An automated test system for subway DC protection devices based on embedded systems

Abstract

The present invention provides an automated test system for subway DC protection devices based on an embedded system, including a program-controlled DC power supply, an electronic load, a data acquisition card, a communication module, and a host computer; the program-controlled DC power supply provides a power basis for the entire test system, and the DC voltage and current output by the program-controlled DC power supply act on the DC protection device of the subway under test; the electronic load works in coordination with the program-controlled DC power supply, and according to the test requirements, it dynamically adjusts its own resistance value to change the current drawn from the power supply, thereby simulating the load changes of the subway in different operating periods and different working conditions; the data acquisition card is connected to the program-controlled DC power supply, the electronic load, and the protection device under test at the same time, and the data acquisition card also collects the feedback signal of the protection device under test; the data acquisition card is connected to the host computer through the communication module. The present invention can realize all-round, high-precision and high-efficiency testing of subway DC protection devices, and ensure the safe and stable operation of the subway DC power supply system.

Description

Technical Field

[0001] The invention relates to the technical field of subway equipment testing, and in particular to an automatic testing system for a subway DC protection device based on an embedded system. Background Art

[0002] As a fast, high-capacity rail transit mode in the city with electric traction, the subway has significant external economic benefits in promoting the development of related industries such as industry and transportation, stimulating employment, and expanding urban development space. It occupies an important position in the national economy and people's lives. The subway DC protection device is a key device to ensure the normal operation of the equipment in the subway DC comprehensive protection substation. In order to ensure the safe operation of the rail transit DC comprehensive protection control system, the subway DC protection device needs to be inspected regularly.

[0003] Traditional subway DC protection device testing methods have many drawbacks. Early testing work mainly relied on manual operation. Testers manually adjusted the output voltage and current of the test power supply, used conventional instruments to monitor data, and relied on experience to determine whether the protection device was operating normally. This manual testing method is not only extremely inefficient, but also difficult to ensure the accuracy of the test results due to the inevitable subjectivity and instability of manual operation. For example, in long-term, multi-batch overcurrent protection tests, it is difficult to achieve complete consistency in the manual adjustment of the current rise rate, which can easily introduce test errors, resulting in deviations in the performance evaluation of the protection device, and thus burying safety hazards in the actual operation of the subway.

[0004] Although the subsequent semi-automatic test system has been improved and introduced some simple program-controlled equipment to assist in testing, it still has not fundamentally solved the problem. On the one hand, there is a lack of deep integration and coordination between the various test devices, each hardware device operates independently, the communication protocols are incompatible, and data interaction is difficult, making it impossible to form an efficient and unified test system. For example, when conducting complex fault simulation tests, since the power supply and load cannot be accurately coordinated, it is impossible to quickly create test conditions that meet actual working conditions, which prolongs the test cycle; on the other hand, since the test data analysis method is single and mostly based on simple threshold comparisons, it is difficult to dig out the deep-level performance characteristics of the protection device from the massive data, and it is impossible to warn of potential fault hazards in advance.

[0005] With the continuous innovation of subway technology, the performance requirements for DC protection devices are becoming increasingly stringent. Not only are they required to be able to cope with conventional overcurrent, overvoltage and other faults, but they must also be able to operate reliably under extreme conditions such as complex harmonic interference and multiple fault superposition. However, the existing test system is difficult to simulate these complex and changeable operating environments due to poor hardware coordination and low software intelligence, and is far from meeting the current high-precision test requirements for subway DC protection devices. Summary of the invention

[0006] The technical problem to be solved by the present invention is to provide an automated testing system for subway DC protection devices based on an embedded system in view of the deficiencies in the prior art, so as to realize all-round, high-precision and high-efficiency testing of subway DC protection devices and ensure the safe and stable operation of the subway DC power supply system.

[0007] In order to solve the above technical problems, the technical solution of the present invention is: An automated test system for subway DC protection devices based on an embedded system, characterized in that it includes a programmable DC power supply, an electronic load, a data acquisition card, a communication module and a host computer; the programmable DC power supply provides a power basis for the entire test system, and the DC voltage and current output by the programmable DC power supply directly act on the DC protection device of the subway under test, simulating the power supply situation in the actual operation of the subway; the electronic load works in coordination with the programmable DC power supply, and according to the test requirements, it dynamically adjusts its own resistance value to change the current drawn from the power supply, thereby simulating the load changes of the subway under different operating periods and different working conditions; the data acquisition card is simultaneously connected to the programmable DC power supply, the electronic load and the protection device under test, on the one hand, collects the voltage and current signals output by the programmable DC power supply, and monitors in real time whether the programmable DC power supply is operating according to the preset working conditions. On the one hand, the data acquisition card outputs stable or dynamically changing power to ensure the accuracy of the power supply simulation, and on the other hand, it collects the current and voltage changes at the electronic load end, obtains the electrical parameters during the electronic load adjustment process, and provides data support for analyzing the performance of the tested protection device under different load responses; the data acquisition card also collects the feedback signal of the tested protection device; the data acquisition card is connected to the host computer through the communication module, and the data acquisition card transmits the collected electrical parameter data and key information reflecting the operating status of the tested protection device to the host computer through the communication module, and the host computer receives and processes these data and information to realize real-time monitoring and management of the test process; the host computer establishes connection with the programmable DC power supply and the electronic load through the communication module, and sends corresponding control instructions to them according to the test requirements and the preset test process.

