A thyristor level circuit testing method
By installing multiple sensors in the thyristor-level circuit and establishing an intelligent sensor network, combining signal processing and machine learning models, and dynamically adjusting the test strategy, the problems of large environmental factors and slow response speed in traditional testing methods are solved, and high-precision and intelligent testing results are achieved.
Patent Information
- Application Number
- CN202411824079.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-12
- Publication Date
- 2025-09-30
- Estimated Expiration
- 2044-12-12
AI Technical Summary
Traditional thyristor-level circuit testing methods fail to fully consider the impact of environmental factors, resulting in low measurement accuracy, slow response speed, difficulty in identifying complex abnormal patterns, and inability to achieve intelligent diagnosis.
By installing multiple sensors at key locations in the thyristor-level circuit, establishing an intelligent sensing network, integrating environmental perception modules, using signal processing algorithms and machine learning models for data diagnosis, and dynamically adjusting measurement models and test strategies to generate test reports.
It realizes real-time monitoring and intelligent judgment of the thyristor working status, improves test accuracy and response speed, enhances fault early warning capability, and ensures the credibility and long-term stability of measurement results.
Smart Images
Figure CN119535146B_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of thyristor level loop testing, in particular to a thyristor level loop testing method. Background Art
[0002] Thyristors are semiconductor devices widely used in power electronics to control and regulate AC or DC power. Thyristor-level loop testing involves testing a circuit containing one or more thyristors to ensure that its performance meets expectations and to promptly detect potential faults. However, traditional testing methods often rely on fixed measurement parameters and fail to fully consider the impact of environmental factors on measurement results, resulting in low measurement accuracy. Existing systems also respond slowly to environmental changes and cannot quickly adjust to optimal operating conditions, affecting the reliability and consistency of test results. Existing testing systems also often use static thresholds to determine faults, making it difficult to identify complex abnormal patterns and unable to implement intelligent diagnosis. Summary of the Invention
[0003] The purpose of this section is to summarize some aspects of the embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section and the abstract and title of this application to avoid obscuring the purpose of this section, the abstract and the title of the invention, and such simplifications or omissions should not be used to limit the scope of the present invention.
[0004] To achieve the above object, the present invention provides the following technical solutions:
[0005] A thyristor-level circuit testing method comprises: setting initial test conditions, selecting original parameters of a thyristor and circuit under test, and constructing a measurement model based on the original thyristor parameters; after initialization, installing multiple sensors at key locations in the thyristor-level circuit, connecting the multiple sensor nodes to a central control unit via wireless communication technology, and integrating an environmental perception module to establish an intelligent sensor network; collecting data in the intelligent sensor network to obtain a data set, applying a signal processing algorithm to extract data feature quantities, diagnosing the data using a machine learning model to obtain a data diagnosis result, and executing an early warning strategy based on the diagnosis result; calibrating the multiple sensors using a standard source based on the collected data set, inputting current operating conditions into the measurement model, calculating theoretical measurement values, collecting data as actual measurement values via the intelligent sensor network, and evaluating the theoretical and actual measurement values to obtain an evaluation result; dynamically adjusting measurement model parameters based on the evaluation result to obtain an updated measurement model; and dynamically adjusting the test strategy based on temperature and humidity changes and electromagnetic interference information provided by the environmental perception module; integrating the diagnosis result, evaluation result, and data set, and using a statistical analysis method to evaluate the overall health of the thyristor and its circuit to generate a test report.
[0006] As a further solution of the present invention, the initial test conditions are set, and original parameters of the thyristor and circuit under test are selected, and a measurement model is constructed based on the original thyristor parameters. The specific steps are: starting the test system and executing the self-test program of the hardware components to ensure that all devices are working properly; setting the sampling frequency and initial test conditions of the data transmission protocol based on the original thyristor parameters; obtaining the rated voltage, current, and temperature range of the thyristor and inputting them into the test system as a reference; and constructing the measurement model based on the original thyristor parameters, the expression being:
[0007] ;
[0008] in, is the minimum current required for the thyristor to start conducting, is the minimum voltage required for the thyristor to start conducting, is the resistance value of the thyristor in the on state, is the temperature of the thyristor when it is working, is a constant, is the difference between the current time and the start time, Indicates that the thyristor is in ideal working condition.
