An on-line monitoring system for high-voltage motor insulation

Abstract

This invention relates to the field of detection technology, specifically disclosing an online monitoring system for high-voltage motor insulation. This system addresses problems such as delayed monitoring response, weak anti-interference capabilities, and narrow test coverage in existing systems. The system includes a monitoring center, a monitoring management module, a monitoring mode switching module, a fault simulation module, a multi-source sensor acquisition module, a signal processing module, a data fusion and analysis module, an alarm and storage module, and a communication module. This invention achieves online simulation testing of sensor channel anomalies, high loads, and customized frequency band signals through fault simulation command injection. It achieves real-time acquisition and preprocessing of broadband signals through multi-source parallel accelerated processing. Data fusion and intelligent judgment improve early warning accuracy and system robustness. Mode switching and online calibration ensure low false alarm rates and reliability throughout the testing process.

Description

Technical Field

[0001] This invention relates to the field of detection technology, and in particular to an online monitoring system for the insulation of high-voltage motors. Background Technology

[0002] Online monitoring of high-voltage motor insulation can issue early warnings when the insulation performance of high-voltage motor shows initial signs of deterioration or when partial discharge signals accumulate, thus avoiding high-voltage motor shutdown, production accidents and safety risks caused by insulation failure. In order to verify the anti-interference performance of online monitoring of high-voltage motor insulation, the existing practices are often divided into two categories: (1) Sensor performance verification. This method is mostly used in the early stage of system development. At this time, various partial discharge coupling sensors, ultrasonic sensors and high-frequency current transformers have not yet been integrated into the actual high-voltage motor, and it is difficult to obtain signals under real working conditions. Usually, in a controlled laboratory environment, standard partial discharge signals or simulated interference pulses are injected into the sensor unit under test through a pulse signal generator or broadband noise source. Then, the anti-interference ability under different interference amplitudes is evaluated by observing the deviation between the collected and expected waveforms; (2) System integration test. At this time, all sensor-level data acquisition modules are assembled on the high-voltage test bench and connected in parallel with the motor under test. Through the multi-source interference generated by the inverter start-up, switching operation and external resonant device, the monitoring system is monitored online for a long time to test its stability, false alarm rate and missed alarm rate in complex electromagnetic environment. However, both of the above verification methods rely on a controllable simulation environment in the laboratory or test bench, resulting in low testing efficiency. Furthermore, they are mainly designed for preset partial discharge or interference model signals and cannot fully cover various coupling interference conditions in the field, leading to inaccurate anti-interference performance evaluation and the tendency for false alarms or missed detections to occur during field applications. Summary of the Invention

[0003] The technical problem to be solved by this invention is to provide an online monitoring system for the insulation of high-voltage motors. By using online fault injection and multi-source parallel accelerated processing, the system improves the sensitivity and broadband anti-interference capability of online monitoring of the insulation status of high-voltage motors, reduces the false alarm rate, enables customized testing of various fault scenarios, and improves the timeliness and operational safety of online early warning of high-voltage motor insulation.

[0004] To achieve the above objectives, the present invention provides the following technical solution:

[0005] A high-voltage motor insulation online monitoring system includes a monitoring center, a monitoring management module, a monitoring mode switching module, a fault simulation module, a multi-source sensor acquisition module, a signal processing module, a data fusion analysis module, an alarm and storage module, and a communication module. The monitoring management module is communicatively connected to the monitoring module and also connected to the fault simulation module to receive and parse fault simulation commands issued by the monitoring center, and obtain the simulation type and parameters. The monitoring mode switching module switches to a recording-only mode according to the simulation type. The fault simulation module generates and injects simulated partial discharge / electromagnetic interference signals into the corresponding channels of the multi-source sensor acquisition module according to the type and parameters. The multi-source sensor acquisition module and the signal processing module work together to collect and preprocess signals. The data fusion analysis module performs feature fusion and status determination on the preprocessed signals. The alarm and storage module triggers an early warning in non-simulation mode or records and stores data in simulation mode. The communication module is used to feed back the monitoring results to the monitoring center.

[0006] As a further technical solution of the present invention, the fault simulation module constructs a simulated signal input based on the fault simulation type and related parameters, including:

[0007] The fault simulation type is either Type I or Type II. Based on the signal characteristics, signal type and amplitude in the relevant parameters, a simulated partial discharge signal / electromagnetic interference signal corresponding to the first signal sequence is constructed. Type I is the partial discharge anomaly type, and Type II is the electromagnetic interference type.

[0008] The fault simulation type is the third type, which uses the target signal parameters determined by the signal generation monitoring and management module, and constructs the simulated set frequency band signal corresponding to the second signal sequence based on the target signal parameters. The third type is the set frequency band signal type, and the target signal parameter is a signal parameter that is greater than the amplitude threshold in the relevant parameters and is the same as the corresponding signal type in the relevant parameters in the monitoring and management module.