[0008] Preferably, during the overcurrent protection test, the electronic load increases its resistance according to a preset rule, causing the loop current to rise, prompting the protection device under test to enter a corresponding test state, and verifying its overcurrent protection function; and in the thermal overload protection test, a large current load is maintained for a long time to simulate the heating process of the equipment, and together with the programmable DC power supply output, a complete test scenario is constructed.

[0009] Preferably, when the data acquisition card acquires the feedback signal of the protection device under test, the sampling frequency is adaptively adjusted according to the signal characteristics.

[0010] Preferably, the host computer includes a test management module, a data analysis and processing module, a prediction module and a human-computer interaction interface module; the communication module is connected to the test management module, and the test management module is respectively connected to the data analysis and processing module, the prediction module and the human-computer interaction interface module; the human-computer interaction interface module is used to provide an operation entry for the tester to set the test parameters and operation instructions, and pass this information to the test management module, and the test management module determines the control strategy for the programmable DC power supply and the electronic load according to the received information, and sends the control instructions to them; the test management module feeds back various status information generated during the test to the human-computer interaction interface module, and the human-computer interaction interface module updates the control strategy according to the received information. A new interface is displayed; during the test process, the test management module transmits the data obtained from the programmable DC power supply and the electronic load to the data analysis and processing module. After analyzing and processing the data, the data analysis and processing module generates analysis results and feeds them back to the test management module. The test management module adjusts the control instructions for the programmable DC power supply and the electronic load according to the data analysis results to optimize the test process; the test management module sends the current test parameters and historical data to the prediction module as input of the prediction model. The prediction module predicts various parameters in the test process and returns the prediction results to the test management module. The test management module optimizes the test strategy according to the prediction results and adjusts the output parameters of the programmable DC power supply and the electronic load in advance.

[0011] Preferably, the data analysis and processing module is connected to the prediction module, and the data analysis and processing module provides the processed data to the prediction module to help the prediction module build and adjust the prediction model; the prediction module then feeds back its prediction results to the data analysis and processing module for comparison with the actual analysis results to further optimize the data analysis and processing algorithm.

[0012] Preferably, the test management module dynamically allocates resources according to the priority weight and impact of the test task: 1) Determine the priority weight of the test task based on the importance of the test task to the safe operation of the subway DC system as an initial guide for resource allocation tendencies; 2) Determine the impact of the test task based on the depth and breadth of the impact on the subsequent test process after the test task fails; 3) Multiply the priority weight of the test task by the impact to obtain a comprehensive influence index, and allocate more resources to test tasks with large comprehensive influence. 4) Continuously update the priority weight and impact of the test task during the test, dynamically calculate the comprehensive influence of the test task, and adjust resource allocation in real time.

[0013] Preferably, the test management module adopts a reinforcement learning algorithm to construct a test strategy optimization model, and regards the test process as a Markov decision process MDP, in which the state space includes the model of the protection device under test, the current test item, the hardware equipment status and historical test data; the action space covers adjusting test parameters, switching test items and controlling hardware equipment. By allowing the intelligent agent to continuously explore and learn in the MDP environment, the test strategy is optimized according to the reward function, so as to achieve personalized and efficient testing for different protection devices under test.

[0014] Preferably, the data analysis and processing module uses a generative adversarial network and a convolutional neural network algorithm to perform deep feature mining. The generator in the generative adversarial network is used to simulate various fault current and voltage waveforms, and the discriminator is constructed based on the convolutional neural network to distinguish the collected real waveform from the waveform generated by the generator, and identify abnormal patterns in the waveform; and uses random process theory to construct a performance degradation model of the tested protection device, and regards the performance indicators of the tested protection device as random processes. Based on historical test data, a probability distribution function is used to fit the degradation trajectory of the performance indicators. Through real-time updating and prediction of distribution parameters, the remaining life of the tested protection device is evaluated, and potential failure risks are warned in advance.

[0015] Preferably, the instruction processing and feedback of the human-computer interaction interface module are as follows: 1) a high-speed shared memory channel is established between the human-computer interaction interface module and the test management module, and instructions are directly written into the shared memory area when issued. The test management module monitors in real time and responds in seconds, directly driving the hardware action; 2) a unified data feedback scheduling center is established, and the hardware devices and software modules send feedback information to this center in a unified manner in a standard format. The center gives priority to key test process feedback based on priority queues and multi-threaded processing technology; the information display layout of the human-computer interaction interface module is as follows: the key information that the tester is most concerned about is given a high visual weight and placed in the most conspicuous and easily accessible area of ​​the human-computer interaction interface module, while relatively minor information is given a lower visual weight and placed in the surrounding auxiliary area. At the same time, the visual weight and layout of the information are dynamically adjusted according to the test process to ensure that the tester can quickly grasp the core points by scanning the interface at a glance, thereby improving operational efficiency and emergency response speed.

