High and low voltage electrical equipment operation monitoring and early warning system

Abstract

The invention discloses a high-low voltage electrical equipment operation monitoring and early warning system, and relates to the technical field of equipment management, and the system comprises a high-low voltage electrical equipment operation data collection module which collects the operation parameters of high-low voltage electrical equipment in real time; the fault diagnosis module is used for analyzing the real-time operation data of the high and low voltage electrical equipment based on the trained fault diagnosis model; the fault prediction module is used for evaluating the health state of the current equipment by using a clustering algorithm based on the real-time diagnosis result data and predicting health state data generated at the next moment; the early warning module is used for analyzing and processing the prediction result of the health state data; and the remote monitoring module is used for monitoring the state of the equipment in real time by an operator. Through real-time data acquisition, fault diagnosis and a prediction model, accurate prediction of equipment faults is realized, an early warning mechanism is optimized, equipment damage is reduced, the equipment management and maintenance decision accuracy is improved, and the reliability of fault prediction is improved by utilizing a machine learning technology.

Description

Technical Field

[0001] This invention relates to the field of equipment management technology, specifically to a high and low voltage electrical equipment operation monitoring and early warning system. Background Technology

[0002] With the widespread application of high and low voltage electrical equipment in modern industry, the reliability and safety of these devices have become paramount. Electrical equipment is often affected by factors such as high temperature, overload, short circuit, and aging, leading to sudden and unpredictable equipment failures. This poses challenges to the normal operation of industrial production and the long-term management of equipment lifespan. Therefore, real-time monitoring and early warning of the operating status of electrical equipment are crucial for preventing failures and extending equipment lifespan. Existing high and low voltage electrical equipment monitoring systems are often limited to basic parameter monitoring and simple alarm mechanisms, failing to provide effective fault prediction and early warning. This results in delayed intervention when equipment failures occur, causing production interruptions, equipment damage, and even safety accidents. Consequently, research on high and low voltage electrical equipment monitoring and early warning systems based on big data analysis, machine learning, and intelligent diagnostic technologies has gradually become a key focus in the industry.

[0003] This invention provides a high and low voltage electrical equipment operation monitoring and early warning system. By integrating a fault diagnosis model, a fault prediction module, and an early warning mechanism, it achieves accurate diagnosis and prediction of equipment faults, improves the accuracy of equipment fault prediction, and provides enterprises with a more intelligent equipment management solution. Summary of the Invention

[0004] To solve the above-mentioned technical problems, a monitoring and early warning system for the operation of high and low voltage electrical equipment is provided. This technical solution solves the above problems.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0006] A high- and low-voltage electrical equipment operation monitoring and early warning system, comprising:

[0007] High and low voltage electrical equipment operation data acquisition module: used to collect various operating parameters of high and low voltage electrical equipment in real time, including voltage, current, temperature, humidity, vibration and power parameters, and transmit them to the fault diagnosis module;

[0008] Fault diagnosis module: Based on the trained fault diagnosis model, it analyzes the real-time operating data of high and low voltage electrical equipment to determine whether the equipment has failed. If a fault has occurred, it will perform equipment alarm processing. If no fault has occurred, it will transmit the data to the fault prediction module.

[0009] Fault prediction module: Based on real-time diagnostic results data, it uses a clustering algorithm to assess the current health status of the device, obtains the current health status data of the device, and uses a time series analysis model to predict the health status data generated in the next moment based on the current health status data. The prediction results of the health status data are then input into the early warning module.

[0010] Early warning module: used to analyze and process the predicted results of health status data, including fault diagnosis, health assessment and early warning analysis, and to issue early warnings based on preset thresholds;

[0011] Remote monitoring module: Enables remote viewing and control of equipment operation data, allowing operators to monitor equipment status in real time.

[0012] Preferably, the fault diagnosis module specifically includes:

[0013] Data preprocessing unit: Cleans the real-time operating data collected from high and low voltage electrical equipment, removes noise data, and extracts statistical and time-series features from the raw data;

[0014] Fault diagnosis model inference unit: Based on the trained fault diagnosis model, load model weights and parameters, take statistical features and time series features as input, take equipment diagnosis results as output, input preprocessed real-time operating data into the fault diagnosis model for inference, and diagnose whether the equipment is currently faulty;

[0015] Anomaly detection and judgment unit: Based on the model output, obtain the probability of the model output, and judge whether the equipment has failed based on the probability of the model output. If the model judges that the equipment has failed, it will enter the fault alarm processing process. If no fault has occurred, the data will be transmitted to the fault prediction module.

