Urban safety risk assessment method and system based on digital standard

Abstract

The invention discloses an urban safety risk assessment method and system based on a digital standard, and the method comprises the steps: obtaining urban safety risk monitoring data and historical data, constructing a risk index event library according to the historical data, carrying out the abnormal screening and classification of the monitoring data, and obtaining risk indexes, matching the risk indicator with the risk indicator event library to determine a first risk event, predicting an uncertain risk event according to the first risk event, and constructing a city safety risk assessment model according to the first risk event, the uncertain risk event and the risk event weight, and inputting to-be-assessed city safety risk monitoring data and historical data into the city safety risk assessment model to obtain a city safety risk assessment result. The method not only can improve the efficiency and accuracy of urban safety risk assessment, but also has good interpretability, and can be directly applied to an urban safety risk assessment system.

Description

Technical Field

[0001] The present invention relates to the field of quality evaluation, and in particular to a method and system for evaluating urban safety risks based on digital standards. Background Art

[0002] The global urbanization process is accelerating, and its scale and complexity are also accompanied by a series of security risk challenges. Therefore, the management and assessment of urban security risks are of great significance to protecting the lives and property of urban residents, maintaining social stability and promoting sustainable development. The rapid development of information technology, especially the application of big data, cloud computing and artificial intelligence, has provided new opportunities for urban security risk assessment.

[0003] Traditional assessment methods often rely on manual experience and judgment, lack of data support and scientific analysis, resulting in inaccurate assessment results and difficulty in real-time updating. At the same time, urban safety risks involve multiple fields and levels. These factors are intertwined and influence each other, making risk assessment more complicated. Although there are some urban safety risk assessment methods based on digital technology, most of these methods remain in the primary stage of data collection and analysis, lacking a systematic and comprehensive assessment model. Therefore, it is important to build a systematic standard database through historical risk data, effectively pre-process monitoring data to identify risk events, predict uncertain events caused by risk events, and build an urban safety risk assessment model based on machine learning technology to score urban safety risks, predict risk events and the probability of occurrence. Design an urban safety risk assessment method and system based on digital standards to overcome the shortcomings of existing evaluation models and promote the development of urban safety risk management in the direction of digitalization and intelligence, which is of great significance to building a harmonious and safe urban environment. Summary of the invention

[0004] The purpose of the present invention is to provide a city safety risk assessment method and system based on digital standards.

[0005] To achieve the above object, the present invention is implemented according to the following technical solutions: The present invention comprises the following steps: Acquire urban safety risk monitoring data and historical data, and pre-process the monitoring data and the historical data; Building a risk indicator event library based on the historical data and determining risk event weights; Performing abnormal screening and classification on the monitoring data to obtain risk indicators, matching the risk indicators with the risk indicator event library to determine a first risk event, and predicting an uncertain risk event based on the first risk event; the uncertain risk event includes a concurrent risk event and a probabilistic risk event; A city safety risk assessment model is constructed according to the first risk event, the uncertain risk event and the risk event weight, and the city safety risk monitoring data and historical data to be assessed are input into the city safety risk assessment model to obtain a city safety risk assessment result.

[0006] Furthermore, the method of constructing a risk indicator event library based on the historical data and determining the risk event weights includes: Use clustering method to identify risk events of different categories in historical data, extract features of risk events to obtain corresponding risk indicators, and build a risk indicator library based on risk events and corresponding risk indicators; Divide the safety risk levels according to risk categories, risk events and risk indicators, and determine the weights of the assessment units at each level by combining the analytic hierarchy process and expert consultation method.

[0007] Furthermore, the method of performing abnormal screening and classification on the monitoring data to obtain risk indicators includes: Determine the safety risk projects according to the regional risk assessment standards, obtain the monitoring data of the safety risk projects, standardize the monitoring data and substitute it into the fuzzy clustering algorithm, divide the n sample space into clusters according to the similarity between the monitoring data, and determine the clustering objective function, which is expressed as: , in is the clustering objective function, m is the category dimension of samples within the cluster, for j Cluster monitoring data samples, for j Cluster monitoring data mean, For the sample i The cluster center of the dimension, For monitoring data samples belong i The degree of membership of the dimension; Calculate the density value of the data object in the monitoring data sample, select the data point with the largest density value as the first cluster center point, and calculate the density value of the remaining data points. The expression is: , , in is the density value of the cluster center point, is the density value of data points outside the cluster center point, is the area radius, Update the value for the area radius, is the density value of the initial maximum cluster center; Update the cluster center and the membership of the monitoring data sample, and iterate until the maximum number of iterations or the objective function is met, and then stop iterating to obtain the data divided into different categories. K clusters; Perform multi-hypersphere learning, perform macroscopic hypersphere learning based on all monitoring data, and perform mesoscopic hypersphere learning on each cluster monitoring data based on the monitoring data clustering results. The expression is: , , , in is the loss function for learning the macroscopic hypersphere, is the mesoscopic hypersphere learning loss function, is the contrastive learning loss function, is the node of the training set, is the set of training nodes, is the number of training nodes, For Node Representation in space, is the center of the macroscopic hypersphere, for norm, is the regularization parameter, For Node The center of the hypersphere, For Node The cluster index, is the cosine similarity function, For the k The query vector of samples, For the k positive sample enhancement vector, Enhance the vector set for all positive samples; The multi-sphere learning target is determined according to the macroscopic hypersphere learning loss function, the mesoscopic hypersphere learning loss function and the contrastive learning loss function, and the risk index is obtained by abnormal judgment. The expression is: , , in L For multi-sphere learning objectives, Learn the parameters for the mesoscopic hypersphere, Compare the learning parameters, For Node Abnormal score of Label risk indicators by data category and data collection location.