[0016] Preferably, the host computer adopts message transmission and event mechanism. When the data acquisition card collects new data, a data update event is triggered. When the test management module subscribes to this event, after this event occurs, the data acquisition card passes the new data to the data analysis and processing module; when the prediction module completes a prediction update, a prediction update event is issued. When the human-computer interaction interface module and the test management module subscribe to this event, they can obtain the latest prediction information in time and perform corresponding processing.

[0017] The beneficial effects of the present invention are: The present invention is based on the hardware architecture design of the embedded system to achieve deep integration of various hardware devices. The program-controlled DC power supply, intelligent electronic load, data acquisition card and communication module work closely together and are scheduled by a unified embedded controller. When simulating the peak load period in subway operation, the embedded system can accurately coordinate the power supply to output high-power DC power within milliseconds, the electronic load synchronously simulates complex dynamic loads, and the data acquisition card captures key electrical parameters at high speed, ensuring that the test scene is highly realistic and the test data is accurate and reliable. Compared with the previous loosely connected equipment combination, the data accuracy is greatly improved.

[0018] The test management module can dynamically plan the optimal test path based on multi-dimensional status information, including the model parameters of the tested protection device, real-time test progress, real-time status of hardware equipment, and historical test data mining results. When faced with complex and changeable subway DC power supply condition simulation requirements, it can quickly make decisions to adjust the order of test items, accurately control the coordinated work of program-controlled DC power supply and intelligent electronic load, and greatly shorten the overall test cycle.

[0019] The prediction model combines the physical model with the machine learning algorithm, and can predict the voltage and current change trends in advance. Before conducting short-circuit fault simulation tests, it can accurately predict key parameters such as voltage sag and current impact peak that may occur, and provide forward-looking guidance for the test management module to optimize the test strategy, effectively avoid test interruptions or data loss due to emergencies, and improve the test success rate.

[0020] The data analysis and processing module combines the generative adversarial network and convolutional neural network algorithms to conduct in-depth mining of the collected data. It can not only accurately identify whether the common over-current and over-voltage protection actions are normal, but also keenly capture the subtle abnormal waveform characteristics caused by weak harmonic interference, potential poor contact, etc. The accuracy of fault diagnosis is greatly improved, and potential fault hazards can be discovered in advance, providing a strong basis for the maintenance and inspection of subway DC protection devices.

[0021] The human-computer interaction interface module can provide testers with an intuitive and convenient operating experience. By displaying the test progress, dynamic changes of electrical parameters, equipment status and other information in real time and visually, it is convenient for testers to quickly grasp the overall situation. Based on the database sharing function, historical test data can be traced back at any time, and the test results of different batches can be compared and analyzed, which is convenient for summarizing rules and optimizing subsequent test processes, greatly improving the efficiency and convenience of test management. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 It is a structural schematic diagram of the present invention. DETAILED DESCRIPTION

[0023] To facilitate understanding 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 understood by those skilled in the art that the embodiments are only intended to help understand the present invention and should not be regarded as specific limitations of the present invention.

[0024] In the description of the following embodiments, specific details such as specific system structures, technologies, etc. are provided for illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application may also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present application.

[0025] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or their collections. It should also be understood that the term "and / or" used in the present specification and the appended claims refers to any combination of one or more of the associated listed items and all possible combinations, and includes these combinations.

[0026] As used in the specification and appended claims of this application, the term "if" can be interpreted as "when" or "uponce" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" can be interpreted as meaning "uponce it is determined" or "in response to determining" or "uponce [described condition or event] is detected" or "in response to detecting [described condition or event]", depending on the context.

[0027] In addition, in the description of the present application specification and the attached claims, the terms "first", "second", "third", etc. are only used to distinguish the description and cannot be understood as indicating or implying relative importance. The reference to "one embodiment" or "some embodiments" described in the present application specification means that one or more embodiments of the present application include specific features, structures or characteristics described in conjunction with the embodiment. Therefore, the statements "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. that appear in the different parts of this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways. Example 1

[0028] like Figure 1 As shown, the present invention provides an automated test system for subway DC protection devices based on an embedded system, including a programmable DC power supply, an electronic load, a data acquisition card, a communication module and a host computer; the programmable DC power supply provides a power basis for the entire test system, and the DC voltage and current output by the programmable DC power supply directly act on the subway DC protection device under test, simulating the power supply situation in the actual operation of the subway; the electronic load works in coordination with the programmable DC power supply, and according to the test requirements, it dynamically adjusts its own resistance value to change the current drawn from the power supply, thereby simulating the load changes of the subway in different operating periods and different working conditions; the data acquisition card is connected to the programmable DC power supply, the electronic load and the protection device under test at the same time, on the one hand, it collects the voltage and current signals output by the programmable DC power supply, and monitors in real time whether the programmable DC power supply is in accordance with the test requirements. The preset working conditions output stable or dynamically changing power to ensure the accuracy of the power supply simulation. On the other hand, the current and voltage changes at the electronic load end are collected to obtain the electrical parameters during the electronic load adjustment process, providing data support for analyzing the performance of the protection device under test under different load responses; the data acquisition card also collects the feedback signal of the protection device under test; the data acquisition card is connected to the host computer through the communication module, and the data acquisition card transmits the collected electrical parameter data and key information reflecting the operating status of the protection device under test to the host computer through the communication module. The host computer receives and processes this data and information to realize real-time monitoring and management of the test process; the host computer establishes a connection with the programmable DC power supply and the electronic load through the communication module, and sends corresponding control instructions to them according to the test requirements and the preset test process.

