An automatic optimization method and system for spatial form of urban high-temperature risk area

Abstract

The application discloses an automatic optimization method and system for urban high-temperature risk area space form, first, according to the vector building form data, taking the neighborhood as the spatial analysis unit, a city scale basic space sand table is built; in each spatial analysis unit, temperature and humidity data are obtained, and the heat risk index of the city layout is calculated; then, according to the heat risk index, the urban high-temperature risk area is zoned, and the importance index of the space form element is calculated through a clustering algorithm; finally, each single body in the city is taken as an agent to establish a multi-agent system, the agents are stimulated to take actions capable of reducing the high-temperature risk through the interaction and iteration of the multi-agents until the optimization condition is met, and the optimization result of the urban high-temperature risk area space form is output. 2 The large-scale high-temperature risk identification and form automatic optimization in the above range meet the needs, and the accuracy and scientificity of subsequent form element adjustment are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of urban planning, and particularly to an automatic optimization method and system for spatial form of urban high-temperature risk area. BACKGROUND

[0002] Urban high temperature has significant harm to people's physical and mental health and quality of life, such as causing heat stroke, respiratory diseases and a series of potential risks; for the city itself, it will also lead to the intensification of heat island effect, affecting the sustainable development of urban ecological environment. Spatial form has a close relationship with urban high temperature, and factors such as the greening rate of urban neighborhood and building layout will directly affect the urban heat environment. By optimizing the spatial form, the urban heat island effect can be effectively alleviated, and the adaptability of the city can be improved, thereby improving the livability and sustainable development. Therefore, in research and practice, the optimization of spatial form of urban high-temperature risk area is considered as a regulatory measure that can effectively respond to the heat island effect in urban planning and construction.

[0003] However, in the traditional method of dealing with urban high-temperature risk space, on the one hand, professionals often pay too much attention to the early high-temperature identification, and judge the current situation through measurement or simulation means, but pay insufficient attention to the later spatial form optimization adjustment, and also fail to effectively capture the mechanism and internal relationship between urban spatial form and high-temperature risk. On the other hand, professionals often propose universal climate adaptation strategies for spatial form based on subjective experience, lack scientific data support and systematic analysis methods, and it is difficult to propose practical optimization schemes for large-scale high-temperature risk areas in a short time. Under this background, it is urgent to study the automatic optimization method of spatial form of high-temperature risk area to provide more scientific, systematic and operable urban renewal schemes to cope with the challenges of urban high-temperature risk. SUMMARY

[0004] The purpose of the present application is to provide an automatic optimization method and system for spatial form of urban high-temperature risk area, so as to realize the automatic adjustment of spatial layout of green land, industrial buildings, residential buildings and commercial buildings in urban high-temperature risk area, and promote the high-temperature risk investigation, hidden danger elimination and sustainable healthy development of urban space.

[0005] The technical scheme of the present application is as follows:

[0006] According to the vector building form data, the neighborhood unit enclosed by the city road network is taken as the spatial analysis unit, and a city scale basic spatial sand table is built;

[0007] Temperature and humidity data are obtained in each spatial analysis unit, and the heat risk index of urban layout is calculated;

[0008] The city's high-temperature risk zone is divided according to the heat risk index and displayed on the city-scale basic spatial sand table.

[0009] Based on the spatial morphological elements and thermal risk index of each spatial analysis unit, the importance index of the spatial morphological elements is calculated using a clustering algorithm.

[0010] A multi-agent system is established by treating each individual unit in the city as an agent. The heat risk index is evaluated by fitting a reward function based on the importance index of the spatial morphology elements. Through the interaction and iteration of the multi-agent system, the agents are incentivized to take actions that can reduce the high temperature risk until the optimization conditions are met. The optimization results of the spatial morphology of the high temperature risk area in the city are then output. The actions of the agents include moving out, adding, displacement and scaling. The state of the agents is the spatial morphology element.

[0011] The results of the spatial morphology optimization of the urban high-temperature risk area are displayed on the urban-scale basic spatial sand table.

[0012] Furthermore, dividing the city into high-temperature risk zones based on the aforementioned thermal risk index includes: calculating the average thermal risk index of each spatial analysis unit. and standard deviation σ, The spatial analysis unit is a general risk area. The spatial analysis unit is the key monitoring area. The spatial analysis unit is the area that urgently needs optimization; x i Let be the thermal risk index of the i-th spatial analysis unit.

[0013] Furthermore, the thermal risk index of each spatial analysis unit is the sum of the thermal risk indices of all land use type plots within that spatial analysis unit, where the thermal risk index H of the nth land use type plot is... n for:

[0014]

[0015] Where τ is the temperature coefficient, γ is the humidity coefficient, and t n T represents the duration of the highest temperature for the nth land use type plot. nmax The highest surface temperature, T′ nmax The highest air temperature, RH is t n The average air humidity over a period of time, and V is the wind speed of the nth land use type plot.