[0009] As a further solution of the present invention: after the initialization is completed, multiple sensors are installed at key positions of the thyristor-level loop, the multiple sensor nodes are connected to the central control unit through wireless communication technology, and an environmental perception module is integrated to establish an intelligent sensing network. The specific steps are: determining the monitoring position according to the initial parameters of the thyristor and its loop; installing multiple sensors at the determined key positions, and packaging the multiple sensors into independent nodes, each node has data acquisition, preprocessing and wireless communication functions, and a microcontroller is integrated inside the node to perform preliminary data processing tasks; using the wireless communication technology ZigBee, a stable communication link is established between the multi-sensor node and the central control unit; integrating the environmental perception module in the intelligent sensing network to monitor environmental changes; after completing the installation and configuration of the multi-sensor nodes, the central control unit uniformly manages each node to form an intelligent sensing network covering the entire thyristor-level loop; starting the intelligent sensing network, performing preliminary data acquisition and transmission tests, and verifying the communication quality and data accuracy between each node.
[0010] As a further solution of the present invention: the data in the intelligent sensor network is collected to obtain a data set, a signal processing algorithm is applied to extract the characteristic quantity of the data, and the data is diagnosed through a machine learning model to obtain a data diagnosis result, and an early warning strategy is executed based on the diagnosis result. The specific steps are: starting a central control unit, sending instructions to multiple sensor nodes to collect multi-sensor data; pre-processing the collected raw data by filtering and normalization; and analyzing the pre-processed data by applying a signal processing algorithm to extract a characteristic quantity reflecting the working state of the thyristor. The expression is:
[0011] ;
[0012] in, is the weighted average and normalized feature quantity, Reflects the importance of each feature; The data points are obtained from different sensors. Based on the collected historical data and the theoretical calculation values generated by the measurement model, the machine learning model is trained and the extracted feature quantities are input into the machine learning model. The machine learning model evaluates whether the current feature quantities are normal or abnormal based on the pattern recognition ability learned during training and outputs the diagnosis result. The expression is:
[0013] ;
[0014] in, Used to evaluate the working status of thyristors, is the activation function, reflects the importance of each feature function, is the comprehensive eigenvalue A function for nonlinear transformation; setting the threshold L, based on the diagnosis results ,When the diagnosis index exceeds the threshold L, the alarm mechanism is triggered, ,and protection actions are automatically taken and the power is cut off.
[0015] As a further solution of the present invention: according to the collected data set, the multi-sensor is calibrated using a standard source, the current operating conditions are input into the measurement model, the theoretical measurement value is calculated, the data is collected as the actual measurement value through the intelligent sensor network, the theoretical measurement value and the actual measurement value are evaluated to obtain an evaluation result. The specific steps are: using a high-precision standard source, including a standard current source and a voltage source, to automatically calibrate the multi-sensor node installed on the thyristor level loop; obtaining the current operating parameters of the test system, including temperature, humidity and electromagnetic interference information, and the working status of the thyristor from the environmental perception module and the central control unit; inputting the current operating conditions into the previous measurement model, calculating the theoretical measurement value, and collecting data from the multi-sensor in real time through the intelligent sensor network to form the actual measurement value; comparing the theoretical measurement value with the actual measurement value, calculating the deviation, and evaluating the deviation. The expression is:
[0016] ;
[0017] in, Used to measure the difference between theoretical measurement value and actual measurement value, and are theoretical measurement values and actual measurement values respectively.
[0018] As a further solution of the present invention, the measurement model parameters are dynamically adjusted based on the evaluation results to obtain an updated measurement model. The specific steps are: based on the evaluation results, several key parameters that have the greatest impact on the deviation are identified, and the expressions are as follows:
[0019] ;
[0020] in, is the adjusted parameter, is the parameter currently used, is a constant used to control the adjustment amplitude. is the difference between the theoretical measurement value and the actual measurement value; the adjusted parameter Applied to the measurement model to recalculate the theoretical measurement value ; Use the updated measurement model to recalculate the theoretical measurement values and compare them with the latest actual measurement values to verify the adjustment effect.
