A park management system based on digital twin

Through digital twin models and deep learning technology, combined with blockchain and the Internet of Things, the problem of low data integration and analysis efficiency in traditional park management is solved, real-time monitoring and prediction is realized, response speed and user satisfaction are improved, and user satisfaction is reduced. Operating cost.

CN118734418BActive Publication Date: 2024-11-22JINGCAI FUTURE INTELLIGENT TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411232295.5
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-09-04
Publication Date
2024-11-22
Estimated Expiration
2044-09-04

AI Technical Summary

Technical Problem

The lack of intelligent platforms in traditional park management has led to low data integration and analysis efficiency, and the inability to share resources and information and real-time statistics, which has increased operating costs and customer service difficulties.

Method used

The park management system based on digital twins is built by building the target park digital twin model, using deep learning to build a dynamic element feature prediction model, combining blockchain and Internet of Things monitoring nodes, monitoring and analyzing the dynamic element features of the park in real time, providing the best transportation Route and emergency repair operations.

Benefits of technology

Real-time monitoring and prediction of dynamic elements in the park is realized, which improves response speed, reduces unnecessary inspection and maintenance work, reduces costs, and improves user convenience and satisfaction.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118734418B_ABST
    Figure CN118734418B_ABST
Patent Text Reader

Abstract

A park management system based on digital twins relates to the field of park management technology. A digital twin model of a target park is constructed according to a dynamic element feature set in the target park and functional characteristics of several buildings and infrastructures. A dynamic element feature estimation model is constructed based on deep learning, and an estimated numerical time series sequence corresponding to each type of dynamic element feature of each digital three-dimensional model in the current monitoring period is output. A park comprehensive management platform is constructed. The park comprehensive management platform feeds back the best traffic route to the user according to the user's behavioral feature data and the estimated numerical time series sequence corresponding to each type of dynamic element feature of each digital three-dimensional model, and monitors each type of dynamic element feature of each digital three-dimensional model in the digital twin model in real time. According to the real-time monitoring results, emergency maintenance operations are performed on the buildings or infrastructure corresponding to the digital three-dimensional models, thereby improving the overall response speed of the park.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of park management, and in particular to a park management system based on digital twins. Background Art

[0002] Prior art CN117196480B "A smart logistics park management system based on digital twin" includes a data acquisition module, a data processing module, a data storage module, and a data output module; the data acquisition module is used to collect real-time data of the warehouse; the data storage module is used to store the three-dimensional model data of the warehouse pre-established by the user and the historical data of the warehouse; the data processing module is used to establish a mapping relationship between the historical data and the three-dimensional model data, import the cargo location information into the three-dimensional model data, and finally prepare a target location report for cargo storage based on the three-dimensional model data, and generate a path planning report;

[0003] The prior art CN116227752B "A park facility management system based on the Internet of Things" includes: a model building module, which is used to collect park spatial data and establish a park digital twin model based on the park spatial data; an environmental monitoring module, which is set at a preset location point in the park and is used to monitor the environmental data of the preset location point in the park; a patrol module, which is used to obtain monitored patrol environmental data according to a preset patrol route; an analysis and management module, which is used to predict the environmental data of a preset patrol route based on the park digital twin model, the environmental data of the preset location point and the preset patrol route, obtain predicted patrol environmental data, and judge the status of the environmental monitoring module based on the comparison between the patrol environmental data and the predicted patrol environmental data;

[0004] At present, the problems faced by traditional park management include the lack of a management platform. Due to the lack of an overall plan for traditional parks, the system is isolated, making it difficult to manage and count park construction in a targeted manner, resulting in low work efficiency. Park operations generate a large amount of data, which managers cannot efficiently integrate, analyze and count. Without an intelligent platform, employees spend a lot of manpower and material resources on internal information management and statistics, which undoubtedly increases the operating costs of the park and cannot better serve customers. The management methods are low. Taking the rental park office area as an example, traditional parks lack the support of Internet technology. When the rent is about to expire, business personnel need to manually send expiration reminders to customers. If the customer scale is large, it will inevitably consume a lot of energy. There is no intelligent management method such as the Internet of Things. When the company is overdue, it is necessary to manually take power-limiting measures in the office area, and the efficiency of collecting rent is low. How to share and integrate resources and information, obtain park data in real time, and conduct statistics and analysis to achieve unified scheduling and management of business within the park is an issue that we urgently need to solve. Summary of the invention

[0005] In order to solve the above technical problems, the purpose of the present invention is to provide a park management system based on digital twins, comprising the following steps:

[0006] Step s1: construct a digital twin model of the target park based on the dynamic element feature set and the functional characteristics of several buildings and infrastructure in the target park;

[0007] Step s2: construct a dynamic element feature estimation model based on deep learning, and output an estimated numerical time series sequence corresponding to each type of dynamic element feature of each digital three-dimensional model in the current monitoring period;

[0008] Step s3: Build a comprehensive park management platform. The park comprehensive management platform will provide users with feedback on the best traffic routes based on the user's behavioral characteristic data and the estimated numerical time series corresponding to the characteristics of each type of dynamic elements of each digital three-dimensional model, and conduct real-time monitoring of the characteristics of each type of dynamic elements of each digital three-dimensional model in the digital twin model, and perform emergency maintenance operations on the buildings or infrastructure corresponding to the digital three-dimensional model based on the real-time monitoring results.