[0016] Furthermore, through multi-agent interaction and iteration, agents are incentivized to take actions that reduce the risk of high temperatures until the optimization conditions are met. The output of the optimized spatial morphology of urban high-temperature risk areas includes:

[0017] In the interactive iteration process, it is judged whether the loss function of the multi-agent system converges to an optimal solution, and when it converges to the optimal solution, the urban high-temperature risk area is re-partitioned according to the thermal risk index, and when the number of areas in urgent need of optimization and key monitoring areas is zero, the optimization condition is met, and the optimization result of the spatial form of the urban high-temperature risk area is output.

[0018] Further, the reward function is:

[0019]

[0020] Wherein, R(s, a) represents the total reward obtained by taking action a in state s, gamma is a discount factor, r t represents the immediate reward obtained at time step t.

[0021] Further, according to the spatial form elements and the thermal risk index of each spatial analysis unit, the importance index of the spatial form elements is calculated by a clustering algorithm, including: the spatial form elements as the nodes of the decision tree, calculating the average impurity of each node as the importance index of the node.

[0022] Further, the vector building form data includes a city road network SHP file, a current closed building and a layer or height SHP file, a green land contour or area SHP file.

[0023] Further, the spatial form elements include building height, building density, building orientation, building form, building function mixing degree, building material and building color index on the building level, street height-width ratio, street direction, road area ratio and line fitting rate index on the street level, green rate, water area, building vegetation coverage rate and hard rate index on the green water body level, and sky view factor.

[0024] The automatic optimization system of the urban high-temperature risk area spatial form provided by the application comprises:

[0025] The city scale basic space sand table establishment unit is used for taking the street block unit enclosed by the city road network as a spatial analysis unit, and building a city scale basic space sand table according to the vector building form data.

[0026] The thermal risk index calculation unit is used for acquiring temperature and humidity data in each spatial analysis unit, and calculating the thermal risk index of the city layout.

[0027] The urban high-temperature risk area partition unit is used for partitioning the urban high-temperature risk area according to the thermal risk index, and displaying on the city scale basic space sand table.

[0028] a spatial form element importance calculation unit configured to calculate an importance index of the spatial form element according to the spatial form element and the thermal risk index of each of the spatial analysis units by using a clustering algorithm;

[0029] a city high-temperature risk area spatial form optimization unit configured to establish a multi-agent system by taking each single body in the city as an intelligent agent, fit a reward function according to the importance index of the spatial form element to evaluate the change of the thermal risk index, encourage the intelligent agent to take an action capable of reducing the high-temperature risk through the interaction and iteration of the multi-agent, until an optimization condition is met, and output a city high-temperature risk area spatial form optimization result, wherein the action of the intelligent agent includes moving out, increasing, displacing and scaling, and the state of the intelligent agent is the spatial form element;

[0030] a spatial form optimization result display unit configured to display the city high-temperature risk area spatial form optimization result on the city scale basic space sand table.

[0031] The electronic device provided by the application comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, and the computer program realizes the automatic optimization method of the city high-temperature risk area spatial form when loaded into the processor.

[0032] Advantages: Compared with the prior art, the application has the advantages that the measurement and grading of the high-temperature risk area are realized based on the city space sand table, the correlation between the city spatial form element index and the thermal risk index under the spatial analysis unit is established, the automatic adjustment and holographic display of the spatial form scheme meeting the city high-temperature risk control requirements are realized through the training and interaction iteration of the multi-agent, and the specific advantages include the following:

[0033] 1. The application realizes the measurement and grading of the high-temperature risk area based on the city space sand table by establishing a high-temperature index measurement system, and in addition to the small-scale spatial form optimization, the discrimination result of the city high-temperature risk area zoning condition is expanded to the city scale, so that the large-scale high-temperature risk identification and automatic form optimization in the range of 100km 2 are realized.

[0034] 2. The application establishes the correlation between the city spatial form element index and the thermal risk index under the spatial analysis unit by constructing a thermal risk index spatial form prediction model, breaks through the deficiency that only expert experience is used for judgment in the traditional city planning and design layout, and improves the accuracy and scientificity of the subsequent form element adjustment.

[0035] 3、The present application realizes automatic adjustment and holographic display of spatial form scheme meeting the requirements of urban high-temperature risk control through the training and interaction iteration of four types of multi-agent, green space, industrial building, residential building and commercial building, shortens the work cycle from more than one month to within one day, and improves the standardization of the scheme and the standardized output. BRIEF DESCRIPTION OF DRAWINGS

[0036] Figure 1 The automatic optimization method of the present application is shown in the flowchart.

