Relay fault detection method and device, electronic device, and storage medium

Abstract

The application provides a relay fault detection method and device, electronic equipment and storage medium, and belongs to the technical field of fault detection. The method comprises the following steps: acquiring electrical data, action time data, shell temperature data and environmental humidity data collected by a relay checker; the electrical data, the action time data, the shell temperature data and the environmental humidity data are respectively preprocessed to obtain a standard feature data set; the electrical data comprises power, current and voltage, and the action time data comprises attraction time and release time; abnormal mark data is determined based on the standard feature data set and rated reference data; correlation feature analysis is performed on the standard feature data set to obtain correlation abnormal mark data; fault risk data is obtained based on the abnormal mark data and the correlation abnormal mark data; and a fault determination result is obtained based on the fault risk data. The application can improve the accuracy of relay fault detection.

Description

Technical Field

[0001] This application belongs to the field of fault detection technology, and more specifically, relates to a relay fault detection method and device, electronic equipment, and storage medium. Background Technology

[0002] Relays are core components in power systems, industrial control, and power supply equipment, performing on / off control and fault protection. Their operational reliability directly affects the overall safety and stability of the system. Incomplete phase relays, intermediate relays, and protection output relays are prone to abnormalities such as coil aging, contact oxidation, mechanical jamming, and parameter drift during long-term operation, leading to faults such as failure to operate, malfunction, and overheating. Therefore, periodic condition monitoring and fault diagnosis of relays are essential to ensuring reliable equipment operation.

[0003] Existing detection technologies are prone to missed or false detections when relays are in the early stages of potential faults or when parameters are slightly drifting. They are difficult to predict potential faults in advance, resulting in insufficient accuracy of detection results and failing to meet the actual needs of high-precision and high-reliability fault detection for relays in the field. Summary of the Invention

[0004] To address the aforementioned technical problems, this application provides a relay fault detection method and apparatus, electronic device, and storage medium to solve the problem of insufficient accuracy in the detection results of existing relay fault detection technologies.

[0005] The embodiments of this application disclose the following technical solutions: Firstly, a relay fault detection method is provided, including: The system acquires electrical data, actuation time data, housing temperature data, and ambient humidity data collected by the relay calibrator. It then preprocesses these data separately to obtain a standard feature dataset. The electrical data includes power, current, and voltage, while the actuation time data includes pull-in and release times. Anomaly labeling data is determined based on standard feature datasets and baseline data. Correlation feature analysis is then performed on the standard feature datasets to obtain correlated anomaly labeling data. Fault risk data is obtained based on anomaly marker data and associated anomaly marker data, and fault determination results are obtained based on fault risk data.

[0006] Secondly, a relay fault detection device is provided, comprising: The data processing module is used to acquire electrical data, actuation time data, housing temperature data, and ambient humidity data collected by the relay calibrator. It preprocesses the electrical data, actuation time data, housing temperature data, and ambient humidity data to obtain a standard feature dataset. The electrical data includes power, current, and voltage, and the actuation time data includes pull-in time and release time. The anomaly analysis module is used to determine anomaly marker data based on standard feature datasets and baseline data, and to perform correlation feature analysis on the standard feature datasets to obtain correlated anomaly marker data. The fault determination module is used to obtain fault risk data based on anomaly marker data and associated anomaly marker data, and to obtain fault determination results based on fault risk data.

[0007] Thirdly, embodiments of this application also provide an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the relay fault detection method provided in any possible implementation of the first aspect.

[0008] Fourthly, embodiments of this application also provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the relay fault detection method provided by any possible implementation of the first aspect.

[0009] The beneficial effects of the technical solution provided in this application are as follows: Compared with related technologies, the relay fault detection method, apparatus, electronic device, and storage medium provided in this application are as follows: Early relay failures and minor parameter drifts often do not lead to significant anomalies in a single parameter. Existing technologies, lacking comprehensive monitoring and correlation analysis, cannot capture such subtle anomalies, leading to missed or false diagnoses. This method, however, acquires multi-dimensional data from a relay calibrator, including electrical parameters, operating time, housing temperature, and ambient humidity, covering the core operational status information of the relay and avoiding the limitations of single-dimensional monitoring. Furthermore, the embodiments of this application preprocess multiple types of data to ensure accuracy, providing a reliable foundation for subsequent judgment.