[0021] As a further solution of the present invention, the test strategy is dynamically adjusted based on the temperature and humidity changes and electromagnetic interference information provided by the environmental sensing module. The specific steps are: collecting environmental parameters, including temperature, humidity, and electromagnetic interference intensity, in real time from the environmental sensing module integrated in the intelligent sensor network; and evaluating the impact of the environmental parameters on the working state of the thyristor based on the collected environmental parameters. The expression is:
[0022] ;
[0023] in, To consider the test strategy after environmental factors, For the initial test strategy, 、 and Used to control the impact of various environmental factors, 、 and Respectively represent the specific impact of environmental factors on the test strategy; the calculated adjusted test strategy Application in the actual testing process includes adjusting the sampling frequency, changing the warning threshold and taking additional protection measures; based on the effect of the adjusted test strategy, if abnormal conditions still exist, continue to optimize the data collection and analysis methods of the environmental perception module and further improve the test strategy.
[0024] As a further solution of the present invention, the diagnostic results, evaluation results, and data sets are integrated, and a statistical analysis method is used to evaluate the overall health of the thyristor and its circuit to generate a test report. The specific steps are: collecting all diagnostic results from the output of the machine learning model, including labels of normal and abnormal states; incorporating the deviation between theoretical and actual measurement values, as well as temperature and humidity changes and electromagnetic interference information provided by the environmental perception module into the evaluation system; summarizing all collected data sets, including current, voltage, and temperature sensor data, to form a complete data record library; and using a statistical analysis method to comprehensively evaluate the overall health of the thyristor and its circuit, expressed as follows:
[0025] ;
[0026] in, Used to evaluate the health of thyristors and their circuits, Reflects the importance of each diagnostic result, Status label for the output of the machine learning model, Reflects the importance of each assessment result, is the deviation between the theoretical measurement value and the actual measurement value, Used to standardize different types of input data; based on the calculated overall health index , combining historical data and preset thresholds to generate detailed test reports.
[0027] Compared with the prior art, the present invention has the following beneficial effects:
[0028] By setting the initial test conditions and selecting the original parameters of the thyristors and circuits under test, the test system is fully initialized, ensuring the normal operation of all hardware components and providing a reliable reference benchmark for subsequent tests, thereby ensuring that the entire test process is stable, reliable and the data is accurate. By installing multiple sensors at key positions in the thyristor-level circuit and establishing an intelligent sensor network, real-time monitoring of the thyristor working status is achieved, the intelligence level of the system is improved, and the test accuracy and response speed are improved. By collecting and processing data in the intelligent sensor network and using machine learning models for diagnosis, intelligent judgment of the thyristor working status is achieved, the fault warning capability is enhanced, and potential problems are discovered in advance and preventive measures are taken. By using standard sources to calibrate multiple sensors and inputting current operating conditions into the measurement model for evaluation, continuous optimization of measurement accuracy is achieved, the credibility of test results is improved, and long-term stability and high-precision measurement are ensured. BRIEF DESCRIPTION OF THE DRAWINGS
[0029] Figure 1 The figure is a flow chart of a thyristor level circuit testing method. DETAILED DESCRIPTION
[0030] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.
[0031] 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. Those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.
[0032] Secondly, the term "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in various places throughout this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive of other embodiments.
[0033] Example 1
[0034] See also Figure 1 , which is the first embodiment of the present invention, provides a thyristor-level loop testing method, comprising:
[0035] S1. Set the initial test conditions and select the original parameters of the thyristor and circuit to be tested. Based on the original thyristor parameters, build a measurement model.