[0016] Preferably, the fault diagnosis model inference unit specifically includes:

[0017] A fault diagnosis model is established based on a fully connected neural network. The model formula is as follows:

[0018]

[0019] In the formula, h i For the output of the i-th layer, W i Let x be the weight of the i-th layer. i-1 b is the input to the previous layer. i σ is the bias term, σ is the ReLU activation function, and y is the result of the output layer, representing the predicted probability of the fault;

[0020] Faults are determined by the results output by the fault diagnosis model. If the model output is 1, an alarm is triggered and a corresponding alarm signal is issued.

[0021] If the model output is 0, the equipment is determined to be operating normally, and the data will be transmitted to the fault prediction module.

[0022] Preferably, the fault prediction module includes:

[0023] Real-time data receiving and processing unit: Receives data from the fault diagnosis module, extracts health status features, uses clustering algorithms to perform a preliminary assessment of the equipment health status based on the real-time diagnosis results data, classifies the health status data in the dataset according to similarity, and assesses the current health status of the equipment.

[0024] Time series analysis unit: Based on the current health status data of the equipment, it predicts the health status of the next moment through a time series analysis model. Based on the current health status of the equipment, it predicts the health status of the next moment and assesses the probability of failure.

[0025] Preferably, the real-time data receiving and processing unit specifically includes:

[0026] It receives data from the fault diagnosis module, takes the device's voltage, current, and temperature as input features, uses the Euclidean distance calculation formula to calculate the similarity between these features, and groups data points with similar operating behaviors into the same category, thereby assigning them to the same cluster;

[0027] The data is clustered using the K-means clustering algorithm to obtain a cluster label for each data point. The clustering results are then mapped to the health status of the equipment, and the clusters are marked as normal, minor fault, and serious fault. If the current status of the equipment belongs to the cluster of minor fault or serious fault, the equipment is marked as having a fault risk.

[0028] Preferably, the time series analysis unit specifically includes:

[0029] Collect the health status data of the equipment, arrange the data in chronological order, represent it in the form of a time series, and perform smoothing and differential processing on the data;

[0030] The ARIMA model is selected for time series analysis. It predicts future values ​​based on autoregression and moving average of historical data. The model is trained using historical equipment health status data. The autoregression part uses the health status of the past few time points to predict the current health status. The differencing part ensures that the time series data is stationary. The moving average part considers past errors to correct the predicted values. The trained model is used to predict the equipment health status at the next moment.

[0031] Based on the predicted health status, a logistic regression model is used to calculate the probability of failure. A health status threshold is set according to historical data. When the predicted health status is lower than the threshold, it is assessed as having a high risk of failure. Based on the predicted equipment health status and the assessed failure probability, the system outputs the predicted equipment health status for the next moment and provides a judgment on whether there is a risk of failure.

[0032] Preferably, the calculation of the probability of failure based on the predicted health status using a logistic regression model specifically includes:

[0033] The formula for the logistic regression model is as follows:

[0034]

[0035] In the formula, P is the probability of equipment failure, β0 is the constant term of logistic regression, and β1 is the regression coefficient related to health status. For predicted health status.

[0036] Preferably, setting a health status threshold based on historical data specifically includes: setting a threshold using a threshold calculation formula based on the mean and standard deviation of the device's historical health status data;

[0037] Preferably, the early warning module specifically includes:

[0038] Early warning generation unit: acquires the health status prediction results output by the time series model, determines whether the equipment has potential faults, compares the predicted health status results of the equipment with the set threshold, determines whether an early warning needs to be issued, and if the health status exceeds the threshold, the system will trigger an early warning.

[0039] Early warning output and feedback unit: Triggers the early warning mechanism based on the fault prediction results and provides feedback on the health status of the equipment to the operator. It issues alarms to the operator via audible and visual alarms, SMS, and email, and records detailed information of the early warning event.

[0040] Preferably, the early warning output and feedback unit specifically includes: recording and storing information on all early warning events, including the operating status of the equipment, fault prediction results, health status, occurrence time and alarm handling status, providing operators with the function of viewing and managing early warning events, supporting operators to view early warning history records, current equipment status, fault prediction analysis reports, and enabling real-time response and control operations.

