Electrical operation risk assessment and early warning system based on big data analysis

Abstract

The invention relates to the technical field of electrical safety, in particular to an electrical operation risk assessment and early warning system based on big data analysis, which comprises a data acquisition module, a data fusion module, a feature extraction module, a risk assessment module, a real-time monitoring and early warning module, a historical data storage and analysis module and a visual monitoring module. Multi-source data are collected in real time by deploying multiple sensors, risk assessment is carried out through Kalman filtering and support vector regression algorithm fusion processing, feature variable extraction and combination of a support vector machine and a random forest algorithm, real-time monitoring and risk level judgment of an electrical working environment are achieved, long-term risk trend prediction is supported, and the risk assessment accuracy is improved. The visual monitoring module provides a basis for optimizing alarm rules and decision support, and displays an evaluation result, early warning information and an analysis result through a visual interface; the system has the advantages of being accurate in data processing, efficient in risk assessment, timely in early warning and the like, and can be widely applied to risk monitoring and safety management of electrical operation.

Description

Technical Field

[0001] The present invention relates to the field of electrical safety technology, and in particular to an electrical operation risk assessment and early warning system based on big data analysis. Background Art

[0002] Electrical work is an indispensable part of industrial production and daily life, involving power transmission and distribution, equipment maintenance, operation control and other fields. However, the electrical work environment is complex and is often affected by multiple factors such as temperature, humidity, and voltage fluctuations, which can easily lead to electrical failures, equipment damage and even serious safety accidents. In order to ensure the safety and reliability of electrical work, real-time monitoring of risk factors in the working environment, accurate assessment of potential risks and timely warning have become key research directions.

[0003] Although the existing technology has made certain progress in electrical operation risk assessment, there are still the following technical problems that need to be solved: first, the efficiency of multi-source data processing and fusion is low, the collected data is noisy and redundant, and it is difficult to provide consistent input data for the assessment model; second, the traditional risk assessment method is insufficient in the ability to analyze the complex relationships of risk factors and cannot accurately identify high-risk states in dynamic environments; third, the existing alarm mechanism lacks a hierarchical design, which can easily lead to false alarms or missed alarms, resulting in operators being unable to respond effectively. Summary of the invention

[0004] The present invention provides an electrical operation risk assessment and early warning system based on big data analysis.

[0005] An electrical operation risk assessment and early warning system based on big data analysis, including a data acquisition module, a data fusion module, a feature extraction module, a risk assessment module, a real-time monitoring and early warning module, a historical data storage and analysis module and a visual monitoring module, wherein;

[0006] The data acquisition module deploys a variety of sensors to collect multi-source data at the electrical operation site in real time, and the multi-source data includes temperature, humidity, current, voltage, leakage current and ground resistance;

[0007] The data fusion module uses a fusion algorithm based on Kalman filtering and support vector regression to fuse multi-source data;

[0008] The feature extraction module extracts feature variables related to electrical operation risks based on the data fusion results using feature engineering technology, and the feature variables include temperature rise rate, voltage fluctuation rate, leakage current increment and equipment operation time change rate;

[0009] The risk assessment module uses support vector machine and random forest algorithm to assess the risk of electrical operations based on the extracted characteristic variables;

[0010] The real-time monitoring and early warning module monitors the electrical working environment in real time based on the risk assessment results and sets a multi-level alarm mechanism. The alarm levels include low risk, medium risk and high risk. When the risk factor exceeds the preset threshold, an alarm signal is generated and real-time early warning information is sent to the operator.

[0011] The historical data storage and analysis module is used to store historical multi-source data and evaluation results of electrical operations, and regularly analyze the long-term trends of the data;

[0012] The visual monitoring module presents the evaluation results, warning information and historical data analysis results in a visual manner.

[0013] Optionally, the data acquisition module includes:

[0014] Sensor layout planning: Plan the layout of sensors according to the characteristics of the electrical working environment and the distribution of equipment. Sensors include temperature sensors, humidity sensors, current sensors, voltage sensors, leakage current sensors, and ground resistance sensors.

[0015] Data collection strategy: Set the dynamic collection frequency according to the characteristics of the physical quantity monitored by the sensor;

[0016] Data processing: Process the collected multi-source data, including noise filtering and outlier detection.