[0036] Specifically, the initial test conditions are set, and the original parameters of the thyristor and circuit under test are selected. Based on the original thyristor parameters, a measurement model is constructed. The specific steps are as follows: start the test system and execute the self-test program of the hardware components to ensure that all devices are working properly; according to the original thyristor parameters, the sampling frequency and the initial test conditions of the data transmission protocol are set; the rated voltage, current and temperature range of the thyristor are obtained and input into the test system as a reference benchmark; based on the original thyristor parameters, a measurement model is constructed, and the expression is:
[0037] ;
[0038] in, is the minimum current required for the thyristor to start conducting, is the minimum voltage required for the thyristor to start conducting, is the resistance value of the thyristor in the on state, is the temperature of the thyristor when it is working, is a constant, is the difference between the current time and the start time, Indicates that the thyristor is in ideal working condition.
[0039] It should be noted that this initialization process not only ensures that all hardware devices are in the best working condition, but also provides reliable benchmark conditions for subsequent data acquisition and analysis. The sampling frequency and data transmission protocol set in the steps are key factors in ensuring data quality and communication efficiency, and the measurement model constructed based on the original thyristor parameters provides a scientific basis for comparing theoretical values with actual values, thereby improving the accuracy and reliability of the entire test system.
[0040] S2. After initialization is completed, multiple sensors are installed at key locations of the thyristor-level circuit, the multiple sensor nodes are connected to the central control unit via wireless communication technology, and the environmental perception module is integrated to establish an intelligent sensor network;
[0041] Specifically, after initialization is completed, multiple sensors are installed at key positions of the thyristor-level loop, the multiple sensor nodes are connected to the central control unit through wireless communication technology, and an environmental perception module is integrated to establish an intelligent sensing network. The specific steps are: determine the monitoring position according to the initial parameters of the thyristor and its loop; install multiple sensors at the determined key positions, and encapsulate the multiple sensors into independent nodes. Each node has data acquisition, preprocessing and wireless communication functions, and a microcontroller is integrated inside the node to perform preliminary data processing tasks; use the wireless communication technology ZigBee to establish a stable communication link between the multi-sensor node and the central control unit; integrate the environmental perception module in the intelligent sensing network to monitor environmental changes; after completing the installation and configuration of the multi-sensor nodes, the central control unit uniformly manages each node to form an intelligent sensing network covering the entire thyristor-level loop; start the intelligent sensing network, conduct preliminary data acquisition and transmission tests, and verify the communication quality and data accuracy between each node.
[0042] It should be noted that the precise installation and configuration of multi-sensor nodes, combined with the application of wireless communication technologies such as ZigBee, have achieved extensive coverage and real-time monitoring of thyristor-level circuits, improving the intelligence level of the system. The integration of environmental perception modules enables the system to respond to changes in the external environment in real time, ensuring the accuracy and adaptability of test results. At the same time, it simplifies wiring and maintenance work and improves the flexibility and ease of use of the system.
[0043] S3. Collect data from the intelligent sensor network to obtain a data set, apply signal processing algorithms to extract data features, diagnose the data through machine learning models, obtain data diagnosis results, and implement early warning strategies based on the diagnosis results;
[0044] Specifically, data from the intelligent sensor network is collected to obtain a data set. Signal processing algorithms are applied to extract data feature quantities. The data is diagnosed through a machine learning model to obtain data diagnosis results. An early warning strategy is then implemented based on the diagnosis results. The specific steps are: starting the central control unit, sending instructions to multiple sensor nodes, and collecting multi-sensor data; pre-processing the collected raw data by filtering and normalization; and applying signal processing algorithms to analyze the pre-processed data to extract feature quantities reflecting the working status of the thyristor. The expression is:
[0045] ;
[0046] in, is the weighted average and normalized feature quantity, Reflects the importance of each feature; The data points are obtained from different sensors. Based on the collected historical data and the theoretical calculation values generated by the measurement model, the machine learning model is trained and the extracted feature quantities are input into the machine learning model. The machine learning model evaluates whether the current feature quantities are normal or abnormal based on the pattern recognition ability learned during training and outputs the diagnosis result. The expression is:
[0047] ;
[0048] in, Used to evaluate the working status of thyristors, is the activation function, reflects the importance of each feature function, is the comprehensive eigenvalue A function for nonlinear transformation; setting the threshold L, based on the diagnosis results ,When the diagnosis index exceeds the threshold L, the alarm mechanism is triggered, ,and protection actions are automatically taken and the power is cut off.