[0009] As a further technical solution of the present invention, when the fault type is a first type or a second type, the fault simulation module constructs an analog input and output corresponding to a first signal sequence based on the signal characteristics, signal type, and amplitude parameters in the relevant parameters, including:

[0010] When the fault simulation type is the first type, the fault simulation module constructs the first signal sequence only based on the signal characteristics, signal type, and amplitude parameters;

[0011] The fault simulation module also constructs a first signal sequence with a pulse count based on the pulse count in the relevant parameters, and on the basis of signal characteristics, signal type and amplitude parameters;

[0012] The first signal sequence includes a description of the signal characteristics corresponding to at least one sensing channel.

[0013] As a further technical solution of the present invention, after determining the target signal parameters using the monitoring and management module, it further includes:

[0014] Pause the execution of analog inputs and outputs corresponding to the target signal parameters in the monitoring and management module;

[0015] After injecting analog inputs and outputs into the target monitoring channel to perform a response test on the channel, the execution of the analog inputs and outputs corresponding to the target signal parameters in the monitoring management module is resumed.

[0016] As a further technical solution of the present invention, in the fault simulation module, an analog signal input and output corresponding to the second signal sequence is constructed according to the target signal parameters, including:

[0017] If there are multiple target signal parameters, select the signal parameter corresponding to the target monitoring channel from the target signal parameters to construct the analog signal input and output corresponding to the second signal sequence.

[0018] As a further technical solution of the present invention, after parsing the monitoring and management commands to obtain the fault type and related parameters, the monitoring and management module further includes:

[0019] The relevant parameters are moved to the storage medium inside the monitoring and management module;

[0020] The fault simulation module reads relevant parameters from the internal storage medium of the monitoring and management module, and constructs analog signal input and output based on the fault simulation type and relevant parameters.

[0021] As a further technical solution of the present invention, the monitoring and management module encodes the fault type and related parameters into the corresponding fields of the extended naming format of the high-voltage motor monitoring protocol to generate monitoring and management commands; it parses the monitoring and management commands according to the fault simulation definition table to obtain the fault simulation type and related parameters. The fault simulation definition table includes an operation code field representing the fault simulation type and a related parameter field. The related parameter field includes signal description, signal type, number of pulses, signal amplitude, frequency band range, sensor channel identifier, and expected response effect. After injecting the simulated signal into the target monitoring channel to perform a response test on the target monitoring channel, it obtains the actual response of the target monitoring channel and compares the actual response with the expected response result in the related parameters.

[0022] As a further technical solution of the present invention, the monitoring task instruction issued by the monitoring center is obtained, and the execution unit is determined according to the monitoring task instruction. The execution unit is the monitoring management module. The monitoring task instruction is sent to the monitoring management module. The monitoring management module splits the monitoring task instruction and sends the split sub-instructions to the corresponding fault simulation module / multi-source sensor acquisition module for test execution. If the execution unit is the multi-source sensor acquisition module, the monitoring task is directly sent to the corresponding multi-source sensor acquisition module for test execution.

[0023] As a further technical solution of the present invention, the monitoring center is used to send monitoring and management commands to the system and obtain monitoring results.

[0024] As a further technical solution of the present invention, the monitoring center is also used to: acquire the test result set of each test channel, the test result set including the result subsets corresponding to each of the multiple fault simulation types; draw the system frequency response Bode plot for each fault type according to the test frequency corresponding to each test result in each result subset; determine the performance score for each fault simulation type based on the gain and phase margin of the system frequency response Bode plot; determine the performance of each monitoring channel based on the performance scores corresponding to multiple fault types; the performance score is a comprehensive quantitative index obtained by normalizing the gain margin, phase margin, and -3dB bandwidth of the system frequency response Bode plot and then weighting and summing them according to preset weights; the preset weights are calculated by a statistical learning model based on the relative contributions of gain margin, phase margin, and -3dB bandwidth to fault detection accuracy and safety risk, combined with historical fault data analysis and expert evaluation.

[0025] Technical advantages of the online insulation monitoring system for high-voltage motors of the present invention:

[0026] This invention issues and parses fault simulation commands sent by the monitoring center through a monitoring management module. The fault simulation module generates and injects controllable partial discharge or electromagnetic interference signals into the corresponding channels of the multi-source sensor acquisition module according to the commands. The monitoring mode switching module switches to a recording-only mode to avoid false alarms. The signal processing module and the data fusion analysis module work together to filter, preprocess, fuse features, and determine the status of the acquired signals. The alarm and storage module triggers early warnings in non-simulation mode or records and stores data in simulation mode. The communication module feeds back the monitoring results to the monitoring center in real time. The monitoring center automatically calculates performance scores and evaluates the performance of each monitoring channel based on the Bode plot parameters of the system frequency response under different fault simulation types. This invention enables online monitoring and testing of early degradation of the insulation state of high-voltage motor windings and abnormal partial discharge, improves the efficiency and accuracy of anti-interference performance evaluation, and overcomes the shortcomings of low testing efficiency, narrow coverage, and easy omissions and false alarms in existing simulation environments. It provides reliable technical support for the intelligent operation and maintenance and safe and stable operation of high-voltage motors, and provides key data support for online monitoring and fault prediction of high-voltage motor insulation. Attached Figure Description

[0027] Figure 1 This is a system block diagram of the present invention;

[0028] Figure 2 This is a flowchart illustrating the current testing status of online insulation detection for high-voltage motors in accordance with the prior art of this invention.