[0017] Optionally, the data fusion module includes:

[0018] Multi-source data preprocessing: Before multi-source data fusion, multi-source data is preprocessed, including timestamp alignment and unit standardization;

[0019] Data cleaning: Use the Kalman filter algorithm to perform preliminary cleaning on multi-source data, filter noise and eliminate random errors;

[0020] Support vector regression fusion model construction: Based on the cleaned multi-source data, the support vector regression algorithm is used to build a fusion model to fuse the multi-source data;

[0021] Fusion data quality assessment: Perform consistency assessment and quality verification on the fused data.

[0022] Optionally, the feature extraction module specifically includes:

[0023] Feature variable extraction: feature engineering technology is used to extract feature variables related to electrical operation risks from the fused multi-source data;

[0024] Feature dimensionality reduction processing: For the extracted feature variable set, the principal component analysis method is used to reduce the dimensionality;

[0025] Feature evaluation and output: The importance weight of each feature variable is calculated using the random forest algorithm to identify the features that contribute most to risk assessment.

[0026] Optionally, the risk assessment model building module includes:

[0027] Data preparation: Receive high-quality feature variable data output by the feature extraction module, including temperature rise rate, voltage fluctuation rate, leakage current increment, and equipment operation time change rate;

[0028] The data is divided into training set, validation set and test set with a ratio of 70%, 15% and 15% respectively for model training, parameter tuning and performance evaluation;

[0029] Classification model construction: The support vector machine algorithm is used to build a support vector machine model to identify risk states in electrical operations;

[0030] Random forest model construction: Based on the random forest algorithm, a random forest model is constructed to generate the final classification results through the voting mechanism of multiple trees;

[0031] Model fusion: The outputs of the support vector machine model and the random forest model are weighted and integrated to combine the classification results of the support vector machine model and the random forest model.

[0032] Optionally, the real-time monitoring and early warning module specifically includes:

[0033] Multi-dimensional risk factor analysis: Calculate the comprehensive risk index based on the temperature rise rate, voltage fluctuation rate, leakage current increment and equipment operation time change rate;

[0034] Risk level determination: Based on the comprehensive risk index, the current risk is divided into low risk, medium risk and high risk;

[0035] Alarm mechanism: set up a multi-level alarm mechanism;

[0036] Generation and sending of early warning information: Based on the results of risk factor analysis, an early warning report containing the current risk level, influencing factors and recommended measures is generated, and the early warning information is sent to the visual monitoring module in real time for operators to view and analyze. At the same time, it is sent to relevant responsible personnel via SMS or email to ensure timely communication of information.

[0037] Optionally, the real-time monitoring and early warning module further includes:

[0038] Dynamic optimization of alarm thresholds: Dynamically adjust alarm thresholds based on the long-term trend analysis results of the historical data storage and analysis module;

[0039] Optimization of multi-level alarm mechanism: adjust the priority of alarm response according to the fluctuation range and duration of risk factors;

[0040] Multi-channel alarm linkage: optimize the alarm notification method to ensure that in high-risk situations, multi-channel linkage alarms cover all relevant operators;

[0041] Alarm record: After each alarm is generated, the alarm type, trigger time, associated risk factors and response status are automatically recorded.

[0042] Optionally, the historical data storage and analysis module includes:

[0043] Data collection and storage: Collect and store historical multi-source data and evaluation results of electrical operations;

[0044] K-Means cluster analysis: Use the K-Means clustering algorithm to perform cluster analysis on historical multi-source data to identify risk patterns under different working conditions;

[0045] Long-term risk trend forecast: Long-term trend forecast based on historical multi-source data analysis results.

[0046] Optionally, the visual monitoring module:

[0047] Data reception: The visual monitoring module receives the assessment results of the risk assessment module, the warning information generated by the real-time monitoring and warning module, and the analysis results of the historical data storage and analysis module;

[0048] Visualization interface: display the evaluation results, warning information and historical data analysis results through the visualization interface;

[0049] Interactive function design: Provide interactive functions to allow users to operate through a visual interface, including data query and view switching;

[0050] Real-time alarm notification: Provides multiple ways to display real-time alarm information.