[0010] Meanwhile, this application's embodiments not only determine anomalies based on standard feature datasets and rated benchmark data, but also perform correlation feature analysis on the data. This allows for the identification of both obvious anomalies in single parameters and subtle, coordinated anomalies between different parameters, accurately identifying early potential faults and minor parameter drifts. This effectively avoids missed or false diagnoses, enabling early warning of potential faults. This application's embodiments rely on a relay calibrator, balancing practicality and economy, and effectively ensuring the reliable operation of relays, thereby maintaining the overall safety and stability of power systems and industrial control equipment. Attached Figure Description

[0011] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1 A schematic flowchart illustrating the relay fault detection method provided in this application embodiment; Figure 2 This is a structural block diagram of the relay fault detection device provided in the embodiments of this application; Figure 3 A schematic block diagram of a computer device provided in an embodiment of this application; Figure 4 Another schematic block diagram of the computer device provided in the embodiments of this application. Detailed Implementation

[0013] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0014] It should be noted that the terms "first," "second," and similar terms used in this application do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Unless the context clearly indicates otherwise, the singular forms "a," "one," or "the," etc., do not indicate a quantity limitation, but rather indicate the presence of at least one. The quantities of "multiple" or "multiple copies" mentioned in the embodiments of this application all refer to a quantity of "at least two," for example, "multiple" means "at least two," and "multiple copies" means "at least two copies." The terms "comprising" and "having," and any variations thereof, as used in this application, are intended to cover non-exclusive inclusion. The term "and / or" as used in this application refers to one of the embodiments, or any combination of multiple embodiments.

[0015] like Figure 1 As shown in the embodiments of this application, the relay fault detection method can be executed by a computer device, which refers to an electronic device with data calculation, processing, and storage capabilities. The method may include: S101: Acquire electrical data, actuation time data, housing temperature data, and ambient humidity data collected by the relay calibrator. Preprocess the electrical data, actuation time data, housing temperature data, and ambient humidity data respectively to obtain a standard feature dataset. The electrical data includes power, current, and voltage, and the actuation time data includes pull-in time and release time.

[0016] In this embodiment, electrical data, actuation time data, casing temperature data, and ambient humidity data are preprocessed to obtain a standard feature dataset, including: Based on electrical data, action time data, casing temperature data, and ambient humidity data, multi-dimensional time-series characteristic data are obtained; Steady-state intervals are extracted from multi-dimensional time-series feature data to obtain effective steady-state feature data. The steady-state effective feature data is standardized to obtain a standard feature dataset.

[0017] In this embodiment, electrical data refers to parameters characterizing the electrical characteristics of the relay operation, including power, current, and voltage. Action time data refers to parameters characterizing the relay's action response characteristics, including pull-in time and release time. Housing temperature data is a physical quantity characterizing the relay's heating state. Ambient humidity data is a physical quantity characterizing the humidity level of the relay's operating environment. Multi-dimensional time-series feature data is a collection of data formed by arranging multiple types of collected parameters in chronological order. Steady-state interval extraction is the process of filtering stable operating phase data from the time-series data. Steady-state effective feature data is stable operating parameter data after removing fluctuation interference. Standardization processing is the operation of unifying parameters with different dimensions to a comparable range. The standard feature dataset is a standardized feature set used for subsequent fault analysis.

[0018] For example, this embodiment utilizes a handheld non-full-phase relay calibrator to complete the entire process. First, the calibrator's magnetic test leads are quickly and stably connected to the relay terminals, relying on contact conduction to complete signal transmission, avoiding the time-consuming and risky wiring issues associated with traditional disconnection and connection methods. The calibrator collects electrical data in real time through a non-isolated sampling circuit. Current and voltage are directly acquired by the hardware circuit, while power is calculated in real time by a built-in algorithm based on the product of current and voltage. The action time data is calculated by the calibrator's internal timing unit, starting from the moment the relay receives the drive signal and continuing until the contacts close to complete the engagement action; similarly, the release time is calculated by starting the timing at the moment of power failure and continuing until the contacts open to complete the release action. The housing temperature data is continuously collected once per second by a temperature sensing unit attached to the relay body surface; ambient humidity data is synchronously collected by the calibrator's built-in humidity sensing unit. These data are continuously generated during the acquisition process and arranged in chronological order, forming multi-dimensional time-series characteristic data.

