Cold region city walking space layout generation design method based on dynamic thermal comfort prediction

Abstract

The invention provides a cold region city walking space layout generation design method based on dynamic thermal comfort prediction. According to the invention, based on big data and IoT data, a cold region walking space way-finding agent model is constructed to obtain pedestrian trajectories of all travel periods in cold region walking spaces with different layouts; based on the pedestrian physiological index data, constructing a mapping relationship between the pedestrian thermal sensation and the thermal environment of the walking space in the cold region so as to obtain pedestrian thermal sensation conditions under different thermal environment changes; the cold region walking space layout is optimized based on pedestrian heat feeling prediction results in typical pedestrian routes in all time periods, and a cold region walking space layout design decision model guided by the decision preference of a designer is obtained based on a machine learning model and decision feedback of the designer on a visual scheme. Compared with the prior art, the method provided by the invention has the advantages of higher simulation degree and pertinence and high efficiency in allusion to pedestrian thermal feeling prediction in the walking space in the cold region.

Description

Technical Field

[0001] The invention belongs to the technical field of thermal comfort prediction in cold-region cities, and in particular relates to a generation and design method for the layout of walking spaces in cold-region cities based on dynamic thermal comfort prediction. Background Art

[0003] With the advancement of relevant research, the concept of thermal comfort is constantly evolving to better describe the thermal feelings of users in real thermal environments. Initial studies believed that thermal neutrality is thermal comfort, so thermal comfort at this stage is used to describe the state in which users feel neither cold nor hot. This concept focuses on the direct stimulation of the thermal environment on the sensory system, while ignoring the subjective influence of psychological factors on thermal sensation. Thermal comfort under this concept is steady-state discrete. However, the research of de Dear et al. shows that thermal sensation is not only related to the current thermal state, but also affected by thermal history, thermal expectations, thermal adaptation level, etc. Therefore, thermal sensation is actually dynamic, continuous and personalized, and is affected by thermal alliesthesia. In this case, thermal neutrality does not necessarily bring the most positive thermal experience. Their research found that a short-term deviation from thermal neutrality within a certain range may bring a pleasant feeling. Therefore, thermal pleasure is included in the consideration of thermal comfort. On this basis, the dynamic thermal comfort theory considering the influence of continuous environment is proposed to describe the subjective thermal feelings of users in non-steady-state environment.

[0004] At present, the prediction method of thermal comfort is still mainly based on the steady-state model, and its prediction results only consider the degree of deviation from thermal neutrality, not the thermal pleasure. The thermal comfort models currently used include single-node thermal comfort model, double-node thermal comfort model and multi-node thermal comfort model. The single-node thermal comfort model regards the human body as a single node to represent the temperature perception of the whole body. This model usually evaluates thermal comfort based on the average temperature of the whole body and the sweat evaporation rate. It mainly includes Fanger's PMV and PPD models, and the widely used thermal comfort standards ASHRAE 55-2016 and ISO 7730 are based on this. It predicts people's average vote on the ambient temperature based on ambient temperature, humidity, wind speed, radiant temperature, clothing index and activity level. This model is a steady-state model that does not consider thermal history, nor does it consider the changes in temperature and comfort over time in the thermal equilibrium state of the human body. Double-node thermal comfort model The double-node thermal comfort model divides the human body into two nodes: core and skin. Compared with the single-node model, this model can more accurately simulate the heat conduction and heat regulation process of the human body, such as heat exchange between the body and the skin. It mainly includes the SET model of Gagge et al., which uses physiological equivalent temperature for thermal comfort prediction and evaluation. Although the model takes into account the heat exchange of the human body, it is generally still used under the steady-state assumption. At the same time, the model does not consider the influence of thermal history. The multi-node thermal comfort model further divides the human body into multiple nodes such as the head, trunk, and limbs to simulate the temperature and thermal perception of different parts. It can simulate the temperature distribution and thermal regulation process of various parts of the human body more finely. It mainly includes the UTCI model, which converts the heat stress level under different environmental conditions into an equivalent temperature index through complex physiological and thermal regulation models. Although this model can simulate non-steady-state environments, it also measures the thermal comfort level according to the degree of deviation from thermal neutrality and does not consider thermal pleasure. Therefore, there is a problem of insufficient consideration of the thermal perception of users' psychological factors. In summary, the thermal comfort model currently used does not take into account the impact of continuous dynamic environment and the subjective feelings of users, and has defects in the accuracy of user thermal comfort prediction in real environments.