[0041] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0042] The high and low voltage electrical equipment operation monitoring and early warning system of this invention achieves early detection and accurate prediction of equipment faults by collecting equipment operation data in real time and combining it with advanced fault diagnosis and prediction models. The system uses clustering algorithms to assess the health status of equipment, optimizes the early warning mechanism, ensures timely warnings and reduces equipment damage, and the remote monitoring function enhances the convenience of equipment management. Intelligent feedback and historical data management improve the accuracy of equipment maintenance decisions. By introducing machine learning technology, this system significantly improves the reliability of fault prediction and the safety of equipment operation. Attached Figure Description

[0043] Figure 1 This is a system framework diagram of the present invention;

[0044] Figure 2 This is a flowchart illustrating the steps of the present invention. Detailed Implementation

[0045] The following description is intended to disclose the invention and enable those skilled in the art to implement it. The preferred embodiments described below are merely examples, and other obvious variations will occur to those skilled in the art.

[0046] Reference Figure 1 As shown, a high- and low-voltage electrical equipment operation monitoring and early warning system includes:

[0047] High and low voltage electrical equipment operation data acquisition module: used to collect various operating parameters of high and low voltage electrical equipment in real time, including voltage, current, temperature, humidity, vibration and power parameters, and transmit them to the fault diagnosis module;

[0048] Fault diagnosis module: Based on the trained fault diagnosis model, it analyzes the real-time operating data of high and low voltage electrical equipment to determine whether the equipment has failed. If a fault has occurred, it will perform equipment alarm processing. If no fault has occurred, it will transmit the data to the fault prediction module.

[0049] Fault prediction module: Based on real-time diagnostic results data, it uses a clustering algorithm to assess the current health status of the device, obtains the current health status data of the device, and uses a time series analysis model to predict the health status data generated in the next moment based on the current health status data. The prediction results of the health status data are then input into the early warning module.

[0050] Early warning module: used to analyze and process the predicted results of health status data, including fault diagnosis, health assessment and early warning analysis, and to issue early warnings based on preset thresholds;

[0051] Remote monitoring module: Enables remote viewing and control of equipment operation data, allowing operators to monitor equipment status in real time.

[0052] Reference Figure 2 As shown, the fault diagnosis module specifically includes:

[0053] Data preprocessing unit: Cleans the real-time operating data collected from high and low voltage electrical equipment, removes noise data, and extracts statistical and time-series features from the raw data;

[0054] Fault diagnosis model inference unit: Based on the trained fault diagnosis model, load model weights and parameters, take statistical features and time series features as input, take equipment diagnosis results as output, input preprocessed real-time operating data into the fault diagnosis model for inference, and diagnose whether the equipment is currently faulty;

[0055] Anomaly detection and judgment unit: Based on the model output, obtain the probability of the model output, and judge whether the equipment has failed based on the probability of the model output. If the model judges that the equipment has failed, it will enter the fault alarm processing process. If no fault has occurred, the data will be transmitted to the fault prediction module.

[0056] The innovation of the fault diagnosis module lies in its inclusion of a data preprocessing unit and a fault diagnosis model inference unit. The data preprocessing unit cleans noisy data and extracts features to ensure the quality of input data, effectively improving the accuracy of fault diagnosis. The fault diagnosis model inference unit, based on a trained model, performs fault judgment on real-time data, further enhancing the system's intelligence level.

[0057] The fault diagnosis model inference unit specifically includes:

[0058] A fault diagnosis model is established based on a fully connected neural network. The model formula is as follows:

[0059]

[0060] In the formula, h i For the output of the i-th layer, W i Let x be the weight of the i-th layer. i-1 b is the input to the previous layer. i σ is the bias term, σ is the ReLU activation function, and y is the result of the output layer, representing the predicted probability of the fault;

[0061] Faults are determined by the results output by the fault diagnosis model. If the model output is 1, an alarm is triggered and a corresponding alarm signal is issued.

[0062] If the model output is 0, the equipment is determined to be operating normally, and the data will be transmitted to the fault prediction module.

[0063] The innovation of the fault diagnosis module lies in its inclusion of a data preprocessing unit and a fault diagnosis model inference unit. The data preprocessing unit cleans up noisy data and extracts features to ensure the quality of input data and effectively improve the accuracy of fault diagnosis. The fault diagnosis model inference unit, based on a trained model, performs fault judgment on real-time data, further enhancing the intelligence level of the system.

[0064] The fault prediction module includes:

[0065] Real-time data receiving and processing unit: Receives data from the fault diagnosis module, extracts health status features, uses clustering algorithms to perform a preliminary assessment of the equipment health status based on the real-time diagnosis results data, classifies the health status data in the dataset according to similarity, and assesses the current health status of the equipment.