[0019] This embodiment can perform steady-state interval extraction on multi-dimensional time-series feature data. Specifically, the calibrator has a built-in dual judgment logic based on timestamps and data volatility, with a preset time window of 300 milliseconds and a data volatility threshold set at 5%. The calibrator traverses the time-series data, identifying unstable data intervals caused by drastic data fluctuations at the beginning of power-on and contact jitter at the end of power-off. It also removes transient abnormal jumps caused by electromagnetic interference, retaining continuously stable data segments without drastic fluctuations, thus obtaining steady-state effective feature data that truly reflects the actual operating state of the relay. This extraction process effectively eliminates interference and improves the reliability of subsequent analysis.

[0020] This embodiment can perform standardization processing on steady-state effective characteristic data. Because the units and numerical ranges of electrical data, operating time data, casing temperature data, and ambient humidity data differ significantly—for example, current might be in the ampere range, temperature in the Celsius range, and humidity in the percentage range—direct comparison would lead to an imbalance in numerical weights. Therefore, the calibrator employs a maximum-minimum standardization method, pre-setting the maximum and minimum values ​​of each type of parameter under normal operating conditions to form a standardized reference interval. The calculation method is: (steady-state effective characteristic data - corresponding parameter minimum value) / (corresponding parameter maximum value - corresponding parameter minimum value). Through this calculation, all parameters are uniformly mapped to a comparable interval of zero to one, eliminating dimensional differences and ensuring that each parameter has equal representativeness in subsequent weighted calculations. The processing is completed by the calibrator's internal microcontroller, with rapid calculation response and no manual intervention required.

[0021] This embodiment integrates all standardized data to form a standard feature dataset, which serves as the foundational input for subsequent anomaly detection and correlation analysis. The entire acquisition and processing flow is automated, eliminating the need for manual reading, recording, or calculation, further shortening the verification time and keeping the verification time for a single relay within 5 minutes, meeting the requirements for efficient power outage maintenance. The data processing is stable and reliable throughout, with high sampling accuracy, accurately reflecting the true operating status of the relay and providing solid data support for fault diagnosis.

[0022] Traditional calibration and measurement processes involve cumbersome manual wiring, and the measurement and collection of relay-related data by personnel is time-consuming and labor-intensive, with low measurement error and accuracy. The non-full-phase relay calibrator, through modification, uses magnetic test leads for wiring and contact conduction, reducing the time wasted and the risk of incorrect wiring caused by frequent disconnection and reconnection of relays on-site. Table 1 shows the procurement and installation of neodymium iron boron magnetic test leads.

[0023] Table 1 Procurement and Installation of Neodymium Iron Boron Magnetic Adsorption Test Wire

[0024] This non-full-phase relay calibrator, through chip and hardware structure modifications, achieves real-time automatic acquisition of current and voltage during relay calibration. It calculates the relay's power in real-time using an acquisition algorithm until the relay actuates, locking the current, voltage, and power at the time of actuation. The power at actuation is then used to determine whether the relay meets standard requirements. The non-full-phase relay calibrator can also obtain the actual actuation time of the relay by applying its rated voltage, thereby determining whether the relay selection is correct, whether the relay performance meets requirements, and whether there are any faults.

[0025] By using a non-full-phase relay calibrator, the non-full-phase state of the relay can be accurately identified, and faulty lines can be quickly disconnected, preventing equipment damage and systemic collapse. The calibrator can optimize protection selectivity, preventing power outages on fault-free lines due to malfunctions of other protection devices. Regular calibration can also detect component aging or logic deviations, providing a basis for preventative maintenance, reducing the risk of protection failure, and ultimately achieving safe, economical, and efficient operation of the power system. Specifically, the non-full-phase relay calibrator can achieve the following objectives: (1) Ensure stable power supply: The non-full-phase relay calibrator can quickly and accurately calibrate non-full-phase relays, ensuring that the relays operate stably in the power grid, reducing the risk of power outages caused by relay failures, providing continuous and reliable power supply for all sectors of society, and reducing the adverse effects of power outages on production and life.

[0026] (2) Reduce enterprise operating costs: Non-full-phase relay calibrators can detect potential problems of relays in a timely manner, which facilitates targeted maintenance, avoids large-scale maintenance and replacement caused by equipment failure, reduces the operating costs of power companies, and promotes the sustainable development of the power industry.

[0027] (3) Improve maintenance efficiency: The non-full-phase relay tester is easy to carry and operate. It can complete complex test procedures in a short time. Compared with traditional test methods, it greatly shortens the test time, allowing power maintenance personnel to complete the testing and maintenance of more equipment in the same amount of time, thus significantly improving work efficiency.