[0009] Furthermore, the process of constructing a digital twin model of the target park according to the dynamic element feature set and the functional characteristics of several buildings and infrastructures in the target park includes:

[0010] Acquire physical entities of several buildings and infrastructure in the target park in the physical space, perform three-dimensional modeling on the physical entities to obtain a three-dimensional model, set an API interface on the three-dimensional model, acquire functional characteristics of several buildings and infrastructure, assign attributes to the three-dimensional models corresponding to several buildings according to the functional characteristics, and acquire a digital three-dimensional model;

[0011] Construct a three-dimensional stereogram of the target park, obtain the location characteristics of several buildings and infrastructure in the target park and the dynamic element feature set in the target park by means of GIS, and map the digital three-dimensional models corresponding to several buildings and infrastructure to the three-dimensional stereogram according to the location characteristics;

[0012] Deploy IoT monitoring nodes on several buildings and infrastructures in the target park according to the dynamic element feature set and the functional characteristics of several buildings and infrastructures in the target park, obtain the data monitoring type of each building and infrastructure by using data retrieval according to the functional characteristics of each building and infrastructure, and the IoT monitoring nodes collect data and mark the monitoring time according to the data monitoring type, and set the monitoring cycle;

[0013] The data collected by the Internet of Things monitoring nodes corresponding to the digital three-dimensional models are obtained for data format preprocessing to generate twin data, and the twin data are matched with the corresponding digital three-dimensional models to obtain the digital twin model.

[0014] Furthermore, the process of deploying IoT monitoring nodes on several buildings and infrastructures in the target park according to the dynamic element feature set and the functional characteristics of several buildings and infrastructures in the target park includes:

[0015] Obtain a dynamic element feature set of each building and infrastructure in the target park, use each type of dynamic element feature in the dynamic element feature set of each building and infrastructure as an evaluation index, wherein each type of dynamic element feature includes pedestrian flow feature data, vehicle flow feature data, environmental feature data, energy feature data, and equipment status feature data, obtain an index weight matrix of the evaluation index according to the functional characteristics of each building and infrastructure, preset a complexity level, and obtain a membership matrix of each building and infrastructure to the complexity level through fuzzy comprehensive evaluation;

[0016] The complexity level of each building and infrastructure is obtained according to the membership matrix and the indicator weight matrix, the number of IoT monitoring nodes deployed for each building and infrastructure is obtained according to the complexity level, and the coverage of the IoT monitoring nodes is obtained. According to the digital three-dimensional models of each building and infrastructure and the coverage of the IoT monitoring nodes, the layout positions of the IoT monitoring nodes in each digital three-dimensional model are determined by the particle swarm algorithm.

[0017] Furthermore, the process of constructing a dynamic element feature estimation model based on deep learning and outputting an estimated numerical time series sequence corresponding to each type of dynamic element feature of each digital three-dimensional model in the current monitoring period includes:

[0018] Construct a dynamic element feature prediction model based on deep learning, obtain dynamic element feature sets in several historical monitoring cycles of each digital three-dimensional model as training sets and test sets, input the training sets into the dynamic element feature prediction model for training until the loss function training is stable, and save the model parameters, test the dynamic element feature prediction model through the test set until it meets the preset requirements, and output the dynamic element feature prediction model;

[0019] The dynamic element feature estimation model outputs the estimated numerical time series corresponding to each type of dynamic element feature of each digital three-dimensional model within the current monitoring period.

[0020] Furthermore, the process of building a comprehensive park management platform includes:

[0021] A comprehensive park management platform is constructed. The comprehensive park management platform is communicatively connected to several blockchain nodes and the API interfaces of each digital three-dimensional model in the digital twin model. The blockchain nodes are linked to each other to form a blockchain network. The blockchain nodes are communicatively connected to user terminals and Internet of Things monitoring nodes. The user terminals are used to collect user behavior feature data and encrypt and upload them to the blockchain nodes. The behavior feature data includes the user's target digital three-dimensional model data and location features.

[0022] Furthermore, the process of the park integrated management platform feeding back the best traffic route to the user based on the user's behavioral characteristic data and the estimated numerical time series corresponding to the characteristics of each type of dynamic elements of each digital three-dimensional model includes:

[0023] Mapping the user's location features and the target digital three-dimensional model into the digital twin model, extracting all path trajectories between the user's location features and the target digital three-dimensional model according to the digital twin model, extracting the digital three-dimensional model corresponding to each path trajectory, and marking the digital three-dimensional model as the digital three-dimensional model to be compared;

[0024] Obtain a numerical time series sequence corresponding to the traffic characteristic data collected during the current monitoring period of each digital three-dimensional model to be compared and an estimated numerical time series sequence corresponding to the traffic characteristic data in the remaining time period of the current monitoring period; obtain a full-cycle numerical time series sequence corresponding to the traffic characteristic data of each digital three-dimensional model to be compared based on the numerical time series sequence corresponding to the traffic characteristic data and the estimated numerical time series sequence; obtain a traffic congestion estimation coefficient of each digital three-dimensional model to be compared based on the full-cycle numerical time series sequence;

[0025] Similarly, the full-cycle numerical time series sequence corresponding to the pedestrian characteristic data of the current monitoring period of each digital three-dimensional model is obtained, and the pedestrian congestion coefficient of each digital three-dimensional model is obtained according to the full-cycle numerical time series sequence, and the traffic impact coefficient between each digital three-dimensional model and each digital three-dimensional model to be compared is obtained. According to the vehicle congestion estimation coefficient, the pedestrian congestion coefficient and the traffic impact coefficient, the traffic priority coefficient of each digital three-dimensional model to be compared is obtained, and the path trajectory corresponding to the digital three-dimensional model to be compared with the largest traffic priority coefficient is screened out and fed back to the user.

