A risk assessment method for power field operations based on a multi-dimensional operation risk assessment model

Abstract

The present invention discloses a method for assessing the risk of electric power field operations based on a multi-dimensional operation risk assessment model. The method collects real-time data through sensors and IoT devices to obtain historical operation data, accident records, maintenance logs and operator information; constructs a multi-dimensional operation risk assessment model through a fuzzy comprehensive evaluation algorithm based on a support vector machine; inputs the collected real-time data and historical data into the multi-dimensional operation risk assessment model to obtain risk information of current electric power field operations; integrates a risk response rule engine to output corresponding response strategies according to risk information; creates a risk heat map through a visualization tool to intuitively display high-risk areas; and regularly generates and stores risk assessment reports and risk control effect reports. The present invention realizes multi-dimensional and high-precision real-time risk assessment of electric power field operations.

Description

Technical Field

[0001] The present invention relates to risk assessment technology, and more specifically to a method for risk assessment of power field operations based on a multi-dimensional operation risk assessment model. Background Art

[0002] With the increasing complexity and scale of power systems, the safety risks of power field operations are also rising. Traditional risk assessment methods mainly rely on manual experience and single-dimensional data, ignoring the comprehensive analysis of multi-dimensional data, resulting in incomplete risk assessment and reliance on post-analysis. It is impossible to monitor and assess the risks of field operations in real time, and it is difficult to issue early warnings and responses in a timely manner. Due to the lack of advanced data processing and analysis technology, traditional methods have major deficiencies in the precision and accuracy of risk assessment, and are prone to misjudgment or omissions. In addition, risk assessment results are usually presented in the form of text or simple charts, which makes it difficult to intuitively display the risk distribution and is not conducive to managers quickly identifying and dealing with high-risk areas.

[0003] In recent years, with the development of Internet of Things (IoT) technology and big data analysis technology, multi-dimensional data fusion and real-time risk assessment methods have gradually become research hotspots. However, existing methods still have shortcomings in data fusion, real-time and visualization, and it is difficult to meet the needs of actual applications. Summary of the invention

[0004] In view of the shortcomings of the above-mentioned technology, the present invention provides a power field operation risk assessment method based on a multi-dimensional operation risk assessment model, which improves the safety of power field operations through multi-dimensional data fusion and real-time risk assessment.

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

[0006] A method for risk assessment of power field operations based on a multi-dimensional operation risk assessment model comprises the following steps:

[0007] Step 1: Use sensors and IoT devices to collect real-time data at the power site, including equipment status and operating environment, and obtain historical operation data, accident records, maintenance logs, and operator information from the database;

[0008] Step 2: construct a multidimensional operation risk assessment model through a fuzzy comprehensive evaluation algorithm model based on a support vector machine, wherein the multidimensional operation risk assessment model includes a data receiving module, a data preprocessing module, an equipment status assessment module, an environmental risk assessment module, a personnel risk assessment module, a multidimensional risk association module, a comprehensive risk assessment module and a risk information output module;

[0009] Step 3: Input the collected real-time data and historical data into a multi-dimensional operation risk assessment model to obtain the risks and risk information of the current power field operation, wherein the risk information includes risk type, risk assessment value and risk level;

[0010] Step 4: Integrate the risk response rule engine to automatically output the corresponding response strategy based on the risk information output by the model, and notify relevant personnel via SMS, email or APP (mobile software);

[0011] Step 5: Based on the risks and risk information of the power site operation, a risk heat map is created through a visualization tool. The risk heat map can distribute the risks on a grid map of the power site and intuitively display high-risk areas.

[0012] Step 6. Generate risk assessment reports and risk control effectiveness reports regularly through the automatic report generation tool JasperReports, and store them in the MySQL database for easy query and viewing by relevant personnel.

[0013] As a further embodiment of the present invention, in step one, the sensors and IoT devices include but are not limited to surveillance cameras, smoke detectors, temperature sensors, humidity sensors and vibration sensors, and the sensors and IoT devices can monitor the equipment status and changes in the working environment at the power site from multiple dimensions.

