Search

CN-121981552-A - Dynamic simulation and agile response method for major risk of oversized city base layer

CN121981552ACN 121981552 ACN121981552 ACN 121981552ACN-121981552-A

Abstract

The invention discloses a dynamic simulation and agile response method for major risks of an oversized city base layer, which relates to the field of data analysis and comprises the steps of taking city base layer treatment units (including streets, villages and towns and communities and villages to which the streets and the villages belong) as division basis, deploying a distributed network comprising self-adaptive mobile sensing nodes and static monitoring terminals, collecting multisource heterogeneous risk data related to population flow, infrastructure operation, environmental parameters and social activity characteristics in real time, taking the city base layer treatment units as objects, collecting multisource heterogeneous risk data in real time through deploying the distributed network, generating a standardized data set through space-time alignment and characteristic extraction, accurately capturing base layer risk key information, and reducing data redundancy and noise interference.

Inventors

  • Shang Qiujin
  • YANG YANYING
  • ZHENG JIANCHUN
  • LIU MENGTING

Assignees

  • 北京市科学技术研究院

Dates

Publication Date
20260505
Application Date
20260203

Claims (9)

  1. 1. The dynamic simulation and agile response method for the major risk of the oversized city base layer is characterized by comprising the following steps of: Taking an urban basic level treatment unit as a division basis, deploying a distributed network comprising self-adaptive mobile sensing nodes and static monitoring terminals, and collecting multi-source heterogeneous risk data related to population flow, infrastructure operation, environmental parameters and social activity characteristics in real time; Performing space-time alignment and feature extraction processing on the collected multi-source heterogeneous risk data, creating a risk data association rule base corresponding to a basic level treatment unit, cleaning data redundancy and noise through preset data fusion logic, and generating a standardized feature data set containing risk probability, influence range and evolution rate indexes; Building a major risk evolution dynamics model, inputting a standardized feature data set into the model, setting evolution parameters under different conditions, simulating diffusion paths of risks in different time dimensions, influencing strength changes and cross-unit conduction effects to output dynamic risk evolution patterns, intuitively presenting risk states, and setting intelligent early warning conditions or mechanisms to realize early warning and research and judgment on risks; Establishing a response priority evaluation matrix with a dynamic risk evolution map according to population density, infrastructure vulnerability and emergency resource distribution parameters of a basic level treatment unit, and generating a layered emergency response scheme adapting to different risk levels and different influence ranges through a risk-resource matching logic; Monitoring the risk evolution state and the response scheme execution progress of the basic level treatment unit in real time, and automatically triggering a scheme adjustment instruction based on a dynamic matching mechanism of a preset risk state and a response scheme when the risk evolution exceeds a preset simulation range or the scheme execution effect is not expected; After the scheme execution is finished, a response effect evaluation model is constructed according to the deviation of actual risk treatment data of the base layer treatment unit and a pre-dynamic simulation result, the validity of the emergency scheme and the accuracy of the evolution model are quantitatively evaluated, and a risk data association rule base and evolution dynamics model parameters are updated according to the evaluation result.
  2. 2. The method for dynamic simulation and agile response to major risks of ultra-large city base layer according to claim 1, wherein the distributed network is subjected to the following steps in a deployment stage: Acquiring population thermodynamic diagram data and an infrastructure point location distribution diagram in a preset historical time interval of a target base layer treatment unit, and dividing the target base layer treatment unit into a plurality of deployment subareas with identical adjacent sides and identical specifications by taking streets/villages as grid units; For each deployment subarea, if the population thermodynamic value is more than or equal to T1 or the quantity of infrastructures is more than or equal to T2, T1 and T2 are preset thresholds, deploying at least two static monitoring terminals in the subarea, and the distance between adjacent static monitoring terminals is not more than half of the side length of the deployment subarea, otherwise, deploying a unique static monitoring terminal in the subarea; The self-adaptive mobile sensing node performs path planning by taking the static monitoring terminal as an anchor point, synchronously sets the cruising period of the mobile node, and triggers two adjacent mobile sensing nodes of the anchor point to be close when the data precision acquired by any static monitoring terminal does not reach 90%, until the data acquisition precision of the static monitoring terminal is recovered to more than 95%.