[0051] Beneficial effects of the present invention:

[0052] The present invention uses a feature extraction module to extract characteristic variables (such as temperature rise rate, voltage fluctuation rate, leakage current increment, etc.) that are highly related to electrical operation risks from fused data, and combines principal component analysis technology to perform dimensionality reduction processing, effectively reducing data complexity while retaining key information. Through the analysis of the changing trend of characteristic variables, the module can accurately identify potential risk factors, provide high-quality input data for subsequent risk assessment modules, and significantly improve the overall system's ability to handle complex data environments.

[0053] The present invention adopts a model construction method combining support vector machine and random forest. Through the complementary advantages of the two models, it can not only accurately classify the high, medium and low risk states in electrical operations, but also explain the contribution of key feature variables based on the random forest algorithm, thereby improving the transparency and reliability of the model. At the same time, the performance of the model is comprehensively evaluated and dynamically optimized through multiple indicators to ensure its robustness and adaptability in complex electrical operation environments.

[0054] The present invention, based on the real-time output of the risk assessment module, can dynamically monitor the electrical operation environment and set up a multi-level alarm mechanism of low risk, medium risk and high risk. The module sends early warning information to the operator in a timely manner through multi-channel linkage (such as sound prompts, mobile device notifications), and provides detailed alarm reasons and recommended measures. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only for the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0056] Figure 1 is a system schematic diagram of an embodiment of the present invention;

[0057] Figure 2 Schematic diagram of the risk assessment module process of an embodiment of the present invention. DETAILED DESCRIPTION

[0058] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. At the same time, it is explained here that in order to make the embodiments more detailed, the following embodiments are the best and preferred embodiments, and those skilled in the art may also adopt other alternatives to implement some known technologies; and the accompanying drawings are only for more specific description of the embodiments, and are not intended to specifically limit the present invention.

[0059] It should be noted that the references to "one embodiment", "an embodiment", "an exemplary embodiment", "some embodiments" and the like in the specification indicate that the embodiments described may include specific features, structures or characteristics, but not every embodiment may include the specific features, structures or characteristics. In addition, when a specific feature, structure or characteristic is described in conjunction with an embodiment, it should be within the knowledge of a person skilled in the art to implement such feature, structure or characteristic in conjunction with other embodiments (whether or not explicitly described).

[0060] In general, a term can be understood, at least in part, from its use in context. For example, depending, at least in part, on the context, the term "one or more" as used herein can be used to describe any feature, structure, or characteristic in the singular sense, or can be used to describe a combination of features, structures, or characteristics in the plural sense. Additionally, the term "based on" can be understood as not necessarily intended to convey an exclusive set of factors, but can instead, depending, at least in part, on the context, allow for the presence of other factors that are not necessarily explicitly described.

[0061] like Figure 1-Figure 2 As shown, an electrical operation risk assessment and early warning system based on big data analysis includes a data acquisition module, a data fusion module, a feature extraction module, a risk assessment module, a real-time monitoring and early warning module, a historical data storage and analysis module and a visual monitoring module, wherein;

[0062] The data acquisition module deploys a variety of sensors to collect multi-source data at the electrical operation site in real time. The multi-source data includes temperature, humidity, current, voltage, leakage current and ground resistance.

[0063] The data fusion module uses a fusion algorithm based on Kalman filtering and support vector regression to fuse multi-source data to eliminate data redundancy and improve data consistency;

[0064] Based on the data fusion results, the feature extraction module uses feature engineering technology to extract feature variables related to electrical operation risks. The feature variables include temperature rise rate, voltage fluctuation rate, leakage current increment, and equipment operation time change rate.

[0065] The risk assessment module uses support vector machine and random forest algorithm to assess the risk of electrical operations based on the extracted feature variables;

[0066] The real-time monitoring and early warning module monitors the electrical operating environment in real time based on the risk assessment results and sets up a multi-level alarm mechanism. The alarm levels include low risk, medium risk and high risk. When the risk factor exceeds the preset threshold, an alarm signal is generated and real-time early warning information is sent to the operator.

[0067] The historical data storage and analysis module is used to store historical multi-source data and evaluation results of electrical operations, and regularly analyze the long-term trends of the data;

[0068] The visual monitoring module presents the evaluation results, warning information and historical data analysis results in a visual way.