[0014] As a further embodiment of the present invention, in the step 2, the data receiving module communicates with sensors and IoT devices through the MQTT protocol to receive real-time data and obtain historical operation data, accident records, maintenance logs and operator information of power field operations;

[0015] The data preprocessing module filters the noise in the received data through Kalman filtering, fills the missing values ​​through KNN, and converts the data to a uniform scale using the Z-Score standardization method;

[0016] The equipment status assessment module diagnoses the operating status of the equipment based on historical data and real-time data through an equipment status assessment model constructed by a support vector machine;

[0017] The environmental risk assessment module assesses the operating environment risk based on the environmental sensor data through an environmental risk assessment model constructed by a support vector machine;

[0018] The personnel risk assessment module identifies potential personnel risks based on the behavior data of the operating personnel by using a personnel risk assessment model constructed by a support vector machine;

[0019] The multi-dimensional risk association module integrates the risk data of equipment, environment and personnel through DS evidence theory, and generates a risk association matrix by analyzing the association relationship between different risks through a priori knowledge algorithm;

[0020] The comprehensive risk assessment module comprehensively evaluates the risk information of each dimension based on the fused data and the risk association matrix based on the fuzzy comprehensive evaluation method to generate the final risk assessment result;

[0021] The risk information output module outputs the final risk assessment results through a compatible RESTful API interface;

[0022] The output end of the data receiving module is connected to the input end of the data preprocessing module, the output end of the data preprocessing module is connected to the input end of the equipment status assessment module, the environmental risk assessment module and the personnel risk assessment module, the output end of the equipment status assessment module, the environmental risk assessment module and the personnel risk assessment module are connected to the input end of the multidimensional risk association module, the output end of the multidimensional risk association module is connected to the input end of the comprehensive risk assessment module, and the output end of the comprehensive risk assessment module is connected to the input end of the risk information output module.

[0023] As a further embodiment of the present invention, the multi-dimensional risk association module includes a risk data fusion unit and a risk relationship analysis unit;

[0024] The risk data fusion unit fuses the risk data of equipment, environment and personnel through DS evidence theory to generate a comprehensive risk data set;

[0025] The risk relationship analysis unit mines frequent item sets and association rules from the comprehensive risk data set through an association rule mining method based on a priori knowledge algorithm, analyzes the association relationship between risks of different dimensions and generates a risk association matrix;

[0026] The output end of the risk data fusion unit is connected to the input end of the risk relationship analysis unit.

[0027] As a further embodiment of the present invention, the working method of the risk data fusion unit is:

[0028] Step s11, obtaining risk assessment data of the equipment status assessment module, the environmental risk assessment module and the personnel risk assessment module, respectively recorded as E 1. E 2 and E 3;

[0029] Step s12: Calculate the basic probability distribution of each risk assessment data by using a basic probability distribution function, the formula of the basic probability distribution function is:

[0030] ,in, m i ( i ) is the i The basic probability distribution function for each risk dimension is E i It is i Risk assessment data for each risk dimension, i is the set of risk states, k is the slope adjustment parameter, t is the threshold value, α is the amplitude of the sine function, β is the frequency of the sine function, e is a natural constant of mathematics;

[0031] Step s13: synthesize the basic probability distribution of each risk assessment data through the synthesis rule of DS evidence theory to obtain a comprehensive basic probability distribution. The synthesis formula is:

[0032] ,in, m ( i ) is the comprehensive basic probability distribution function, i is the set of risk states, K is the standardized coefficient, A , B and C They are subsets corresponding to different risk dimensions. , and are the basic probability distribution of subsets corresponding to different risk dimensions, c is the amplitude of the cosine function, d is the frequency of the cosine function;

[0033] Step s14: normalize the comprehensive basic probability distribution to obtain a comprehensive risk assessment value. The normalization formula is:

[0034] ,in, R is a comprehensive risk assessment value. is the set of all risk states, m ( i ) is the comprehensive basic probability distribution function, i is the set of risk states, l is the magnitude of the tangent function, m is the frequency of the tangent function.

[0035] As a further embodiment of the present invention, the working method of the risk relationship analysis unit is:

[0036] Step s21, preprocessing the comprehensive risk data set by using Kalman filtering and data cleaning tool OpenRefine to remove noise and outliers from the comprehensive risk data set;

[0037] Step s22, mining frequent item sets in the comprehensive risk data set through a priori knowledge algorithm;

[0038] Step s23, extracting association rules from frequent item sets through a priori knowledge algorithm to obtain support and confidence of the association rules;

[0039] Step s24: Filter high-credibility association rules according to support and confidence thresholds, and obtain the association strength of the association rules through the Lift indicator;

[0040] Step s25: Fill the association strength of the association rule into the risk association matrix, where the elements of the risk association matrix represent the association strength between different risk dimensions.