[0008] Furthermore, the method for determining the first risk event includes: It is defined that multiple risk indicators have the probability of causing one or more risk events. The principal component analysis is performed on the risk indicators to obtain the principal component characteristics. One principal component characteristic corresponds to a group of risk indicators. The risk indicator group corresponding to the principal component characteristic is determined by calculating the correlation coefficient between the principal component characteristic and the risk indicator. The expression is: , , in For the i Principal component eigenvalues With j Risk indicator value The correlation coefficient of For the i The absolute value mean of the correlation coefficient between the principal component eigenvalues ​​and all risk index values, is the mean of all principal component eigenvalues, is the mean value of all risk indicators, is the number of principal component features, is the number of risk indicators, is the nonlinear term weight, is a nonlinear function used to capture and Non-linear relationships, is the regularization coefficient, is the regularization strength; choose The risk indicators consist of i The risk indicator group of principal component characteristics is defined as , is the feature vector element, the risk event vector set , is the risk event vector element, h The risk indicator threshold corresponding to each risk event is composed of s is the number of risk events in the risk indicator event library, and the correlation between multiple risk indicator combinations and risk events in the risk indicator event library is calculated to determine the first risk event. The expression is: , , , in Principal component features i and risk events h The distance For the iThe mean vector of the principal component feature samples, For the h The mean vector of risk event samples, For the i The covariance matrix of the principal component feature samples, For the h The covariance matrix of the risk event sample threshold, is the matrix determinant; Setting distance threshold ,when Determine the principal component characteristics i The corresponding risk indicator group may trigger a risk event h , otherwise the principal component characteristics i The corresponding risk indicator group does not trigger risk events; Label the first risk event by data category and data collection location.

[0009] Furthermore, the method for predicting a concurrent risk event based on the first risk event includes: Construct the FP-growth algorithm module, explore the correlation between risk events, and determine concurrent risk events through known risk events. The construction steps of the FP-growth algorithm module are as follows: Scan the frequency of statistical items in historical data, set the minimum support to select frequent items, and construct the item header table based on frequent items and frequencies; Filter and sort the items in the transaction according to the item header table, and build the FP-tree based on the filtering results; Determine the conditional pattern base and transaction item set of frequent items according to the item header table, establish a conditional FP-tree according to the conditional pattern base, use historical data to train FP-tree conditional instances, use deep mining and recursive methods to explore common item sets, and output FP-tree; The first risk event is input into the FP-growth algorithm module to obtain its concurrent risk events, and the concurrent risk events are labeled according to the data category and data collection location.

[0010] Furthermore, the method for predicting a probabilistic risk event according to the first risk event includes: Define the characteristic vector of the risk indicator corresponding to the first risk event as , define the characteristic vector of the risk indicator threshold corresponding to the target risk event in the risk indicator event library as , calculate the comprehensive similarity between the first risk event feature vector and the target risk event feature vector, the expression is: , in For the f The first risk event feature vector With g target risk event feature vector The comprehensive similarity of is the weight coefficient, is the feature vector A set of risk indicators, Eigenvector The risk indicator threshold set, t is the number of risk indicators for all projects, is the maximum number of project risk indicators, is the feature vector The number of risk indicators, is the feature vector The number of risk indicator thresholds, , is the feature vector Risk indicator mean and variance, , is the feature vector The mean and variance of the risk indicator thresholds, K is the vector dimension, is the feature vector The element value of is the feature vector The element value of is the resolution coefficient; A comprehensive similarity threshold is set, and the probabilistic risk event corresponding to the first risk event is determined based on the comprehensive similarity, and the probabilistic risk event is labeled according to the data category and data collection location.