[0037] Figure 2 The high-temperature risk classification result display diagram of the embodiment of the present application is shown.

[0038] Figure 3 The multi-agent system schematic diagram of the embodiment of the present application is shown.

[0039] Figure 4 The spatial form optimization result display diagram of the embodiment of the present application is shown. DETAILED DESCRIPTION

[0040] The technical solutions of the present application will be further described below with reference to the accompanying drawings.

[0041] As shown in Figure 1 The automatic optimization method of the urban high-temperature risk area spatial form includes the following steps.

[0042] S01, build a city scale basic space sand table, based on Landsat8 remote sensing image and site observation, obtain high-precision land use remote sensing monitoring map to obtain land use type data. Based on GIS geographic information platform and surveying and mapping unmanned aerial vehicle equipped with Beidou satellite navigation system, obtain vector building form data, and determine the spatial distribution of existing green space, industrial building, residential building and commercial building in city space. Take the street unit enclosed by city road network as the spatial analysis unit, label each spatial analysis unit, and match it with land use data, vector building form data in coordinate and elevation, unify data format and input into geographic information system platform, and generate city scale basic space sand table.

[0043] Among them, the vector building form data includes road network SHP file, current closed building and layer (height) SHP file, green space contour (area) SHP file.

[0044] The city scale basic space sand table converts data coordinates by using the projection operation, adjusts city three-dimensional vector data to be unified to the 2000 national geodetic coordinate system, uses the Layer3DToFeatureClass operation, combines building height information, stretches the building to form a three-dimensional model, and matches the spatial analysis unit attribute information with the land use data and vector building shape data in coordinates and elevation, and finally generates the city scale basic space sand table.

[0045] S02, high temperature risk area measurement and classification, setting and establishing a high temperature index measurement system in the spatial analysis unit, using temperature and humidity sensors to obtain spatial temperature data and air humidity data in the measurement area; combining the infrared channel of the satellite-borne sensor to obtain the land surface temperature data, forming a high temperature index database. Calling the corresponding formula to calculate the heat risk index of the existing layout, and constructing a heat risk index dataset.

[0046] The heat risk index of the city space area is standardized and imported into the city scale basic space sand table, the average value and standard deviation of the heat risk index are calculated, and three numerical intervals are constructed. Based on the standard deviation classification, the heat risk index is divided into low risk (index value < average value), medium risk (average value ≤ index value ≤ average value + standard deviation) and high risk (index value > average value + standard deviation) three grades, and the city high temperature disaster risk area is divided into general risk area, key monitoring area and urgent optimization area according to the standard, and the city high temperature risk area partition condition is displayed based on the space sand table.

[0047] The high temperature index measurement system includes measurement point layout, measurement height setting, time resolution setting, spatial resolution setting, wind environment measurement setting and high temperature index calibration. The measurement point layout is to uniformly set temperature and humidity sensors and unmanned aerial vehicles in the measurement area with a minimum measurement unit of 50Mx50M, which meets the full coverage of the target area; the measurement height setting is to set the temperature and humidity measurement height to 2m for temperature and humidity sensors and the wind environment measurement height to 5m for unmanned aerial vehicle flight route. The time resolution setting is to set the time resolution to 30 minutes; the wind environment measurement facility is to use unmanned aerial vehicles equipped with intelligent wind measurement instruments to measure the wind speed and direction values in the target area at a height of 5m. The land surface temperature data is obtained by combining the infrared channel of the satellite-borne sensor.

[0048] The heat risk index, the statistical spatial scale of the heat risk index is a spatial analysis unit, the calculation spatial scale of the heat risk index is a single land use type plot, and the heat risk index of the spatial analysis unit is the sum of the heat risk indexes of the single land use type plots. The heat risk index is calculated by comprehensively considering the factors of air temperature, ground temperature, high temperature duration, air humidity and wind speed, a calculation formula positively correlated with the heat risk index is constructed, and a heat risk index data set is constructed, and the formula is:

[0049]

[0050] Wherein, G is the heat risk index of the i th spatial analysis unit, H n is the heat risk index of the n th land use type plot in the i th spatial analysis unit;

[0051]

[0052] Wherein, t n is the duration of the highest temperature of the n th land use type plot, T nmax is the ground maximum temperature, T' nmax is the air maximum temperature, RH is the average air humidity in t n , and V is the wind speed value of the n th land use type plot.