[0041] As a further embodiment of the present invention, the working method of the comprehensive risk assessment module is:

[0042] Step s31, obtaining a comprehensive risk data set and a risk association matrix generated by the multi-dimensional risk association module;

[0043] Step s32: Calculate the membership of each risk item by using the fuzzy comprehensive evaluation method, and construct a fuzzy evaluation matrix to map each risk item in the comprehensive risk data set to a fuzzy set. The membership of each risk item is calculated as:

[0044] ,in, It is a risk item i The membership degree of x is the risk assessment value, c i It is a risk item i The central value of is the width of the fuzzy set, is the adjustment parameter, exp It is an exponential function with the natural constant e as the base;

[0045] Step s33: According to the risk association matrix and the membership of each risk item, the comprehensive membership of risk items of different dimensions is calculated by Sugeno fuzzy reasoning. The comprehensive membership is calculated as:

[0046] ,in, is the comprehensive membership degree,w i It is a risk item i The weight of It is a risk item i The membership degree of Lift i It is a risk item i The strength of association, β i is the correlation strength influence coefficient, n Indicates the number of risk items;

[0047] Step s34, generating the final risk assessment result through fuzzy decision making according to the comprehensive membership of risk items of different dimensions;

[0048] Step s35: sort the severity of each risk item by fuzzy analytic hierarchy process to identify high-risk areas.

[0049] As a further embodiment of the present invention, in step five, the risk heat map is generated by the data visualization tool Matplotlib, and the data visualization tool Matplotlib divides the power field operation area into a number of grid points, and maps the calculated severity of the risk items to the grid points to generate a risk heat map; the risk heat map uses red to represent high-risk areas, and the color depth of each grid point represents the severity of the risk at the grid point, intuitively displaying the risk distribution of power field operations.

[0050] The beneficial effects of the present invention are: by collecting real-time data at the power site through sensors and IoT devices, including equipment status and operating environment, and combining historical data for evaluation, real-time risk monitoring of power site operations can be performed. A multi-dimensional operation risk assessment model is constructed by a fuzzy comprehensive evaluation algorithm model based on a support vector machine, and the collected real-time data and historical data are input into the multi-dimensional operation risk assessment model to obtain the risks and risk information existing in the current power site operations. By using DS evidence theory and prior knowledge algorithm, the risk data of equipment, environment and personnel are integrated to generate a comprehensive risk data set, thereby improving the comprehensiveness and accuracy of risk assessment. A risk heat map is generated through a data visualization tool to intuitively display the risk distribution of power site operations and help managers quickly identify high-risk areas. The integrated risk response rule engine and automatic report generation tool realize the automation of risk response and report generation, improve management efficiency, and realize multi-dimensional and high-precision real-time risk assessment of power site operations. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 It is a schematic diagram of the working process of the method of the present invention;

[0052] Figure 2It is a schematic diagram of the structure of the multi-dimensional operation risk assessment model of the present invention;

[0053] Figure 3 It is a schematic diagram of the work flow of the risk data fusion unit of the present invention;

[0054] Figure 4 This is a schematic diagram of the work flow of the risk relationship analysis unit of the present invention;

[0055] Figure 5 Schematic diagram of the workflow of the comprehensive risk assessment module of the present invention. DETAILED DESCRIPTION

[0056] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0057] like Figure 1 As shown, the present invention provides a method for risk assessment of power field operations based on a multi-dimensional operation risk assessment model, comprising the following steps:

[0058] Step 1: Use sensors and IoT devices to collect real-time data at the power site, including equipment status and operating environment, and obtain historical operation data, accident records, maintenance logs, and operator information from the database;

[0059] Step 2: construct a multidimensional operation risk assessment model through a fuzzy comprehensive evaluation algorithm model based on a support vector machine, wherein the multidimensional operation risk assessment model includes a data receiving module, a data preprocessing module, an equipment status assessment module, an environmental risk assessment module, a personnel risk assessment module, a multidimensional risk association module, a comprehensive risk assessment module and a risk information output module;

[0060] Step 3: Input the collected real-time data and historical data into a multi-dimensional operation risk assessment model to obtain the risks and risk information of the current power field operation, wherein the risk information includes risk type, risk assessment value and risk level;

[0061] Step 4: Integrate the risk response rule engine to automatically output the corresponding response strategy based on the risk information output by the model, and notify relevant personnel via SMS, email or APP;

[0062] Step 5: Based on the risks and risk information of the power site operation, a risk heat map is created through a visualization tool. The risk heat map can distribute the risks on a grid map of the power site and intuitively display high-risk areas.

[0063] Step 6. Generate risk assessment reports and risk control effectiveness reports regularly through the automatic report generation tool JasperReports, and store them in the MySQL database for easy query and viewing by relevant personnel.

