed, resulting in poor dynamic warning effects for different power equipment. At the same time, voltage withstand test warning analysis often relies on manual experience, and the practicability of warning analysis is poor.

[0146] Based on this, the present invention builds LSTM-based prediction models for the three dimensions of equipment voltage, equipment current, and leakage current in the test process of the power equipment. During the test, the corresponding power data of the equipment is converted into serialized data. The conversion process is based on the time information dimension, and the serialized data of the first week is used to learn and train the power data trend and change law of the power equipment during the test process. At the same time, the three models are data coupled to form a data twin model, which effectively realizes the information monitoring and prediction of the power equipment, improves the information analysis capability of the withstand voltage test, and reduces the situation of manpower participation in the test analysis.

[0147] Further, in an embodiment of the present invention, a high-voltage withstand voltage test is performed in the second cycle through a digital twin model. In the high-voltage withstand voltage test, data of power equipment in different time periods are collected for prediction, and early warning analysis is performed based on the real-time prediction results of the data in the current time period. The early warning analysis includes comparing the predicted data with the power data in the normal state. If it exceeds the warning range, the early warning information is recorded to obtain the early warning time node and the early warning prediction data. Through the early warning time node and the early warning prediction data, the test plan for the target power equipment can be reset to improve the test efficiency and the accuracy of the test. For example, in a certain time period in the second cycle, the predicted data of the next time period of the target power equipment under the preset voltage and current conditions is predicted, and an early warning analysis is performed based on the predicted data to determine whether there is a situation exceeding the warning range. If so, the early warning information is recorded. Through the early warning information, the test plan in the current time period can be further adjusted, such as increasing the voltage or adjusting the duration, adjusting the number of polarity conversions and the time node, etc. The adjustment of the test plan is aimed at the predicted data in the current time period, and the target test plan based on the complete cycle is formed through the analysis of each predicted time period. Through the predictive analysis and target test scheme of the present invention, it is possible to effectively discover the fault warnings and equipment safety hazards hidden in the power test data, and to achieve accurate fault warnings and accurate test evaluation of leakage current for power equipment.

[0148] This target test plan can be applied to the test evaluation of the third cycle, and conduct targeted early warning analysis on the target power equipment, reduce the involvement of human experience, and improve information analysis capabilities.

[0149] It should be noted that the target test plan includes the test set voltage and current for the target power equipment, the test duration corresponding to the set voltage and current, etc.

[0150] The third aspect of the present invention also provides a computer-readable storage medium, which includes a polarity conversion program for a vehicle-mounted ultra-high voltage DC withstand voltage test device. When the polarity conversion program for the vehicle-mounted ultra-high voltage DC withstand voltage test device is executed by a processor, the steps of the polarity conversion method for the vehicle-mounted ultra-high voltage DC withstand voltage test device as described in any one of the above items are implemented.

[0151] The present invention discloses a polarity conversion method for a vehicle-mounted ultra-high voltage direct current withstand voltage test device. The low-voltage scheme is preliminarily measured and data is collected, and the LSTM model is used to predict the voltage, current and leakage current of the equipment. Then, a digital twin model is constructed to collect data in real time during the high-voltage test phase, and the model is imported to predict the power state to achieve early warning analysis. Based on the early warning results, the test scheme is dynamically adjusted, including voltage, time and monitoring strategy, and finally an optimized target test scheme is generated to improve the accuracy and safety of power equipment testing.

[0152] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.

[0153] The units described above as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units; they may be located in one place or distributed on multiple network units; some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0154] In addition, all functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may be separately used as a unit, or two or more units may be integrated into one unit; the above-mentioned integrated units may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0155] Those skilled in the art can understand that: all or part of the steps of implementing the above method embodiments can be completed by hardware related to program instructions, and the aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps of the above method embodiments; and the aforementioned storage medium includes: mobile storage devices, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), disks or optical disks, and other media that can store program codes.

[0156] Alternatively, if the above-mentioned integrated unit of the present invention is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present invention can be essentially or partly reflected in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as mobile storage devices, ROM, RAM, magnetic disks or optical disks.