[0053] τ is the temperature coefficient (when the maximum temperature T is less than 30℃, τ takes the value of 0; when the maximum temperature 30℃≤T≤35℃, τ takes the value of 1; when the maximum temperature T is greater than or equal to 35℃, τ takes the value of 2); γ is the humidity coefficient (when the coincidence degree of the maximum humidity and the maximum temperature time is more than 70%, γ takes the value of 2; when the coincidence degree of the maximum humidity and the maximum temperature is more than 40%, γ takes the value of 1; when the coincidence degree of the maximum humidity and the maximum temperature is 0%, γ takes the value of 0).

[0054] The standard deviation is classified, and the heat risk index data x i is extracted, wherein the data with the heat risk index of 0 does not participate in the calculation of the average value and the standard deviation, the average value of the heat risk index is calculated n is the number of spatial analysis units, i is the spatial analysis unit label, and the standard deviation of the heat risk index is calculated According to the average value and the standard deviation, the heat risk index is divided into low risk , medium risk and high risk three levels, and the urban high temperature disaster risk area is divided into general risk area, key monitoring area and area in urgent need of optimization according to the standard.

[0055] The urban high-temperature risk zoning condition is used to import the spatial analysis unit heat risk index value into a geographic information platform, perform spatial analysis, and set different gray color to express different degrees of urban high-temperature heat risk conditions.

[0056] S03, a heat risk index spatial form prediction model is constructed, the spatial form element index set of different spatial analysis units and the heat risk index calculation results are input into the heat risk index spatial form element prediction relationship database, the extracted feature data is subjected to data cleaning and standardization processing, and is divided into 70% training set and 30% test set data samples, a random forest algorithm is used to train the target city data set, and the root mean square error (RMSE) and the mean absolute error (MAE) are used to evaluate the prediction performance of the model. Based on the average reduction of impurity, the impurity reduction amount for dividing nodes in all decision trees is calculated, and the average value is further calculated to measure the importance of spatial form element features.

[0057] The spatial form element index set includes building level, street level, green space and water body level, and sky visibility angle coefficient. The building level includes building height, building density, building orientation, building form, building function mixing degree, building material and building color index; the street level includes street height-width ratio, street direction, road area ratio and line fitting rate index; the green space and water body level includes green space rate, water body area, building vegetation coverage rate and hard rate index; the sky visibility angle coefficient is calculated by using fisheye photography method with the aid of image processing technology. The spatial form element index set is constructed by comprehensively considering the four levels of indexes. Some indexes are obtained by city control regulation indexes.

[0058] The average reduction of impurity refers to that the feature data extracted from the heat risk index spatial form element prediction relationship database is divided into nodes in the construction of decision tree, and the probability that two samples in the node belong to different categories, i.e. impurity, is calculated. The higher the probability, the lower the purity of the node, and the importance of the node to urban high temperature is ranked later. At the same time, the impurity of the node in other decision trees is calculated, and the average value is obtained. The importance of the feature data corresponding to the node to urban high temperature is obtained, and the importance of each feature data can be obtained according to the importance.

[0059] S04, a spatial form automatic optimization multi-agent system is defined, in a DGX computer workstation, a multi-agent system is defined, each single body of green space, industrial building, residential building and commercial building (including green space or building single body) is defined as each agent, and a multi-agent deep reinforcement learning (RIAL) is used to solve the spatial form optimization problem of urban high-temperature risk area in different neighborhoods of the target city.

[0060] The state space of the environment system at different times is defined as the spatial form element index parameter of the neighborhood space unit, the action space of the four types of agents is defined as "moving out", "increasing", "displacement" and "scaling" of the single body under the requirements of the target city urban design management, the action probability distribution of each agent is output by using the neural network, the reward function is fitted according to the influence factor contribution value of the target city heat risk index spatial form prediction model in step S3 to evaluate the change of heat risk index, the neural network of the agent is trained, the loss function of the neural network is set, the gradient of the loss function is calculated, and the weights and biases of the neural network are updated along the opposite direction of the gradient.

[0061] The action space of the agent, "moving out" means removing the green space or building single body from the current position; "increasing" means copying to generate new green space and building single body at the current position, and the new green space and building single body inherit the neural network parameters of the original agent; "displacement" means moving the planar position coordinates of the green space or building single body; "scaling" means changing the size of the green space or building single body, and the scaled green space or building single body cannot be overlapped with other agents.

[0062] The reward function is fitted according to the influence factor contribution value of the heat risk index spatial form prediction model of the target city different spatial analysis unit, the greater the contribution value of the spatial form element index to the negative feedback of the heat risk index, the greater the characteristic weight of the set reward function, so as to encourage the agent to take actions that can reduce the high temperature risk. The specific formula is:

[0063]

[0064] Wherein, R(s,a) represents the total reward obtained by taking action a in state s, γ is the discount factor, r t represents the immediate reward obtained at time step t. By calculating the cumulative reward, the overall benefit obtained by the agent after taking a specific action can be evaluated.