[0049] It should be noted that by applying advanced signal processing algorithms and machine learning models, the steps have realized intelligent diagnosis of the working status of thyristors, greatly improving the timeliness and accuracy of fault detection. The extraction of feature quantities has simplified the complex data processing process, and the learning ability of the machine learning model has ensured the effective identification of abnormal conditions. By setting reasonable thresholds and triggering alarm mechanisms, preventive measures can be taken before problems occur, reducing potential risks and losses.
[0050] S4. Calibrate the multi-sensor using a standard source based on the collected data set, input the current operating conditions into the measurement model, calculate the theoretical measurement value, collect data as the actual measurement value through the intelligent sensor network, evaluate the theoretical measurement value and the actual measurement value, and obtain the evaluation result;
[0051] Specifically, based on the collected data set, a standard source is used to calibrate the multi-sensor, the current operating conditions are input into the measurement model, the theoretical measurement value is calculated, and data is collected as the actual measurement value through the intelligent sensor network. The theoretical measurement value and the actual measurement value are evaluated to obtain the evaluation result. The specific steps are: using high-precision standard sources, including standard current sources and voltage sources, to automatically calibrate the multi-sensor nodes installed on the thyristor level loop; obtaining the current operating parameters of the test system from the environmental perception module and the central control unit, including information on temperature, humidity, and electromagnetic interference, as well as the working status of the thyristor; inputting the current operating conditions into the previous measurement model, calculating the theoretical measurement value, and collecting data from the multi-sensor in real time through the intelligent sensor network to form the actual measurement value; comparing the theoretical measurement value with the actual measurement value, calculating the deviation, and evaluating the deviation. The expression is:
[0052] ;
[0053] in, Used to measure the difference between theoretical measurement value and actual measurement value, and are theoretical measurement values and actual measurement values respectively.
[0054] It should be noted that the use of high-precision standard sources to calibrate multiple sensors ensures the long-term stability and accuracy of the measurement results and eliminates errors caused by factors such as time drift. By inputting the current operating conditions into the measurement model to calculate the theoretical measurement values and comparing them with the actual measurement values, the deviation can be quantified and the measurement model can be adjusted to ensure that it is always close to the actual situation. This step is crucial to maintaining system reliability and improving test accuracy.
[0055] S5. Dynamically adjust measurement model parameters based on the evaluation results to obtain an updated measurement model;
[0056] Specifically, the measurement model parameters are dynamically adjusted based on the evaluation results to obtain an updated measurement model. The specific steps are as follows: Based on the evaluation results, several key parameters with the greatest impact on the deviation are identified. The expressions are as follows:
[0057] ;
[0058] in, is the adjusted parameter, is the parameter currently used, is a constant used to control the adjustment amplitude. is the difference between the theoretical measurement value and the actual measurement value; the adjusted parameter Applied to the measurement model to recalculate the theoretical measurement value ; Use the updated measurement model to recalculate the theoretical measurement values and compare them with the latest actual measurement values to verify the adjustment effect.
[0059] It should be noted that the process of dynamically adjusting the measurement model parameters reflects the system's adaptive ability and continuous optimization characteristics. Through in-depth analysis of the evaluation results, the key parameters affecting the deviation are identified, and the measurement model is adjusted accordingly, which can significantly improve the accuracy of the prediction. The theoretical measurement values are recalculated and compared with the latest actual measurement values to verify the adjustment effect, ensuring that the model improvement is practical and effective, thereby enhancing the overall performance of the system.