[0066] Time series analysis unit: Based on the current health status data of the equipment, it predicts the health status of the next moment through a time series analysis model, and assesses the probability of failure based on the current health status of the equipment.

[0067] The fault prediction module combines clustering algorithms and time series analysis models to not only assess the health status of equipment in real time, but also to predict faults based on historical data. This multi-level prediction method not only improves the accuracy of fault prediction, but also reduces the risk of equipment failure through early warning.

[0068] The real-time data receiving and processing unit specifically includes:

[0069] It receives data from the fault diagnosis module, takes the device's voltage, current, and temperature as input features, uses the Euclidean distance calculation formula to calculate the similarity between these features, and groups data points with similar operating behaviors into the same category, thereby assigning them to the same cluster;

[0070] The data are clustered using the K-means clustering algorithm to obtain the cluster label for each data point. The clustering results are then mapped to the health status of the equipment, and the clusters are marked as normal, minor fault, and serious fault. If the current status of the equipment belongs to the cluster of minor fault or serious fault, the equipment is marked as having a fault risk.

[0071] The real-time data receiving and processing unit uses the K-means clustering algorithm to classify the health status of equipment according to similarity. By clustering the data using the Euclidean distance calculation formula, it not only enhances the system's ability to monitor the health status of equipment, but also accurately identifies equipment status with a high risk of failure, providing more precise early warning information.

[0072] The time series analysis unit specifically includes:

[0073] Collect the health status data of the equipment, arrange the data in chronological order, represent it in the form of a time series, and perform smoothing and differential processing on the data;

[0074] The ARIMA model is selected for time series analysis. It predicts future values ​​based on autoregression and moving average of historical data. The model is trained using historical equipment health status data. The autoregression part uses the health status of the past few time points to predict the current health status. The differencing part ensures that the time series data is stationary. The moving average part takes into account past errors to correct the predicted values.

[0075] The trained model is used to predict the device health status at the next moment, where the model formula is:

[0076]

[0077] In the formula, It represents the predicted health status, where α is a constant term and p is the order of the autoregressive component. For each lagged term g t-s Contribution to the predicted value, g t-s Let q be the past value of the device health status at the current time t, θ be the order of the moving average component, and t be the value of the device health status at the current time t. j For each lag term ∩ t-k Contribution to the predicted value, ∈ t-j This is the error term at time point t;

[0078] Based on the predicted health status, a logistic regression model is used to calculate the probability of failure. A health status threshold is set according to historical data. When the predicted health status is lower than the threshold, it is assessed as having a high risk of failure. Based on the predicted equipment health status and the assessed failure probability, the system outputs the predicted equipment health status for the next moment and provides a judgment on whether there is a risk of failure.

[0079] The time series analysis unit combines the ARIMA model with autoregressive and moving average analysis methods to predict the health status of equipment. This time series-based data analysis method can effectively capture the changing trends of equipment health status, providing more stable and scientific prediction results, which helps to improve the timeliness and accuracy of fault early warning.

[0080] Based on the predicted health status, the probability of failure is calculated using a logistic regression model, specifically including:

[0081] The formula for the logistic regression model is as follows:

[0082]

[0083] In the formula, P is the probability of equipment failure, β0 is the constant term of logistic regression, and β1 is the regression coefficient related to health status. For predicted health status;

[0084] The probability of failure is calculated using a logistic regression model, which is based not only on the health status data of the equipment but also on the trends of historical data. The application of the logistic regression model enables the system to accurately assess the likelihood of failure, further improving the reliability of the prediction.

[0085] Setting health status thresholds based on historical data specifically includes:

[0086] Based on the mean and standard deviation of the equipment's historical health status data, a threshold is set using a threshold calculation formula, where the threshold calculation formula is:

[0087] θ=μ-kσ

[0088] In the formula, μ is the mean of the equipment health status, σ is the standard of the equipment health status, and the difference k is the position of the mean minus twice the standard deviation;

[0089] The system calculates the health status threshold by using the mean and standard deviation of historical health status data, and judges the risk of failure by setting the threshold. This method enhances the system's sensitivity to abnormal fluctuations in equipment health status and provides a more scientific and accurate threshold reference for the early warning module.

[0090] The early warning module specifically includes:

[0091] Early warning generation unit: acquires the health status prediction results output by the time series model, determines whether the equipment has potential faults, compares the predicted health status results of the equipment with the set threshold, determines whether an early warning needs to be issued, and if the health status exceeds the threshold, the system will trigger an early warning.