  3. 3. The dynamic simulation and agile response method for major risks of a basic level of a super large city according to claim 1, wherein the processing flow of space-time alignment and feature extraction of the multi-source heterogeneous risk data is as follows: The space-time alignment process converts the space data collected by all the sensing nodes into coordinate values under a national 2000 geodetic coordinate system by taking the coordinate system as a space reference, and simultaneously uses the regional management server clock of the base layer treatment unit as a time reference, so that the time of all the sensing nodes is synchronized through an NTP protocol, and the synchronization precision is not more than 10ms; in the feature extraction processing stage, a multi-mode fusion extraction strategy is adopted for different types of multi-source heterogeneous risk data: the population flow data adopts a long-period memory network to extract sequential characteristics of population inflow/outflow rate and retention time within a continuous preset time threshold; The infrastructure operation data adopts a sliding window method to extract the statistical characteristics of the maximum value, the minimum value and the fluctuation variance in the preset time threshold; the environmental parameters adopt a convolutional neural network to extract the spatial characteristics of pollutant concentration distribution and temperature and humidity gradient in a square spatial range with preset size; Splicing the time sequence features, the statistical features and the space features into initial feature vectors with consistent dimensions, and then reducing the dimensions to preset dimensions through principal component analysis to obtain basic feature data for constructing a risk data association rule base; The infrastructure operation data comprise power supply load, water supply pipe pressure, unit time power supply quantity and unit time water supply quantity, the environment parameters comprise PM2.5, temperature and humidity, and the preset dimension is preferably 1/3-1/2 of the dimension of the initial feature vector.
  4. 4. The dynamic simulation and agile response method for major risk of oversized city base layer according to claim 1, wherein the expression of the major risk evolution dynamics model is: ; wherein: the risk intensity value of the ith base layer treatment unit at the moment t; A risk endophytic growth coefficient for the ith treatment unit; A risk bearing threshold value for the ith abatement unit; The number of base treatment units adjacent to the ith treatment unit; A risk cross-cell conductivity coefficient for the jth abatement unit to the ith abatement unit; the risk intensity value of the jth base layer treatment unit at the moment t; A risk cross-cell conductivity coefficient for the jth abatement unit to the ith abatement unit; the geometric center distance of the ith treatment unit and the jth treatment unit.
  5. 5. The method for dynamically simulating and agilely responding to the major risk of the basic level of the ultra-large city according to claim 1, wherein the step of creating the response priority evaluation matrix is as follows: constructing a calculation model of response priority index, wherein the expression is as follows: ; wherein: for response priority index; Is a risk probability; is the influence range; Is population density; is infrastructure vulnerability; reserve for emergency resources; the maximum emergency resource reserve is reserved for the urban area where the treatment unit is located; Allocating time length for emergency resources; the weight coefficients are risk dimension, vulnerability dimension and resource dimension; in response to the priority index, the base layer treatment unit number is taken as the row dimension of the matrix The grading interval of the matrix is the column dimension of the matrix, a response priority evaluation matrix is constructed, and each element in the matrix is a matching result of a corresponding row pointing to the base layer treatment unit and a corresponding column pointing to the priority grading interval; Wherein, the The sum is 1, the value range is restricted to 0.4-0.5, 0.3-0.4, 0.1-0.2, 、 、 、 、 、 All are normalized, and the value ranges are 0-1.
  6. 6. The method for dynamically simulating and agilely responding to the major risk of the basic level of the ultra-large city according to claim 4, wherein the dynamic risk evolution map is a three-dimensional visualization map, and the construction and output are as follows: Step1, using a vector boundary diagram of a basic level treatment unit as a base diagram, and adopting tone mapping to identify real-time risk intensity of each unit; Step2, setting a time axis scale by taking an hour as a unit, supporting switching and viewing of at least three time dimensions, and marking 3 base layer treatment unit numbers and risk values with the maximum risk intensity at the moment by each time node; Step3, connecting the treatment units with risk cross-unit conduction through the line segments with arrows, wherein the thickness of the line segments is equal to that of the line segments Positive correlation, application arrow direction indicates risk conduction direction; Step4 according to Dynamically adjusting, controlling units The larger the map updating frequency is, the faster the map updating frequency is, and conversely, the slower the map updating frequency is, The risk evolution rate is noted.