[0065] S05, multi-agent training and interactive iteration, in the interactive iteration process, it is judged whether the loss function converges to the optimal solution, when the convergence condition is met, the number of each step action occurrence is checked according to the numerical interval standard determined by the average value and standard deviation of the present situation heat risk index in step (3) and the number of the areas in urgent need of optimization and the key monitoring areas, when the number of the above two types of risk areas is 0 at the same time, the iteration is stopped and the urban high temperature disaster risk space optimization scheme is output, otherwise the multi-agent interactive iteration process is executed again until the risk area number condition is met.

[0066] The convergence condition refers to that the loss function tends to be stable after a plurality of continuous iterations and no longer changes significantly. When the loss function converges to below 0.1 and the reward function converges to an optimal solution, it is considered that an optimal solution or a state close to the optimal solution is reached, and iteration is stopped to proceed to the next step.

[0067] S06, the spatial optimization scheme holographic display and output, input the spatial optimization scheme of the urban high temperature risk area to the city space holographic sand table, the user carries the MR mixed reality glasses to carry out the comparative observation of the present situation scheme and the spatial optimization scheme, and the planning and design and management personnel hold the control handle to carry out the detailed adjustment and the check, finally, the heat risk index of different space analysis units of the city before and after optimization and the high temperature risk area are printed to the table and output.

[0068] The following will take Nanjing Jiangbei New District as an example to explain the technical scheme of the application in detail.

[0069] (1) Taking Nanjing Jiangbei New District as the target city, land use data and urban vector building form data are obtained, and a street unit is taken as a spatial analysis unit for labeling, height and coordinate matching are carried out, and a Nanjing Jiangbei New District basic space sand table is generated, specifically including:

[0070] (1.1) Through Landsat8 remote sensing image and station observation, land use remote sensing monitoring map of Nanjing Jiangbei New District is obtained, and is translated into land use type data.

[0071] (1.2) Through GIS geographic information platform and Beidou satellite navigation system, a resolution of 1920*1080 or above surveying and mapping unmanned aerial vehicle is obtained, vector building form data and road data of Nanjing Jiangbei New District are obtained, including road network SHP file, present situation closed building and layer (height) SHP file, green land contour (area) SHP file, and the existing green land, industrial building, residential building, commercial building and road of Nanjing Jiangbei New District are determined,

[0072] (1.3) The road network SHP file, the present situation closed building and the layer (height) SHP file, the green land contour (area) SHP file of the present situation Nanjing Jiangbei New District are imported into the geographic information system software, the building closed surface and the building layer point are spatially associated, and each building is attached with its layer (height) information; and according to the street unit divided by the road network as a spatial analysis unit, attribute labeling processing is carried out in the geographic information system software, and it is divided into 46 small scale space units.

[0073] (1.4) Projection operation is used to convert the data coordinates, adjust the city three-dimensional vector data to be unified to the 2000 national geodetic coordinate system; Layer3DToFeatureClass operation is used, combined with building height information, to stretch the building to form a three-dimensional model; and the attribute information of the spatial analysis unit is matched with the land use data and vector building shape data in coordinates, and finally the three-dimensional base model of Nanjing Jiangbei New Area is generated.

[0074] (2) Select each spatial analysis unit in Nanjing Jiangbei New Area, set up a high temperature index measurement system in it, obtain the high temperature index and calculate the heat risk index, realize the measurement and classification of high temperature risk area in Nanjing Jiangbei New Area, which specifically includes:

[0075] (2.1) Set up a high temperature index measurement system in each spatial analysis unit, set up temperature and humidity sensors and unmanned aerial vehicles with 30 minutes as the time resolution and 50Mx50M as the minimum measurement unit, set up temperature and humidity sensors with 2m as the temperature and humidity measurement height, and arrange unmanned aerial vehicle flight routes with 5m as the wind environment measurement height, adopt unmanned aerial vehicles equipped with intelligent wind measuring instruments to measure the wind speed and direction value in the target area with 5m as the reference to meet the full coverage of Nanjing Jiangbei New Area; obtain the spatial temperature data and air humidity data in the spatial analysis unit,

[0076] (2.2) Use the thermal infrared sensor of Landsat satellite to obtain the infrared channel data of satellite-borne sensor, and perform atmospheric correction and inversion algorithm to obtain accurate ground temperature. Combine the air temperature and humidity in step (1) to construct the high temperature index database of Nanjing Jiangbei New Area.