[0005] The thermal environment of cold-region walking spaces varies greatly in time and space. Therefore, the temporal and spatial changes experienced by pedestrians during their walking will bring about changes in the thermal environment, which will have a dynamic impact on pedestrian thermal sensation, causing fluctuations in pedestrian skin temperature, thus bringing positive or negative thermal sensations. However, the current prediction methods do not fully consider its impact when evaluating pedestrian thermal comfort. Therefore, it is necessary to develop a dynamic thermal comfort prediction method for pedestrians in cold-region walking spaces based on thermal synesthesia, and to design the overall layout of cold-region walking spaces based on this. Summary of the invention

[0006] The purpose of the present invention is to solve the problem that the existing pedestrian thermal comfort prediction methods do not take environmental changes and pedestrians' subjective feelings into consideration sufficiently and the prediction accuracy is not high, to improve the prediction accuracy of pedestrian dynamic thermal comfort in cold-region urban walking spaces, to break through the prediction efficiency bottleneck of existing prediction methods, to strengthen its supporting role in the preliminary design of cold-region urban walking spaces, to construct a cold-region walking space layout generation and design method based on dynamic thermal comfort prediction, to solve the problem that the existing cold-region urban and rural walking space design decisions do not take dynamic thermal comfort into consideration sufficiently, and to propose a cold-region urban walking space layout generation and design method based on dynamic thermal comfort prediction.

[0007] The present invention is implemented by the following technical scheme. The present invention proposes a method for generating and designing the layout of walking space in cold-region cities based on dynamic thermal comfort prediction. The method comprises the following steps:

[0008] S1. Construction of a pathfinding agent model for cold-region walking space;

[0009] The step S1 is specifically as follows:

[0010] S1.1: Pedestrian group trajectory data collection at different travel times in typical cold-region walking spaces based on IoT;

[0011] S1.2: Construction of gray box model for path finding in cold-region walking space at different travel periods;

[0012] S1.3: Extract accurate pedestrian trajectories at different travel times in typical cold-region walking spaces based on drone image data;

[0013] S1.4: Calibration of the cold-region walking space pathfinding agent model based on typical spatial IoT perception data;

[0014] S2, construction of thermal environment data of cold-region walking space and pedestrian dynamic thermal comfort mapping;

[0015] The step S2 is specifically as follows:

[0016] S2.1: Pedestrian thermal sensation acquisition in typical cold-weather walking spaces based on ecological instantaneous assessment;

[0017] S2.2: Construction of thermal sensation model of pedestrians in cold-region walking spaces under hot winter and summer conditions;

[0018] S2.3: Calibration of thermal sensation model of pedestrians in winter and summer in cold-region walking space based on reinforcement learning;

[0019] S3, Generative design of cold-region walking space layout driven by dynamic thermal comfort data;

[0020] The step S3 is specifically as follows:

[0021] S3.1: Obtaining the layout plan of cold-region walking space driven by generation rules;

[0022] S3.2: Optimal design of cold-region walking space layout driven by dynamic thermal comfort data;

[0023] S3.3: Decision support for cold-region walking space layout design based on human-computer interaction.

[0024] Furthermore, the step S2.1 is specifically as follows:

[0025] S2.1.1: Pedestrian position acquisition in cold-weather walking spaces;

[0026] S2.1.2: Real-time thermal sensation collection of pedestrians in cold-weather walking spaces under changing pedestrian positions;

[0027] S2.1.3: Real-time collection of thermal environment data of typical cold-region walking spaces.

[0028] Furthermore, in step S2.1.2, real-time skin temperature, heart rate and EDA skin electrical signal data of pedestrians in cold-region walking spaces are collected based on wearable devices as their positions change, and subjective evaluation data of pedestrians in cold-region walking spaces as their positions change are obtained. Based on this, a mapping relationship between the physiological indicator data and thermal sensation of pedestrians in cold-region walking spaces is constructed, thereby obtaining real-time thermal sensation information of pedestrians in cold-region walking spaces as their positions change.

[0029] Furthermore, in step S3.1, the cold-region block layout model data is obtained and voxelized, and the layout and building features are extracted using a three-dimensional convolutional neural network; based on the scale characteristics of the cold-region block, the predefined cold-region pedestrian space is divided into three-dimensional units, and based on the design conditions of the cold-region pedestrian space and the extracted features, constraints are set on the three-dimensional matrix so that its generation scheme meets the geometric and topological requirements; under the constraints, the three-dimensional units are allocated based on the multi-agent system to realize the generation of the cold-region pedestrian space layout, thereby obtaining the optimized prototype of the cold-region pedestrian space layout.