[0011] Furthermore, the method for obtaining the urban safety risk assessment result includes: The first risk event, concurrent risk event, probability risk event, risk index and risk event weight are combined into a safety risk data set, and the random forest algorithm is used to divide the safety risk set into a training set and a test set in a ratio of 7:3; Constructing a city safety risk assessment model, wherein the city safety risk assessment model includes a Bayesian network, a multimodal prediction model, and a strategy layer; The Bayesian network estimates the probability relationship of risk events based on the historical data of the known first risk event, concurrent risk events, and probabilistic risk events, and calculates the probability of each risk event given other risk events; The multimodal prediction model is used to learn the training set data and the probability of risk events for safety risk assessment, including the genetic-BP neural network base model, the support vector machine regression base model and the random forest base model; The genetic-BP neural network is used to predict the risk score based on the input data, the mean square error loss function is used to evaluate the difference between the predicted value and the actual value, the genetic algorithm is used to optimize the network structure and parameters, and the Adam optimizer is used to adjust the learning rate; Support vector machine regression finds a hyperplane by inputting data to distinguish events of different risk levels and make risk score predictions. Insensitive loss function The score error is predicted within the range, and the SMO algorithm is used to solve the optimal hyperplane; Random forest improves the accuracy and stability of risk scoring by constructing decision trees for voting, using the mean square error loss function to evaluate the difference between the predicted value and the actual value, and selecting the best split point based on information gain; The strategy layer uses a weighted voting method to integrate the base model prediction risk score, and the output layer extracts the category and location of risk events, and finally outputs the security risk assessment results. The test set is used to evaluate the urban security risk assessment model; The safety risk monitoring data and historical data of the city to be assessed are input into the urban safety risk assessment model to obtain the safety risk assessment results.

[0012] Second, the urban safety risk assessment system based on digital standards includes: Data acquisition module: used to obtain urban safety risk monitoring data and historical data, and pre-process the monitoring data and historical data; Data processing module: used to determine the risk event weight, to perform abnormal screening and classification on the monitoring data to obtain risk indicators, to match the risk indicators with the risk indicator event library, and to predict uncertain risk events; Assessment model module: constructing a city safety risk assessment model according to the first risk event, the uncertain risk event and the risk event weight, and inputting the city safety risk monitoring data and historical data to be assessed into the city safety risk assessment model to obtain the city safety risk assessment result; Intelligent supervision module: used to store, view and manage the monitoring data, the historical data and the city safety risk assessment results, used to determine the safety risk level and provide risk warnings and suggestions to users based on the city safety risk assessment results, and used to mark risk events and their categories, safety risk levels and probability of occurrence on the safety risk map based on location information.

[0013] The beneficial effects of the present invention are: The present invention is a method and system for evaluating urban safety risks based on digital standards. Compared with the prior art, the present invention has the following technical effects: The present invention can improve the data preprocessing capabilities and enhance the model adaptability in urban safety risk assessment by determining risk event weights, anomaly screening and classification of monitoring data, determining the first risk event through data matching, predicting uncertain risk events, and constructing model steps, thereby improving the efficiency and accuracy of urban safety risk assessment, optimizing urban safety risk assessment technology, greatly saving resources, and improving work efficiency, and can realize the assessment of urban safety risks. It can comprehensively, objectively, and accurately obtain the safety risk status of urban areas, provide precise scientific guidance for urban safety risk management, and can adapt to different urban safety risk assessment systems based on digital standards and the terminal assessment needs of urban safety risk assessment systems based on digital standards of different users, and has a certain degree of universality. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 The present invention is a flowchart of the steps of the urban safety risk assessment method based on digital standards. DETAILED DESCRIPTION

[0015] The present invention is further described below by means of specific embodiments. The illustrative embodiments and descriptions of the present invention are used to explain the present invention but are not intended to limit the present invention.

[0016] The urban safety risk assessment method and system based on digital standards of the present invention include the following steps: like Figure 1 As shown, in this embodiment, the following steps are included: Acquire urban safety risk monitoring data and historical data, and pre-process the monitoring data and the historical data; Building a risk indicator event library based on the historical data and determining risk event weights; Performing abnormal screening and classification on the monitoring data to obtain risk indicators, matching the risk indicators with the risk indicator event library to determine a first risk event, and predicting an uncertain risk event based on the first risk event; the uncertain risk event includes a concurrent risk event and a probabilistic risk event; A city safety risk assessment model is constructed according to the first risk event, the uncertain risk event and the risk event weight, and the city safety risk monitoring data and historical data to be assessed are input into the city safety risk assessment model to obtain a city safety risk assessment result.