[0029] Figure 3 An improved flowchart for online monitoring and fault testing of high-voltage motor insulation according to the present invention;

[0030] Figure 4 This is a schematic diagram of the task chain of the present invention;

[0031] Figure 5 Flowchart of a method for testing the system provided by the present invention;

[0032] Figure 6 This is a description of the signal processing and a schematic diagram of the corresponding parameter analysis of the present invention;

[0033] Figure 7 This is a Bode plot of the magnitude response of the test group system in this invention.

[0034] Figure 8 This is a Bode plot of the phase response of the test group system of this invention. Detailed Implementation

[0035] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0036] In existing technologies, the test workstations in the system integration testing phase verify the functionality from the user's perspective. To ensure that the test scenario is consistent with the application scenario of high-voltage motor insulation monitoring, standard partial discharge and electromagnetic interference signals representing the operating conditions of high-voltage motor insulation monitoring are typically injected into the system using test tools. The simulated signals sent out are mostly fault models under normal user maintenance scenarios. The entire test is a closed-loop environment, making it difficult to inject customized scenarios. This testing method also has the following problems: it is difficult to trigger the real response of multi-source sensor acquisition and signal processing under abnormal conditions such as initial insulation degradation or multi-channel interference superposition; it is difficult to make all or some channels withstand high-load scenarios such as high-frequency bandwidth pulses and frequency conversion interference for online verification; especially in the bench simulation environment, the overall simulation speed is slow, and the various components of the system are in a low-load state for a long time. After the test, it is impossible to determine whether the alarm, recording and processing logic of the system under high load and high interference conditions in the field meets the expected design goals; it is difficult to test the performance of the system under special input conditions such as non-aligned frequency bands and phase mismatch.

[0037] In existing technologies, the testing work in the sensor performance verification stage requires a lot of bench piling and signal simulation to make a single channel work normally, including preset partial discharge models and noise modes. This testing method differs a lot from the actual field environment. Similar to front-end simulation verification, it is difficult to directly evaluate whether the overall performance of multi-source sensor acquisition, signal processing and data fusion analysis meets expectations in real operating scenarios.

[0038] Online monitoring of high-voltage motor insulation generally uses multiple monitoring units to process sensor signals in parallel to accelerate real-time monitoring. Specifically, for a certain insulation status detection, the processing is generally divided into signal acquisition, filtering and gain adjustment, feature extraction, data fusion, status determination, early warning triggering and result storage. Each step is one or more monitoring units.

[0039] To address the shortcomings of the existing technologies described above and the characteristics of online monitoring of high-voltage motor insulation, this invention proposes an online monitoring system for high-voltage motor insulation. By autonomously issuing fault simulation commands within the monitoring management module, the fault simulation module injects customizable partial discharge and electromagnetic interference simulation signals into the multi-source sensor acquisition channels, thereby realizing a customized online monitoring system for high-voltage motor insulation that is geared towards the development stage.

[0040] For details, see Figure 1 , Figure 1 This is a system block diagram of an online monitoring system for high-voltage motor insulation provided in an embodiment of the present invention. It includes a monitoring center, a monitoring management module, a monitoring mode switching module, a fault simulation module, a multi-source sensor acquisition module, a signal processing module, a data fusion and analysis module, an alarm and storage module, and a communication module. Solid lines represent control-level interactions, and dashed lines represent monitoring data interactions. The functional modules and their descriptions in the system involved in this invention are as follows:

[0041] The monitoring center is used to issue fault commands to the monitoring and management module and receive monitoring results from the system.

[0042] The monitoring management module communicates with the monitoring module and connects with the fault simulation module to receive and parse fault simulation commands issued by the monitoring center and obtain the simulation type and parameters.

[0043] The monitoring mode switching module switches to the recording-only mode according to the simulation type, and the fault simulation module generates and injects simulated partial discharge / electromagnetic interference signals into the corresponding channel of the multi-source sensor acquisition module according to the type and parameters.

[0044] The multi-source sensor acquisition module and the signal processing module work together to collect and preprocess signals;

[0045] The data fusion and analysis module performs feature fusion and state determination on the preprocessed signal;

[0046] The alarm and storage module triggers warnings in non-analog mode or records and stores data in analog mode, while the communication module is used to send monitoring results back to the monitoring center.

[0047] like Figure 2 As shown, in the existing technology, the monitoring center issues monitoring instructions according to... Figure 2 The process execution involves two steps: Step 1 and Step 2, which are management commands executed at the control level for system configuration and mode switching. These will not be elaborated on here.