[0077] (2.3) Calculate the heat risk index of the spatial analysis unit in the high temperature index database in step (2.2) to construct the heat risk index set; according to the standard deviation classification, the heat risk index is divided into three levels of low risk, medium risk and high risk, which correspond to the general risk area, key monitoring area and urgent optimization area of Nanjing Jiangbei New Area respectively, and are imported into the basic space sand table of Nanjing Jiangbei New Area for heat point analysis, the degree of urban high temperature risk is expressed through different depths of red, and the status quo map of high temperature risk area of Nanjing Jiangbei New Area is formed, the general risk area H n (2.8≤H n ≤6.5), the key monitoring area H n (6.5≤H n ≤10.5), and the urgent optimization area H n (H n ≥10.5). As shown in the high temperature risk area classification result display diagram of the embodiment. Figure 2

[0078] ​(3) Constructing the spatial form prediction model of the heat risk index of the Jiangbei New Area of Nanjing, specifically including:

[0079] (3.1) Obtaining the spatial form elements of different spatial analysis units in the Jiangbei New Area of Nanjing, including the building level, street level, green space and water body level, and sky visibility angle coefficient. Through remote sensing images, building height, building density, building orientation, building form index, green space rate, water area, building vegetation coverage rate, and hard rate index are obtained; through street view photo collection and analysis technology, building material and building color index are obtained; through vector building data and road network data, street height-width ratio, street direction, road area ratio, and line rate index are obtained; sky visibility angle coefficient is calculated by using fisheye photography method with image processing technology. The spatial form element index set of each spatial analysis unit in the Jiangbei New Area of Nanjing is constructed by integrating the four levels of indexes.

[0080] (3.2) The spatial form element index set of different spatial analysis units and the heat risk index calculation results are input into the heat risk index spatial form element prediction relationship database, the extracted feature data is subjected to data cleaning, standardization processing, and divided into 70% training set and 30% test set data samples.

[0081] (3.3) The training set in step two is trained by random forest algorithm on the target city data set, and the root mean square error (RMSE) and mean absolute error (MAE) are used to evaluate the prediction performance of the heat risk index spatial form prediction model of the Jiangbei New Area of Nanjing.

[0082] (3.4) The feature data extracted from the heat risk index spatial form element prediction relationship database, the importance of the feature data to the urban high temperature, and the importance ranking of each feature data can be obtained to measure the importance of the spatial form element features.

[0083] (4) As shown in Figure 3 , the spatial form automatic optimization multi-agent system of the Jiangbei New Area of Nanjing is constructed, in the DGX computer workstation, the multi-agent system is defined, each single body of green space, industrial building, residential building and commercial building in the Jiangbei New Area of Nanjing is defined as each agent, and the multi-agent deep reinforcement learning (RIAL) is used to solve the spatial form optimization problem of urban high temperature risk area of different spatial analysis units in the Jiangbei New Area of Nanjing, specifically including:

[0084] (4.1) Defining the state space of the environment system of green space, industrial building, commercial building and residential building at different times as the spatial form element index parameters of the spatial analysis unit;

[0085] (4.2) Define the action space of the four types of agents as "moving out", "adding", "displacement" and "scaling" of green space or building monomer under the requirements of Nanjing Jiangbei New Area Urban Design Management Regulations, and use neural network to output the action probability distribution of each agent,

[0086] (4.3) According to the influence factor contribution value fitting of the spatial form prediction model of thermal risk index of different spatial analysis units in Nanjing Jiangbei New Area, the reward function is obtained. The greater the contribution value of the spatial form element index to the negative feedback of the thermal risk index, the greater the characteristic weight of the set reward function, so as to encourage the agent to take actions that can reduce the high temperature risk.

[0087] (4.4) According to the influence factor contribution value fitting of the spatial form prediction model of thermal risk index of different spatial analysis units in Nanjing Jiangbei New Area, the reward function is obtained. The greater the contribution value of the spatial form element index to the negative feedback of the thermal risk index, the greater the characteristic weight of the set reward function, so as to encourage the agent to take actions that can reduce the high temperature risk.

[0088] (5) Train and interact the multi-agent system of Nanjing Jiangbei New Area, and output the high temperature disaster risk space optimization scheme of Nanjing Jiangbei New Area when the convergence condition is met and the number of urgent optimization areas and key monitoring areas is zero. Specifically, it includes:

[0089] (5.1) In the interactive iteration process, judge whether the loss function converges to the optimal solution. When the loss function converges to 0.1 or less, and the reward function converges to the optimal solution, it is considered to reach the optimal solution or the state close to the optimal solution, and the convergence condition is met.