[0029] The core control of the programmable DC power supply adopts the model predictive control (MPC) algorithm, which predicts the voltage and current output states at multiple future sampling moments by establishing a discrete time model of the converter inside the power supply. In each control cycle, a constrained optimization problem is solved to minimize the error between the output voltage and current and the preset reference value, while meeting the physical limitations of the power supply hardware, such as the maximum power, maximum current change rate and other constraints, so as to maintain the output accuracy at ±0.05%. The programmable DC power supply is equipped with a high-speed dynamic response module. Compared with traditional silicon-based devices, the switching loss and on-resistance are significantly reduced by using silicon carbide (SiC) power devices, and the millisecond-level switching from rated working conditions to fault simulation states is achieved. During the switching process, the voltage change rate (dV / dt) can be controlled within 1000V / μs, which can ensure the smoothness and accuracy of the transition of the test scenario and avoid interference with the protection device test due to power switching. During the overcurrent protection test, the electronic load increases its resistance according to the preset rules, causing the loop current to rise, prompting the protection device under test to enter the corresponding test state to verify its overcurrent protection function; in the thermal overload protection test, a large current load is maintained for a long time to simulate the heating process of the equipment, and together with the programmable DC power supply output, a complete test scenario is constructed.

[0030] The load curve simulation of the electronic load in the intelligent mode is based on the deep neural network (DNN) model. First, a large amount of load current data of the actual operation of the subway is collected as training samples, including the current waveforms in the morning and evening peak hours, off-peak hours, and special working conditions (such as emergency braking energy feedback). The DNN model is used to learn the timing characteristics and amplitude change laws in these samples, and a prediction model that can accurately reproduce the load curve is constructed. During the test, according to the preset test script, load control instructions are generated in real time to drive the electronic load to dynamically adjust the resistance value to simulate the real load change. The built-in real-time temperature monitoring module of the electronic load uses a high-precision thermistor array distributed in key heating parts, such as power resistors, heat sinks, etc. By measuring the temperature values ​​at different positions, a partial differential equation (PDE) model of heat conduction is constructed, and the PDE model is solved by the finite difference method to infer the overall temperature distribution of the load and feedback to the host computer to provide accurate temperature field information for the thermal overload protection test.

[0031] When the data acquisition card collects the feedback signal of the protection device under test, it adaptively adjusts the sampling frequency according to the signal characteristics. Combined with the time-frequency domain characteristics of the signal, the voltage and current signals are analyzed by short-time Fourier transform (STFT), and the Hanning window is used as the window function to monitor the frequency component changes of the signal in real time; when a high-frequency transient signal is detected (such as pulse interference, switching transient), the sampling frequency is automatically increased to 5MHz to capture the complete signal details; in the steady-state stage, it is restored to 2MHz sampling frequency to effectively balance the data volume and information integrity. In this way, both the rapidly changing transient signal and the relatively stable steady-state signal can have high-fidelity characteristics, providing a solid data foundation for subsequent data analysis. The data transmission of the data acquisition card is based on the reliable transmission protocol of the high-speed fiber channel (such as the FC-AE-1553 protocol), which uses the cyclic redundancy check (CRC) and automatic retransmission request (ARQ) mechanism at the link layer to ensure error-free data transmission. At the same time, data priority management is introduced in the upper application layer protocol to give priority to the transmission of key electrical parameter data with high real-time requirements, ensuring that the host computer can process the most important information in a timely manner.

[0032] The communication module can convert the control instructions of the host computer into a format that can be recognized by the protection device under test, and convert the information fed back by the protection device under test into a form that can be processed by the host computer. The communication module adopts protocol parsing and reconstruction technology. According to the protocol specification, a dedicated protocol stack code is written to achieve accurate parsing of the Profibus-DP frame structure, including the start byte, address field, data field, checksum and other parts. When communicating with the protection device under test, it can dynamically reconstruct the data packet that conforms to the protocol according to the other party's communication requirements to ensure accurate information interaction.

[0033] The communication module has the functions of communication fault self-diagnosis and backup link switching based on the Markov chain model. A Markov chain for communication link state transfer is established, and the link bit error rate, signal strength and other indicators are monitored in real time as the probability basis for state transfer; when the probability of main link failure is detected to exceed the preset threshold (such as 0.8), the backup link switching operation is immediately triggered, and the switching process is completed within 10ms, ensuring uninterrupted data transmission during the test, avoiding test failure or data loss due to communication interruption, and maintaining the stable operation of the entire test system. Example 2

[0034] The host computer includes a test management module, a data analysis and processing module, a prediction module and a human-computer interaction interface module; the communication module is connected to the test management module, and the test management module is connected to the data analysis and processing module, the prediction module and the human-computer interaction interface module respectively. The human-computer interaction interface module is used to provide an operation entry for testers to set test parameters (such as selecting test items, setting current rise rate thresholds, etc.) and operation instructions, and pass this information to the test management module. The test management module determines the control strategy for the programmable DC power supply and electronic load based on the received information, and sends the control instructions to them; the test management module feeds back various status information generated during the test to the human-computer interaction interface module, and the human-computer interaction interface module updates the interface display based on the received information.