[0048] Step 3: The monitoring center issues monitoring tasks;

[0049] Step 4: The signal distribution module splits or routes the task according to the characteristics of the monitoring task instruction or the configuration of the online monitoring system. Monitoring subtasks that can be fully processed by the hardware monitoring engine are directly routed to the hardware monitoring engine.

[0050] Step 5: For monitoring subtasks that still need to be processed by the internal engine of the monitoring management module, route them to the internal engine of the monitoring management module.

[0051] Step 6: If the monitoring subtask processed by the internal engine of the monitoring management module still meets the hardware acceleration conditions, it will continue to be routed to the hardware monitoring engine.

[0052] Step 7: If the monitoring subtask previously routed to the hardware monitoring engine in step 4 encounters complex abnormal signals or scenarios beyond the capabilities of the hardware engine during hardware processing, it will be routed again to the internal engine of the monitoring management module for processing.

[0053] It should be noted that the principle of the system proposed in this invention is as follows: Figure 3 As shown, the following four steps are added to the above process:

[0054] (1) After step 3, after the monitoring and management module parses the "fault injection start" test management command, it first calls the fault simulation module to construct the required simulation signal;

[0055] (2) Before the signal distribution in step 4, the analog signal is injected into the corresponding multi-source sensor acquisition channel to replace or superimpose the conventional monitoring task signal;

[0056] (3) Between step 5 and step 6, automatically switch to recording mode according to the current simulation type to ensure that only simulation phase signals are recorded and no alarms are triggered during the processing of the hardware monitoring engine or the internal engine of the monitoring management module;

[0057] (4) After step 7, when all sub-task tests are completed and data is collected, the monitoring management module triggers a fault injection stop command to restore the normal monitoring mode and compare the actual response with the expected result to complete a round of fault simulation test verification.

[0058] Step 8: The monitoring and management module routes the fault simulation management command issued by the monitoring center to the fault simulation module;

[0059] Step 9: After the fault simulation module switches to the recording-only mode, it generates and injects simulated partial discharge or electromagnetic interference signals into the corresponding channel of the multi-source sensor acquisition module based on the fault simulation type and related parameters.

[0060] Step 10: When the fault simulation module needs to use real-time acquired signals for simulation, the monitoring and management module suspends the normal data processing of the corresponding channel. The multi-source sensor acquisition module provides the current real sensing signal. The fault simulation module constructs and injects the simulation signal into the channel based on the acquired real-time signal. After the simulation is completed, the fault simulation module notifies the monitoring and management module to resume real-time data processing and returns the processing result of the channel to the data fusion and analysis module.

[0061] All signals generated by the fault simulation module are directly sent to the corresponding channels of the multi-source sensor acquisition module to verify the response capability and processing flow of the hardware monitoring engine.

[0062] It should be specifically noted that in the system of this invention, the main function of the multi-source sensor acquisition module is to perform parallel acceleration of online processing of the acquired multi-channel sensor signals. To achieve the acceleration goal, the signal processing flow of a single monitoring task needs to be broken down into multiple steps, and each step is executed by a different hardware monitoring engine, thereby achieving parallel processing. Signal processing based on online monitoring algorithms is generally broken down into multiple task descriptions, corresponding to signal filtering engines, gain adjustment engines, spectrum analysis engines, and feature extraction engines, respectively. These different multi-task descriptions corresponding to the same monitoring task together form a task chain. When the task chain in the multi-task description is executed sequentially, the entire monitoring task is completed. Each hardware monitoring engine only perceives the task chain it receives and is not aware of the overall structure of the multi-task description. Figure 4 As shown, Figure 4 This is a general illustration of a task chain in an embodiment of the present invention, which includes the sequential execution of multiple task descriptions. It is for illustrative purposes only and does not represent the specific design of task descriptions in an actual system.

[0063] like Figure 5 As shown, the method for testing the system provided by this invention includes:

[0064] Step S1: During the process of monitoring the insulation status of the high-voltage motor based on the online monitoring program, the monitoring management command sent by the monitoring center is obtained. The monitoring management command is the monitoring behavior management command injected by the fault.

[0065] In this embodiment, the online monitoring program is a self-developed monitoring script or an existing monitoring tool, used to continuously collect multi-source sensor signals; the monitoring management command is actively issued under the high-voltage motor operating environment to simulate or trigger specific fault scenarios in order to detect the response of each sensor channel and signal processing module under abnormal conditions. The monitoring center sends the command to the monitoring management module through the communication module and can perform format and content verification after receiving it.

[0066] Step S2: The monitoring and management module parses the monitoring and management command to obtain the fault simulation type and the corresponding relevant parameters, and switches the monitoring mode according to the fault simulation type. The switched recording mode is used to indicate that when an abnormality is encountered in the test, only the signal is recorded and no warning is triggered.

[0067] In this embodiment, the fault simulation types include: Type 1 (sensor hardware malfunction), Type 2 (high load operation), and Type 3 (specific frequency band signal); the corresponding parameters include: channel identifier, signal characteristics, number of pulses, amplitude threshold, and frequency band range; the error handling mode is defined as "only record without processing when encountering an anomaly"; the monitoring management module parses the command field by field through the predefined fault definition table and completes the dynamic switching of the monitoring mode configuration.