[0030] Furthermore, the step S3.2 is specifically as follows:

[0031] S3.2.1: Construction of optimization model for cold-region walking space layout;

[0032] S3.2.2: Dynamic thermal comfort of pedestrians during travel periods under cold-region pedestrian space layout;

[0033] S3.2.3: Optimization of cold-region walking space layout under the guidance of dynamic thermal comfort.

[0034] Furthermore, the step S3.2.2 is specifically as follows:

[0035] S3.2.2.1: Construction of typical thermal environment prediction model for each travel period in winter and summer under the spatial layout of cold-region pedestrian space;

[0036] S3.2.2.2: Thermal environment acquisition of pedestrians along typical routes during winter and summer travel periods in cold-region pedestrian spaces;

[0037] S3.2.2.3: Pedestrian thermal sensation acquisition during typical travel periods on a cold-region pedestrian space layout.

[0038] Furthermore, in step S3.2.2.1, the thermal environment images of each travel period in winter and summer under the cold-region walking space layout are obtained, and the thermal environment images of each travel period are clustered respectively, and then a mapping relationship between the cold-region walking space layout and the clustered images of each travel period in winter and summer is constructed.

[0039] Furthermore, in step S3.3, after obtaining the optimal layout plan of cold-region walking space under the guidance of dynamic thermal comfort, a random forest model is used to construct a cold-region walking space layout decision model based on the dynamic thermal comfort data of the cold-region walking space layout plan and the designer's morphological evaluation data of the layout plan, and a preliminary decision plan is obtained; VR equipment and environmental control equipment are used to provide users with a simulation of the walking experience of the plan, obtain their evaluation of the plan, and use it as feedback to the decision model, and further adjust the cold-region walking space layout decision model, thereby providing designers with efficient and comprehensive decision support.

[0040] The present invention also proposes an electronic device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of a method for generating a design method for a cold-region urban walking space layout based on dynamic thermal comfort prediction are implemented.

[0041] The present invention also proposes a computer-readable storage medium for storing computer instructions, which, when executed by a processor, implement the steps of a method for generating a design for a cold-region urban pedestrian space layout based on dynamic thermal comfort prediction.

[0042] Beneficial effects of the present invention:

[0043] The present invention provides a cold-region walking space layout generation design method based on dynamic thermal comfort prediction, and provides designers with efficient design decision support for cold-region walking space layout based on pedestrian thermal perception prediction results that are more in line with the actual situation. The present invention constructs a cold-region walking space pathfinding agent model based on big data and IoT data to obtain pedestrian trajectories in different travel periods under cold-region walking spaces with different layouts; based on pedestrian physiological index data, a mapping relationship between pedestrian thermal perception and the thermal environment of cold-region walking spaces is constructed to obtain pedestrian thermal perception under different thermal environment changes; based on the pedestrian thermal perception prediction results of pedestrians on typical routes of pedestrians in each period, the cold-region walking space layout is optimized, and based on the machine learning model and the designer's decision feedback on the visualization scheme, a cold-region walking space layout design decision model guided by the designer's decision preference is obtained. For the prediction of pedestrian thermal perception in cold-region walking spaces, the present invention is more simulated and targeted than existing technologies, and is also efficient. For the decision-making of cold-region walking space layout, the present invention can better take into account the designer's preferences on the basis of ensuring the stability of the decision-making than existing technologies, so it is more conducive to application in actual engineering scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0045] Figure 1 The present invention is a flow chart of a method for generating a design for a cold-region urban walking space layout based on dynamic thermal comfort prediction. DETAILED DESCRIPTION

[0046] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0047] Combination Figure 1 The present invention proposes a method for generating and designing the layout of walking space in cold-region cities based on dynamic thermal comfort prediction, and the method comprises the following steps:

[0048] S1. Construction of a pathfinding agent model for cold-region walking space;

[0049] The step S1 is specifically as follows:

[0050] S1.1: Pedestrian group trajectory data collection at each travel period in a typical cold-region walking space based on IoT; In step S1.1, select some typical cold-region spaces, deploy multiple Wi-Fi and Bluetooth access points or beacons in the target area, record the MAC address and signal strength of each device at different travel periods, and use triangulation and data fusion technology to accurately estimate the location and trajectory of each device, so as to achieve pedestrian positioning and trajectory tracking at each travel period in the typical cold-region walking space;