[0028] (4) Improve the technical level of the industry: As an advanced power testing equipment, the application of the non-full-phase relay calibrator prompts power technicians to continuously learn and master new skills, promotes technical exchanges and progress in the industry, and helps the power industry develop towards intelligence and efficiency.

[0029] This embodiment improves data reliability and detection accuracy through multi-dimensional automatic data collection and standardized processing, while shortening verification time and increasing maintenance efficiency. This embodiment can accurately identify potential relay faults, ensure stable operation of the power system, reduce operation and maintenance costs, and enhance the intelligence level of power maintenance work.

[0030] S102: Determine the anomaly labeling data based on the standard feature dataset and the rated benchmark data, and perform correlation feature analysis on the standard feature dataset to obtain the correlated anomaly labeling data.

[0031] In this embodiment, anomaly marker data is determined based on a standard feature dataset and a nominal baseline data, including: Parameter deviation feature data are obtained based on the standard feature dataset and the rated benchmark data; Acquire relay load data and determine an anomaly detection threshold set based on the relay load data; Parameter anomaly identification data are determined based on parameter deviation feature data and anomaly judgment threshold set; Perform continuous status verification on the parameter anomaly identification data, and use the parameter anomaly identification data that passes the verification as anomaly marker data.

[0032] In this embodiment, correlation feature analysis is performed on the standard feature dataset to obtain correlation anomaly marker data, including: Based on the power and current data in the standard feature dataset, the linear fitting slope data of power and current is obtained; based on the linear fitting slope data of power and current, the first correlation deviation data is obtained. Based on the suction time data and shell temperature data in the standard feature dataset, we obtain the action time and temperature change rate data; based on the action time and temperature change rate data, we obtain the second correlation deviation data. Impedance and voltage matching data are obtained based on coil impedance and voltage data in the standard feature dataset; third correlation deviation data are obtained based on impedance and voltage matching data. The associated anomaly marker data is determined based on the first associated deviation data, the second associated deviation data, and the third associated deviation data.

[0033] In this embodiment, the linear fitting slope data of power and current is obtained based on the power data and current data in the standard feature dataset, including: Based on the power and current data in the standard feature dataset, a two-dimensional data point set for power and current is constructed. The least-squares linear fit is performed on the two-dimensional data point set of power and current to obtain the fitted line equation; The linear fitting slope data of power and current are calculated based on the fitted linear equation.

[0034] In this embodiment, the standard feature dataset is a set of relay operating features with unified dimensions and scales after preprocessing. Rated reference data are the factory-set reference values ​​for the rated operating parameters of the relay. Anomaly marker data are identification data indicating whether a single parameter exceeds the normal range. Associated anomaly marker data are identification data indicating whether the collaborative relationship between multiple parameters is abnormal. Parameter deviation feature data is data on the degree of difference between the standard feature data and the rated reference data. Relay load data is operating data reflecting the current load state of the relay. The anomaly judgment threshold set is a set of anomaly judgment boundaries set based on the load state. Parameter anomaly marker data is marker data for preliminary judgment of single-parameter anomalies. Continuous state verification is a process of continuously verifying the anomaly markers. Power data are operating parameters characterizing the relay's functional capacity. Current data are operating parameters characterizing the circuit conduction strength.

[0035] Linear fitting slope data reflects the trend of power change with current. A two-dimensional data point set is a set of coordinate points consisting of corresponding power and current values. Least squares linear fitting is a method of fitting straight lines to data points. The fitted line equation is a mathematical expression describing the relationship between power and current changes. Pull-in time data is the duration from energization to contact closure of the relay. Housing temperature data is the temperature value of the relay body surface. Rate of change data reflects how quickly the pull-in time changes with temperature. Coil impedance data is the impedance parameter of the relay coil. Voltage data is the electrical parameter applied across the relay. Matching degree data is a value characterizing the reasonableness of impedance and voltage coordination. The first correlation deviation data, second correlation deviation data, and third correlation deviation data are the deviation values ​​between the corresponding correlation characteristics and the normal state, respectively.

[0036] For example, this embodiment can perform difference calculations on each item based on the standard feature dataset and the rated reference data to obtain parameter deviation feature data. The calculation method is to subtract the absolute value of the rated reference data from the standard feature data. This value directly represents the deviation of the real-time operating parameters from the rated factory parameters. The calibrator collects the current operating load information of the relay in real time through its built-in sampling circuit, and takes the arithmetic mean of ten sets of data to obtain stable relay load data.