[0068] Step S3: Based on the fault simulation type and the relevant parameters, construct the simulated signal input and output; inject the simulated signal into the target monitoring channel to test the response behavior of the target monitoring channel, wherein the target monitoring channel is the sensing channel specified in the relevant parameters.

[0069] In this embodiment, the fault simulation module generates simulated partial discharge or electromagnetic interference signals based on the parsed type and parameters. This signal, as a signal processing chain, contains several signal processing descriptions and is sequentially routed to hardware monitoring engine modules such as the filtering engine, gain engine, and spectrum engine. After injection, the actual response data of each channel can be obtained to determine whether the monitoring function indicators are unaffected, whether the performance changes meet expectations, whether the channel processing quality and anomaly reporting meet the design goals, and to verify the overall performance and robustness of the system under specific fault scenarios.

[0070] One implementation of this invention is as follows: S3, the fault simulation module constructs analog signal input and output based on the fault simulation type and related parameters, including:

[0071] If the fault simulation type is the first type or the second type, then the analog signal input and output corresponding to the first signal processing chain are constructed according to the signal description, signal type and amplitude parameters; where the first type represents the sensor channel abnormality type and the second type represents the channel high load type.

[0072] If the fault simulation type is the third type, the target signal parameters are determined by the engine submodule inside the monitoring and management module, and the analog signal input and output corresponding to the second signal processing chain are constructed according to the target signal parameters; wherein, the third type represents a specific frequency band signal type, and the target signal parameters are signal parameters in the engine submodule inside the monitoring and management module that are greater than the amplitude threshold in the relevant parameters and have the same signal type.

[0073] In this embodiment, the first type is used to simulate sensor channel anomalies. If the fault simulation type is the first type, then the simulated signal input and output corresponding to the first signal processing chain are constructed according to the signal description, signal type, and amplitude parameters. The first signal processing chain includes signal processing descriptions corresponding to at least one sensor channel, enabling the setting and simulation of abnormal tasks for multiple sensor channels when constructing the simulated signal input and output. This allows for a comprehensive test of the collaborative capabilities of each channel and the system response under specific fault scenarios, achieving more realistic and reliable multi-channel fault injection testing.

[0074] Sensor channel anomalies typically occur when a situation is encountered that does not conform to design expectations during the parsing of the signal processing description or the data pointer specified therein. This anomaly may trigger an interrupt notification to the internal engine of the monitoring and management module to handle the anomaly, or it may trigger an event notification to the internal engine to perform anomaly correction.

[0075] For the first type of analog signal input and output, the first signal processing chain is as follows: Figure 6 As shown, Figure 6 This is a schematic diagram of a signal simulation for a simulated sensing channel provided in an embodiment of the present invention. This signal processing chain consists of a simulated channel identifier, signal processing description 1, and parameter data transmitted by the fault simulation module based on the simulation command issued by the monitoring and management module. Alternatively, the fault simulation module can autonomously create the signal processing description and parameter data to generate a signal processing task description, which can trigger anomaly reporting or correction processes. If the reported anomaly matches the expectation, it indicates that the anomaly handling function of the hardware monitoring engine has passed the test and verification in a real operating environment. Figure 6 This is merely an example. Depending on the specific testing objectives or monitoring engine, the task chain composition can be expanded as needed. For example, it may sequentially include signal processing description 1, signal processing description 2 and its corresponding parameter 1 and corresponding parameter 2. Signal processing description 1 points to hardware monitoring engine 1, and signal processing description 2 points to monitoring engine 2. This allows monitoring engine 1 to parse and execute signal processing task 1 / related parameter 1 after receiving the task description and trigger an exception. Engine 2 can parse and execute signal processing task 2 / related parameter 2 and trigger an exception. This embodiment will not elaborate further.

[0076] After injecting the analog signal processing chain into the target monitoring channel to perform response testing on the hardware monitoring engine module, the method further includes: after all signal processing descriptions in the signal processing chain have been executed sequentially, the signal simulation module notifies the internal engine submodule of the monitoring management module to resume processing of the original real-time monitoring signal; specifically, the analog signal processing chain is equivalent to the test signal injected in the normal monitoring process. After execution, if the function and performance of the normal monitoring process are not affected or only fluctuate within the expected range, the test is considered passed; in this embodiment of the invention, by suspending the processing of real-time signals by the internal engine submodule of the monitoring management module during the injection test phase, it can be ensured that the monitoring management module will not interfere with the analog test process, allowing the test to focus on the impact of the analog signal on the target channel; resuming the execution of the engine submodule after the test ensures that the monitoring management module can continue to perform signal processing based on the monitoring tasks issued by the current online monitoring program, ensuring the stability and accuracy of the test environment, thereby improving the reliability of the test results.