[0051] S1.2: Construction of gray-box pathfinding model for each travel period in cold-region walking space; In step S1.2, the pedestrian trajectory information is converted into pedestrian path graph structure data, and the pedestrian path selection model is constructed based on this application graph convolutional neural network model to achieve efficient prediction of pedestrian path node selection; On this basis, the social force model, multi-agent system and particle filter algorithm are integrated to construct a pedestrian pathfinding gray-box model, and the model parameters are optimized and adjusted based on the pedestrian trajectory data, and the mathematical model and data-driven method are integrated to achieve the prediction of the specific trajectory between pedestrian nodes;

[0052] S1.3: Extract accurate pedestrian trajectories in each travel period of a typical cold-region walking space based on drone image data; In step S1.3, multi-angle image data of a typical cold-region walking space in each time period is obtained based on drone low-altitude photography, a high-precision three-dimensional information model of a cold-region block is constructed based on the multi-angle images, and the positions of pedestrians in typical frames are marked in the high-precision three-dimensional information model. At the same time, the YOLO algorithm is used to perform target detection in continuous drone image frames, and the CSRT algorithm is used to track each detected pedestrian separately to obtain the position change of each pedestrian, and the pedestrian position in the typical frame and the pedestrian position change information are integrated to reconstruct the pedestrian trajectory with high precision;

[0053] S1.4: Calibration of the cold-region walking space pathfinding agent model based on typical spatial IoT perception data;

[0054] S2, construction of thermal environment data of cold-region walking space and pedestrian dynamic thermal comfort mapping;

[0055] The step S2 is specifically as follows:

[0056] S2.1: Acquisition of thermal sensation of pedestrians in a typical cold-region walking space based on ecological instantaneous assessment; the step S2.1 is specifically as follows:

[0057] S2.1.1: Pedestrian position acquisition in cold-weather walking spaces;

[0058] S2.1.2: Real-time thermal sensation collection of pedestrians in cold-region walking spaces under changing positions of pedestrians; In step S2.1.2, real-time skin temperature, heart rate and EDA skin electrical signal data of pedestrians in cold-region walking spaces under changing positions of pedestrians are collected based on wearable devices, and subjective evaluation data of pedestrians in cold-region walking spaces under changing positions of pedestrians are obtained, based on which a mapping relationship between physiological index data of pedestrians in cold-region walking spaces and thermal sensation is constructed, thereby obtaining real-time thermal sensation information of pedestrians in cold-region walking spaces under changing positions of pedestrians;

[0059] S2.1.3: Real-time collection of thermal environment data of typical cold-region walking spaces;

[0060] S2.2: Construction of thermal sensation model of pedestrians in cold-region walking spaces under hot winter and summer conditions;

[0061] S2.3: Calibration of thermal sensation model of pedestrians in winter and summer in cold-region walking space based on reinforcement learning;

[0062] S3, Generative design of cold-region walking space layout driven by dynamic thermal comfort data;

[0063] The step S3 is specifically as follows:

[0064] S3.1: Acquisition of cold-region pedestrian space layout scheme driven by generation rules; In step S3.1, the cold-region block layout model data is obtained and voxelized, and the layout and building features are extracted using a three-dimensional convolutional neural network; Based on the scale characteristics of the cold-region block, the predefined cold-region pedestrian space is divided into three-dimensional units, and constraints are set on the three-dimensional matrix based on the cold-region pedestrian space design conditions and the extracted features so that its generation scheme meets the geometric and topological requirements. Under the constraints, the three-dimensional units are allocated based on the multi-agent system to realize the generation of the cold-region pedestrian space layout, thereby obtaining the cold-region pedestrian space layout optimization prototype;

[0065] S3.2: Optimal design of cold-region walking space layout driven by dynamic thermal comfort data; the step S3.2 is specifically as follows:

[0066] S3.2.1: Construction of optimization model for cold-region walking space layout;

[0067] S3.2.2: Acquisition of dynamic thermal comfort of pedestrians during travel periods under cold-region walking space layout; the step S3.2.2 is specifically as follows:

[0068] S3.2.2.1: Construction of a typical thermal environment prediction model for each travel period in winter and summer under the cold-region pedestrian space layout; In step S3.2.2.1, the thermal environment images for each travel period in winter and summer under the cold-region pedestrian space layout are obtained, and the thermal environment images for each travel period are clustered respectively, and then a mapping relationship between the cold-region pedestrian space layout and the clustered images for each travel period in winter and summer is constructed;