[0069] The data acquisition module includes:

[0070] Sensor layout planning: Plan the layout of sensors according to the characteristics of the electrical working environment and the distribution of equipment. Sensors include temperature sensors, humidity sensors, current sensors, voltage sensors, leakage current sensors and ground resistance sensors, among which;

[0071] Temperature sensors are placed on the shell surface of key heat-generating equipment and near heat sources to monitor equipment temperature changes in real time;

[0072] Humidity sensors are installed in electrical cabinets or working environments to monitor changes in ambient humidity;

[0073] Current sensors and voltage sensors are installed on the power distribution lines, focusing on monitoring the current and voltage status of the power supply lines;

[0074] Leakage current sensors are installed at key nodes of the power distribution line to detect current leakage;

[0075] The grounding resistance sensor is arranged at the grounding device to monitor the resistance change of the grounding system in real time;

[0076] Data collection strategy: To ensure the real-time and validity of the collected data, a dynamic collection frequency is set according to the physical quantity characteristics monitored by the sensor, and a data cache mechanism is set at the sensor node to temporarily store the collected raw data to reduce the risk of data loss;

[0077] The sampling frequency of high-frequency parameters (such as current and voltage) is set to above 10Hz to ensure real-time capture of sudden abnormal fluctuations;

[0078] The sampling frequency of low-frequency parameters (such as temperature and humidity) is set to 1Hz-2Hz to monitor slow-changing trends;

[0079] Data processing: Process the collected multi-source data, including noise filtering and outlier detection;

[0080] (1) Noise filtering: The collected raw data is denoised using a Kalman filter to remove random errors caused by environmental interference;

[0081] (2) Outlier detection: The statistically based triple standard deviation method is used to detect and mark outliers in the collected data;

[0082] The data acquisition module can acquire multi-source data from the electrical operation site in real time, efficiently and stably, provide high-quality input data for the data fusion module, and ensure the operational reliability and accuracy of the entire system.

[0083] The data fusion module includes:

[0084] Multi-source data preprocessing: Before multi-source data fusion, multi-source data is preprocessed, including timestamp alignment and unit standardization;

[0085] Timestamp alignment: Align the timestamps of asynchronously collected data from different sensors, and use linear interpolation to fill the inconsistency of data collection time intervals to ensure that the input data has a consistent time base;

[0086] Unit standardization: Convert data of different physical quantities (such as temperature, current, and voltage) into the International System of Units to ensure that multi-source data has a unified dimension for subsequent fusion;

[0087] Data cleaning: Use the Kalman filter algorithm to perform preliminary cleaning on multi-source data, filter noise and eliminate random errors, including:

[0088] State space definition: Define multi-source data as a state space model. Let x t is the current state, z t is the observed value, and the state transfer equation is:

[0089] x t =Ax t-1 +w t ;

[0090] Among them, A is the state transfer matrix, w t is the process noise, and the observation equation is:

[0091] z t =Hx t +v t ;

[0092] Among them, H is the observation matrix, v t is the observation noise;

[0093] Filter update: through the Kalman gain K t Calculate state estimates The formula is:

[0094]

[0095] Among them, K t is the gain matrix, defined as:

[0096] K t =P t-1 H T (HP t-1 H T +R) -1 ;

[0097] Among them, P t-1 is the prediction covariance matrix, R is the observation noise covariance;

[0098] Support vector regression fusion model construction: Based on the cleaned multi-source data, the support vector regression algorithm is used to build a fusion model to fuse the multi-source data, including:

[0099] Input feature selection: The multi-source data after Kalman filter cleaning is used as the input features of the support vector regression model. The feature variables include temperature rise rate, voltage fluctuation rate, leakage current increment, etc.

[0100] Objective function definition: The optimization objective function of support vector regression is defined as:

[0101]

[0102] Among them, w is the regression weight, C is the regularization parameter, ξ i is the loss function; fusion result calculation: the output of the fusion data is the predicted value of the support vector regression model, expressed as:

[0103]

[0104] Among them, X i is single sensor data, w i is the fusion weight, b is the bias term;

[0105] Fusion data quality assessment: Consistency assessment and quality verification of fused data, including:

[0106] (1) Data consistency detection: Compare the fused data with the original multi-source data to determine whether the fusion result meets the set error range. The error calculation formula is:

[0107]

[0108] When Δ exceeds a set threshold (e.g., 5%), the fusion model parameters are readjusted;

[0109] (2) Data integrity check: Ensure that all sensor data are included in the fusion model without omission or loss.