[0064] In the above embodiment, the sensors and IoT devices in step one include but are not limited to surveillance cameras, smoke detectors, temperature sensors, humidity sensors and vibration sensors. The sensors and IoT devices can monitor the equipment status and changes in the working environment at the power site from multiple dimensions.

[0065] In a specific embodiment, the model of the surveillance camera is Hikvision DS-2CD2185FWD-I, with a resolution of 3840×2160, a frame rate of 30fps, and a night vision distance of 30 meters. It is installed in key equipment, work areas, and entrances and exits to cover the entire work site. The model of the smoke detector is Honeywell 5808W3, the sensor type is a photoelectric sensor, the sensitivity is 0.5% to 4% / ft, and the alarm response time is less than 10 seconds. It is installed in areas prone to fire such as substations, distribution rooms, and cable trenches. The model of the temperature sensor is Fluke 561, with a measurement range of -40℃ to 550℃, an accuracy of ±1.0℃, and a response time of less than 500ms. It is installed on equipment prone to heat such as transformers, switch cabinets, and cable joints. The model of the humidity sensor is Honeywell HIH-4000, with a measurement range of 0% to 100% RH, an accuracy of ±3.5%RH, and a response time of less than 15 seconds. It is installed in areas where humidity needs to be monitored, such as distribution rooms, substations, and cable trenches. The vibration sensor model is SKF CMSS 2200, with a measurement range of 0 to 100 mm / s, an accuracy of ±5%, and a frequency range of 10 Hz to 1 kHz. It is installed on equipment that needs to monitor vibration, such as motors, generators, and transformers. The data of all sensors and IoT devices are collected to the central control system through gateway devices for unified management, and further processing and analysis.

[0066] In the above embodiment, if Figure 2 As shown, in step 2, the data receiving module communicates with sensors and IoT devices through the MQTT protocol to receive real-time data and obtain historical operation data, accident records, maintenance logs and operator information of power field operations;

[0067] The data preprocessing module filters the noise in the received data through Kalman filtering, fills the missing values ​​through KNN, and converts the data to a uniform scale using the Z-Score standardization method;

[0068] The equipment status assessment module diagnoses the operating status of the equipment based on historical data and real-time data through an equipment status assessment model constructed by a support vector machine;

[0069] The environmental risk assessment module assesses the operating environment risk based on the environmental sensor data through an environmental risk assessment model constructed by a support vector machine;

[0070] The personnel risk assessment module identifies potential personnel risks based on the behavior data of the operating personnel by using a personnel risk assessment model constructed by a support vector machine;

[0071] The multi-dimensional risk association module integrates the risk data of equipment, environment and personnel through DS evidence theory, and generates a risk association matrix by analyzing the association relationship between different risks through a priori knowledge algorithm;

[0072] The comprehensive risk assessment module comprehensively evaluates the risk information of each dimension based on the fused data and the risk association matrix based on the fuzzy comprehensive evaluation method to generate the final risk assessment result;

[0073] The risk information output module outputs the final risk assessment results through a compatible RESTful API interface;

[0074] The output end of the data receiving module is connected to the input end of the data preprocessing module, the output end of the data preprocessing module is connected to the input end of the equipment status assessment module, the environmental risk assessment module and the personnel risk assessment module, the output end of the equipment status assessment module, the environmental risk assessment module and the personnel risk assessment module are connected to the input end of the multidimensional risk association module, the output end of the multidimensional risk association module is connected to the input end of the comprehensive risk assessment module, and the output end of the comprehensive risk assessment module is connected to the input end of the risk information output module.

[0075] In a specific embodiment, taking the equipment status assessment module as an example, the working method of the support vector machine to construct the equipment status assessment model is:

[0076] H1. Obtaining a data set of equipment operation status after Kalman filtering, missing value filling and standardization from a data preprocessing module, wherein the data set includes the temperature, vibration and historical fault records of the equipment;

[0077] H2. According to the needs of equipment status assessment, features related to risk assessment are selected and extracted through principal component analysis, and the data set is annotated according to the extracted features;

[0078] H3, divide the labeled data set into a training set and a test set for training and verifying the constructed equipment status assessment model;

[0079] H4, the support vector machine model selects the RBF kernel function, and defines the candidate value ranges of hyperparameters M and N as M=[0.1, 1, 10, 100], N=[0.001, 0.01, 0.1, 1];

[0080] H5, use 5-fold cross validation combined with grid search to traverse all kernel function and hyperparameter combinations, calculate the average performance index of each combination, and select the combination with the best performance to accurately identify equipment risks;

[0081] H6. Use the selected kernel function and hyperparameter combination to train the support vector machine model through the training set data to obtain the equipment status assessment model;

[0082] H7. Verify the equipment status assessment model through test set data and evaluate the performance indicators of the model to ensure the effectiveness of the model;

[0083] H8. Apply the trained equipment status assessment model to real-time data and historical data to evaluate the equipment status and generate corresponding risk assessment results.