[0017] In this embodiment, the method of constructing a risk indicator event library based on the historical data and determining the risk event weights includes: Use clustering method to identify risk events of different categories in historical data, extract features of risk events to obtain corresponding risk indicators, and build a risk indicator library based on risk events and corresponding risk indicators; Divide the safety risk levels according to risk categories, risk events and risk indicators, and determine the weights of the assessment units at each level by combining the analytic hierarchy process and expert consultation method; In the actual assessment, the safety risk levels are divided according to risk categories, risk events and risk indicators, and the first-level indicator risk categories and corresponding weights are obtained: natural disasters - 0.3 (earthquakes, heavy rains, typhoons, high temperatures); accidents and disasters - 0.25 (chemical accidents, fireworks accidents, civilian explosive accidents, industrial and trade accidents, transportation accidents, construction accidents); public health - 0.2 (infectious disease outbreaks, food safety incidents, occupational hazards); public safety - 0.15 (criminal activities, other safety incidents); infrastructure - 0.1 (collapse of old buildings, large passenger flows, building fires, urban elevator operations, transportation, power operations, gas leaks and explosions, bridge operations, and water supply and drainage network failures).

[0018] In this embodiment, the method for performing abnormal screening and classification on the monitoring data to obtain risk indicators includes: Determine the safety risk projects according to the regional risk assessment standards, obtain the monitoring data of the safety risk projects, standardize the monitoring data and substitute it into the fuzzy clustering algorithm, divide the n sample space into clusters according to the similarity between the monitoring data, and determine the clustering objective function, which is expressed as: , in is the clustering objective function, m is the category dimension of samples within the cluster, for j Cluster monitoring data samples, for j Cluster monitoring data mean, For the sample i The cluster center of the dimension, For monitoring data samples belong i The degree of membership of the dimension; Calculate the density value of the data object in the monitoring data sample, select the data point with the largest density value as the first cluster center point, and calculate the density value of the remaining data points. The expression is: , , in is the density value of the cluster center point, is the density value of data points outside the cluster center point, is the area radius, Update the value for the area radius, is the density value of the initial maximum cluster center; Update the cluster center and the membership of the monitoring data sample, and iterate until the maximum number of iterations or the objective function is met, and then stop iterating to obtain the data divided into different categories. K clusters; Perform multi-hypersphere learning, perform macroscopic hypersphere learning based on all monitoring data, and perform mesoscopic hypersphere learning on each cluster monitoring data based on the monitoring data clustering results. The expression is: , , , in is the loss function for learning the macroscopic hypersphere, is the mesoscopic hypersphere learning loss function, is the contrastive learning loss function, is the node of the training set, is the set of training nodes, is the number of training nodes, For Node Representation in space, is the center of the macroscopic hypersphere, for norm, is the regularization parameter, For Node The center of the hypersphere, For Node The cluster index, is the cosine similarity function, For the k The query vector of samples, For the k positive sample enhancement vector, Enhance the vector set for all positive samples; The multi-sphere learning target is determined according to the macroscopic hypersphere learning loss function, the mesoscopic hypersphere learning loss function and the contrastive learning loss function, and the risk index is obtained by abnormal judgment. The expression is: , , in L For multi-sphere learning objectives, Learn the parameters for the mesoscopic hypersphere, Compare the learning parameters, For Node Abnormal score of Label risk indicators by data category and data collection location; In the actual assessment, the safety risk monitoring data of a certain city was screened and classified for anomalies to obtain risk indicators: natural disasters (daily rainfall of 250 mm, riverbed height of +6 meters, number of waterlogged areas of 15, waterlogged depth of 0.3 m, number of discussions about rainstorms on social media of 10,000 per hour); accidents and disasters (1 chemical leak in an industrial area); public safety (2 public transportation failures, 5 thefts, 1 fight); infrastructure (traffic congestion index of 8.5, number of emergency service calls of 1,500 times, 40 reports of power failures, mobile network traffic increased by 50% compared to usual, network attacks increased by 30% compared to usual, and 1 public facility was damaged).