[0026] Furthermore, the park integrated management platform monitors the characteristics of various types of dynamic elements of each digital three-dimensional model in the digital twin model in real time, and performs emergency maintenance operations on the buildings or infrastructure corresponding to the digital three-dimensional model according to the real-time monitoring results. The process includes:

[0027] Mark the environmental feature data, energy feature data and equipment status feature data of each digital three-dimensional model as key feature data, use the estimated numerical time series sequence corresponding to the key feature data of each digital three-dimensional model in the current monitoring period as the judgment data, extract the numerical time series sequence corresponding to the key feature data of each digital three-dimensional model from the data collected by the Internet of Things monitoring node received by each blockchain node, compare the numerical time series sequence corresponding to the key feature data with the corresponding judgment data, obtain the deviation rate information corresponding to the key feature data, preset the deviation rate threshold, compare the deviation rate information with the deviation rate threshold, if the deviation rate information of the key feature data of a certain digital three-dimensional model is greater than the deviation rate threshold, then mark the key feature data as abnormal feature data, generate an abnormal feature alarm signal of the certain digital three-dimensional model and feed it back to the manager of the park comprehensive management platform, and the manager performs emergency maintenance operations on the building or infrastructure corresponding to the certain digital three-dimensional model according to the abnormal feature alarm signal;

[0028] When abnormal feature alarm signals of several digital three-dimensional models are generated, importance coefficients of the several digital three-dimensional models are obtained, the several digital three-dimensional models are sorted according to the importance coefficients, and the operation sequence of emergency maintenance operations for buildings or infrastructures corresponding to the several digital three-dimensional models by management personnel is determined according to the sorting results.

[0029] Furthermore, the process of obtaining the importance coefficients of several digital three-dimensional models includes:

[0030] The deviation rate information of the abnormal feature data of each digital three-dimensional model is obtained, and the deviation amplitude between the deviation rate threshold, the pedestrian congestion coefficient and the vehicle congestion coefficient are obtained; and the importance coefficient of each digital three-dimensional model is obtained according to the deviation amplitude, the pedestrian congestion coefficient and the vehicle congestion coefficient.

[0031] Compared with the prior art, the present invention has the following beneficial effects:

[0032] 1. Through real-time monitoring and prediction of the characteristics of dynamic elements in the park, it is possible to respond to emergencies in a timely manner, such as traffic congestion, facility failures, etc., and improve the overall response speed of the park.

[0033] 2. Provide customized services based on the user's personal behavioral characteristics. By optimizing traffic flow and resource scheduling, it can effectively reduce the user's waiting time and improve user convenience and satisfaction.

[0034] 3. Through real-time monitoring and analysis of the dynamic characteristics of facilities and accurate positioning of facilities that need maintenance, unnecessary inspection and maintenance work can be reduced, thereby reducing costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1This is a schematic diagram of a digital twin-based campus management system according to an embodiment of the present application. DETAILED DESCRIPTION

[0036] The following is a clear and complete description of the technical solutions in the embodiments of the present application in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present application.

[0037] like Figure 1 As shown, a park management system based on digital twins includes the following steps:

[0038] Step s1: construct a digital twin model of the target park based on the dynamic element feature set and the functional characteristics of several buildings and infrastructure in the target park;

[0039] Step s2: construct a dynamic element feature estimation model based on deep learning, and output an estimated numerical time series sequence corresponding to each type of dynamic element feature of each digital three-dimensional model in the current monitoring period;

[0040] Step s3: Build a comprehensive park management platform. The park comprehensive management platform will provide users with feedback on the best traffic routes based on the user's behavioral characteristic data and the estimated numerical time series corresponding to the characteristics of each type of dynamic elements of each digital three-dimensional model, and conduct real-time monitoring of the characteristics of each type of dynamic elements of each digital three-dimensional model in the digital twin model, and perform emergency maintenance operations on the buildings or infrastructure corresponding to the digital three-dimensional model based on the real-time monitoring results.