[0110] The feature extraction module specifically includes:

[0111] Feature variable extraction: Feature engineering technology is used to extract feature variables related to electrical operation risks from the fused multi-source data, including the following:

[0112] Temperature rise rate: Calculate the temperature change rate based on time series data. The formula is:

[0113]

[0114] Among them, ΔT is the change of temperature per unit time, and Δt is the time interval;

[0115] Voltage fluctuation rate: By calculating the ratio of the standard deviation of the voltage per unit time to the mean, the formula is expressed as:

[0116]

[0117] Among them, σ V is the standard deviation of voltage, μ V is the mean value of voltage;

[0118] Leakage current increment: By comparing the leakage current values ​​at consecutive time points, its increment is calculated and expressed as:

[0119] Leakage current increment = I t -I t-1 ;

[0120] Among them, I t and I t-1 Respectively represent the leakage current value at the current time and the previous time;

[0121] Equipment running time change rate: By counting the change trend of equipment running time, the change rate characteristics are extracted. The calculation formula is expressed as:

[0122]

[0123] Feature dimensionality reduction processing: For the extracted feature variable set, the principal component analysis method is used for dimensionality reduction processing. The specific steps are as follows:

[0124] (1) Covariance matrix calculation: Calculate the covariance matrix of feature variables to measure the correlation between different features;

[0125] (2) Principal component selection: By calculating eigenvalues ​​and eigenvectors, the principal components with a cumulative contribution rate of more than 95% are selected to reduce the dimension of the feature data;

[0126] (3) Feature mapping: Mapping the original feature variables to a low-dimensional space to ensure data simplification while retaining the main information;

[0127] Feature evaluation and output: The importance weight of each feature variable is calculated using the random forest algorithm to identify the features that contribute most to risk assessment;

[0128] The feature extraction module can efficiently extract key feature variables related to electrical operation risks, provide high-quality input data for the risk assessment model building module, and ensure the accuracy and reliability of the overall system.

[0129] The risk assessment model building modules include:

[0130] Data preparation: Receive high-quality feature variable data output by the feature extraction module, including temperature rise rate, voltage fluctuation rate, leakage current increment, and equipment operation time change rate;

[0131] The data is divided into training set, validation set and test set with a ratio of 70%, 15% and 15% respectively for model training, parameter tuning and performance evaluation;

[0132] Classification model construction: The support vector machine algorithm is used to build a support vector machine model to identify the risk status in electrical operations. The objective function is expressed as:

[0133]

[0134] Among them, w is the weight vector of the classification hyperplane, C is the regularization parameter, ξ i is the slack variable;

[0135] Kernel function selection: According to the nonlinear distribution of feature variables, select the radial basis kernel function (RBF):

[0136] K(x i ,x j )=exp(-γ||x i -x j || 2 );

[0137] Among them, γ is the kernel parameter, which is used to control the nonlinear mapping ability of the model;

[0138] Model training: Based on the training set data, the objective function is optimized using the gradient descent algorithm to determine model parameters such as w, b, and γ;

[0139] Random forest model construction: Based on the random forest algorithm, a random forest model is constructed to generate the final classification results through the voting mechanism of multiple trees, which effectively avoids the overfitting problem that may occur in a single decision tree model and provides efficient and stable support for risk prediction, including:

[0140] Model initialization: The random forest model consists of N decision trees, where each tree is generated by introducing random sampling of features and random sampling of samples (out-of-bag method, OOB);

[0141] Splitting rule: In the splitting process of each tree, information gain or Gini coefficient is used as the node splitting criterion; the information gain formula is:

[0142]

[0143] Among them, H(D) is the information entropy of data set D, D iis the branch dataset;

[0144] Feature importance calculation: By measuring the contribution of each feature in all tree splits, the feature importance weight is calculated to explain the criticality of the risk factor;