[0035] During the test, the test management module passes the data obtained from the programmable DC power supply and the electronic load to the data analysis and processing module. After analyzing and processing the data, the data analysis and processing module generates analysis results (such as the judgment result of whether the protection device is activated, whether the current rise rate exceeds the standard, etc.) and feeds them back to the test management module. The test management module adjusts the control instructions for the programmable DC power supply and the electronic load according to the data analysis results to optimize the test process.

[0036] The test management module sends the current test parameters and historical data to the prediction module as the input of the prediction model. The prediction module predicts various parameters in the test process and returns the prediction results to the test management module. The test management module optimizes the test strategy according to the prediction results and adjusts the output parameters of the programmable DC power supply and electronic load in advance to better simulate various working conditions in actual operation and improve the comprehensiveness and effectiveness of the test.

[0037] The data analysis and processing module is connected to the prediction module. The data analysis and processing module provides the processed data to the prediction module. The data includes some characteristic information, such as current harmonic components, statistical characteristics of voltage fluctuations, etc., which are used to help the prediction module to more accurately build and adjust the prediction model. The prediction module then feeds back its prediction results to the data analysis and processing module for comparison with the actual analysis results, and further optimizes the data analysis and processing algorithm, such as evaluating the accuracy of the prediction model by comparing the predicted current rise rate with the current rise rate obtained by actual analysis, and then adjusting the filtering parameters used for data processing.

[0038] The test management module adopts one-click automated testing. Based on the workflow engine technology, the test process is abstracted into a series of task nodes with sequential and logical relationships, including hardware device initialization, test parameter configuration, test process execution, test report generation, etc. Each task node is implemented by a corresponding software module or hardware driver. The workflow engine automatically coordinates the execution order of each node according to the preset workflow template to achieve fully automated testing from beginning to end. At the same time, the transaction processing mechanism is introduced to ensure that in the event of unexpected situations (such as hardware failures, software crashes), it can automatically roll back to the previous stable state to ensure the reliability and integrity of the test.

[0039] The host computer also includes a database; the test management module, data analysis and processing module, prediction module and human-computer interaction interface module are all connected to the database, storing relevant data in the database or reading data from the database. The test management module stores the configuration parameters, test time and other information of each test in the database, and the data analysis and processing module stores the result data obtained by analysis (such as the protection device action time series, harmonic analysis results, etc.) in the database; the human-computer interaction interface module reads historical test data from the database to display historical test records; the prediction module uses the historical data in the database for model training and optimization.

[0040] The host uses message passing and event mechanisms. When the data acquisition card collects new data, a data update event is triggered. When the test management module subscribes to this event, after this event occurs, the data acquisition card passes the new data to the data analysis and processing module. When the prediction module completes a prediction update, it publishes a prediction update event. When the human-computer interaction interface module and the test management module subscribe to this event, they can obtain the latest prediction information in time and perform corresponding processing. This event-based message passing mechanism can improve the coupling between modules, making the system more flexible and easy to expand. Example 3

[0041] The test management module can grasp the status of test tasks in real time. It interacts closely with hardware (programmable current power supply, electronic load, data acquisition card, etc.) and other software modules, and can know in real time whether each test task is in the stage of data collection, execution or analysis, providing real-time basis for resource allocation. The test management module dynamically allocates resources according to the priority weight and impact degree of the test task to help achieve high-precision testing of subway DC protection devices: 1) According to the importance of the test task to the safe operation of the subway DC system, the priority weight of the test task is determined as an initial guide for resource allocation tendency. For example, the overcurrent protection test is very important and is given a high priority weight, while auxiliary tasks such as daily equipment self-inspection have a lower weight. 2) According to the depth and breadth of the impact of the test task on the subsequent test process after the test task fails, the impact degree of the test task is determined. For example, the overcurrent protection test fault troubleshooting is complex, the impact is wide, and the impact degree is large, while the simple appearance inspection has little impact and a small impact degree. 3) The test task priority weight is multiplied by the impact degree to obtain a comprehensive influence index, and more resources are allocated to test tasks with a large comprehensive influence. Taking CPU time allocation as an example, for test tasks with large comprehensive influence, such as thermal overload protection test, more computing time slices are allocated per unit time to ensure the efficient advancement of complex calculations, while test tasks with small comprehensive influence are allocated less. 4) During the test, the priority weight and influence of test tasks are continuously updated, the comprehensive influence of test tasks is dynamically calculated, and resource allocation is adjusted in real time. Based on the real-time progress of test tasks, once a secondary test task is abnormal and involves key links, its priority weight is immediately increased; if the estimated impact of the test task changes due to technical improvements, the impact value is adjusted in time to ensure that the weight reflects the actual needs of the current test task; in the event of hardware performance fluctuations such as the decrease in the accuracy of the programmable DC power supply, or external environmental changes such as the approaching compression time window of the subway operation peak, the test management module responds quickly, recalculates the comprehensive influence of the test task, and adjusts the resource allocation ratio such as CPU time and data transmission bandwidth in real time to ensure that key tasks can still be completed on time and with high quality in changing situations, and maintain system stability.