[0084] In the above embodiment, the multi-dimensional risk association module includes a risk data fusion unit and a risk relationship analysis unit;

[0085] The risk data fusion unit fuses the risk data of equipment, environment and personnel through DS evidence theory to generate a comprehensive risk data set;

[0086] The risk relationship analysis unit mines frequent item sets and association rules from the comprehensive risk data set through an association rule mining method based on a priori knowledge algorithm, analyzes the association relationship between risks of different dimensions and generates a risk association matrix;

[0087] The output end of the risk data fusion unit is connected to the input end of the risk relationship analysis unit.

[0088] In a specific embodiment, Figure 3 As shown, the working method of the risk data fusion unit is:

[0089] Step s11, obtaining risk assessment data of the equipment status assessment module, the environmental risk assessment module and the personnel risk assessment module, respectively recorded as E 1. E 2 and E 3;

[0090] Step s12: Calculate the basic probability distribution of each risk assessment data by using a basic probability distribution function, the formula of the basic probability distribution function is:

[0091] ,in,m i ( i ) is the i The basic probability distribution function of each risk dimension reflects the current risk assessment data. E i The probability of risk status;

[0092] E i It is i The risk assessment data of each risk dimension is used as input data, representing the current risk assessment value. e is a natural constant of mathematics;

[0093] i is a risk state set, representing all possible risk states, and the basic probability distribution function m i ( i ) Assign probabilities to these risk states;

[0094] k is the slope adjustment parameter, which is used to adjust the slope of the function and affects the sensitivity of the basic probability distribution. k The value will make the function E i is more sensitive to changes in k A value of 0 makes the function smoother;

[0095] t is the threshold used to adjust the benchmark of risk assessment data by subtracting t , you can adjust the input data E i The benchmark allows the function to be calculated at different benchmark points;

[0096] α is the amplitude of the sine function, which determines the maximum value of the basic probability distribution function. α The value will make the output range of the function larger, and smaller α The value makes the output range smaller.

[0097] β is the frequency of the sine function, which affects the periodicity of the basic probability distribution function. β The value will make the function change faster, lower β The value makes the function change more slowly;

[0098] Step s13: synthesize the basic probability distribution of each risk assessment data through the synthesis rule of DS evidence theory to obtain a comprehensive basic probability distribution. The synthesis formula is:

[0099] ,in, m ( i ) is a comprehensive basic probability allocation function, which is obtained by synthesizing the basic probability allocation functions of different risk dimensions;

[0100] i is the risk state set, representing all possible risk states;

[0101] K is the normalization coefficient, which is used to ensure that the probability distribution after synthesis satisfies the constraint that the sum of probabilities is 1;

[0102] A , B and C They are subsets corresponding to different risk dimensions, which are used to perform intersection operations in the synthesis rules of DS evidence theory;

[0103] , and They are the basic probability distributions of subsets corresponding to different risk dimensions, which are used to calculate the comprehensive basic probability distribution in the synthesis process;

[0104] c It is the amplitude of the cosine function, which determines the maximum value of the cosine function and affects the weight of each risk dimension in the synthesis process;

[0105] d is the frequency of the cosine function, affects the periodicity of the cosine function, and adjusts the weight changes in the synthesis rule;

[0106] Step s14: normalize the comprehensive basic probability distribution to obtain a comprehensive risk assessment value. The normalization formula is:

[0107] ,in, R It is a comprehensive risk assessment value, reflecting the overall risk level; is the set of all risk states, representing the set of all possible risk states, and the comprehensive risk assessment value R It is calculated for these risk states;

[0108] m ( i ) is a comprehensive basic probability distribution function, representing the risk state set i The basic probability of is obtained by synthesizing the basic probability distribution functions of different risk dimensions;

[0109] i is the risk state set, representing all possible risk states, Indicates traversal and calculation of all possible risk states;

[0110] l It is the amplitude of the tangent function, which determines the maximum value of the tangent function and affects the weight of the comprehensive risk assessment value;

[0111] m It is the frequency of the tangent function, which affects the periodicity of the tangent function and adjusts the weight change of the comprehensive risk assessment value.