[0019] In this embodiment, the method for determining the first risk event includes: It is defined that multiple risk indicators have the probability of causing one or more risk events. The principal component analysis is performed on the risk indicators to obtain the principal component characteristics. One principal component characteristic corresponds to a group of risk indicators. The risk indicator group corresponding to the principal component characteristic is determined by calculating the correlation coefficient between the principal component characteristic and the risk indicator. The expression is: , , in For the i Principal component eigenvalues With j Risk indicator value The correlation coefficient of For the i The absolute value mean of the correlation coefficient between the principal component eigenvalues ​​and all risk index values, is the mean of all principal component eigenvalues, is the mean value of all risk indicators, is the number of principal component features, is the number of risk indicators, is the nonlinear term weight, is a nonlinear function used to capture and Non-linear relationships, is the regularization coefficient, is the regularization strength; choose The risk indicators consist of i The risk indicator group of principal component characteristics is defined as , is the feature vector element, the risk event vector set , is the risk event vector element, h The risk indicator threshold corresponding to each risk event is composed ofs is the number of risk events in the risk indicator event library, and the correlation between multiple risk indicator combinations and risk events in the risk indicator event library is calculated to determine the first risk event. The expression is: , , , in Principal component features i and risk events h The distance For the i The mean vector of the principal component feature samples, For the h The mean vector of risk event samples, For the i The covariance matrix of the principal component feature samples, For the h The covariance matrix of the risk event sample threshold, is the matrix determinant; Setting distance threshold ,when Determine the principal component characteristics i The corresponding risk indicator group may trigger a risk event h , otherwise the principal component characteristics i The corresponding risk indicator group does not trigger risk events; Label the first risk event by data category and data collection location; In the actual assessment, two risk indicator groups were determined by calculating the correlation of risk indicators (daily rainfall, riverbed height, number of waterlogged areas, water depth of waterlogging, number of social media discussions about heavy rain, correlation coefficient 0.88> correlation coefficient absolute value mean 0.62), (number of waterlogged areas, public transportation failures, traffic congestion index, power failure reports, mobile network traffic increased compared to usual, correlation coefficient 0.75> correlation coefficient absolute value mean 0.51); According to the distance between the correlation principal component characteristics and the risk event (0.08, 0.15, distance threshold 0.2), the first risk event is determined to be heavy rain, extreme weather and traffic congestion.

[0020] In this embodiment, the method for predicting a concurrent risk event according to the first risk event includes: Construct the FP-growth algorithm module, explore the correlation between risk events, and determine concurrent risk events through known risk events. The construction steps of the FP-growth algorithm module are as follows: Scan the frequency of statistical items in historical data, set the minimum support to select frequent items, and construct the item header table based on frequent items and frequencies; Filter and sort the items in the transaction according to the item header table, and build the FP-tree based on the filtering results; Determine the conditional pattern base and transaction item set of frequent items according to the item header table, establish a conditional FP-tree according to the conditional pattern base, use historical data to train FP-tree conditional instances, use deep mining and recursive methods to explore common item sets, and output FP-tree; Input the first risk event into the FP-growth algorithm module to obtain its concurrent risk events, and label them as concurrent risk events according to data category and data collection location; In the actual assessment, the risk indicators corresponding to extreme weather such as rainstorms and traffic congestion are input into the FP-growth algorithm module, and the concurrent risk events are determined to be public transportation delays and drainage system obstructions.

[0021] In this embodiment, the method for predicting a probabilistic risk event according to the first risk event includes: Define the characteristic vector of the risk indicator corresponding to the first risk event as , define the characteristic vector of the risk indicator threshold corresponding to the target risk event in the risk indicator event library as , calculate the comprehensive similarity between the first risk event feature vector and the target risk event feature vector, the expression is: , in For the f The first risk event feature vector With g target risk event feature vector The comprehensive similarity of is the weight coefficient, is the feature vector A set of risk indicators, Eigenvector The risk indicator threshold set, t is the number of risk indicators for all projects, is the maximum number of project risk indicators, is the feature vector The number of risk indicators, is the feature vector The number of risk indicator thresholds, , is the feature vector Risk indicator mean and variance, , is the feature vector The mean and variance of the risk indicator thresholds, Kis the vector dimension, is the feature vector The element value of is the feature vector The element value of is the resolution coefficient; Set a comprehensive similarity threshold, determine the probabilistic risk event corresponding to the first risk event based on the comprehensive similarity, and label the probabilistic risk event according to the data category and data collection location; In the actual assessment, the comprehensive similarity threshold is set to 0.6, and the comprehensive similarity between the characteristic vector of the risk indicator corresponding to the first risk event and the characteristic vector of the risk indicator threshold corresponding to the target risk event in the risk indicator event library is calculated. The probabilistic risk events are determined according to the comprehensive similarity threshold: damage to public equipment, damage to houses, water accumulation in underground facilities, public health issues, cancellation of public activities, interruption of public services, damage to network communications, agricultural economic losses, and supply chain interruptions.