[0077] In this embodiment of the invention, the fault simulation module parsing process includes: the fault simulation module first parses the monitoring management command (extended command format), and then moves the fault simulation type and its related parameters to the internal storage unit of the monitoring management module through the interaction interface between the monitoring system and the monitoring center. It can then move signal type parameters and amplitude thresholds to this storage unit to build the conditions for generating the simulated signal processing chain for the signal simulation module. Furthermore, the fault simulation module can verify the integrity of the signal processing chain and parameter data, and set a signal processing task completion flag and anomaly handling flag, which are then executed by the signal simulation module. The signal processing chain runs in the hardware monitoring engine, and upon completion, it produces two results: either an anomaly report is triggered, or the signal processing task description ends normally. Therefore, it needs to be pre-configured so that when an anomaly is reported, the corresponding anomaly handling logic is executed, and when the signal processing chain ends, subsequent recovery operations are performed and test results are output.

[0078] Specifically, after parsing the monitoring and management commands to obtain the fault simulation type and related parameters, the process further includes: moving the related parameters to the internal storage medium of the monitoring and management module; correspondingly, constructing simulated signal input and output based on the fault simulation type and related parameters, including: reading the related parameters from the internal storage medium and constructing a signal processing task description based on the fault simulation type and parameters. In this embodiment, moving the parameters to the internal storage medium reduces communication latency, and subsequently reading the parameters directly from the internal storage to construct the simulated signal improves efficiency; moreover, the internal storage environment is more stable, reducing the impact of external interference on the parameters and ensuring the accuracy and integrity of the parameters.

[0079] Specifically, after parsing the monitoring and management commands to obtain the fault simulation type and related parameters, the process also includes:

[0080] The relevant parameters are moved to the internal storage medium of the monitoring and management module; accordingly, analog signal inputs and outputs are constructed based on the fault simulation type and relevant parameters, including:

[0081] Relevant parameters are read from the internal storage medium of the monitoring and management module, and an analog signal processing chain is constructed based on the fault simulation type and relevant parameters.

[0082] In this embodiment, after moving the relevant parameters to the internal storage unit, the method further includes:

[0083] If the relevant parameters include signal processing descriptions and parameter data, then perform an integrity check on the signal processing descriptions and parameter data;

[0084] Accordingly, the monitoring mode is switched according to the fault simulation type, including:

[0085] If both the signal processing description and parameter data pass the integrity check, then switch to "Record Only Mode" or "Warning Mode" depending on the fault simulation type.

[0086] In this embodiment, the integrity check signal processing description and parameter data contain all elements in terms of content, structure and format. Existing verification algorithms can be used to verify their hash values ​​to ensure that the parameters are not tampered with or lost during transmission and storage.

[0087] Only after the signal processing description and parameter data have passed the integrity check will the system switch the monitoring mode and construct the signal processing chain, thereby ensuring that the parameters on which the subsequent analog signal generation and injection are based are accurate and complete, avoiding test deviations or failures caused by parameter errors or missing parameters, and improving the reliability of the entire verification process.

[0088] In this embodiment, the monitoring and management commands are generated by the monitoring center through the online monitoring protocol to extend the command fields. When the monitoring and management module parses the commands, it extracts the fault simulation type and parameters field by field according to the fault simulation definition table (including operation code field, signal processing description, parameter data field, etc.) to achieve accurate parsing and execution of the fault simulation commands.

[0089] In one possible implementation of this invention, after injecting the analog signal processing chain into the target monitoring channel to perform response testing on the hardware monitoring engine module, the method further includes:

[0090] Obtain the execution results of the hardware monitoring engine module;

[0091] The execution result is compared with the expected response result defined by the relevant parameters in the fault simulation command.

[0092] Specifically, after all signal processing descriptions in the analog signal processing chain have been executed sequentially, the signal simulation module notifies the internal engine submodule of the monitoring and management module to resume processing of the original real-time acquired signal. It can be understood that the analog signal processing chain is equivalent to a test signal injected into the normal monitoring process. After its execution, if the functionality and performance of the normal monitoring process are not affected or only fluctuate within the expected range, the test is considered passed.

[0093] As can be seen, in this embodiment of the invention, by pausing the processing of real-time signals by the engine submodule within the monitoring and management module during the execution phase of the analog signal processing chain, it can be ensured that the engine will not interfere with the simulation test process, allowing the test to focus on the impact of the analog signal processing chain on the hardware monitoring engine module; after the test is completed, resuming the execution of the engine submodule ensures that the monitoring and management module can continue to perform signal processing based on the monitoring tasks issued by the current online monitoring program, ensuring the stability and accuracy of the test environment, thereby improving the reliability of the test results.

[0094] In addition, the system mentioned in the embodiments of the present invention also includes a monitoring center, which is used to send monitoring and management commands and obtain monitoring results.