[0157] The above description is only a specific implementation mode of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A polarity conversion method for a vehicle-mounted ultra-high voltage direct current withstand voltage test device, characterized in that: include: Based on the vehicle-mounted UHV DC withstand voltage test device, the target power equipment is tested. In the first cycle, the test is performed based on the preset low-voltage solution, and different polarity conditions are tested through the automated polarity conversion method, and the first power test data of the target power equipment is collected; Based on the time information of the first cycle, the first power test data is serialized to obtain serialized data, three LSTM-based prediction models are set based on the three dimensions of device voltage, device current, and leakage current, and the three prediction models are trained through the serialized data; The three prediction models are combined to form a digital twin model. In the second cycle, the target power equipment is tested based on the preset high-voltage scheme. In multiple preset time periods in the second cycle, the polarity conversion is performed automatically, and the power data is collected in real time to obtain the second power test data. In a plurality of preset time periods within the second cycle, based on a preset time period, data corresponding to the second power test data is imported into the digital twin model to perform power data prediction, obtain power prediction data, perform early warning analysis on the prediction data, and record the early warning time node and the early warning prediction data; Through the early warning time nodes and early warning prediction data, the test plan of the target power equipment is regulated and analyzed in terms of test voltage, test time, and test monitoring, and a target test plan is generated.

2. The polarity conversion method of a vehicle-mounted ultra-high voltage direct current withstand voltage test device according to claim 1, characterized in that: The vehicle-mounted UHV DC withstand voltage test device tests the target power equipment. In the first cycle, the test is performed based on a preset low-voltage solution, and different polarity conditions are tested through an automated polarity conversion method. The first power test data of the target power equipment is collected, specifically: Start the vehicle-mounted UHV DC withstand voltage test device to connect and test the target power equipment; In the first cycle, the test is performed based on the preset low-voltage solution, different polarity conditions are tested through an automated polarity conversion method, and the first power test data of the target power equipment is collected in real time; The first power test data includes equipment voltage, equipment current, and leakage current parameters during the test; The preset low voltage scheme includes test voltage and current, polarity switching time, number of switching times and continuous test time before and after switching.

3. The polarity conversion method of a vehicle-mounted ultra-high voltage direct current withstand voltage test device according to claim 2, characterized in that: Based on the time information of the first cycle, the first power test data is serialized to obtain serialized data, three LSTM-based prediction models are set based on the three dimensions of device voltage, device current, and leakage current, and the three prediction models are trained through the serialized data, specifically: Based on the time information of the first cycle, serializing the first power test data to obtain serialized data; The serialized data includes device voltage, device current, and leakage current data; Build three LSTM-based prediction models; The serialized data is split based on the three dimensions of device voltage, device current, and leakage current, and imported into three prediction models for training respectively. The back propagation algorithm is used to perform feedback analysis and parameter tuning of the model prediction results.

4. The polarity conversion method of a vehicle-mounted ultra-high voltage direct current withstand voltage test device according to claim 3, characterized in that: The three prediction models are combined to form a digital twin model. In the second cycle, the target power equipment is tested based on the preset high-voltage scheme. In multiple preset time periods in the second cycle, the polarity conversion is performed automatically, and the power data is collected in real time to obtain the second power test data, which is specifically: The data interfaces in the three prediction models are standardized by unified data format settings and interfaces, and unified communication protocols are set for data interaction to achieve model coupling and form a digital twin model with unified input and output; The second cycle is divided into a plurality of time periods, and the target power equipment is tested based on a preset high voltage scheme; During a plurality of preset time periods in the second cycle, automatic polarity conversion is performed, and power data is collected in real time to obtain second power test data.

5. The polarity conversion method of a vehicle-mounted UHV DC withstand voltage test device according to claim 4, characterized in that: In the multiple preset time periods within the second cycle, based on a preset time period, the data corresponding to the second power test data is imported into the digital twin model to perform power data prediction, obtain power prediction data, perform early warning analysis on the prediction data, and record the early warning time node and the early warning prediction data, specifically: Selecting a current preset time period from among a plurality of preset time periods in the second cycle; From the second power test data, obtain the power data corresponding to the current preset time period and import it into the digital twin model for data prediction and analysis, the prediction time is the next time period, and obtain the power data prediction; Conduct equipment operation early warning analysis on the forecast data, and record the early warning time nodes and early warning forecast data; Perform early warning analysis for each time period, and integrate and count all early warning time nodes and early warning forecast data to obtain the early warning data set in the second cycle.