[0037] This embodiment can divide relay load data into three intervals—light load, normal load, and heavy load—according to preset rules. Light load is 30% or less of the rated load, normal load is 30% to 70% of the rated load, and heavy load is 70% or more of the rated load. Each interval is matched with a corresponding anomaly detection boundary, which is then combined to form an anomaly detection threshold set. Specifically, under normal operating conditions, the parameter deviation from the upper limit is 10% of the rated baseline data; under light load conditions, it is 15%; and under heavy load conditions, it is 8%. This allows anomaly detection to adapt to different load conditions, improving accuracy and applicability.

[0038] In this embodiment, the parameter deviation feature data can be compared one by one with the corresponding boundary of the anomaly judgment threshold set. When the parameter deviation feature data is less than the corresponding threshold, it is judged as normal and marked as no anomaly; when the parameter deviation feature data is greater than or equal to the corresponding threshold, it is judged as parameter anomaly and parameter anomaly identification data is obtained.

[0039] To avoid misjudgments caused by transient electromagnetic interference or power-on fluctuations, this embodiment performs continuous state verification on the parameter anomaly identification data. The verification time window is set to five consecutive sampling periods, with each sampling period being one hundred milliseconds. Only when the parameter remains in a state exceeding the threshold for five consecutive sampling periods is it determined to be a continuously valid anomaly. The parameter anomaly identification data confirmed by continuous state verification is used as the final anomaly marker data.

[0040] After completing the single-parameter anomaly determination, this embodiment extracts synchronously acquired power and current data from the standard feature dataset. Power and current values ​​at the same sampling time are paired one-to-one to construct a two-dimensional data point set for power and current. This embodiment can perform least-squares linear fitting on the two-dimensional data point set to calculate the slope and intercept of the fitted line, forming a complete fitted line equation. The linear fitting slope data for power and current can then be directly extracted from the fitted line equation. This embodiment can compare this linear fitting slope data with the allowable slope range under normal operating conditions. The normal slope range is set to 0.5 to 1.5; exceeding this range calculates the first correlation deviation data.

[0041] This embodiment extracts adsorption time data and shell temperature data from a standard feature dataset, calculates the difference between two adjacent sets of sampled data, and divides the change in adsorption time by the change in shell temperature to obtain the rate of change data of action time and temperature. This embodiment can compare this rate of change data with a preset allowable fluctuation range, which is set to 0.02s to 0.08s per degree Celsius. If the fluctuation exceeds the range, a second correlation deviation data is calculated.

[0042] In this embodiment, coil impedance data and voltage data are extracted. The matching degree value is obtained by dividing the coil impedance value by the voltage value. The matching degree value is compared with the reasonable matching range, which is set to 10 ohms to 30 ohms per volt. If the value is outside the range, the third correlation deviation data is calculated.

[0043] This embodiment comprehensively judges the first correlation deviation data, the second correlation deviation data, and the third correlation deviation data. When any one of the correlation deviation data exceeds the corresponding allowable range, it is judged as a correlation anomaly. Simultaneously, multiple correlation deviation data are comprehensively judged and integrated to finally obtain the correlation anomaly marker data.

[0044] This embodiment combines single-parameter anomaly detection with multi-parameter correlation analysis to accurately identify early potential faults and minor parameter drifts in relays, effectively solving the problems of missed and false diagnoses in traditional detection methods. This embodiment utilizes a handheld calibrator to achieve fully automated processing, employing magnetic wiring to improve operational efficiency, shorten calibration time, and reduce human error and labor intensity. This embodiment features high sampling accuracy and reliable judgment logic, providing a basis for preventative relay maintenance, ensuring stable power system operation, and improving the intelligence and efficiency of maintenance work.

[0045] S103: Obtain fault risk data based on anomaly marker data and associated anomaly marker data, and obtain fault determination results based on fault risk data.

[0046] In this embodiment, fault risk data is obtained based on anomaly marker data and associated anomaly marker data, including: Acquire relay load data, determine the scenario type based on the relay load data, and determine the single parameter weight and associated feature weight based on the scenario type; A single-dimensional risk score is obtained by weighting the anomaly-marked data based on single-parameter weights. The association risk score is obtained by weighting the associated anomaly labeled data based on the associated feature weights. The fault risk data is obtained by weighting the single-dimensional risk score data and the associated risk score data.