[0112] In a specific embodiment, Figure 4 As shown, the working method of the risk relationship analysis unit is:

[0113] Step s21, preprocessing the comprehensive risk data set by using Kalman filtering and data cleaning tool OpenRefine to remove noise and outliers from the comprehensive risk data set;

[0114] Step s22, converting the preprocessed data into a transaction database, mining the frequent itemsets in the comprehensive risk data set through a priori knowledge algorithm, and setting the minimum support threshold to 0.05;

[0115] Step s23, extracting association rules from frequent item sets through a priori knowledge algorithm to obtain support and confidence of the association rules;

[0116] Step s24, set the confidence threshold to 0.6, and select high-confidence association rules according to the support and confidence threshold of the association rules. Lift The indicator obtains the association strength of the association rule;

[0117] Step s25: Fill the association strength of the association rule into the risk association matrix, where the elements of the risk association matrix represent the association strength between different risk dimensions.

[0118] In a specific embodiment, Figure 5 As shown, the working method of the comprehensive risk assessment module is:

[0119] Step s31, obtaining a comprehensive risk data set and a risk association matrix generated by the multi-dimensional risk association module;

[0120] Step s32: Calculate the membership of each risk item by using the fuzzy comprehensive evaluation method, and construct a fuzzy evaluation matrix to map each risk item in the comprehensive risk data set to a fuzzy set. The membership of each risk item is calculated as:

[0121] ,in, It is a risk item i The membership degree of i The degree of belonging to a certain fuzzy set, the value range is [0,1]. The closer the value is to 1, the more the risk item conforms to the fuzzy set. exp It is an exponential function with the natural constant e as the base;

[0122] x It is the risk assessment value, which reflects the size of the current risk;

[0123] c i It is a risk item i The central value of i An ideal or typical value in a fuzzy set;

[0124] It is the width of the fuzzy set, indicating the degree of expansion of the fuzzy set and reflecting the fuzziness of the fuzzy set. The larger the width, the looser the fuzzy set is and the more it can accommodate different risk assessment values.

[0125] is an adjustment parameter used to adjust the calculation of membership and control the shape of the fuzzy set;

[0126] Step s33: According to the risk association matrix and the membership of each risk item, the comprehensive membership of risk items of different dimensions is calculated by Sugeno fuzzy reasoning. The comprehensive membership is calculated as:

[0127] ,in, n represents the number of risk items, is the comprehensive membership, which indicates the final risk assessment result, integrating the impact of each risk item, and its value range is [0,1];

[0128] w i It is a risk item i The larger the weight value, the greater the proportion of the risk item in the comprehensive assessment;

[0129] It is a risk item i The membership degree of i The degree of belonging to a fuzzy set, the value range is [0,1];

[0130] Lift i It is a risk item i The correlation strength of i The correlation strength between the risk items and other risk items. The larger the correlation strength value, the greater the risk item. i The greater the impact on other risk items;

[0131] β i is the correlation strength influence coefficient, which is used to adjust the risk item i The impact of the strength of association on the comprehensive membership;

[0132] Step s34, generating the final risk assessment result through fuzzy decision making according to the comprehensive membership of risk items of different dimensions;

[0133] Step s35: sort the severity of each risk item by fuzzy analytic hierarchy process to identify high-risk areas.

[0134] In a specific embodiment, the accuracy of the multi-dimensional operation risk assessment model of the present invention and the traditional method for evaluating the risk of power field operations is compared as shown in Table 1 below:

[0135] Table 1 Comparison of the accuracy of risk assessment for power field operations

[0136]

[0137] In the above embodiment, in step five, the risk heat map is generated by the data visualization tool Matplotlib, which divides the power field operation area into a number of grid points, and maps the calculated risk item severity to the grid points to generate a risk heat map; the risk heat map uses red to represent high-risk areas, and the color depth of each grid point represents the severity of the risk at the grid point, intuitively displaying the risk distribution of power field operations.

[0138] In a specific embodiment, the risk heat map is applied as follows:

[0139] S1. The data visualization tool Matplotlib divides a power field operation area with an area of ​​100 m × 100 m into 10 × 10 grids, with each grid point covering an area of ​​10 m × 10 m;

[0140] S2. Perform risk assessment on the real-time data and historical data input at each grid point through a multi-dimensional operation risk assessment model to obtain a risk assessment value and risk level;

[0141] S3, mapping the risk assessment values ​​to the corresponding grid points, generating a heat map, and using red to represent high-risk areas, where darker colors indicate higher risks;

[0142] S4. Display the generated risk heat map on the large screen of the central control system to monitor the risk distribution of the power site in real time;

[0143] S5. The risk heat map will be synchronized to the dedicated APP on the terminal devices of relevant personnel to facilitate timely adoption of corresponding risk control measures.