[0090] (5.2) Check the number of urgent optimization areas and key monitoring areas after each action according to the numerical interval determined by the average value and standard deviation of the current thermal risk index in step (3). When the number of the above two types of risk areas is 0, stop iteration and output the high temperature disaster risk space optimization scheme of Nanjing Jiangbei New Area. Otherwise, re-execute the multi-agent interactive iteration process until the risk area number condition is met.

[0091] (6) Perform holographic display and output of Nanjing Jiangbei New Area spatial optimization scheme, including:

[0092] As shown in Figure 4 , input the high temperature risk area space optimization scheme of Nanjing Jiangbei New Area into the Nanjing Jiangbei New Area spatial holographic sand table, and the user carries MR mixed reality glasses to compare and observe the current scheme and the spatial optimization scheme, and the planning and management personnel hold the control handle to make detailed adjustment and check. Finally, the thermal risk index and high temperature risk area of different spatial analysis units of the city before and after optimization are printed to the table for output.

[0093] The automatic optimization system of the spatial form of the urban high-temperature risk area comprises:

[0094] The city-scale basic space sand table establishment unit is used for establishing a city-scale basic space sand table according to the vector building form data and the street block unit enclosed by the city road network.

[0095] The heat risk index calculation unit is used for obtaining temperature and humidity data in each space analysis unit and calculating the heat risk index of the city layout.

[0096] The city high-temperature risk area partition unit is used for partitioning the city high-temperature risk area according to the heat risk index and displaying the city high-temperature risk area on the city-scale basic space sand table.

[0097] The spatial form element importance calculation unit is used for calculating the importance index of the spatial form element by a clustering algorithm according to the spatial form element and the heat risk index of each space analysis unit.

[0098] The city high-temperature risk area spatial form optimization unit is used for establishing a multi-agent system by taking each single body in the city as an agent, fitting a reward function according to the importance index of the spatial form element to evaluate the heat risk index change, and through the interaction and iteration of the multi-agent, encouraging the agent to take actions capable of reducing the high-temperature risk until the optimization condition is met, and outputting the city high-temperature risk area spatial form optimization result, wherein the actions of the agent include moving out, increasing, displacing and scaling.

[0099] The spatial form optimization result display unit is used for displaying the city high-temperature risk area spatial form optimization result on the city-scale basic space sand table.

[0100] The electronic device comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, and the computer program realizes the automatic optimization method of the spatial form of the urban high-temperature risk area when loaded into the processor.

[0101] The computer-readable storage medium can include RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage or other magnetic storage devices, flash memory or any other medium that can be used to store desired program codes in the form of instructions or data structures and can be accessed by a computer.

[0102] The processor is used for executing the computer program stored in the memory to realize each step in the method involved in the above embodiments.

Claims

1. An automatic optimization method for urban high-temperature risk area spatial form, characterized in that, Comprising the following steps: According to the vector building shape data, the neighborhood unit enclosed by the city road network is taken as the spatial analysis unit to build a city scale basic spatial sand table; In each of the spatial analysis unit, temperature and humidity data are obtained, and the heat risk index of the city layout is calculated; The thermal risk index of each spatial analysis unit is the sum of the thermal risk indices of all land use type plots within the spatial analysis unit, and the thermal risk index H of the nth land use type plot is: n Hn=∑i=1nXi·Ri Where τ is the temperature coefficient, γ is the humidity coefficient, and t n T represents the duration of the highest temperature for the nth land use type plot. n max The highest surface temperature, T′ n max The highest air temperature, RH is t n The average air humidity over a period of time, and V is the wind speed of the nth land use type plot; According to the heat risk index, the high temperature risk area of the city is divided, and the city scale basic spatial sand table is displayed; The spatial form element index set of different spatial analysis units and the heat risk index calculation results are input into the heat risk index spatial form element prediction relationship database; and the training set and the test set are divided, the random forest algorithm is used to train the target city data set; the impurity reduction amount for dividing nodes in all decision trees is calculated based on the average reduction of impurity, and the average value is further calculated as the importance index of spatial form element; Each single body in the city is taken as an intelligent agent to build a multi-agent system, a reward function is fitted according to the importance index of the spatial form element to evaluate the change of the heat risk index, the intelligent agent is encouraged to take actions that can reduce the high temperature risk through the interaction and iteration of the multi-agent, until the optimization condition is met, the spatial form optimization result of the high temperature risk area of the city is output, the actions of the intelligent agent include moving out, increasing, displacement and scaling, and the state of the intelligent agent is the spatial form element; the spatial form element index set includes building level, street level, green space and water body level, and sky visibility angle coefficient; The spatial form optimization result of the high temperature risk area of the city is displayed on the city scale basic spatial sand table.