[0092] Early warning output and feedback unit: Triggers the early warning mechanism based on the fault prediction results and provides feedback on the health status of the equipment to the operator. It issues alarms to the operator through audible and visual alarms, SMS and email, and records detailed information of the early warning events.

[0093] The early warning module automatically determines whether an early warning needs to be issued by acquiring the health status prediction results and comparing them with preset thresholds. The innovation of this process lies in the fact that by combining real-time health status prediction with automated early warning triggering, the response speed and efficiency of fault early warning are greatly improved.

[0094] The early warning output and feedback unit specifically includes: recording and storing information on all early warning events, including the equipment's operating status, fault prediction results, health status, occurrence time, and alarm handling status; providing operators with the ability to view and manage early warning events; supporting operators to view early warning history records, current equipment status, and fault prediction analysis reports; and enabling real-time response and control operations.

[0095] The early warning output and feedback unit can not only issue alarms to operators through various means such as audible and visual alarms, SMS, and email, but also provides detailed recording and management functions for early warning events. Operators can view early warning history records, current equipment status, and fault prediction reports in real time, and perform necessary real-time response and control operations, improving the operability and manageability of the system.

[0096] In summary, the advantages of this invention are:

[0097] This system collects various operating data from high and low voltage electrical equipment in real time and combines them with advanced fault diagnosis models to comprehensively analyze the operating status of the equipment, ensuring early detection and timely warning of equipment faults.

[0098] The system uses a clustering algorithm-based equipment health status assessment and time series analysis model to accurately predict the future health status of equipment, further improving the reliability of fault prediction, identifying potential fault risks in advance, and reducing the occurrence of sudden faults.

[0099] The early warning module combines health status data with set thresholds to accurately assess the risk of equipment failure and promptly issue early warnings to operators, thereby enabling them to take effective preventative measures and reduce losses from equipment damage and production interruptions.

[0100] The remote monitoring module enables operators to view and control the equipment status in real time, realizing remote and intelligent equipment management and providing enterprises with a more convenient operating experience.

[0101] The early warning feedback system records detailed information on all early warning events and provides operators with viewing and management functions, facilitating historical data analysis, fault diagnosis and retrospection, and optimization of equipment maintenance decisions.

[0102] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the claimed invention. The scope of protection claimed by the appended claims and their equivalents is defined.

Claims

1. A high- and low-voltage electrical equipment operation monitoring and early warning system, characterized in that, include: High and low voltage electrical equipment operation data acquisition module: used to collect various operating parameters of high and low voltage electrical equipment in real time, including voltage, current, temperature, humidity, vibration and power parameters, and transmit them to the fault diagnosis module; Fault diagnosis module: Based on the trained fault diagnosis model, it analyzes the real-time operating data of high and low voltage electrical equipment to determine whether the equipment has failed. If a fault has occurred, it will perform equipment alarm processing. If no fault has occurred, it will transmit the data to the fault prediction module. Fault prediction module: Based on real-time diagnostic results data, it uses a clustering algorithm to assess the current health status of the device, obtains the current health status data of the device, and uses a time series analysis model to predict the health status data generated in the next moment based on the current health status data. The prediction results of the health status data are then input into the early warning module. Early warning module: used to analyze and process the predicted results of health status data, including fault diagnosis, health assessment and early warning analysis, and to issue early warnings based on preset thresholds; Remote monitoring module: Enables remote viewing and control of equipment operation data, allowing operators to monitor equipment status in real time.

2. The high and low voltage electrical equipment operation monitoring and early warning system according to claim 1, characterized in that, The fault diagnosis module specifically includes: Data preprocessing unit: Cleans the real-time operating data collected from high and low voltage electrical equipment, removes noise data, and extracts statistical and time-series features from the raw data; Fault diagnosis model inference unit: Based on the trained fault diagnosis model, load model weights and parameters, take statistical features and time series features as input, take equipment diagnosis results as output, input preprocessed real-time operating data into the fault diagnosis model for inference, and diagnose whether the equipment is currently faulty; Anomaly detection and judgment unit: Based on the model output, obtain the probability of the model output, and judge whether the equipment has failed based on the probability of the model output. If the model judges that the equipment has failed, it will enter the fault alarm processing flow. If no fault has occurred, the data will be transmitted to the fault prediction module.