[0095] In one feasible approach, the monitoring center is further configured to: acquire a test result set for each monitoring channel, which includes result subsets corresponding to multiple fault simulation types; plot a system frequency response curve for each fault simulation type according to the test time of each test result in the result subset; determine a performance score for each fault simulation type based on the key parameters of the system frequency response curve, such as gain margin, phase margin, and -3dB bandwidth, according to a preset score calculation rule; and determine the comprehensive performance score of each monitoring channel by weighted or averaged calculation based on the performance scores corresponding to multiple fault simulation types, so as to intuitively evaluate the channel's response capability and anti-interference performance.

[0096] Specifically, the same monitoring channel can perform multiple tests on various fault simulation types to obtain a subset of results for each type. For a specific fault simulation type, a Bode plot of the system frequency response is drawn for each fault simulation type according to the test frequency corresponding to each test result in each subset. Based on the gain margin and phase margin in each Bode plot, the gain margin, phase margin, and -3dB bandwidth are normalized and weighted and summed according to preset weights to calculate the performance score for that fault type. Finally, the comprehensive performance score of each monitoring channel is determined by weighting or averaging the performance scores of each fault type.

[0097] The preset weights are determined by the relative contributions of each key parameter (gain margin, phase margin, -3dB bandwidth) to the accuracy of fault detection and safety risks. They are calculated through a statistical learning model in combination with historical fault data analysis and expert evaluation, providing a quantitative basis for system optimization and operation and maintenance.

[0098] The following is a setting example of an actual test group, and taking the "channel high load" fault simulation type as an example, its system frequency response Bode diagram is drawn to illustrate the specific implementation of the above technical solution.

[0099] The test group settings are shown in Table 1:

[0100] Table 1 Test Group Setting Parameters

[0101]

[0102] Channel 2 was selected, and the amplitude and phase frequency responses of the hardware under high load were tested by increasing the injected signal load (80%) and repeating the injection 10 times, covering the 100Hz-100kHz frequency band, as follows:

[0103] from Figure 7 (Amplitude response Bode plot) It can be seen that under the "high channel load" scenario, the monitoring channel of this system maintains good broadband response characteristics to the input signal: in the frequency band from DC to about 1kHz, the amplitude-frequency curve is basically flat and the gain is close to 0dB, indicating that the high load injection does not cause significant gain attenuation; the -3dB bandwidth is about 1kHz, covering the typical frequency band of most partial discharge and electromagnetic interference signals, which can ensure high sensitivity detection of fault signals;

[0104] from Figure 8 (Phase response Bode plot) shows that at the gain crossover frequency (about 1 kHz), the phase is about -45°, and the phase margin is about 45°-50°, indicating that the system still has sufficient phase margin and good dynamic stability under high load injection. As the frequency increases, the phase change is smooth, without abrupt changes or excessive delay, indicating that the signal processing chain can still maintain synchronization and timing consistency under high frequency interference conditions.

[0105] Therefore, this technical solution, through fault simulation injection and multi-source parallel acceleration processing, not only ensures the broadband response capability of the monitoring channel under high load testing, but also... Figure 7 It can also maintain sufficient phase margin and system stability. Figure 8 This enables high-precision, low-latency, stable and reliable online monitoring under real high-load fault scenarios, effectively improving the system's early warning capability for insulation degradation and partial discharge of high-voltage motors.