[0022] In this embodiment, the method for obtaining the city safety risk assessment result includes: The first risk event, concurrent risk event, probability risk event, risk index and risk event weight are combined into a safety risk data set, and the random forest algorithm is used to divide the safety risk set into a training set and a test set in a ratio of 7:3; Constructing a city safety risk assessment model, wherein the city safety risk assessment model includes a Bayesian network, a multimodal prediction model, and a strategy layer; The Bayesian network estimates the probability relationship of risk events based on the historical data of the known first risk event, concurrent risk events, and probabilistic risk events, and calculates the probability of each risk event given other risk events; The multimodal prediction model is used to learn the training set data and the probability of risk events for safety risk assessment, including the genetic-BP neural network base model, the support vector machine regression base model and the random forest base model; The genetic-BP neural network is used to predict the risk score based on the input data, the mean square error loss function is used to evaluate the difference between the predicted value and the actual value, the genetic algorithm is used to optimize the network structure and parameters, and the Adam optimizer is used to adjust the learning rate; Support vector machine regression finds a hyperplane by inputting data to distinguish events of different risk levels and make risk score predictions. Insensitive loss function The score error is predicted within the range, and the SMO algorithm is used to solve the optimal hyperplane; Random forest improves the accuracy and stability of risk scoring by constructing decision trees for voting, using the mean square error loss function to evaluate the difference between the predicted value and the actual value, and selecting the best split point based on information gain; The strategy layer uses a weighted voting method to integrate the base model to predict the risk score, and the output layer extracts the category and location of risk events, and finally outputs the safety risk assessment results. The test set is used to evaluate the urban safety risk assessment model; the safety risk assessment results include the overall risk score, risk events and their categories, probability of occurrence and location of occurrence; the risk events include major risk events and uncertain risk events; Input the city safety risk monitoring data and historical data to be assessed into the city safety risk assessment model to obtain the safety risk assessment results; In the actual assessment, the probability of risk events is calculated using the Bayesian network according to risk events, risk indicators and corresponding category weights: (risk events: rainstorm extreme weather 90%, traffic congestion 80%), (concurrent risk events: public transportation delays 70%, drainage system obstacles 60%), (uncertain risk events: public equipment damage 30%, housing damage 20%, underground facility water accumulation 25%, public health issues 15%, public event cancellation 10%, public service interruption 5%, network communication damage 10%, agricultural economic losses 20%, supply chain interruption 15%). The city safety risk monitoring data and historical data to be assessed are input into the city safety risk assessment model to obtain the city safety risk assessment results: the city safety risk score is 47 points, which is a public transportation congestion / delay event caused by extreme rainstorm weather events, which belongs to the natural disaster and accident disaster risk events, and is accompanied by the occurrence of accident disaster, public health and infrastructure risk events.

[0023] Second, the urban safety risk assessment system based on digital standards includes: Data acquisition module: used to obtain urban safety risk monitoring data and historical data, and pre-process the monitoring data and historical data; Data processing module: used to determine the risk event weight, to perform abnormal screening and classification on the monitoring data to obtain risk indicators, to match the risk indicators with the risk indicator event library, and to predict uncertain risk events; Assessment model module: constructing a city safety risk assessment model according to the first risk event, the uncertain risk event and the risk event weight, and inputting the city safety risk monitoring data and historical data to be assessed into the city safety risk assessment model to obtain the city safety risk assessment result; Intelligent supervision module: used for storing, viewing and managing the monitoring data, the historical data and the city safety risk assessment results, for determining the safety risk level according to the city safety risk assessment results and providing risk warnings and suggestions to users, and for marking risk events and their categories, safety risk levels and occurrence probabilities on the safety risk map according to location information; In the actual assessment, the city safety risk assessment results are imported through the supervision module, the location of the risk indicator is marked in the safety risk map, and the corresponding risk events, risk event categories, risk event probabilities and city safety risk scores are annotated. The city safety risk score of 47 points is determined to be a medium risk through the risk threshold, and early warning information and response suggestions are sent to the user end.

[0024] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. Urban safety risk assessment method based on digital standards, characterized by: The following steps are involved: S1. Acquire urban safety risk monitoring data and historical data, and pre-process the monitoring data and historical data; S2. Build a risk indicator event library based on the historical data and determine the risk event weights; S3, performing abnormal screening and classification on the monitoring data to obtain risk indicators, matching the risk indicators with the risk indicator event library to determine a first risk event, and predicting an uncertain risk event based on the first risk event; The uncertain risk events include concurrent risk events and probabilistic risk events; S4. Construct a city safety risk assessment model based on the first risk event, the uncertain risk event and the risk event weight, and input the city safety risk monitoring data and historical data to be assessed into the city safety risk assessment model to obtain a city safety risk assessment result.