[0041] It should be further explained that, in the specific implementation process, the process of constructing the digital twin model of the target park according to the dynamic element feature set and the functional characteristics of several buildings and infrastructure in the target park includes:

[0042] Obtain physical entities of several buildings and infrastructure in the target park in the physical space, perform three-dimensional modeling on the physical entities to obtain a three-dimensional model, set an API interface on the three-dimensional model, obtain functional characteristics of several buildings and infrastructure, the functional characteristics include commercial functions, residential functions, educational functions, medical functions, transportation hub functions, water supply and drainage functions, energy functions, etc., assign attributes to the three-dimensional models corresponding to several buildings according to the functional characteristics, and obtain a digital three-dimensional model;

[0043] Construct a three-dimensional stereogram of the target park, obtain the location features of several buildings and infrastructure in the target park and the dynamic element feature set in the target park by means of GIS, wherein the location features include the longitude and latitude coordinates covered by the buildings and infrastructure and the elevation data of the buildings and infrastructure, and map the digital three-dimensional models corresponding to the several buildings and infrastructure to the three-dimensional stereogram according to the location features;

[0044] Deploy IoT monitoring nodes on several buildings and infrastructures in the target park according to the dynamic element feature set and the functional characteristics of several buildings and infrastructures in the target park, obtain the data monitoring type of each building and infrastructure by using data retrieval according to the functional characteristics of each building and infrastructure, and the IoT monitoring nodes collect data and mark the monitoring time according to the data monitoring type, and set the monitoring cycle;

[0045] The data monitoring types include:

[0046] Building data monitoring types: environmental monitoring (e.g. temperature and humidity, air quality (CO2, VOCs, PM2.5, etc.)), energy management (e.g. power consumption, water consumption, energy efficiency indicators (e.g. EUI), structural health monitoring (e.g. structural vibration, tilt, strain and stress, crack monitoring, etc.), safety monitoring (e.g. fire alarm, smoke detection, etc.), equipment performance monitoring (e.g. HVAC system performance, lighting system status, elevator and escalator status, fire equipment status, etc.);

[0047] Infrastructure data monitoring types: transportation infrastructure (e.g. traffic flow, road conditions (such as road damage), bridge health monitoring, etc.), water supply and drainage system (e.g. water quality monitoring, water pressure and flow, pump station operation status, drainage network blockage, etc.), energy infrastructure (e.g. power generation facility operation status, transmission and distribution line status, grid load, etc.), communication infrastructure (e.g. signal strength, network latency, communication tower health, fiber optic line status, etc.);

[0048] The data collected by the Internet of Things monitoring nodes corresponding to the digital three-dimensional models are obtained for data format preprocessing to generate twin data, and the twin data are matched with the corresponding digital three-dimensional models to obtain the digital twin model.

[0049] It should be further explained that, in the specific implementation process, the process of deploying IoT monitoring nodes on several buildings and infrastructures in the target park according to the dynamic element feature set and the functional characteristics of several buildings and infrastructures in the target park includes:

[0050] The dynamic element feature set of each building and infrastructure in the target park is obtained, and each type of dynamic element feature in the dynamic element feature set of each building and infrastructure is used as an evaluation index. The various types of dynamic element features include pedestrian feature data, vehicle feature data, environmental feature data, energy feature data, and equipment status feature data. The indicator weight matrix of the evaluation index is obtained according to the functional characteristics of each building and infrastructure. It should be noted that the indicator weight of the evaluation index is determined according to the experience of experts, and the purpose is to reduce the uncertainty in the fuzzy evaluation process. The significance of obtaining the indicator weight matrix of the evaluation index according to the functional characteristics of each building and infrastructure is that, for example, if the functional characteristics of the infrastructure are transportation hub functions , then for transportation infrastructure, the dynamic element characteristics with higher weights for the complexity of transportation infrastructure include traffic flow characteristics. Traffic flow characteristics are crucial for transportation infrastructure because they are directly related to issues such as traffic smoothness, safety and traffic congestion. The dynamic element characteristics with lower weights for the complexity of transportation infrastructure include environmental characteristics, such as air quality degradation. Although it has an impact on people's health, it has little impact on the transportation infrastructure itself. Of course, in extreme cases, severe air pollution may affect vision and thus affect traffic safety, but this is usually not the norm. Then, the complexity level is preset, and the membership matrix of each building and infrastructure to the complexity level is obtained through fuzzy comprehensive evaluation;

[0051] The complexity level of each building and infrastructure is obtained according to the membership matrix and the indicator weight matrix, the number of IoT monitoring nodes deployed for each building and infrastructure is obtained according to the complexity level, and the coverage of the IoT monitoring nodes is obtained. According to the digital three-dimensional models of each building and infrastructure and the coverage of the IoT monitoring nodes, the layout positions of the IoT monitoring nodes in each digital three-dimensional model are determined by the particle swarm algorithm.

[0052] It should be further explained that, in the specific implementation process, the process of obtaining the complexity level of each building and infrastructure according to the membership matrix and the indicator weight matrix includes:

[0053] The indicator weight matrix and the membership matrix of the evaluation indicators are integrated through a formula to obtain a fuzzy comprehensive evaluation matrix of the evaluation indicators, the membership of each building and infrastructure to different complexity levels is obtained according to the fuzzy comprehensive evaluation matrix, the complexity level with the highest membership corresponding to each building and infrastructure is screened out, and the complexity level with the highest membership corresponding to each building and infrastructure is used as the complexity level of each building and infrastructure;

[0054] Wherein, the formula is:

[0055] ;

[0056] in, is the fuzzy comprehensive evaluation matrix of the evaluation index, is the indicator weight matrix of the evaluation indicator, is the membership matrix, " represents the multiplication of the weight matrix of the evaluation index and the elements at the corresponding positions of the membership matrix, It is a weighted parameter for controlling the balance between the weight matrix and the membership matrix in the fuzzy comprehensive evaluation matrix of the evaluation index.