[0144] In a specific embodiment, a power site operation risk assessment method based on a multi-dimensional operation risk assessment model of the present invention is used to monitor the risk of the power site for seven days. The recorded data information is data information output from more than a dozen different substations. The database is large and there are many operation and maintenance points. The monitoring records are as follows:

[0145] Table 2 Power site risk monitoring record

[0146]

[0147] Through the above experiments, the power field operation risk assessment method of the present invention significantly improves the accuracy and real-time performance of risk assessment, and provides an effective risk management tool for power field operations.

Claims

1. A method for risk assessment of power field operations based on a multi-dimensional operation risk assessment model, characterized in that: The following steps are involved: Step 1: Use sensors and IoT devices to collect real-time data at the power site, including equipment status and operating environment, and obtain historical operation data, accident records, maintenance logs, and operator information from the database; Step 2: construct a multidimensional operation risk assessment model through a fuzzy comprehensive evaluation algorithm model based on a support vector machine, wherein the multidimensional operation risk assessment model includes a data receiving module, a data preprocessing module, an equipment status assessment module, an environmental risk assessment module, a personnel risk assessment module, a multidimensional risk association module, a comprehensive risk assessment module and a risk information output module; Step 3: Input the collected real-time data and historical data into a multi-dimensional operation risk assessment model to obtain the risks and risk information of the current power field operation, wherein the risk information includes risk type, risk assessment value and risk level; Step 4: Integrate the risk response rule engine to automatically output the corresponding response strategy based on the risk information output by the model, and notify relevant personnel via SMS, email or APP; Step 5: Based on the risks and risk information of the power site operation, a risk heat map is created through a visualization tool. The risk heat map can distribute the risks on a grid map of the power site and intuitively display high-risk areas. Step 6: Generate risk assessment reports and risk control effectiveness reports regularly through the automatic report generation tool JasperReports, and store them in the MySQL database for easy query and review by relevant personnel; The working method of the comprehensive risk assessment module is: Step s31, obtaining a comprehensive risk data set and a risk association matrix generated by the multi-dimensional risk association module; Step s32: Calculate the membership of each risk item by using the fuzzy comprehensive evaluation method, and construct a fuzzy evaluation matrix to map each risk item in the comprehensive risk data set to a fuzzy set. The membership of each risk item is calculated as: ,in, It is a risk item i The membership degree of x is the risk assessment value, c i It is a risk item i The central value of is the width of the fuzzy set, is the adjustment parameter, exp It is an exponential function with the natural constant e as the base; Step s33: According to the risk association matrix and the membership of each risk item, the comprehensive membership of risk items of different dimensions is calculated by Sugeno fuzzy reasoning. The comprehensive membership is calculated as: ,in, is the comprehensive membership degree, w i It is a risk item i The weight of It is a risk item i The membership degree of Lift i It is a risk item i The strength of association, β i is the correlation strength influence coefficient, n Indicates the number of risk items; Step s34, generating the final risk assessment result through fuzzy decision making according to the comprehensive membership of risk items of different dimensions; Step s35: sort the severity of each risk item by fuzzy analytic hierarchy process to identify high-risk areas.

2. The method for risk assessment of power field operations based on a multi-dimensional operation risk assessment model according to claim 1, characterized in that: In step one, the sensors and IoT devices include surveillance cameras, smoke detectors, temperature sensors, humidity sensors, and vibration sensors. The sensors and IoT devices can monitor the equipment status and changes in the working environment at the power site from multiple dimensions.