2. The urban safety risk assessment method based on digital standards according to claim 1 is characterized in that: The method of constructing a risk indicator event library based on the historical data and determining the risk event weights includes: Use clustering method to identify risk events of different categories in historical data, extract features of risk events to obtain corresponding risk indicators, and build a risk indicator library based on risk events and corresponding risk indicators; Divide the safety risk levels according to risk categories, risk events and risk indicators, and determine the weights of the assessment units at each level by combining the analytic hierarchy process and expert consultation method.

3. The urban safety risk assessment method based on digital standards according to claim 1 is characterized in that: The method of performing abnormal screening and classification on the monitoring data to obtain risk indicators includes: Determine the safety risk projects according to the regional risk assessment standards, obtain the monitoring data of the safety risk projects, standardize the monitoring data and substitute it into the fuzzy clustering algorithm, divide the n sample space into clusters according to the similarity between the monitoring data, and determine the clustering objective function, which is expressed as: ， in is the clustering objective function, m is the category dimension of the samples within the cluster, for j Cluster monitoring data samples, for j Cluster monitoring data mean, For the sample i The cluster center of the dimension, For monitoring data samples belong i The degree of membership of the dimension; Calculate the density value of the data object in the monitoring data sample, select the data point with the largest density value as the first cluster center point, and calculate the density value of the remaining data points. The expression is: ， ， in is the density value of the cluster center point, is the density value of data points outside the cluster center point, is the area radius, Update the value for the area radius, is the density value of the initial maximum cluster center; Update the cluster center and the membership of the monitoring data sample, and iterate until the maximum number of iterations or the objective function is met, and then stop iterating to obtain the data divided into different categories. K clusters; Perform multi-hypersphere learning, perform macroscopic hypersphere learning based on all monitoring data, and perform mesoscopic hypersphere learning on each cluster monitoring data based on the monitoring data clustering results. The expression is: ， ， ， in is the loss function for learning the macroscopic hypersphere, is the mesoscopic hypersphere learning loss function, is the contrastive learning loss function, is the node of the training set, is the set of training nodes, is the number of training nodes, For Node Representation in space, is the center of the macroscopic hypersphere, for norm, is the regularization parameter, For Node The center of the hypersphere, For Node The cluster index, is the cosine similarity function, For the k The query vector of samples, For the k positive sample enhancement vector, Enhance the vector set for all positive samples; The multi-sphere learning target is determined according to the macroscopic hypersphere learning loss function, the mesoscopic hypersphere learning loss function and the contrastive learning loss function, and the risk index is obtained by abnormal judgment. The expression is: ， ， in L For multi-sphere learning objectives, Learn the parameters for the mesoscopic hypersphere, Compare the learning parameters, For Node Abnormal score of Label risk indicators by data category and data collection location.

4. The urban safety risk assessment method based on digital standards according to claim 1 is characterized in that: The method for determining the first risk event includes: It is defined that multiple risk indicators have the probability of causing one or more risk events. The principal component analysis is performed on the risk indicators to obtain the principal component characteristics. One principal component characteristic corresponds to a group of risk indicators. The risk indicator group corresponding to the principal component characteristic is determined by calculating the correlation coefficient between the principal component characteristic and the risk indicator. The expression is: ， ， in For the i Principal component eigenvalues With j Risk indicator value The correlation coefficient of For the i The absolute value mean of the correlation coefficient between the principal component eigenvalues ​​and all risk index values, is the mean of all principal component eigenvalues, is the mean value of all risk indicators, is the number of principal component features, is the number of risk indicators, is the nonlinear term weight, is a nonlinear function used to capture and Non-linear relationships, is the regularization coefficient, is the regularization strength; choose The risk indicators consist of i The risk indicator group of principal component characteristics is defined as , is the feature vector element, the risk event vector set , is the risk event vector element, h The risk indicator threshold corresponding to each risk event is composed of s is the number of risk events in the risk indicator event library, and the correlation between multiple risk indicator combinations and risk events in the risk indicator event library is calculated to determine the first risk event. The expression is: ， ， ， in Principal component features i and risk events h The distance For the i The mean vector of the principal component feature samples, For the h The mean vector of risk event samples, For the i The covariance matrix of the principal component feature samples, For the h The covariance matrix of the risk event sample threshold, is the matrix determinant; Setting distance threshold ,when Determine the principal component characteristics i The corresponding risk indicator group may trigger a risk event h , otherwise the principal component characteristics i The corresponding risk indicator group does not trigger risk events; Label the first risk event by data category and data collection location.