[0057] It should be further explained that, in the specific implementation process, the process of determining the layout positions of the IoT monitoring nodes in each digital three-dimensional model through the particle swarm algorithm according to the digital three-dimensional models of each building and infrastructure and the coverage of the IoT monitoring nodes is as follows:

[0058] Obtain a three-dimensional model of a building or infrastructure, define an optimization objective function, the objective function in this embodiment is to minimize the monitoring blind area, determine the size of the particle group, randomly generate an initial position and speed for each particle (representing a deployment plan), set parameters, and then perform an optimization process. After multiple iterations, determine the optimal solution, that is, the optimal position of the group gbest, which represents the optimal deployment position of the IoT monitoring point. The calculation formula of the optimization process is:

[0059] ;

[0060] ;

[0061] in, represents the inertia weight; , Represents the acceleration coefficient, which is used to weight the optimal position of the particle itself and the optimal position of the group; Indicates the current iteration of particle i speed; Indicates the current iteration of particle i location; , is a random number between 0 and 1; represents the optimal position of particle i.

[0062] It should be further explained that, in the specific implementation process, the process of constructing a dynamic element feature estimation model based on deep learning and outputting the estimated numerical time series corresponding to each type of dynamic element features of each digital three-dimensional model in the current monitoring period includes:

[0063] Construct a dynamic element feature prediction model based on deep learning, obtain dynamic element feature sets in several historical monitoring cycles of each digital three-dimensional model as training sets and test sets, input the training sets into the dynamic element feature prediction model for training until the loss function training is stable, and save the model parameters, test the dynamic element feature prediction model through the test set until it meets the preset requirements, and output the dynamic element feature prediction model;

[0064] The dynamic element feature estimation model outputs the estimated numerical time series corresponding to each type of dynamic element feature of each digital three-dimensional model within the current monitoring period.

[0065] It should be further explained that, in the specific implementation process, the process of building a comprehensive park management platform includes:

[0066] A comprehensive park management platform is constructed. The comprehensive park management platform is communicatively connected to several blockchain nodes and the API interfaces of each digital three-dimensional model in the digital twin model. The blockchain nodes are linked to each other to form a blockchain network. The blockchain nodes are communicatively connected to user terminals and Internet of Things monitoring nodes. The user terminals include mobile phones, computers, etc. The user terminals are used to collect user behavior feature data and encrypt and upload them to the blockchain nodes. The behavior feature data includes the user's target digital three-dimensional model data and location features.

[0067] It should be further explained that, in the specific implementation process, the park integrated management platform provides the user with feedback on the best traffic route based on the user's behavioral characteristic data and the estimated numerical time series corresponding to the characteristics of various types of dynamic elements of each digital three-dimensional model, including:

[0068] Mapping the user's location features and the target digital three-dimensional model into the digital twin model, extracting all path trajectories between the user's location features and the target digital three-dimensional model according to the digital twin model, extracting the digital three-dimensional model corresponding to each path trajectory, and marking the digital three-dimensional model as the digital three-dimensional model to be compared;

[0069] Obtain a numerical time series sequence corresponding to the traffic characteristic data collected during the current monitoring period of each digital three-dimensional model to be compared, obtain an estimated numerical time series sequence corresponding to the traffic characteristic data of each digital three-dimensional model to be compared in the remaining time period of the current monitoring period, obtain a full-cycle numerical time series sequence corresponding to the traffic characteristic data of each digital three-dimensional model to be compared based on the numerical time series sequence corresponding to the traffic characteristic data and the estimated numerical time series sequence, and obtain a traffic congestion estimation coefficient of each digital three-dimensional model to be compared based on the full-cycle numerical time series sequence corresponding to the traffic characteristic data;

[0070] Among them, the calculation formula for obtaining the traffic congestion estimation coefficient of each digital three-dimensional model to be compared according to the full-cycle numerical time series corresponding to the traffic characteristic data is:

[0071] ;

[0072] ;

[0073] ;

[0074] in, represents the traffic congestion prediction coefficient of the digital three-dimensional model j to be compared, represents the total number of moments of the numerical time series corresponding to the collected traffic flow characteristic data in the full-cycle numerical time series corresponding to the traffic flow characteristic data of the digital three-dimensional model j to be compared, ts represents the current collection time of the current monitoring cycle, represents the total number of moments of the estimated numerical time series sequence corresponding to the traffic flow characteristic data in the remaining time period of the current monitoring cycle in the full-cycle numerical time series sequence corresponding to the traffic flow characteristic data of the digital three-dimensional model j to be compared, represents the value at the t1th moment corresponding to the collected traffic flow characteristic data in the full-cycle numerical time series corresponding to the traffic flow characteristic data of the digital three-dimensional model j to be compared, represents the estimated value at the t2th moment corresponding to the traffic characteristic data in the remaining time period of the current monitoring cycle in the full-cycle numerical time series corresponding to the traffic characteristic data of the digital three-dimensional model j to be compared, represents the weight factor of the value at the t1th moment corresponding to the collected traffic flow characteristic data in the full-cycle numerical time series corresponding to the traffic flow characteristic data of the digital three-dimensional model j to be compared, represents the weight factor of the estimated value at the t2th moment corresponding to the traffic characteristic data in the remaining time period of the current monitoring cycle in the full-cycle numerical time series corresponding to the traffic characteristic data of the digital three-dimensional model j to be compared, e represents the index, represents the forgetting coefficient, represents the influence coefficient;