3. The method for risk assessment of power field operations based on a multi-dimensional operation risk assessment model according to claim 1, characterized in that: In the step 2, the data receiving module communicates with sensors and IoT devices through the MQTT protocol to receive real-time data and obtain historical operation data, accident records, maintenance logs and operator information of power field operations; The data preprocessing module filters the noise in the received data through Kalman filtering, fills the missing values ​​through KNN, and converts the data to a uniform scale using the Z-Score standardization method; The equipment status assessment module diagnoses the operating status of the equipment based on historical data and real-time data through an equipment status assessment model constructed by a support vector machine; The environmental risk assessment module assesses the operating environment risk based on the environmental sensor data through an environmental risk assessment model constructed by a support vector machine; The personnel risk assessment module identifies potential personnel risks based on the behavior data of the operating personnel by using a personnel risk assessment model constructed by a support vector machine; The multi-dimensional risk association module integrates the risk data of equipment, environment and personnel through DS evidence theory, and generates a risk association matrix by analyzing the association relationship between different risks through a priori knowledge algorithm; The comprehensive risk assessment module comprehensively evaluates the risk information of each dimension based on the fused data and the risk association matrix based on the fuzzy comprehensive evaluation method to generate the final risk assessment result; The risk information output module outputs the final risk assessment results through a compatible RESTful API interface; The output end of the data receiving module is connected to the input end of the data preprocessing module, the output end of the data preprocessing module is connected to the input end of the equipment status assessment module, the environmental risk assessment module and the personnel risk assessment module, the output end of the equipment status assessment module, the environmental risk assessment module and the personnel risk assessment module are connected to the input end of the multidimensional risk association module, the output end of the multidimensional risk association module is connected to the input end of the comprehensive risk assessment module, and the output end of the comprehensive risk assessment module is connected to the input end of the risk information output module.

4. The method for risk assessment of power field operations based on a multi-dimensional operation risk assessment model according to claim 3 is characterized by: The multi-dimensional risk association module includes a risk data fusion unit and a risk relationship analysis unit; The risk data fusion unit fuses the risk data of equipment, environment and personnel through DS evidence theory to generate a comprehensive risk data set; The risk relationship analysis unit mines frequent item sets and association rules from the comprehensive risk data set through an association rule mining method based on a priori knowledge algorithm, analyzes the association relationship between risks of different dimensions and generates a risk association matrix; The output end of the risk data fusion unit is connected to the input end of the risk relationship analysis unit.

5. The method for risk assessment of power field operations based on a multi-dimensional operation risk assessment model according to claim 4 is characterized in that: The working method of the risk data fusion unit is: Step s11, obtaining risk assessment data of the equipment status assessment module, the environmental risk assessment module and the personnel risk assessment module, respectively recorded as E 1. E 2 and E 3; Step s12: Calculate the basic probability distribution of each risk assessment data by using a basic probability distribution function, the formula of the basic probability distribution function is: ,in, m i ( θ ) is the i The basic probability distribution function for each risk dimension is E i It is i Risk assessment data for each risk dimension, θ is the set of risk states, k is the slope adjustment parameter, t is the threshold value, α is the amplitude of the sine function, β is the frequency of the sine function, e is a natural constant in mathematics; Step s13: synthesize the basic probability distribution of each risk assessment data through the synthesis rule of DS evidence theory to obtain a comprehensive basic probability distribution. The synthesis formula is: ,in, m ( θ ) is the comprehensive basic probability distribution function, θ is the set of risk states, K is the standardized coefficient, A , B and C They are subsets corresponding to different risk dimensions. , and are the basic probability distribution of subsets corresponding to different risk dimensions, γ is the amplitude of the cosine function, δ is the frequency of the cosine function; Step s14: normalize the comprehensive basic probability distribution to obtain a comprehensive risk assessment value. The normalization formula is: ,in, R is a comprehensive risk assessment value. is the set of all risk states, m ( θ ) is the comprehensive basic probability distribution function, θ is the set of risk states, λ is the magnitude of the tangent function, μ is the frequency of the tangent function.

6. The method for risk assessment of power field operations based on a multi-dimensional operation risk assessment model according to claim 4 is characterized in that: The working method of the risk relationship analysis unit is: Step s21, preprocessing the comprehensive risk data set by using Kalman filtering and data cleaning tool OpenRefine to remove noise and outliers from the comprehensive risk data set; Step s22, mining frequent item sets in the comprehensive risk data set through a priori knowledge algorithm; Step s23, extracting association rules from frequent item sets through a priori knowledge algorithm to obtain support and confidence of the association rules; Step s24: Filter high-credibility association rules according to support and confidence thresholds, and obtain the association strength of the association rules through the Lift indicator; Step s25: Fill the association strength of the association rule into the risk association matrix, where the elements of the risk association matrix represent the association strength between different risk dimensions.

7. The method for risk assessment of power field operations based on a multi-dimensional operation risk assessment model according to claim 1, characterized in that: In the step 5, the risk heat map is generated by using a data visualization tool Matplotlib, which divides the power field operation area into a number of grid points, and maps the calculated severity of the risk items to the grid points to generate a risk heat map; The risk heat map uses red to represent high-risk areas, and the color depth of each grid point represents the severity of the risk at the grid point, intuitively displaying the risk distribution of power field operations.

Images