5. The urban safety risk assessment method based on digital standards according to claim 1 is characterized in that: The method for predicting a concurrent risk event according to the first risk event includes: Construct the FP-growth algorithm module, explore the correlation between risk events, and determine concurrent risk events through known risk events. The construction steps of the FP-growth algorithm module are as follows: Scan the frequency of statistical items in historical data, set the minimum support to select frequent items, and construct the item header table based on frequent items and frequencies; Filter and sort the items in the transaction according to the item header table, and build the FP-tree based on the filtering results; Determine the conditional pattern base and transaction item set of frequent items according to the item header table, establish a conditional FP-tree according to the conditional pattern base, use historical data to train FP-tree conditional instances, use deep mining and recursive methods to explore common item sets, and output FP-tree; The first risk event is input into the FP-growth algorithm module to obtain its concurrent risk events, and the concurrent risk events are labeled according to the data category and data collection location.

6. The urban safety risk assessment method based on digital standards according to claim 1 is characterized in that: The method for predicting a probabilistic risk event according to the first risk event comprises: Define the characteristic vector of the risk indicator corresponding to the first risk event as , define the characteristic vector of the risk indicator threshold corresponding to the target risk event in the risk indicator event library as , calculate the comprehensive similarity between the first risk event feature vector and the target risk event feature vector, the expression is: ， in For the f The first risk event feature vector With g target risk event feature vector The comprehensive similarity of is the weight coefficient, is the feature vector A set of risk indicators, Eigenvector The risk indicator threshold set, t is the number of risk indicators for all projects, is the maximum number of project risk indicators, is the feature vector The number of risk indicators, is the feature vector The number of risk indicator thresholds, , is the feature vector Risk indicator mean and variance, , is the feature vector The mean and variance of the risk indicator thresholds, K is the vector dimension, is the feature vector The element value of is the feature vector The element value of is the resolution coefficient; A comprehensive similarity threshold is set, and the probabilistic risk event corresponding to the first risk event is determined based on the comprehensive similarity, and the probabilistic risk event is labeled according to the data category and data collection location.

7. The urban safety risk assessment method based on digital standards according to claim 1 is characterized in that: The method for obtaining the urban safety risk assessment result includes: The first risk event, concurrent risk event, probability risk event, risk index and risk event weight are combined into a safety risk data set, and the random forest algorithm is used to divide the safety risk set into a training set and a test set in a ratio of 7:3; Constructing a city safety risk assessment model, wherein the city safety risk assessment model includes a Bayesian network, a multimodal prediction model, and a strategy layer; The Bayesian network estimates the probability relationship of risk events based on the historical data of the known first risk event, concurrent risk events, and probabilistic risk events, and calculates the probability of each risk event given other risk events; The multimodal prediction model is used to learn the training set data and the probability of risk events for safety risk assessment, including the genetic-BP neural network base model, the support vector machine regression base model and the random forest base model; The genetic-BP neural network is used to predict the risk score based on the input data, the mean square error loss function is used to evaluate the difference between the predicted value and the actual value, the genetic algorithm is used to optimize the network structure and parameters, and the Adam optimizer is used to adjust the learning rate; Support vector machine regression finds a hyperplane by inputting data to distinguish events of different risk levels and make risk score predictions. Insensitive loss function The score error is predicted within the range, and the SMO algorithm is used to solve the optimal hyperplane; Random forest improves the accuracy and stability of risk scoring by constructing decision trees for voting, using the mean square error loss function to evaluate the difference between the predicted value and the actual value, and selecting the best split point based on information gain; The strategy layer uses a weighted voting method to integrate the base model prediction risk score, and the output layer extracts the category and location of risk events, and finally outputs the security risk assessment results. The test set is used to evaluate the urban security risk assessment model; The safety risk monitoring data and historical data of the city to be assessed are input into the urban safety risk assessment model to obtain the safety risk assessment results.

8. A city safety risk assessment system based on digital standards, used to implement the method according to any one of claims 1 to 7, characterized in that: include: Data acquisition module: used to obtain urban safety risk monitoring data and historical data, and pre-process the monitoring data and historical data; Data processing module: used to determine the risk event weight, to perform abnormal screening and classification on the monitoring data to obtain risk indicators, to match the risk indicators with the risk indicator event library, and to predict uncertain risk events; Assessment model module: constructing a city safety risk assessment model according to the first risk event, the uncertain risk event and the risk event weight, and inputting the city safety risk monitoring data and historical data to be assessed into the city safety risk assessment model to obtain the city safety risk assessment result; Intelligent supervision module: used to store, view and manage the monitoring data, the historical data and the city safety risk assessment results, used to determine the safety risk level and provide risk warnings and suggestions to users based on the city safety risk assessment results, and used to mark risk events and their categories, safety risk levels and probability of occurrence on the safety risk map based on location information.

Images