[0047] In this embodiment, the fault determination result is obtained based on fault risk data, including: Obtain multiple historical fault risk data for relays, and obtain aging trend data based on these data. Risk assessment data is determined based on fault risk data and aging trend data, and fault determination results are determined based on risk assessment data.

[0048] In this embodiment, the scenario type is an operating condition category categorized based on relay load data. Single-parameter weight is a risk calculation coefficient assigned to a single-parameter abnormal state. Correlation feature weight is a risk calculation coefficient assigned to a multi-parameter correlated abnormal state. The single-dimensional risk score is a weighted individual risk value obtained from anomaly marker data. The correlated risk score is a weighted collaborative risk value obtained from correlated anomaly marker data. Fault risk data is the overall risk value obtained by combining single-dimensional and correlated risks. Historical fault risk data is a risk record obtained from past detections of the same relay. Aging trend data reflects the direction and magnitude of long-term performance degradation of the relay. Risk assessment data is a comprehensive assessment value obtained by combining real-time risk and aging trend. The fault determination result is the final state conclusion derived from the risk assessment data.

[0049] For example, in the fault risk data calculation stage, this embodiment can acquire relay load data through a calibrator, and then compare the load data with a preset range to determine the current scenario type. Scenario types are divided into three categories: light load, normal, and heavy load. Each scenario type corresponds to a fixed set of single-parameter weights and associated feature weights. The weight values ​​are directly retrieved from the calibrator's internal storage unit without manual setting. After the weights are determined, this embodiment can perform weighted calculations on the anomaly-marked data to obtain a single-dimensional risk score. The calculation process involves multiplying each anomaly-marked data item by its corresponding single-parameter weight, and then summing all the product results sequentially. The summed result is directly used as the single-dimensional risk score.

[0050] This embodiment can perform weighted calculations on associated anomaly marker data to obtain associated risk scores. The calculation process involves multiplying the first, second, and third associated deviation data by their corresponding associated feature weights, and then summing the three products. The sum is the associated risk score, which fully reflects the potential risks brought about by multi-parameter collaborative anomalies. This embodiment can also multiply the single-dimensional risk score by its corresponding weight, and the associated risk score by its corresponding weight, then add the two sets of products together. The final sum is the fault risk data, which is a continuous numerical value; its magnitude directly reflects the current fault risk level of the relay.

[0051] After entering the fault determination result generation stage, this embodiment can read multiple sets of historical fault risk data for the same relay from the storage unit. The data is arranged in chronological order of detection time. This embodiment can perform trend calculation on the historical fault risk data to obtain aging trend data. The calculation process involves calculating the difference between two adjacent historical fault risk data points sequentially to obtain multiple sets of changes. Then, all changes are arithmetically averaged to obtain the average change amplitude, while determining whether the change direction is upward or downward. The combination of the average change amplitude and the change direction forms the aging trend data, which is used to characterize the long-term performance degradation rate and degradation state of the relay.

[0052] This embodiment can fuse real-time fault risk data and aging trend data to obtain risk assessment data. The fusion calculation process involves adding a correction amount from the aging trend data to the real-time fault risk data. The correction amount is determined by both the magnitude and direction of change in the aging trend data; the value increases when the risk increases and decreases when the risk decreases. The final result is the risk assessment data. This embodiment can compare the risk assessment data with preset multi-level risk intervals one by one, directly determining the fault level based on the interval in which the data falls. The intervals correspond to normal state, alert state, abnormal state, and fault state, and the final fault determination result is output.

[0053] This embodiment calculates fault risks using a scenario-based weighted approach and combines this with an aging trend analysis for comprehensive evaluation. This enables a complete and accurate identification of both transient relay anomalies and long-term aging hazards, significantly improving fault diagnosis accuracy. The handheld calibrator automates the entire process, reducing calibration time. This embodiment effectively identifies early potential faults, providing a basis for preventative maintenance, reducing the risk of protection failure, ensuring stable power system operation, and improving maintenance efficiency and intelligence.