[0060] S6. Dynamically adjust the test strategy based on the temperature and humidity changes and electromagnetic interference information provided by the environmental perception module;
[0061] Specifically, the test strategy is dynamically adjusted based on the temperature and humidity changes and electromagnetic interference information provided by the environmental perception module. The specific steps are as follows: the environmental parameters, including temperature, humidity, and electromagnetic interference intensity, are collected in real time from the environmental perception module integrated in the intelligent sensor network; based on the collected environmental parameters, the impact of the environmental parameters on the working state of the thyristor is evaluated. The expression is:
[0062] ;
[0063] in, To consider the test strategy after environmental factors, For the initial test strategy, 、 and Used to control the impact of various environmental factors, 、 and Respectively represent the specific impact of environmental factors on the test strategy; the calculated adjusted test strategy Application in the actual testing process includes adjusting the sampling frequency, changing the warning threshold and taking additional protection measures; based on the effect of the adjusted test strategy, if abnormal conditions still exist, continue to optimize the data collection and analysis methods of the environmental perception module and further improve the test strategy.
[0064] It should be noted that the real-time environmental information provided by the environmental perception module is crucial for understanding the working status of thyristors. By evaluating the impact of temperature, humidity and electromagnetic interference on thyristors and adjusting the test strategy accordingly, it can better adapt to the needs under different working conditions. Dynamic adjustment of the test strategy, such as changing the sampling frequency or warning threshold, helps to improve the pertinence and effectiveness of the test, while reducing unnecessary resource consumption and ensuring the efficient operation of the system.
[0065] S7, integrating the diagnosis results, evaluation results and data sets, using statistical analysis methods to evaluate the overall health status of the thyristor and its circuit, and generating a test report;
[0066] Specifically, the diagnostic results, evaluation results, and data sets are integrated, and statistical analysis methods are used to evaluate the overall health of the thyristor and its circuit to generate a test report. The specific steps are as follows: collect all diagnostic results from the output of the machine learning model, including the marks of normal and abnormal states; incorporate the deviation between theoretical and actual measurement values, as well as the temperature and humidity changes and electromagnetic interference information provided by the environmental perception module into the evaluation system; summarize all collected data sets, including current, voltage, and temperature sensor data, to form a complete data record library; use statistical analysis methods to comprehensively evaluate the overall health of the thyristor and its circuit, expressed as follows:
[0067] ;
[0068] in, Used to evaluate the health of thyristors and their circuits, Reflects the importance of each diagnostic result, Status label for the output of the machine learning model, Reflects the importance of each assessment result, is the deviation between the theoretical measurement value and the actual measurement value, Used to standardize different types of input data; based on the calculated overall health index , combining historical data and preset thresholds to generate detailed test reports.
[0069] It should be noted that the comprehensive diagnostic results, evaluation results and data sets are comprehensively evaluated using statistical analysis methods, providing intuitive and quantitative evaluation indicators for the overall health status of thyristors and their circuits. By calculating the overall health index, combining historical data and preset thresholds, a detailed test report is generated, which not only helps maintenance personnel quickly understand the system status, but also provides a scientific basis for future maintenance plans, ultimately achieving the goal of ensuring the safe and stable operation of the power system.
[0070] In summary, by setting the initial test conditions and selecting the original parameters of the thyristors and circuits under test, the test system is fully initialized, ensuring the normal operation of all hardware components and providing a reliable reference benchmark for subsequent tests, thereby ensuring that the entire test process is stable, reliable and the data is accurate. By installing multiple sensors at key positions in the thyristor-level circuit and establishing an intelligent sensor network, real-time monitoring of the thyristor working status is achieved, the intelligence level of the system is improved, and the test accuracy and response speed are improved. By collecting and processing data in the intelligent sensor network and using machine learning models for diagnosis, intelligent judgment of the thyristor working status is achieved, the fault warning capability is enhanced, and potential problems are discovered in advance and preventive measures are taken. By using standard sources to calibrate multiple sensors and inputting current operating conditions into the measurement model for evaluation, continuous optimization of measurement accuracy is achieved, the credibility of test results is improved, and long-term stability and high-precision measurement are ensured.
[0071] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.