[0106] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0107] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A high-voltage motor insulation online monitoring system, characterized in that, It includes a monitoring center, a monitoring management module, a monitoring mode switching module, a fault simulation module, a multi-source sensor acquisition module, a signal processing module, a data fusion analysis module, an alarm and storage module, and a communication module. The monitoring management module communicates with the monitoring module and is connected to the fault simulation module to receive and parse the fault simulation commands issued by the monitoring center and obtain the simulation type and parameters. After the monitoring and management module parses the "fault injection start" test management command, it first calls the fault simulation module to construct the required simulated signal. Before signal distribution, the simulated signal is injected into the corresponding multi-source sensor acquisition channel to replace or superimpose the conventional monitoring task signal. The signal distribution module splits or routes the task according to the characteristics of the monitoring task instruction or the configuration of the online monitoring system. Monitoring subtasks that can be fully processed by the hardware monitoring engine are directly routed to the hardware monitoring engine. Monitoring subtasks that still need to be processed by the internal engine of the monitoring and management module are routed to the internal engine of the monitoring and management module. The monitoring mode switching module switches to the recording-only mode according to the simulation type to ensure that during the processing of the hardware monitoring engine or the internal engine of the monitoring and management module, only the simulation phase signal is recorded and no alarm is triggered. If a monitoring subtask processed by the monitoring management module's internal engine still meets the hardware acceleration conditions, it will continue to be routed to the hardware monitoring engine. If a monitoring subtask routed to the hardware monitoring engine encounters complex abnormal signals or scenarios that exceed the capabilities of the hardware engine during hardware processing, it will be routed back to the monitoring management module's internal engine for processing. Once all subtask tests are completed and data is collected, the monitoring and management module triggers a fault injection stop command to restore normal monitoring mode and compare the actual response with the expected result, thus completing a round of fault simulation test verification. The fault simulation module generates and injects simulated partial discharge / electromagnetic interference signals into the corresponding channel of the multi-source sensor acquisition module according to the type and parameters. The multi-source sensor acquisition module and the signal processing module work together to collect and preprocess the signals. The data fusion and analysis module performs feature fusion and status determination on the preprocessed signals. The alarm and storage module triggers early warning in non-analog mode or records and stores in analog mode. The communication module is used to feed back the monitoring results to the monitoring center. The fault simulation module constructs simulated signal inputs based on the fault simulation type and relevant parameters, including: The fault simulation type is either Type I or Type II. Based on the signal characteristics, signal type and amplitude in the relevant parameters, a simulated partial discharge signal / electromagnetic interference signal corresponding to the first signal sequence is constructed. Type I is the partial discharge anomaly type, and Type II is the electromagnetic interference type. The fault simulation type is the third type, which uses the target signal parameters determined by the signal generation monitoring and management module, and constructs the simulated set frequency band signal corresponding to the second signal sequence according to the target signal parameters. The third type is the set frequency band signal type, and the target signal parameter is a signal parameter that is greater than the amplitude threshold in the relevant parameters and is the same as the corresponding signal type in the relevant parameters in the monitoring and management module. After determining the target signal parameters using the monitoring and management module, the following is also included: Pause the execution of analog inputs and outputs corresponding to the target signal parameters in the monitoring and management module; After injecting analog inputs and outputs into the target monitoring channel to perform a response test on the channel, the execution of the analog inputs and outputs corresponding to the target signal parameters in the monitoring management module is resumed; After parsing the monitoring and management commands to obtain the fault type and related parameters, the monitoring and management module also includes: The relevant parameters are moved to the storage medium inside the monitoring and management module; The fault simulation module reads relevant parameters from the internal storage medium of the monitoring and management module, and constructs simulated signal input and output according to the fault simulation type and relevant parameters; The monitoring and management module encodes the fault type and related parameters into the corresponding fields of the extended naming format of the high-voltage motor monitoring protocol to generate monitoring and management commands. It parses the monitoring and management commands according to the fault simulation definition table to obtain the fault simulation type and related parameters. The fault simulation definition table includes an operation code field representing the fault simulation type and related parameter fields. The related parameter fields include signal description, signal type, number of pulses, signal amplitude, frequency band range, sensor channel identifier, and expected response effect. After injecting the simulated signal into the target monitoring channel to perform a response test on the target monitoring channel, the actual response of the target monitoring channel is obtained, and the actual response is compared with the expected response result in the related parameters. The monitoring center is also used for: acquiring the test result set for each test channel, which includes result subsets corresponding to multiple fault simulation types; drawing a Bode plot of the system frequency response for each fault type according to the test frequency corresponding to each test result in each result subset; determining the performance score for each fault simulation type based on the gain and phase margin of the system frequency response Bode plot; determining the performance of each monitoring channel based on the performance scores corresponding to multiple fault types; the performance score is a comprehensive quantitative index calculated by normalizing the gain margin, phase margin, and -3dB bandwidth of the system frequency response Bode plot and then weighting and summing them according to preset weights; the preset weights are calculated by a statistical learning model based on the relative contributions of gain margin, phase margin, and -3dB bandwidth to fault detection accuracy and safety risk, combined with historical fault data analysis and expert evaluation.

2. The online monitoring system for high-voltage motor insulation according to claim 1, characterized in that, When the fault type is type one or type two, the fault simulation module constructs analog input and output corresponding to the first signal sequence based on the signal characteristics, signal type, and amplitude parameters in the relevant parameters, including: When the fault simulation type is the first type, the fault simulation module constructs the first signal sequence only based on the signal characteristics, signal type, and amplitude parameters; The fault simulation module also constructs a first signal sequence with a pulse count based on the pulse count in the relevant parameters, and on the basis of signal characteristics, signal type and amplitude parameters; The first signal sequence includes a description of the signal characteristics corresponding to at least one sensing channel.

3. The online monitoring system for high-voltage motor insulation according to claim 1, characterized in that, In the fault simulation module, based on the target signal parameters, the analog signal input and output corresponding to the second signal sequence are constructed, including: If there are multiple target signal parameters, select the signal parameter corresponding to the target monitoring channel from the target signal parameters to construct the analog signal input and output corresponding to the second signal sequence.

4. The online monitoring system for high-voltage motor insulation according to claim 1, characterized in that, The system obtains monitoring task instructions issued by the monitoring center, determines the execution unit based on the instructions, and if the execution unit is the monitoring management module, it sends the monitoring task instructions to the monitoring management module. The monitoring management module then breaks down the monitoring task instructions and sends the sub-instructions to the corresponding fault simulation module / multi-source sensor acquisition module for testing. If the execution unit is the multi-source sensor acquisition module, the monitoring task is directly sent to the corresponding multi-source sensor acquisition module for testing.

5. The online monitoring system for high-voltage motor insulation according to claim 1, characterized in that, The monitoring center is used to send monitoring and management commands to the system and obtain monitoring results.

Images