[0054] Based on the same principle as the relay fault detection method provided in the embodiments of this application, the embodiments of this application also provide a relay fault detection device, such as... Figure 2 As shown, the relay fault detection device 20 may specifically include: a data processing module 21, an anomaly analysis module 22, and a fault determination module 23. The data processing module 21 is used to acquire electrical data, operating time data, housing temperature data, and ambient humidity data collected by the relay calibrator, and to preprocess the electrical data, operating time data, housing temperature data, and ambient humidity data respectively to obtain a standard feature dataset. The electrical data includes power, current, and voltage, and the operating time data includes pull-in time and release time. Anomaly analysis module 22 is used to determine anomaly marker data based on standard feature dataset and rated baseline data, and to perform correlation feature analysis on standard feature dataset to obtain correlated anomaly marker data; The fault determination module 23 is used to obtain fault risk data based on the anomaly marker data and associated anomaly marker data, and to obtain fault determination results based on the fault risk data.

[0055] In one embodiment of this application, the data processing module 21 is specifically used to: obtain multi-dimensional time-series feature data based on electrical data, action time data, shell temperature data, and ambient humidity data; Steady-state intervals are extracted from multi-dimensional time-series feature data to obtain effective steady-state feature data. The steady-state effective feature data is standardized to obtain a standard feature dataset.

[0056] In one embodiment of this application, the anomaly analysis module 22 is specifically used to: obtain parameter deviation feature data based on the standard feature dataset and the nominal benchmark data; Acquire relay load data and determine an anomaly detection threshold set based on the relay load data; Parameter anomaly identification data are determined based on parameter deviation feature data and anomaly judgment threshold set; Perform continuous status verification on the parameter anomaly identification data, and use the parameter anomaly identification data that passes the verification as anomaly marker data.

[0057] In one embodiment of this application, the anomaly analysis module 22 is further configured to: obtain linear fitting slope data of power and current based on power data and current data in the standard feature dataset; and obtain first correlation deviation data based on the linear fitting slope data of power and current. Based on the suction time data and shell temperature data in the standard feature dataset, we obtain the action time and temperature change rate data; based on the action time and temperature change rate data, we obtain the second correlation deviation data. Impedance and voltage matching data are obtained based on coil impedance and voltage data in the standard feature dataset; third correlation deviation data are obtained based on impedance and voltage matching data. The associated anomaly marker data is determined based on the first associated deviation data, the second associated deviation data, and the third associated deviation data.

[0058] In one embodiment of this application, the anomaly analysis module 22 is further configured to: construct a two-dimensional data point set of power and current based on the power data and current data in the standard feature dataset; The least-squares linear fit is performed on the two-dimensional data point set of power and current to obtain the fitted line equation; The linear fitting slope data of power and current are calculated based on the fitted linear equation.

[0059] In one embodiment of this application, the fault determination module 23 is specifically used for: acquiring relay load data, determining the scenario type based on the relay load data, and determining the single parameter weight and associated feature weight based on the scenario type; A single-dimensional risk score is obtained by weighting the anomaly-marked data based on single-parameter weights. The association risk score is obtained by weighting the associated anomaly labeled data based on the associated feature weights. The fault risk data is obtained by weighting the single-dimensional risk score data and the associated risk score data.

[0060] In one embodiment of this application, the fault determination module 23 is further configured to: acquire multiple historical fault risk data of the relay, obtain aging trend data based on the multiple historical fault risk data; determine risk assessment data based on the fault risk data and the aging trend data, and determine the fault determination result based on the risk assessment data.

[0061] Each module in the aforementioned relay fault detection device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of a computer device in hardware form or independent of it, or stored in the memory of a computer device in software form, so that the processor can call and execute the corresponding operations of each module.

[0062] In some embodiments, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 3 As shown, this computer device includes a processor, memory, input / output interfaces (I / O), and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operating system and computer programs stored in the non-volatile storage media. The database stores data such as relay parameters. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communicating with external terminals via a network. When the computer program is executed by the processor, it implements a relay fault detection method.

[0063] In some embodiments, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 4 As shown, the computer device includes a processor, memory, input / output interface, communication interface, display unit, and input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interface. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage medium. The input / output interface is used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements a relay fault detection method. The display unit of the computer device is used to form a visually visible image. It can be a display screen, a projection device, or a virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0064] Those skilled in the art will understand that Figure 3 , Figure 4The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0065] In some embodiments, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the relay fault detection method described above.

[0066] In some embodiments, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described relay fault detection method.

[0067] In some embodiments, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps of the above-described relay fault detection method.

[0068] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0069] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, data processing logic devices, etc., and are not limited to these.