Claims
1. A thyristor-level loop testing method, characterized in that: include: Set the initial test conditions, select the original parameters of the thyristor and circuit under test, and build a measurement model based on the original thyristor parameters; After initialization, multiple sensors are installed at key locations in the thyristor-level circuit, connected to the central control unit via wireless communication technology, and an environmental perception module is integrated to establish an intelligent sensor network. Collect data from the intelligent sensor network to obtain a data set, apply signal processing algorithms to extract data features, diagnose the data through machine learning models, obtain data diagnosis results, and implement early warning strategies based on the diagnosis results; Based on the collected data set, multiple sensors are calibrated using a standard source, the current operating conditions are input into the measurement model, theoretical measurement values are calculated, data is collected through the intelligent sensor network as actual measurement values, and the theoretical and actual measurement values are evaluated to obtain evaluation results; Dynamically adjust measurement model parameters based on evaluation results to obtain an updated measurement model; Dynamically adjust the test strategy based on the temperature and humidity changes and electromagnetic interference information provided by the environmental perception module; Integrate diagnostic results, evaluation results, and data sets, use statistical analysis methods to evaluate the overall health of thyristors and their circuits, and generate test reports.
2. A thyristor-level loop testing method according to claim 1, characterized in that: The initial test conditions are set, and the original parameters of the thyristor and the circuit to be tested are selected. Based on the original parameters of the thyristor, a measurement model is constructed. The specific steps are as follows: Start the test system and perform self-test procedures on hardware components to ensure that all devices are working properly; According to the original parameters of the thyristor, set the sampling frequency and the initial test conditions of the data transmission protocol; Obtain the rated voltage, current, and temperature range of the thyristor and input them into the test system as a reference benchmark; Based on the original parameters of the thyristor, a measurement model is constructed, and the expression is: ； in, is the minimum current required for the thyristor to start conducting, is the minimum voltage required for the thyristor to start conducting, is the resistance value of the thyristor in the on state, is the temperature of the thyristor when it is working, is a constant, is the difference between the current time and the start time, Indicates that the thyristor is in ideal working condition.
3. The thyristor level loop testing method according to claim 1, wherein: After the initialization is completed, multiple sensors are installed at key positions of the thyristor-level circuit, the multiple sensor nodes are connected to the central control unit through wireless communication technology, and the environmental perception module is integrated to establish an intelligent sensor network. The specific steps are as follows: Determine the monitoring position based on the initial parameters of the thyristor and its circuit; Install multiple sensors at identified key locations and package them into independent nodes. Each node has data acquisition, preprocessing, and wireless communication functions. A microcontroller is integrated within the node to perform preliminary data processing tasks. Using wireless communication technology ZigBee to establish a stable communication link between multiple sensor nodes and the central control unit; Integrate environmental perception modules into intelligent sensor networks to monitor environmental changes; After the installation and configuration of multiple sensor nodes are completed, each node is managed uniformly through the central control unit to form an intelligent sensor network covering the entire thyristor-level circuit; Start the intelligent sensor network and conduct preliminary data collection and transmission tests to verify the communication quality and data accuracy between nodes.
4. The thyristor-level loop testing method according to claim 1, wherein: The data in the intelligent sensor network is collected to obtain a data set, a signal processing algorithm is applied to extract the characteristic quantity of the data, and the data is diagnosed through a machine learning model to obtain a data diagnosis result, and an early warning strategy is executed based on the diagnosis result. The specific steps are as follows: Start the central control unit, send instructions to the multi-sensor nodes, and collect multi-sensor data; Preprocess the collected raw data by filtering and normalizing; The signal processing algorithm is applied to analyze the pre-processed data and extract the characteristic quantity reflecting the working state of the thyristor. The expression is: ; in, is the weighted average and normalized feature quantity, Reflects the importance of each feature; are the data points obtained from different sensors; Based on the collected historical data and the theoretical calculation values generated by the measurement model, the machine learning model is trained and the extracted feature quantities are input into the machine learning model. The machine learning model evaluates whether the current feature quantities are normal or abnormal based on the pattern recognition capabilities learned during training and outputs the diagnosis results, which are expressed as: ; in, Used to evaluate the working status of thyristors, is the activation function, reflects the importance of each feature function, is the comprehensive eigenvalue Functions that perform nonlinear transformations; Set the threshold L according to the diagnosis results ,When the diagnosis index exceeds the threshold L, the alarm mechanism is triggered, ,and protection actions are automatically taken and the power is cut off.