[0075] The weighted average processing is performed by the numerical time series sequence of the collected traffic flow characteristic data and the estimated numerical time series sequence of the uncollected traffic flow characteristic data in the above formula. The weight of the numerical value of the collected traffic flow characteristic data is dynamically adjusted according to the time interval between the numerical value of the collected traffic flow characteristic data and the current collection time. The smaller the time interval between the numerical value of the collected traffic flow characteristic data and the current collection time, the greater the weight of the numerical value of the collected traffic flow characteristic data. At the same time, the estimated numerical value of the future estimated traffic flow characteristic data is dynamically adjusted according to the time interval between the estimated numerical value of the future estimated traffic flow characteristic data and the current collection time. Weight: the longer the time interval between the estimated value of the future estimated traffic flow characteristics and the current collection time, the more likely the estimated value of the future estimated traffic flow characteristics will deviate, so the weight of the estimated value of the future estimated traffic flow characteristics is smaller. The traffic flow characteristic data collected in the current collection period of each digital three-dimensional model to be compared and the future estimated traffic flow characteristic data are comprehensively considered to obtain the traffic congestion coefficient representing the comprehensive traffic congestion situation of the digital three-dimensional model to be compared in the current collection period, so as to ensure the accuracy and real-time performance of the traffic congestion coefficient in reflecting the comprehensive traffic congestion situation of the digital three-dimensional model to be compared;

[0076] Similarly, the full-cycle numerical time series sequence corresponding to the crowd flow characteristic data of each digital three-dimensional model in the current monitoring period is obtained, and the crowd congestion coefficient of each digital three-dimensional model is obtained according to the full-cycle numerical time series sequence corresponding to the crowd flow characteristic data. The specific process of obtaining the full-cycle numerical time series sequence corresponding to the crowd flow characteristic data is to obtain the numerical time series sequence corresponding to the crowd flow characteristic data collected in the current monitoring period of each digital three-dimensional model, obtain the estimated numerical time series sequence corresponding to the crowd flow characteristic data of each digital three-dimensional model in the remaining time period of the current monitoring period, and obtain the full-cycle numerical time series sequence corresponding to the crowd flow characteristic data of each digital three-dimensional model in the current monitoring period according to the numerical time series sequence corresponding to the crowd flow characteristic data and the estimated numerical time series sequence. The method comprises the following steps: a first step is to obtain a numerical time series sequence of each digital three-dimensional model and a digital three-dimensional model to be compared according to the position characteristics of each digital three-dimensional model and the digital three-dimensional model to be compared, and a second step is to obtain a traffic impact coefficient between each digital three-dimensional model and each digital three-dimensional model to be compared according to the shortest Euclidean distance; a second step is to obtain a traffic priority coefficient of each digital three-dimensional model to be compared according to the vehicle congestion estimation coefficient of each digital three-dimensional model to be compared, the pedestrian congestion coefficient of each digital three-dimensional model and the traffic impact coefficient between each digital three-dimensional model and each digital three-dimensional model to be compared, and a path trajectory corresponding to the digital three-dimensional model to be compared having the largest traffic priority coefficient is screened out and fed back to the user.

[0077] It should be further explained that, in the specific implementation process, the calculation formula for obtaining the traffic impact coefficient between each digital three-dimensional model and each digital three-dimensional model to be compared according to the shortest Euclidean distance is:

[0078] ;

[0079] in, represents the traffic impact coefficient between the digital 3D model z and the digital 3D model j to be compared, represents the conversion coefficient, e represents the exponent, represents the Euclidean distance between the digital three-dimensional model z and the digital three-dimensional model j to be compared;

[0080] According to the traffic congestion estimation coefficient of the digital three-dimensional model to be compared, the pedestrian congestion coefficient of each digital three-dimensional model and the traffic impact coefficient between each digital three-dimensional model and the digital three-dimensional model to be compared, the calculation formula for obtaining the traffic priority coefficient of the digital three-dimensional model to be compared is:

[0081] ;

[0082] in, represents the traffic priority coefficient of the digital 3D model j to be compared, represents the conversion factor, represents the crowd congestion coefficient of the digital 3D model z, represents a set of other digital 3D models except the digital 3D model j to be compared, , Represents the weight factor.