[0070] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0071] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A relay fault detection method, characterized in that, include: The system acquires electrical data, actuation time data, housing temperature data, and ambient humidity data collected by the relay calibrator. It then preprocesses these data separately to obtain a standard feature dataset. The electrical data includes power, current, and voltage, while the actuation time data includes pull-in and release times. Anomaly labeling data is determined based on standard feature datasets and baseline data. Correlation feature analysis is then performed on the standard feature datasets to obtain correlated anomaly labeling data. Fault risk data is obtained based on anomaly marker data and associated anomaly marker data, and fault determination results are obtained based on fault risk data.

2. The relay fault detection method as described in claim 1, characterized in that, The electrical data, actuation time data, casing temperature data, and ambient humidity data are preprocessed to obtain a standard feature dataset, including: Based on the electrical data, the action time data, the housing temperature data, and the ambient humidity data, multi-dimensional time-series feature data is obtained; Steady-state intervals are extracted from the multi-dimensional time-series feature data to obtain steady-state effective feature data; The steady-state effective feature data is standardized to obtain the standard feature dataset.

3. The relay fault detection method as described in claim 1, characterized in that, The determination of anomaly marker data based on standard feature datasets and baseline data includes: Parameter deviation feature data are obtained based on the standard feature dataset and the rated benchmark data; Acquire relay load data, and determine an anomaly detection threshold set based on the relay load data; Based on the parameter deviation feature data and the anomaly determination threshold set, parameter anomaly identification data is determined; The parameter anomaly identification data is continuously verified, and the parameter anomaly identification data that passes the verification is used as the anomaly marker data.

4. The relay fault detection method as described in claim 1, characterized in that, The process of performing correlation feature analysis on the standard feature dataset to obtain correlation anomaly marker data includes: Based on the power and current data in the standard feature dataset, linear fitting slope data of power and current is obtained; based on the linear fitting slope data of power and current, first correlation deviation data is obtained. Based on the suction time data and shell temperature data in the standard feature dataset, the rate of change of action time and temperature is obtained; based on the rate of change of action time and temperature, the second correlation deviation data is obtained. Impedance and voltage matching data are obtained based on the coil impedance and voltage data in the standard feature dataset; third correlation deviation data are obtained based on the impedance and voltage matching data. The associated anomaly marker data is determined based on the first associated deviation data, the second associated deviation data, and the third associated deviation data.

5. The relay fault detection method as described in claim 4, characterized in that, The process of obtaining the linear fitting slope data for power and current based on the power and current data in the standard feature dataset includes: Based on the power and current data in the standard feature dataset, a two-dimensional data point set for power and current is constructed. The least-squares linear fit is performed on the two-dimensional data point set of power and current to obtain the fitted line equation; The linear fitting slope data of the power and current are calculated based on the fitted linear equation.

6. The relay fault detection method as described in claim 1, characterized in that, The fault risk data obtained based on anomaly marker data and associated anomaly marker data includes: Acquire relay load data, determine the scenario type based on the relay load data, and determine the single parameter weight and associated feature weight based on the scenario type; The anomaly marker data is weighted based on the single-parameter weight to obtain a single-dimensional risk score. The association risk score is obtained by weighting the association anomaly marker data based on the association feature weights. The fault risk data is obtained by weighting the single-dimensional risk score data and the associated risk score.

7. The relay fault detection method as described in claim 1, characterized in that, The fault determination result obtained based on fault risk data includes: Obtain multiple historical fault risk data of the relay, and obtain aging trend data based on the multiple historical fault risk data; Risk assessment data is determined based on the fault risk data and the aging trend data, and fault determination results are determined based on the risk assessment data.

8. A relay fault detection device, characterized in that, include: The data processing module is used to acquire electrical data, actuation time data, housing temperature data, and ambient humidity data collected by the relay calibrator. It preprocesses the electrical data, actuation time data, housing temperature data, and ambient humidity data to obtain a standard feature dataset. The electrical data includes power, current, and voltage, and the actuation time data includes pull-in time and release time. The anomaly analysis module is used to determine anomaly marker data based on standard feature datasets and baseline data, and to perform correlation feature analysis on the standard feature datasets to obtain correlated anomaly marker data. The fault determination module is used to obtain fault risk data based on anomaly marker data and associated anomaly marker data, and to obtain fault determination results based on fault risk data.

9. A computer device, characterized in that, The computer device includes a processor and memory: The memory is used to store computer programs; The processor is configured to execute the relay fault detection method according to any one of claims 1-7 according to the computer program.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium is used to store a computer program, which, when executed by a computer device, implements the relay fault detection method according to any one of claims 1-7.

Images