  7. 7. The method for dynamically simulating and agilely responding to the major risk of the basic level of the ultra-large city according to claim 1, wherein the triggering logic of the scheme adjustment instruction is as follows: Presetting risk state threshold value, namely setting risk intensity upper limit threshold value Upper threshold for risk diffusion rate If the real-time risk intensity of any base layer treatment unit Or real-time risk diffusion rate Starting risk early warning; scheme execution effect evaluation index, setting scheme execution progress threshold value Emergency resource utilization threshold If the scheme execution time length reaches 80% of the preset time length, the execution progress does not exceed Or the resource utilization rate is not exceeded Judging that the executing effect of the scheme is not expected; the automatic adjustment instruction is generated, if the risk evolution exceeds the preset simulation range, the adjustment instruction comprises the steps of increasing the response priority of the treatment unit by 1 level, increasing the emergency resource allocation amount of the unit, wherein the allocation amount increase proportion is that And the increment ratio is not lower than 20%, if the scheme execution effect is not expected, the adjustment instruction comprises the steps of re-planning the shortest path based on the real-time traffic data, supplementing the mobile emergency treatment units, wherein the supplementing quantity is the original unit quantity And the number of supplements is not less than 1, Representing the actual execution progress; and (3) verifying the adjustment instruction, namely after the instruction is generated, performing simulation verification through a major risk evolution dynamic model, if a simulation result shows that the risk intensity in a preset time threshold after adjustment is reduced by not less than 10%, issuing the adjustment instruction, otherwise, iterating the adjustment parameters again until the conditions are met.
  8. 8. The dynamic simulation and agile response method for major risks of ultra-large city base according to claim 1, wherein when the response effect evaluation model quantitatively evaluates the validity of an emergency scheme and the accuracy of an evolution model, the method is characterized by following: ; wherein: Is an emergency scheme effectiveness index; The risk intensity value of the base layer treatment unit after the scheme is executed; Performing a risk intensity value for the pre-substrate abatement unit for the plan; the method is an evolution model accuracy index; Number of risk monitors during execution of the protocol; the simulation risk intensity value of the significant risk evolution dynamic model at the moment t is obtained; the actual monitoring risk intensity value at the time t; Setting up 、 Denoted as q1, q2, and a composite evaluation score s=q1×e+q2×a is calculated.
  9. 9. The method for dynamically simulating and agilely responding to major risks of ultra-large city base according to claim 1, wherein the operation control of the adaptive mobile sensing node comprises a dynamic dormancy and wakeup mechanism based on energy consumption and data acquisition requirements: Collecting the residual electric quantity percentage Q and the data collection error of each self-adaptive mobile sensing node in real time , Respectively representing an acquisition value and a true value, wherein the true value is determined by the average value of adjacent static monitoring terminals; If Q is less than or equal to 20 percent If the number of the clock modules is less than or equal to 5%, the node is controlled to enter a dormant state, only the built-in clock modules and the wake-up detection modules are kept to run during the dormant period, and the data acquisition and transmission modules are closed; setting a wake-up triggering condition of the dormant node: The method comprises the following steps that 1, a node receives a data precision shortage signal sent by an adjacent static monitoring terminal through a wake-up detection module, namely, the acquisition precision of the static monitoring terminal is not more than 90%; the node residual electric quantity is supplemented to Q which is more than or equal to 50% through a built-in charging module of the node residual electric quantity; When any one of the conditions is met, the node automatically wakes up and resumes the data acquisition and transmission function; Setting wake-up delay time synchronously , The delay coefficient is represented, and the value is 0.5s/%, namely the higher the residual electric quantity is, the longer the node wake-up delay time is.