[0145] Model training: Generate the final classification result through the voting mechanism of multiple trees to avoid overfitting of a single model;

[0146] Model fusion: The outputs of the support vector machine model and the random forest model are weighted and fused, and the classification results of the support vector machine model and the random forest model are integrated. The fusion formula is expressed as:

[0147] P 融合 =w1P SVM +w2P RF ;

[0148] Among them, P SVM and P RF are the prediction probabilities of the support vector machine model and the random forest model, respectively. w1 and w2 are weight coefficients, which are determined by tuning the validation set;

[0149] According to the fusion probability P 融合 The thresholds are used to divide the risk levels, which are specifically expressed as:

[0150] P 融合 When <0.3, it is low risk;

[0151] 0.3≤P 融合 When <0.7, it is medium risk;

[0152] P 融合 ≥0.7 indicates high risk.

[0153] The real-time monitoring and early warning module specifically includes:

[0154] Multi-dimensional risk factor analysis: Calculate the comprehensive risk index based on the temperature rise rate, voltage fluctuation rate, leakage current increment and equipment operation time change rate, expressed as:

[0155]

[0156] Among them, X i is the characteristic variable value, w i is the feature weight;

[0157] Risk level determination: Based on the comprehensive risk index, the current risk is divided into low risk, medium risk and high risk. The determination conditions are as follows:

[0158] Low risk: comprehensive risk index R < 30;

[0159] Medium risk: 30≤R<70;

[0160] High risk: R ≥ 70;

[0161] Alarm mechanism: Set up a multi-level alarm mechanism, including:

[0162] Low risk alarm: record the low risk status, update the system log, but do not generate active alarm;

[0163] Medium risk alarm: Generates a medium risk alarm signal to alert operators and sends the alarm information to the operator's mobile device via wireless communication;

[0164] High-risk alarm: Trigger a high-risk alarm signal, prompt the operator to take immediate action through a multi-level linkage mechanism of sound, flash and mobile device notification, and record a detailed alarm log;

[0165] Generation and sending of early warning information: Based on the results of risk factor analysis, an early warning report containing the current risk level, influencing factors and recommended measures is generated, and the early warning information is sent to the visual monitoring module in real time for operators to view and analyze. At the same time, it is sent to relevant responsible personnel via SMS or email to ensure timely communication of information.

[0166] The real-time monitoring and early warning module also includes:

[0167] Dynamic optimization of alarm thresholds: Based on the long-term trend analysis results of the historical data storage and analysis module, the alarm thresholds are dynamically adjusted, including:

[0168] (1) Use time series analysis methods to predict the changing trends of future risk factors;

[0169] (2) Adjust the comprehensive risk index range of low risk, medium risk and high risk to make the alarm mechanism adapt to changes in the operating environment. The threshold optimization formula is expressed as:

[0170] R′=R+ΔR;

[0171] Where R is the current alarm threshold, and ΔR is the adjustment amount calculated based on trend prediction;

[0172] Optimization of multi-level alarm mechanism: Adjust the priority of alarm response according to the fluctuation range and duration of risk factors, for example:

[0173] If the risk factor fluctuates quickly, an alarm will be triggered in advance;

[0174] If the high-risk status lasts longer than the set time, the alarm level will be upgraded and higher-level management personnel will be notified;

[0175] Multi-channel alarm linkage: optimize the alarm notification method to ensure that in high-risk situations, multi-channel (such as sound, flash, mobile device message, etc.) linkage alarms cover all relevant operators;

[0176] Alarm record: After each alarm is generated, the alarm type, trigger time, associated risk factors and response status are automatically recorded, and the log is stored in the historical data storage and analysis module.