[0083] It should be further explained that, in the specific implementation process, the park comprehensive management platform monitors the characteristics of various types of dynamic elements of each digital three-dimensional model in the digital twin model in real time, and performs emergency maintenance operations on the buildings or infrastructure corresponding to the digital three-dimensional model based on the real-time monitoring results. The process includes:

[0084] The estimated numerical time series sequence corresponding to the environmental characteristic data, energy characteristic data and equipment status characteristic data of each digital three-dimensional model in the current monitoring period is used as the judgment data, and the numerical time series sequence corresponding to the environmental characteristic data, energy characteristic data and equipment status characteristic data of each digital three-dimensional model is extracted from the data collected by the Internet of Things monitoring node received by each blockchain node, and the numerical time series sequence corresponding to the environmental characteristic data, energy characteristic data and equipment status characteristic data is compared with the corresponding judgment data to obtain the deviation rate information corresponding to the environmental characteristic data, energy characteristic data and equipment status characteristic data, preset the deviation rate threshold, and compare the deviation rate information with the deviation rate threshold. If the deviation rate information of the environmental characteristic data, energy characteristic data or equipment status characteristic data of a certain digital three-dimensional model is greater than the deviation rate threshold, the environmental characteristic data, energy characteristic data or equipment status characteristic data is marked as abnormal characteristic data, and an abnormal characteristic alarm signal of the certain digital three-dimensional model is generated and fed back to the management personnel of the park comprehensive management platform, and the management personnel perform emergency maintenance operations on the building or infrastructure corresponding to the certain digital three-dimensional model according to the abnormal characteristic alarm signal;

[0085] It should be further explained that, in the specific implementation process, the calculation formula for obtaining the deviation rate information of the key feature data is:

[0086] ;

[0087] in, Indicates the deviation rate information of key feature data k, represents the value of key feature data k at the tth moment, represents the value of the interpretation data corresponding to the key feature data k at the tth moment, and n1 represents the total number of moments;

[0088] When abnormal feature alarm signals of several digital three-dimensional models are generated, the importance coefficients of the several digital three-dimensional models are obtained, and the several digital three-dimensional models are sorted according to the importance coefficients. The larger the importance coefficient, the higher the ranking. The operation order of emergency maintenance operations for buildings or infrastructures corresponding to the several digital three-dimensional models by management personnel is determined according to the sorting results.

[0089] It should be further explained that, in the specific implementation process, the process of obtaining the importance coefficients of several digital three-dimensional models includes:

[0090] Obtaining the deviation amplitude between the deviation rate information of the abnormal feature data of each digital three-dimensional model and the deviation rate threshold, the pedestrian congestion coefficient and the vehicle congestion coefficient, and obtaining the importance coefficient of each digital three-dimensional model according to the deviation amplitude, the pedestrian congestion coefficient and the vehicle congestion coefficient;

[0091] The calculation formula for obtaining the importance coefficient of the digital three-dimensional model according to the deviation amplitude, the pedestrian congestion coefficient and the vehicle congestion coefficient is:

[0092] ;

[0093] in, Represents the importance coefficient of the digital 3D model z, represents the deviation amplitude of the digital 3D model z, represents the crowd congestion coefficient of the digital 3D model z, Represents the traffic congestion coefficient of the digital 3D model z.