2. The method for automatic optimization of urban high temperature risk zone spatial morphology according to claim 1, characterized in that, According to the heat risk index, the high temperature risk area of the city is divided, including: calculating the average value of the thermal risk index of each spatial analysis unit and the standard deviation σ, spatial analysis units with a thermal risk index of 0.5 < T < 1 are general risk areas, spatial analysis units with a thermal risk index of 1 < T < 1.5 are areas of special monitoring, spatial analysis units with a thermal risk index of 1.5 < T < 2 are areas in urgent need of optimization; x i T is the thermal risk index of the i-th spatial analysis unit.

3. The method for automatic optimization of urban high temperature risk zone spatial morphology according to claim 2, characterized in that, Through the interaction and iteration of the multi-agent, the intelligent agent is encouraged to take actions that can reduce the high temperature risk, until the optimization condition is met, the spatial form optimization result of the high temperature risk area of the city is output, including: In the process of interaction and iteration, it is judged whether the loss function of the multi-agent system converges to the optimal solution, when it converges to the optimal solution, the high temperature risk area of the city is redivided according to the heat risk index, when the number of areas in urgent need of optimization and key monitoring areas is zero, the optimization condition is met, and the spatial form optimization result of the high temperature risk area of the city is output.

4. The method for automatic optimization of urban high temperature risk zone spatial morphology according to claim 1, characterized in that, The reward function is: where R(s, a) represents the total reward obtained by taking action a in state s, g is a discount factor, and r t represents the immediate reward obtained at time step t. 5.The method of claim 1, wherein, Based on the average reduction of impurity, the impurity reduction amount for dividing nodes in all decision trees is calculated, and the average value is further calculated as the importance index of spatial form element, including: the spatial form element is taken as the node of the decision tree, the average reduction of impurity of each node is calculated as the importance index of the node.

6. The method for automatic optimization of urban high temperature risk zone spatial morphology according to claim 1, characterized in that, The vector building shape data includes the road network SHP file of the city, the current closed building and the number of layers or height SHP file, the green space contour or area SHP file.

7. The method for automatic optimization of urban high temperature risk zone spatial morphology according to claim 1, characterized in that, The spatial form element includes building height, building density, building orientation, building shape, building function mixing degree, building material and building color index at building level, street high-width ratio, street direction, road area ratio and line fitting rate index at street level, green space rate, water area, building vegetation coverage and hard rate index at green space and water body level, and sky visibility angle coefficient.

8. An automatic optimization system for urban high-temperature risk area spatial form, characterized in that, Including: The urban scale basic space sand table establishment unit is configured to take a street block unit enclosed by a city road network formed according to vector building shape data as a spatial analysis unit, and build an urban scale basic space sand table; The heat risk index calculation unit is configured to obtain temperature and humidity data in each spatial analysis unit, and calculate a heat risk index of the urban layout; The thermal risk index of each spatial analysis unit is the sum of the thermal risk indices of all land use type plots within the spatial analysis unit, and the thermal risk index H of the nth land use type plot is: n Hn=∑i=1nXi·Ri where τ is the temperature coefficient, γ is the humidity coefficient, t n is the duration of the maximum temperature for the nth land use type plot, T n max is the surface maximum temperature, T′ n max is the air maximum temperature, RH is the average air humidity over t n is the wind speed for the nth land use type plot; The urban high-temperature risk zoning unit is configured to zone the urban high-temperature risk area according to the heat risk index, and display the zoning on the urban scale basic space sand table; The spatial form element importance calculation unit is configured to input spatial form element index sets of different spatial analysis units and heat risk index calculation results into a heat risk index spatial form element prediction relationship database, divide a training set and a test set, train a target city data set by using a random forest algorithm, calculate an impurity reduction amount for dividing nodes in all decision trees based on average reduction of impurity, and further calculate an average value as an importance index of a spatial form element; The urban high-temperature risk area spatial form optimization unit is configured to take each monomer in the city as an agent to establish a multi-agent system, fit a reward function according to the importance index of the spatial form element to evaluate a heat risk index change, encourage the agent to take an action capable of reducing the high-temperature risk through interaction and iteration of the multi-agent, until an optimization condition is met, and output an urban high-temperature risk area spatial form optimization result, wherein the action of the agent includes moving out, increasing, displacement, and scaling, and the state of the agent is a spatial form element; the spatial form element index set includes a building level, a street level, a green space and water body level, and a sky visibility angle coefficient; The spatial form optimization result display unit is configured to display the urban high-temperature risk area spatial form optimization result on the urban scale basic space sand table.

9. An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The computer program, when loaded into a processor, implements the automatic optimization method of the urban high-temperature risk area spatial form according to any one of claims 1-7.

Images