[0177] The historical data storage and analysis modules include:

[0178] Data collection and storage: Collect and store historical multi-source data and evaluation results of electrical operations, including:

[0179] Data source integration: Receive historical data from the data acquisition module, data fusion module and real-time monitoring and early warning module, including temperature, voltage, leakage current, risk assessment results and alarm records;

[0180] Distributed storage architecture: Use distributed database to efficiently store historical multi-source data, supporting fast data reading and writing and scalability;

[0181] Data classification management: Establish classification labels according to data type (such as physical quantity data, alarm data, operation response data), time and source to ensure the orderliness of data storage and ease of query;

[0182] K-Means cluster analysis: Use the K-Means clustering algorithm to perform cluster analysis on historical multi-source data to identify risk patterns under different working conditions, including:

[0183] Cluster feature selection: Select representative feature variables (such as temperature rise rate, voltage fluctuation rate, leakage current increment, etc.) as input;

[0184] Cluster center initialization: Determine the initial cluster center by random selection or density-based method;

[0185] Clustering process execution: It iteratively optimizes the distance between data points and cluster centers until the clustering results converge, and outputs clustering results of different risk patterns;

[0186] Long-term risk trend forecast: Long-term trend forecast based on historical multi-source data analysis results, including:

[0187] Trend modeling: Use time series analysis techniques to establish a long-term change model for key characteristic variables;

[0188] Risk prediction generation: predict the possible future risk change range based on trends, and provide prediction basis for real-time monitoring and early warning modules;

[0189] Forecast accuracy verification: Verify the reliability of the forecast results based on the actual changes in historical data, and dynamically adjust the trend model parameters.

[0190] Visual monitoring module:

[0191] Data reception: The visual monitoring module receives the assessment results of the risk assessment module, the warning information generated by the real-time monitoring and warning module, and the analysis results of the historical data storage and analysis module;

[0192] Visualization interface: display the evaluation results, warning information and historical data analysis results through the visualization interface;

[0193] Interactive function design: Provide interactive functions to allow users to operate through a visual interface, including data query and view switching;

[0194] Real-time alarm notification: Provides multiple ways to display real-time alarm information, including:

[0195] Interface pop-up notification: Alarm prompts will pop up in real time on the monitoring interface, showing the alarm level, trigger time and handling suggestions;

[0196] Color and sound warning: highlight the current alarm status with a bright color and attract the operator's attention with a sound prompt;

[0197] Multi-device synchronous notification: Synchronize alarm information to mobile devices to ensure that operators can receive information and take countermeasures in time.

[0198] The present invention covers any substitution, modification, equivalent method and scheme made on the essence and scope of the present invention. In order to make the public have a thorough understanding of the present invention, specific details are described in detail in the following preferred embodiments of the present invention, but those skilled in the art can fully understand the present invention without the description of these details. In addition, in order to avoid unnecessary confusion about the essence of the present invention, well-known methods, processes, procedures, components and circuits are not described in detail.

[0199] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. An electrical operation risk assessment and early warning system based on big data analysis, characterized in that: It includes data acquisition module, data fusion module, feature extraction module, risk assessment module, real-time monitoring and early warning module, historical data storage and analysis module and visual monitoring module, among which; The data acquisition module deploys a variety of sensors to collect multi-source data at the electrical operation site in real time, and the multi-source data includes temperature, humidity, current, voltage, leakage current and ground resistance; The data fusion module uses a fusion algorithm based on Kalman filtering and support vector regression to fuse multi-source data; The feature extraction module extracts feature variables related to electrical operation risks based on the data fusion results using feature engineering technology, and the feature variables include temperature rise rate, voltage fluctuation rate, leakage current increment and equipment operation time change rate; The risk assessment module uses support vector machine and random forest algorithm to assess the risk of electrical operations based on the extracted characteristic variables; The real-time monitoring and early warning module monitors the electrical working environment in real time based on the risk assessment results and sets a multi-level alarm mechanism. The alarm levels include low risk, medium risk and high risk. When the risk factor exceeds the preset threshold, an alarm signal is generated and real-time early warning information is sent to the operator. The historical data storage and analysis module is used to store historical multi-source data and evaluation results of electrical operations, and regularly analyze the long-term trends of the data; The visual monitoring module presents the evaluation results, warning information and historical data analysis results in a visual manner.

2. The electrical operation risk assessment and early warning system based on big data analysis according to claim 1 is characterized in that: The data acquisition module comprises: Sensor layout planning: Plan the layout of sensors according to the characteristics of the electrical working environment and the distribution of equipment. Sensors include temperature sensors, humidity sensors, current sensors, voltage sensors, leakage current sensors, and ground resistance sensors. Data collection strategy: Set the dynamic collection frequency according to the characteristics of the physical quantity monitored by the sensor; Data processing: Process the collected multi-source data, including noise filtering and outlier detection.