[0042] The test management module uses a reinforcement learning algorithm to build a test strategy optimization model, and regards the test process as a Markov decision process MDP, in which the state space includes the model of the tested protection device, the current test items, the hardware equipment status and historical test data; the action space covers adjusting test parameters, switching test items and controlling hardware equipment. By allowing the intelligent agent to continuously explore and learn in the MDP environment, the test strategy is optimized according to the reward function (such as improving test efficiency and reducing the misjudgment rate of the protection device), and personalized and efficient testing for different tested protection devices is achieved. The coordinated control with the hardware equipment adopts a distributed real-time system architecture based on a time-triggered protocol. Under this architecture, the test management module, as the central scheduling node of the system, sends synchronization control instructions to each hardware device (DC power supply, electronic load, acquisition card, etc.) according to a pre-defined time schedule to ensure that the operation of the entire test system is precisely synchronized in time, and the error is controlled at the microsecond level, thereby improving the reliability and repeatability of the test.

[0043] The data analysis and processing module uses generative adversarial networks and convolutional neural network algorithms for deep feature mining. The generator in the generative adversarial network is used to simulate various fault current and voltage waveforms, including complex harmonic combinations, pulse interference superposition, etc. The discriminator is built based on the convolutional neural network, and uses its powerful feature extraction capabilities to distinguish the collected real waveform from the waveform generated by the generator, and identify abnormal patterns in the waveform. Through adversarial training between the two, the recognition accuracy of waveform features in complex electromagnetic environments is continuously improved, providing a more accurate basis for the performance evaluation of protection devices.

[0044] The data analysis and processing module uses random process theory to construct a performance degradation model for the tested protection device, and regards the performance indicators of the tested protection device (such as action time and threshold accuracy) as random processes. Based on historical test data, probability distribution functions such as Weibull distribution and gamma distribution are used to fit the degradation trajectory of performance indicators. Through real-time updating and prediction of distribution parameters, the remaining life of the tested protection device is evaluated, and potential failure risks are warned in advance, providing scientific guidance for the equipment maintenance plan of the subway operation department.

[0045] The prediction module combines the fusion strategy of physical model and machine learning model with variational autoencoder (VAE) and physical constraint neural network (PCNN). VAE is used to reduce the dimension of massive historical data, extract key features, and construct low-dimensional latent space representation; PCNN is based on traditional neural network and embeds physical constraint equations derived from the circuit principle of subway power supply system, such as Kirchhoff's law and Ohm's law in discrete time domain. By jointly training VAE and PCNN, the prediction model can not only fully learn the statistical laws in historical data, but also meet the constraints of physical principles, realize high-precision prediction of voltage and current changes, and provide accurate trend prediction for testers in advance. The real-time correction of prediction error by the prediction module is based on Kalman filtering technology. The output of the prediction model is used as the prior estimate of the Kalman filter, and the actual measurement value is used as the observation value. The prediction value is updated in real time through the Kalman filtering algorithm to reduce the prediction error. At the same time, according to the change of the error covariance matrix, the parameters of the prediction model are dynamically adjusted to improve the stability and reliability of the prediction. Example 4

[0046] The instruction processing and feedback of the human-computer interaction interface module are as follows: 1) Optimize the information transmission path: adopt a direct communication architecture, remove the redundant protocol conversion layer in the middle, establish a high-speed shared memory channel between the human-computer interaction interface module and the test management module, and write directly to the shared memory area when the instruction is issued. The test management module monitors in real time and responds in seconds, directly driving the hardware action; 2) Improve the feedback mechanism: establish a unified data feedback scheduling center, and the hardware equipment and software modules send the feedback information to this center in a unified standard format. The center prioritizes the feedback of key test processes based on priority queues and multi-threaded processing technology. For example, in the thermal overload protection test, the temperature and current change data collected by the data acquisition card, the real-time resistance adjustment information of the intelligent electronic load, etc., are immediately pushed to the scheduling center, and after sorting and verification, they are pushed back to the human-computer interaction interface module in real time through a high-bandwidth, low-latency dedicated channel. This can greatly improve the completeness and clarity of the feedback information. All data are accompanied by detailed unit annotations, high-definition waveform display, and adaptive adjustment of the coordinate axis scale. It also shortens the command response time and achieves an efficient balance in the processing of operation commands, enabling testers to experience a smooth and timely interactive experience and avoid operational errors due to information delays or omissions.

[0047] The information display layout of the human-computer interaction interface module is as follows: the key information that testers are most concerned about (such as the real-time status of the protection device, emergency fault warning, etc.) is given a high visual weight and placed in the most conspicuous and easily accessible area of ​​the human-computer interaction interface module, while relatively minor information (such as historical test statistics, etc.) is given a lower visual weight and placed in the surrounding auxiliary area. At the same time, the visual weight and layout of the information are dynamically adjusted according to the test progress to ensure that testers can quickly grasp the core points at a glance at the interface, thereby improving operational efficiency and emergency response speed.