[0094] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A park management system based on digital twins, characterized in that: The following steps are involved: Step s1: Construct a digital twin model of the target park based on the dynamic element feature set and the functional characteristics of several buildings and infrastructures in the target park, including: Acquire physical entities of several buildings and infrastructure in the target park in the physical space, perform three-dimensional modeling on the physical entities to obtain a three-dimensional model, set an API interface on the three-dimensional model, acquire functional characteristics of several buildings and infrastructure, assign attributes to the three-dimensional models corresponding to several buildings according to the functional characteristics, and acquire a digital three-dimensional model; Construct a three-dimensional stereogram of the target park, obtain the location characteristics of several buildings and infrastructure in the target park and the dynamic element feature set in the target park by means of GIS, and map the digital three-dimensional models corresponding to several buildings and infrastructure to the three-dimensional stereogram according to the location characteristics; Deploy IoT monitoring nodes on several buildings and infrastructures in the target park according to the dynamic element feature set and the functional characteristics of several buildings and infrastructures in the target park, obtain the data monitoring type of each building and infrastructure by using data retrieval according to the functional characteristics of each building and infrastructure, and the IoT monitoring nodes collect data and mark the monitoring time according to the data monitoring type, and set the monitoring cycle; Obtain data collected by IoT monitoring nodes corresponding to a number of digital three-dimensional models, perform data format preprocessing, generate twin data, and match the twin data with the corresponding digital three-dimensional models to obtain a digital twin model; The process of deploying IoT monitoring nodes on several buildings and infrastructures in the target park according to the dynamic element feature set and the functional characteristics of several buildings and infrastructures in the target park includes: Obtain a dynamic element feature set of each building and infrastructure in the target park, use each type of dynamic element feature in the dynamic element feature set of each building and infrastructure as an evaluation index, wherein each type of dynamic element feature includes pedestrian flow feature data, vehicle flow feature data, environmental feature data, energy feature data, and equipment status feature data, obtain an index weight matrix of the evaluation index according to the functional characteristics of each building and infrastructure, preset a complexity level, and obtain a membership matrix of each building and infrastructure to the complexity level through fuzzy comprehensive evaluation; Obtain the complexity level of each building and infrastructure according to the membership matrix and the indicator weight matrix, obtain the number of IoT monitoring nodes deployed for each building and infrastructure according to the complexity level, and obtain the coverage of the IoT monitoring nodes, and determine the layout position of the IoT monitoring nodes in each digital three-dimensional model through a particle swarm algorithm according to the digital three-dimensional model of each building and infrastructure and the coverage of the IoT monitoring nodes; Step s2: construct a dynamic element feature estimation model based on deep learning, and output an estimated numerical time series sequence corresponding to each type of dynamic element feature of each digital three-dimensional model in the current monitoring period; Step s3: Construct a comprehensive park management platform. The park comprehensive management platform provides users with feedback on the best traffic route based on the user's behavioral feature data and the estimated numerical time series corresponding to the various types of dynamic element features of each digital three-dimensional model, and performs real-time monitoring of the various types of dynamic element features of each digital three-dimensional model in the digital twin model. According to the real-time monitoring results, emergency maintenance operations are performed on the buildings or infrastructure corresponding to the digital three-dimensional models, including: Mapping the user's location features and the target digital three-dimensional model into the digital twin model, extracting all path trajectories between the user's location features and the target digital three-dimensional model according to the digital twin model, extracting the digital three-dimensional model corresponding to each path trajectory, and marking the digital three-dimensional model as the digital three-dimensional model to be compared; Obtain a numerical time series sequence corresponding to the traffic characteristic data collected during the current monitoring period of each digital three-dimensional model to be compared and an estimated numerical time series sequence corresponding to the traffic characteristic data in the remaining time period of the current monitoring period; obtain a full-cycle numerical time series sequence corresponding to the traffic characteristic data of each digital three-dimensional model to be compared based on the numerical time series sequence corresponding to the traffic characteristic data and the estimated numerical time series sequence; obtain a traffic congestion estimation coefficient of each digital three-dimensional model to be compared based on the full-cycle numerical time series sequence; Similarly, the full-cycle numerical time series sequence corresponding to the pedestrian characteristic data of the current monitoring period of each digital three-dimensional model is obtained, and the pedestrian congestion coefficient of each digital three-dimensional model is obtained according to the full-cycle numerical time series sequence, and the traffic impact coefficient between each digital three-dimensional model and each digital three-dimensional model to be compared is obtained. According to the vehicle congestion estimation coefficient, the pedestrian congestion coefficient and the traffic impact coefficient, the traffic priority coefficient of each digital three-dimensional model to be compared is obtained, and the path trajectory corresponding to the digital three-dimensional model to be compared with the largest traffic priority coefficient is screened out and fed back to the user; Mark the environmental feature data, energy feature data and equipment status feature data of each digital three-dimensional model as key feature data, use the estimated numerical time series sequence corresponding to the key feature data of each digital three-dimensional model in the current monitoring period as the judgment data, extract the numerical time series sequence corresponding to the key feature data of each digital three-dimensional model from the data collected by the Internet of Things monitoring node received by each blockchain node, compare the numerical time series sequence corresponding to the key feature data with the corresponding judgment data, obtain the deviation rate information corresponding to the key feature data, preset the deviation rate threshold, compare the deviation rate information with the deviation rate threshold, if the deviation rate information of the key feature data of a certain digital three-dimensional model is greater than the deviation rate threshold, then mark the key feature data as abnormal feature data, generate an abnormal feature alarm signal of the certain digital three-dimensional model and feed it back to the manager of the park comprehensive management platform, and the manager performs emergency maintenance operations on the building or infrastructure corresponding to the certain digital three-dimensional model according to the abnormal feature alarm signal; When abnormal characteristic alarm signals of the plurality of digital three-dimensional models are generated, importance coefficients of the plurality of digital three-dimensional models are obtained, the plurality of digital three-dimensional models are sorted according to the importance coefficients, and an operation sequence of emergency maintenance operations for the buildings or infrastructures corresponding to the plurality of digital three-dimensional models by the management personnel is determined according to the sorting result; The process of obtaining the importance coefficients of several digital three-dimensional models includes: The deviation rate information of the abnormal feature data of each digital three-dimensional model is obtained, and the deviation amplitude between the deviation rate threshold, the pedestrian congestion coefficient and the vehicle congestion coefficient are obtained; and the importance coefficient of each digital three-dimensional model is obtained according to the deviation amplitude, the pedestrian congestion coefficient and the vehicle congestion coefficient.

2. According to claim 1, a digital twin-based park management system is characterized in that: The process of building a dynamic element feature estimation model based on deep learning and outputting the estimated numerical time series corresponding to each type of dynamic element features of each digital three-dimensional model in the current monitoring period includes: Construct a dynamic element feature prediction model based on deep learning, obtain dynamic element feature sets in several historical monitoring cycles of each digital three-dimensional model as training sets and test sets, input the training sets into the dynamic element feature prediction model for training until the loss function training is stable, and save the model parameters, test the dynamic element feature prediction model through the test set until it meets the preset requirements, and output the dynamic element feature prediction model; The dynamic element feature estimation model outputs the estimated numerical time series corresponding to each type of dynamic element feature of each digital three-dimensional model within the current monitoring period.

3. A digital twin-based park management system according to claim 2, characterized in that: The process of building a comprehensive park management platform includes: A comprehensive park management platform is constructed. The comprehensive park management platform is communicatively connected to several blockchain nodes and the API interfaces of each digital three-dimensional model in the digital twin model. The blockchain nodes are linked to each other to form a blockchain network. The blockchain nodes are communicatively connected to user terminals and Internet of Things monitoring nodes. The user terminals are used to collect user behavior feature data and encrypt and upload them to the blockchain nodes. The behavior feature data includes the user's target digital three-dimensional model data and location features.