3. The electrical operation risk assessment and early warning system based on big data analysis according to claim 2 is characterized in that: The data fusion module comprises: Multi-source data preprocessing: Before multi-source data fusion, multi-source data is preprocessed, including timestamp alignment and unit standardization; Data cleaning: Use the Kalman filter algorithm to perform preliminary cleaning on multi-source data, filter noise and eliminate random errors; Support vector regression fusion model construction: Based on the cleaned multi-source data, the support vector regression algorithm is used to build a fusion model to fuse the multi-source data; Fusion data quality assessment: Perform consistency assessment and quality verification on the fused data.

4. The electrical operation risk assessment and early warning system based on big data analysis according to claim 3 is characterized in that: The feature extraction module specifically includes: Feature variable extraction: feature engineering technology is used to extract feature variables related to electrical operation risks from the fused multi-source data; Feature dimensionality reduction processing: For the extracted feature variable set, the principal component analysis method is used to reduce the dimensionality; Feature evaluation and output: The importance weight of each feature variable is calculated using the random forest algorithm to identify the features that contribute most to risk assessment.

5. The electrical operation risk assessment and early warning system based on big data analysis according to claim 4 is characterized in that: The risk assessment model building module includes: Data preparation: receiving the characteristic variable data output by the feature extraction module, including the temperature rise rate, voltage fluctuation rate, leakage current increment and equipment operation time change rate; Classification model construction: The support vector machine algorithm is used to build a support vector machine model to identify risk states in electrical operations; Random forest model construction: Based on the random forest algorithm, a random forest model is constructed to generate the final classification results through the voting mechanism of multiple trees; Model fusion: The outputs of the support vector machine model and the random forest model are weighted and integrated to combine the classification results of the support vector machine model and the random forest model.

6. The electrical operation risk assessment and early warning system based on big data analysis according to claim 5 is characterized in that: The real-time monitoring and early warning module specifically includes: Multi-dimensional risk factor analysis: Calculate the comprehensive risk index based on the temperature rise rate, voltage fluctuation rate, leakage current increment and equipment operation time change rate; Risk level determination: Based on the comprehensive risk index, the current risk is divided into low risk, medium risk and high risk; Alarm mechanism: set up a multi-level alarm mechanism; Generation and sending of early warning information: Based on the results of risk factor analysis, an early warning report containing the current risk level, influencing factors and recommended measures is generated, and the early warning information is sent to the visual monitoring module in real time, and also sent to relevant responsible personnel via SMS or email.

7. The electrical operation risk assessment and early warning system based on big data analysis according to claim 6 is characterized in that: The real-time monitoring and early warning module also includes: Dynamic optimization of alarm thresholds: Dynamically adjust alarm thresholds based on the long-term trend analysis results of the historical data storage and analysis module; Optimization of multi-level alarm mechanism: adjust the priority of alarm response according to the fluctuation range and duration of risk factors; Multi-channel alarm linkage: optimize the alarm notification method to ensure that in high-risk situations, multi-channel linkage alarms cover all relevant operators; Alarm record: After each alarm is generated, the alarm type, trigger time, associated risk factors and response status are automatically recorded.

8. The electrical operation risk assessment and early warning system based on big data analysis according to claim 7 is characterized in that: The historical data storage and analysis module includes: Data collection and storage: Collect and store historical multi-source data and evaluation results of electrical operations; K-Means cluster analysis: Use the K-Means clustering algorithm to perform cluster analysis on historical multi-source data to identify risk patterns under different working conditions; Long-term risk trend forecast: Long-term trend forecast based on historical multi-source data analysis results.

9. The electrical operation risk assessment and early warning system based on big data analysis according to claim 8, characterized in that: The visual monitoring module: Data reception: The visual monitoring module receives the assessment results of the risk assessment module, the warning information generated by the real-time monitoring and warning module, and the analysis results of the historical data storage and analysis module; Visualization interface: display the evaluation results, warning information and historical data analysis results through the visualization interface; Interactive function design: Provide interactive functions to allow users to operate through a visual interface, including data query and view switching; Real-time alarm notification: Provides multiple ways to display real-time alarm information.

Images