6. The polarity conversion method of a vehicle-mounted UHV DC withstand voltage test device according to claim 5, characterized in that: The test scheme of the target power equipment is regulated and analyzed in terms of test voltage, test time, and test monitoring through the early warning time node and the early warning prediction data, and a target test scheme is generated, specifically: Obtain the warning time node and warning prediction data of a preset time period through the warning data set, perform warning range analysis on the equipment voltage, equipment current and leakage current based on the warning prediction data, and reset the test voltage and test current within the preset time period. Set the test duration of the test voltage and test current based on the warning level of the warning prediction data, and generate the withstand voltage test process of the time period based on the test voltage, test current and test duration; Based on the early warning data set, early warning analysis is performed on multiple preset time periods of the entire second cycle, and multiple voltage withstand test processes are integrated to form a target test plan.

7. A polarity conversion system for a vehicle-mounted ultra-high voltage direct current withstand voltage test device, characterized in that: The system includes: a memory and a processor, wherein the memory includes a polarity conversion program of a vehicle-mounted ultra-high voltage direct current withstand voltage test device, and when the polarity conversion program of the vehicle-mounted ultra-high voltage direct current withstand voltage test device is executed by the processor, the following steps are implemented: Based on the vehicle-mounted UHV DC withstand voltage test device, the target power equipment is tested. In the first cycle, the test is performed based on the preset low-voltage solution, and different polarity conditions are tested through the automated polarity conversion method, and the first power test data of the target power equipment is collected; Based on the time information of the first cycle, the first power test data is serialized to obtain serialized data, three LSTM-based prediction models are set based on the three dimensions of device voltage, device current, and leakage current, and the three prediction models are trained through the serialized data; The three prediction models are combined to form a digital twin model. In the second cycle, the target power equipment is tested based on the preset high-voltage scheme. In multiple preset time periods in the second cycle, the polarity conversion is performed automatically, and the power data is collected in real time to obtain the second power test data. In a plurality of preset time periods within the second cycle, based on a preset time period, data corresponding to the second power test data is imported into the digital twin model to perform power data prediction, obtain power prediction data, perform early warning analysis on the prediction data, and record the early warning time node and the early warning prediction data; Through the early warning time nodes and early warning prediction data, the test plan of the target power equipment is regulated and analyzed in terms of test voltage, test time, and test monitoring, and a target test plan is generated.

8. The polarity conversion system of a vehicle-mounted UHV DC withstand voltage test device according to claim 7, characterized in that: The vehicle-mounted UHV DC withstand voltage test device tests the target power equipment. In the first cycle, the test is performed based on a preset low-voltage solution, and different polarity conditions are tested through an automated polarity conversion method. The first power test data of the target power equipment is collected, specifically: Start the vehicle-mounted UHV DC withstand voltage test device to connect and test the target power equipment; In the first cycle, the test is performed based on the preset low-voltage solution, different polarity conditions are tested through an automated polarity conversion method, and the first power test data of the target power equipment is collected in real time; The first power test data includes equipment voltage, equipment current, and leakage current parameters during the test; The preset low voltage scheme includes test voltage and current, polarity switching time, number of switching times and continuous test time before and after switching.

9. The polarity conversion system of a vehicle-mounted UHV DC withstand voltage test device according to claim 7, characterized in that: Based on the time information of the first cycle, the first power test data is serialized to obtain serialized data, three LSTM-based prediction models are set based on the three dimensions of device voltage, device current, and leakage current, and the three prediction models are trained through the serialized data, specifically: Based on the time information of the first cycle, serializing the first power test data to obtain serialized data; The serialized data includes device voltage, device current, and leakage current data; Build three LSTM-based prediction models; The serialized data is split based on the three dimensions of device voltage, device current, and leakage current, and imported into three prediction models for training respectively. The back propagation algorithm is used to perform feedback analysis and parameter tuning of the model prediction results.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a polarity conversion program for a vehicle-mounted ultra-high voltage DC withstand voltage test device. When the polarity conversion program for the vehicle-mounted ultra-high voltage DC withstand voltage test device is executed by a processor, the steps of the polarity conversion method for a vehicle-mounted ultra-high voltage DC withstand voltage test device as described in any one of claims 1 to 6 are implemented.

Images