5. The thyristor level loop testing method according to claim 1, characterized in that: The method comprises the following steps:
1. calibrating the multi-sensor using a standard source based on the collected data set; 2. inputting the current operating conditions into the measurement model; 3. calculating the theoretical measurement value; 4. collecting data as the actual measurement value through the intelligent sensor network; 5. evaluating the theoretical measurement value and the actual measurement value to obtain the evaluation result; 6. Use high-precision standard sources, including standard current and voltage sources, to automatically calibrate the multi-sensor nodes installed in the thyristor stage loop; Obtain the current operating parameters of the test system from the environmental perception module and the central control unit, including information on temperature, humidity, and electromagnetic interference, as well as the operating status of the thyristors; Input the current operating conditions into the previous measurement model to calculate the theoretical measurement value, and collect data from multiple sensors in real time through the intelligent sensor network to form the actual measurement value; Compare the theoretical measurement value with the actual measurement value, calculate the deviation, and evaluate the deviation. The expression is: ; in, Used to measure the difference between theoretical measurement value and actual measurement value, and are theoretical measurement values and actual measurement values respectively.
6. The thyristor level loop testing method according to claim 1, characterized in that: The method of dynamically adjusting the measurement model parameters based on the evaluation results to obtain an updated measurement model comprises the following specific steps: Based on the evaluation results, several key parameters with the greatest impact on the deviation are identified, as shown below: ; in, is the adjusted parameter, is the parameter currently used, is a constant used to control the adjustment amplitude. is the difference between the theoretical measurement value and the actual measurement value; The adjusted parameters Applied to the measurement model to recalculate the theoretical measurement value ; The theoretical measurement values are recalculated using the updated measurement model and compared with the latest actual measurement values to verify the adjustment effect.
7. The thyristor-level loop testing method according to claim 1, wherein: The test strategy is dynamically adjusted based on the temperature and humidity changes and electromagnetic interference information provided by the environmental perception module. The specific steps are as follows: Real-time collection of environmental parameters, including temperature, humidity, and electromagnetic interference intensity, from the environmental sensing module integrated in the intelligent sensor network; Based on the collected environmental parameters, evaluate the impact of environmental parameters on the working state of thyristors; Apply the calculated adjusted test strategy to the actual test process, including adjusting the sampling frequency, changing the warning threshold, and taking additional protection measures; Based on the effect of the adjusted test strategy, if abnormal conditions still exist, continue to optimize the data collection and analysis methods of the environmental perception module and further improve the test strategy.
8. The thyristor-level loop testing method according to claim 1, wherein: The steps of integrating the diagnosis results, evaluation results and data sets, using statistical analysis methods to evaluate the overall health of the thyristor and its circuit, and generating a test report are as follows: Collect all diagnostic results from the output of the machine learning model, including labels for normal and abnormal states; The deviation between theoretical and actual measurement values, as well as the temperature and humidity changes and electromagnetic interference information provided by the environmental perception module, are incorporated into the evaluation system; Aggregate all collected data sets, including current, voltage, and temperature sensor data, to form a complete data record library; Use statistical analysis methods to comprehensively evaluate the overall health status of thyristors and their circuits; Generate a detailed test report based on the calculated overall health index, combined with historical data and preset thresholds.
9. The thyristor-level loop testing method according to claim 7, characterized in that: According to the collected environmental parameters, the expression for evaluating the influence of the environmental parameters on the working state of the thyristor is: ； in, To consider the test strategy after environmental factors, For the initial test strategy, 、 and Used to control the impact of various environmental factors, 、 and They respectively represent the specific impact of environmental factors on testing strategies.
10. The thyristor level loop testing method according to claim 8, characterized in that: The expression for comprehensively evaluating the overall health status of the thyristor and its circuit using the statistical analysis method is: ; in, Used to evaluate the health of thyristors and their circuits, Reflects the importance of each diagnostic result, Status label for the output of the machine learning model, Reflects the importance of each assessment result, is the deviation between the theoretical measurement value and the actual measurement value, Used to normalize different types of input data.