3. The high and low voltage electrical equipment operation monitoring and early warning system according to claim 2, characterized in that, The fault diagnosis model inference unit specifically includes: A fault diagnosis model is established based on a fully connected neural network. The model formula is as follows: In the formula, h i For the output of the i-th layer, W i Let x be the weight of the i-th layer. i-1 b is the input to the previous layer. i σ is the bias term, σ is the ReLU activation function, and y is the result of the output layer, representing the predicted probability of the fault; The fault is determined by the output of the fault diagnosis model. If the model output is 1, an alarm is triggered and a corresponding alarm signal is issued. If the model output is 0, the equipment is determined to be operating normally, and the data will be transmitted to the fault prediction module.

4. The high and low voltage electrical equipment operation monitoring and early warning system according to claim 3, characterized in that, The fault prediction module includes: Real-time data receiving and processing unit: Receives data from the fault diagnosis module, extracts health status features, uses clustering algorithms to perform a preliminary assessment of the equipment health status based on the real-time diagnosis results data, classifies the health status data in the dataset according to similarity, and assesses the current health status of the equipment. Time series analysis unit: Based on the current health status data of the equipment, it predicts the health status of the next moment through a time series analysis model. Based on the current health status of the equipment, it predicts the health status of the next moment and assesses the probability of failure.

5. A high- and low-voltage electrical equipment operation monitoring and early warning system according to claim 4, characterized in that, The real-time data receiving and processing unit specifically includes: It receives data from the fault diagnosis module, takes the device's voltage, current, and temperature as input features, uses the Euclidean distance calculation formula to calculate the similarity between these features, and groups data points with similar operating behaviors into the same category, thereby assigning them to the same cluster; The data is clustered using the K-means clustering algorithm to obtain a cluster label for each data point. The clustering results are then mapped to the health status of the equipment, and the clusters are marked as normal, minor fault, and serious fault. If the current status of the equipment belongs to the cluster of minor fault or serious fault, the equipment is marked as having a fault risk.

6. A high- and low-voltage electrical equipment operation monitoring and early warning system according to claim 4, characterized in that, The time series analysis unit specifically includes: Collect the health status data of the equipment, arrange the data in chronological order, represent it in the form of a time series, and perform smoothing and differential processing on the data; The ARIMA model is selected for time series analysis. It predicts future values ​​based on autoregression and moving average of historical data. The model is trained using historical equipment health status data. The autoregression part uses the health status of the past few time points to predict the current health status. The differencing part ensures that the time series data is stationary. The moving average part takes into account past errors to correct the predicted values. Based on the predicted health status, a logistic regression model is used to calculate the probability of failure. A health status threshold is set according to historical data. When the predicted health status is lower than the threshold, it is assessed as having a high risk of failure. Based on the predicted equipment health status and the assessed failure probability, the system outputs the predicted equipment health status for the next moment and provides a judgment on whether there is a risk of failure.

7. A high- and low-voltage electrical equipment operation monitoring and early warning system according to claim 6, characterized in that, The predicted health status is specifically calculated using a logistic regression model to determine the probability of failure. include: The formula for the logistic regression model is as follows: In the formula, P is the probability of equipment failure, β0 is the constant term of logistic regression, and β1 is the regression coefficient related to health status. For predicted health status.

8. A high- and low-voltage electrical equipment operation monitoring and early warning system according to claim 6, characterized in that, Setting health status thresholds based on historical data specifically includes: using the mean and standard deviation of the device's historical health status data, and then using a threshold calculation formula to set the threshold.

9. A high- and low-voltage electrical equipment operation monitoring and early warning system according to claim 1, characterized in that, The early warning module specifically includes: Early warning generation unit: acquires the health status prediction results output by the time series model, determines whether the equipment has potential faults, compares the predicted health status results of the equipment with the set threshold, determines whether an early warning needs to be issued, and if the health status exceeds the threshold, the system will trigger an early warning. Early warning output and feedback unit: Triggers the early warning mechanism based on the fault prediction results and provides feedback on the health status of the equipment to the operator. It issues alarms to the operator via audible and visual alarms, SMS, and email, and records detailed information of the early warning event.

10. A high- and low-voltage electrical equipment operation monitoring and early warning system according to claim 9, characterized in that, The early warning output and feedback unit specifically includes: recording and storing information on all early warning events, including the equipment's operating status, fault prediction results, health status, occurrence time, and alarm handling status; providing operators with the function of viewing and managing early warning events; supporting operators to view early warning history records, current equipment status, and fault prediction analysis reports; and enabling real-time response and control operations.

Images