Description

Dynamic simulation and agile response method for major risk of oversized city base layer Technical Field The invention relates to the technical field of data analysis, in particular to a dynamic simulation and agile response method for major risks of a basic level of a super large city. Background The city base layer is the final kilometer of city governance, and takes streets/villages and towns, communities/villages as basic governance bodies to connect governments, social organizations and residents. The system bears functions of civil service, community management, policy landing and the like, is a basic unit for responding to demands of masses and maintaining urban operation, and directly relates to living quality of residents and harmony and stability of cities. The invention discloses a city safety risk monitoring and early warning system based on big data, which comprises a city comprehensive risk monitoring module, a supervision management and cooperative treatment module, a city safety analysis early warning module, a city safety management and cooperative treatment module, a city safety control system and a city safety control system, wherein the city comprehensive risk monitoring module is used for realizing real-time monitoring, early warning of risk prediction for city safety management by collecting, converging and analyzing the big data, the city public service module is used for providing two services of government affair clients and public, reporting and pushing the discovered city safety risk information, sending city public safety knowledge interpretation to target users, the supervision management and cooperative treatment module is used for supervising and managing the discovered city safety risk, dividing departments and checking hidden danger, and carrying out the supervision on the generated city safety risk, and realizing the quick elimination of the cooperative treatment by the cooperative treatment, realizing the quick elimination of the abnormal running condition by the fusion analysis of the big data, and the risk prediction early warning is provided for the city safety management, and the city safety public safety management is provided, and the city public safety risk public safety management information is reported and pushed to target users, and the city public safety risk is discovered. However, from the urban basic public security risk management process, related tasks such as collection, analysis, management and the like of risk information data still depend on manpower to a great extent at present, and the intelligent degree and the disposal efficiency are low. Therefore, we propose a dynamic simulation and agile response method for major risks of the ultra-large city base layer. Disclosure of Invention Aiming at the defects existing in the prior art, the invention provides a dynamic simulation and agile response method for major risks of a basic level of a super large city, which can effectively solve the defects of the prior art. In order to achieve the above object, the present invention is achieved by the following technical scheme; The invention discloses a dynamic simulation and agile response method for major risks of a basic level of an oversized city, which comprises the following steps: Taking an urban basic level treatment unit as a division basis, deploying a distributed network comprising self-adaptive mobile sensing nodes and static monitoring terminals, and collecting multi-source heterogeneous risk data related to population flow, infrastructure operation, environmental parameters and social activity characteristics in real time; Performing space-time alignment and feature extraction processing on the collected multi-source heterogeneous risk data, creating a risk data association rule base corresponding to a basic level treatment unit, cleaning data redundancy and noise through preset data fusion logic, and generating a standardized feature data set containing risk probability, influence range and evolution rate indexes; Building a major risk evolution dynamics model, inputting a standardized feature data set into the model, setting evolution parameters under different conditions, simulating diffusion paths of risks in different time dimensions, influencing strength changes and cross-unit conduction effects to output dynamic risk evolution patterns, intuitively presenting risk states, and setting intelligent early warning conditions or mechanisms to realize early warning and research and judgment on risks; establishing a response priority evaluation matrix according to population density, infrastructure vulnerability, emergency resource distribution (emergency team quantity and equipment condition) and emergency refuge site construction condition of the basic level treatment unit and a dynamic risk evolution map, and generating a layered emergency response scheme adapting to different risk levels and different influence ranges through a risk-resource mat