[0048] The terminal interface of the human-computer interaction interface module adopts a minimalist style, highlighting key information and operation buttons. By streamlining unnecessary functions and interface elements, core functions such as real-time data display, test status monitoring, and operation instruction sending are concentrated in a concise and clear interface, making it convenient for testers to operate quickly on site.

[0049] The real-time dynamic information panel is based on the animation transition effect and real-time data binding function of the data visualization library to realize the dynamic update of the instrument panel pointer, dial, label and other elements. For example, when the voltage amplitude changes, the instrument panel pointer can rotate to a new position with a smooth animation effect, and the numerical label on the dial is also updated accordingly, giving the tester an intuitive and smooth visual experience. The display of the progress bar adopts dynamic segment filling technology, combined with the task decomposition of the test process, and divides the entire test process into multiple subtasks. For example, each protection function test is a subtask, and each subtask is further divided into several steps, such as the preparation stage, the test stage, and the result analysis stage. The progress bar is filled in different colors according to the currently completed subtasks and steps, clearly showing the overall progress of the test and the current stage, so that the tester can understand the progress of the test. The control of the flashing indicator light is based on the event-driven programming model, which defines a series of events related to the test status, such as protection device action events, hardware failure events, data acquisition abnormal events, etc. When these events occur, the flashing logic of the corresponding indicator light is triggered. The flashing frequency and color are set according to the urgency and importance of the event. For example, the hardware fault indicator light flashes red at a high frequency, prompting the tester to handle it immediately.

[0050] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0051] In various embodiments, the hardware implementation of the technology can directly adopt existing intelligent devices, including but not limited to industrial computers, PCs, smart phones, handheld stand-alone computers, floor-standing stand-alone computers, etc. Its input device preferably adopts a screen keyboard, its data storage and calculation module adopts existing memory, calculator, controller, its internal communication module adopts existing communication ports and protocols, and its remote communication adopts existing GPRS network, World Wide Internet, etc.

[0052] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0053] In the embodiments provided by the present invention, it should be understood that the disclosed devices / terminal equipment and methods can be implemented in other ways. For example, the device / terminal equipment embodiments described above are only schematic, for example, the division of modules or units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms. The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the scheme of this embodiment.

[0054] Each functional unit in each embodiment of the present invention may be integrated into a processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of a software functional unit. If the integrated module / unit is implemented in the form of a software functional unit and sold or used as an independent product, it may be stored in a computer-readable storage medium. Based on such an understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and may also instruct the relevant hardware to complete it through a computer program. The computer program may be stored in a computer-readable storage medium, and the computer program may be executed by the processor to implement the steps of the above-mentioned method embodiments. Among them, the computer program includes computer program code, and the computer program code may be in the form of source code, object code, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium, etc.

Claims

1. An automatic test system for subway DC protection devices based on embedded systems, characterized in that: It includes a programmable DC power supply, an electronic load, a data acquisition card, a communication module and a host computer; the programmable DC power supply provides the power basis for the entire test system, and its output DC voltage and current directly act on the DC protection device of the subway under test, simulating the power supply situation in the actual operation of the subway; the electronic load works in coordination with the programmable DC power supply, and according to the test requirements, it dynamically adjusts its own resistance value to change the current drawn from the power supply, thereby simulating the load changes of the subway under different operating periods and different working conditions; the data acquisition card is simultaneously connected to the programmable DC power supply, the electronic load and the protection device under test, on the one hand, it collects the voltage and current signals output by the programmable DC power supply, and monitors in real time whether the programmable DC power supply outputs stable or dynamically changing power according to the preset working conditions, to ensure The accuracy of power supply simulation is improved. On the other hand, the current and voltage changes at the electronic load end are collected to obtain the electrical parameters during the electronic load adjustment process, so as to provide data support for analyzing the performance of the protection device under test under different load responses; the data acquisition card also collects the feedback signal of the protection device under test; the data acquisition card is connected to the host computer through the communication module, and the data acquisition card transmits the collected electrical parameter data and the key information reflecting the operating status of the protection device under test to the host computer through the communication module, and the host computer receives and processes these data and information to realize real-time monitoring and management of the test process; the host computer establishes connection with the programmable DC power supply and the electronic load through the communication module, and sends corresponding control instructions to them according to the test requirements and the preset test process.

2. The automatic test system for subway DC protection device based on embedded system according to claim 1 is characterized in that: During the overcurrent protection test, the electronic load increases its resistance according to the preset rules, causing the loop current to rise, prompting the protection device under test to enter the corresponding test state to verify its overcurrent protection function; in the thermal overload protection test, a large current load is maintained for a long time to simulate the heating process of the equipment, and together with the programmable DC power supply output, a complete test scenario is constructed.

3. The automatic test system for subway DC protection device based on embedded system according to claim 1 is characterized in that: When the data acquisition card acquires the feedback signal of the protection device under test, the sampling frequency is adaptively adjusted according to the signal characteristics.

4. The automatic test system for subway DC protection device based on embedded system according to claim 1 is characterized in that: The host computer includes a test management module, a data analysis and processing module, a prediction module and a human-computer interaction interface module; the communication module is connected to the test management module, and the test management module is respectively connected to the data analysis and processing module, the prediction module and the human-computer interaction interface module; the human-computer interaction interface module is used to provide an operation entrance for the tester to set the test parameters and operation instructions, and pass this information to the test management module. The test management module determines the control strategy for the program-controlled DC power supply and the electronic load according to the received information, and sends the control instructions to them; the test management module feeds back various status information generated during the test to the human-computer interaction interface module, and the human-computer interaction interface module updates the interface display according to the received information; during the test process, the test management module passes the data obtained from the program-controlled DC power supply and the electronic load to the data analysis and processing module. After analyzing and processing the data, the data analysis and processing module generates an analysis result and feeds it back to the test management module. The test management module adjusts the control instructions for the program-controlled DC power supply and the electronic load according to the data analysis result to optimize the test process; The test management module sends the current test parameters and historical data to the prediction module as input of the prediction model. The prediction module predicts various parameters in the test process and returns the prediction results to the test management module. The test management module optimizes the test strategy according to the prediction results and adjusts the output parameters of the programmable DC power supply and electronic load in advance.

5. The automatic test system for subway DC protection device based on embedded system according to claim 4 is characterized in that: The data analysis and processing module is connected to the prediction module, and the data analysis and processing module provides the processed data to the prediction module to help the prediction module build and adjust the prediction model; the prediction module then feeds back its prediction results to the data analysis and processing module for comparison with the actual analysis results to further optimize the data analysis and processing algorithm.

6. The subway DC protection device automatic test system based on embedded system according to claim 4 is characterized in that: The test management module dynamically allocates resources according to the priority weight and impact of the test tasks: 1) according to the importance of the test tasks to the safe operation of the subway DC system, the priority weight of the test tasks is determined as an initial guide for resource allocation tendency; 2) Determine the impact of the test task based on the depth and breadth of the impact on the subsequent test process after the test task fails; 3) Multiply the test task priority weight and the impact degree to obtain the comprehensive influence index, and allocate more resources to the test tasks with large comprehensive influence. 4) Continuously update the test task priority weight and impact degree during the test, dynamically calculate the comprehensive influence of the test task, and adjust resource allocation in real time.

7. The subway DC protection device automatic test system based on embedded system according to claim 4 is characterized in that: The test management module uses a reinforcement learning algorithm to construct a test strategy optimization model, and regards the test process as a Markov decision process MDP, in which the state space includes the model of the protection device under test, the current test item, the hardware equipment status and historical test data; the action space covers adjusting test parameters, switching test items and controlling hardware equipment. By allowing the intelligent agent to continuously explore and learn in the MDP environment, the test strategy is optimized according to the reward function, so as to achieve personalized and efficient testing for different protection devices under test.

8. The subway DC protection device automatic test system based on embedded system according to claim 4 is characterized in that: The data analysis and processing module uses a generative adversarial network and a convolutional neural network algorithm to perform deep feature mining. The generator in the generative adversarial network is used to simulate various fault current and voltage waveforms, and the discriminator is constructed based on the convolutional neural network to distinguish the collected real waveform from the waveform generated by the generator, and identify abnormal patterns in the waveform; and uses the random process theory to construct a performance degradation model of the tested protection device, and regards the performance indicators of the tested protection device as random processes. Based on historical test data, a probability distribution function is used to fit the degradation trajectory of the performance indicators. Through real-time updating and prediction of distribution parameters, the remaining life of the tested protection device is evaluated, and potential failure risks are warned in advance.

9. The automatic test system for subway DC protection device based on embedded system according to claim 4 is characterized in that: The instruction processing and feedback of the human-computer interaction interface module are as follows: 1) a high-speed shared memory channel is established between the human-computer interaction interface module and the test management module. When an instruction is issued, it is directly written into the shared memory area. The test management module monitors in real time and responds in seconds, directly driving the hardware action; 2) a unified data feedback scheduling center is established. The hardware equipment and software modules send feedback information to this center in a unified manner in a standard format. The center gives priority to key test process feedback based on priority queues and multi-threaded processing technology; the information display layout of the human-computer interaction interface module is to give high visual weight to the key information that the tester is most concerned about, and place it in the most conspicuous and easily accessible area of ​​the human-computer interaction interface module, while giving relatively minor information a lower visual weight and placing it in the surrounding auxiliary area, and dynamically adjusting the visual weight and layout of the information according to the test process.

10. The subway DC protection device automatic test system based on embedded system according to claim 4, characterized in that: The host computer adopts message transmission and event mechanism. When the data acquisition card collects new data, a data update event is triggered. When the test management module subscribes to this event, after this event occurs, the data acquisition card passes the new data to the data analysis and processing module. When the prediction module completes a prediction update, it publishes a prediction update event. When the human-computer interaction interface module and the test management module subscribe to this event, they can obtain the latest prediction information in time and perform corresponding processing.

Images