[0069] S3.2.2.2: Thermal environment acquisition of pedestrians along typical routes during winter and summer travel periods in cold-region pedestrian spaces;

[0070] S3.2.2.3: Pedestrian thermal sensation acquisition during typical travel periods on a cold-region walking space layout;

[0071] S3.2.3: Optimization of cold-region walking space layout under the guidance of dynamic thermal comfort;

[0072] S3.3: Decision support for cold-region walking space layout design based on human-computer interaction. In step S3.3, after obtaining the optimal cold-region walking space layout plan under the guidance of dynamic thermal comfort, the cold-region walking space layout decision model is constructed using the random forest model based on the dynamic thermal comfort data of the cold-region walking space layout plan and the designer's morphological evaluation data of the layout plan to obtain a preliminary decision plan; VR equipment and environmental control equipment are used to provide users with a simulated walking experience of the plan, obtain their evaluation of the plan, and use it as feedback to the decision model to further adjust the cold-region walking space layout decision model, thereby providing designers with efficient and comprehensive decision support.

[0073] Example

[0074] according to Figure 1 As shown, the present invention provides a cold-region walking space layout generation design method based on dynamic thermal comfort prediction, comprising the following steps:

[0075] Step 1: Construction of cold-region walking space pathfinding agent model;

[0076] The step 1 is specifically as follows:

[0077] Step 1.1: Collect pedestrian group trajectory data at different travel periods in typical cold-region walking spaces based on IoT;

[0078] Step 1.2: Construction of gray box model for path finding in cold-region walking space at different travel periods;

[0079] Step 1.3: Extract accurate pedestrian trajectories at different travel times in typical cold-region walking spaces based on drone image data;

[0080] Step 1.4: Calibrate the pathfinding agent model of cold-region pedestrian space based on accurate pedestrian trajectory information. Use the accurate pedestrian trajectory data of each travel period in typical cold-region pedestrian space as the fine-tuning dataset to further train the pathfinding gray box model of cold-region pedestrian space in each travel period, and gradually adjust and optimize the model parameters.

[0081] The step 1.1 is specifically as follows:

[0082] Select a representative cold-region walking space as the research area, and deploy multiple Wi-Fi and Bluetooth access points or beacons in the target area to ensure signal coverage throughout the area. The spacing of access points should be reasonably arranged according to environmental conditions and positioning accuracy requirements. During each travel period, record the MAC address and signal strength of all devices entering the target area. Wi-Fi and Bluetooth access points will periodically scan the surrounding environment to capture the MAC address and received signal strength (RSSI) of the device. The collected MAC address and signal strength data are stored in the database. Each record should contain information such as timestamp, access point ID, MAC address, and signal strength for subsequent data processing and analysis.

[0083] Using triangulation technology, the location of the device is calculated based on the signal strength of the same device received at different access points. First, based on the received signal strength indication, the path loss model is used to convert the signal strength into distance. Then, based on the calculated distance and the location of the access point, the location of the device is estimated using triangulation. Since the signal will be affected by difficult-to-predict differentiated obstacles (such as walls, furniture, etc.) in cold walking spaces, the free space path loss model and the ITU indoor path loss model are not used here. Instead, a more flexible empirical model is used, and its calculation formula is as follows:

[0084] RSSI = RSSI0-10nlog 10 (d)

[0085] RSSI is the received signal strength indicator, in dBm. RSSI0 is the signal strength at the reference distance d0, which is usually measured at 1 meter. d is the distance between the receiving point and the transmitting point, in meters (m). n is the path loss index, which depends on the environment and is usually between 2 and 4.

[0086] The step 1.2 is specifically as follows:

[0087] First, the collected pedestrian trajectory data is preprocessed, including noise removal, missing data filling and standardization. Then, the preprocessed trajectory data is converted into path graph structure data, defining the nodes in the graph as key positions in the walking space, the edges as the paths of pedestrian movement, and assigning corresponding weights to each edge (such as pedestrian flow, travel time, etc.).

[0088] Based on the transformed path graph structure data, a graph convolutional neural network (GCN) model is applied to build a pedestrian path selection model. First, the GCN model is designed and trained, taking the path graph structure data as input to predict the probability of pedestrians choosing nodes in the path graph. The model training process includes feature extraction, graph convolution operation and node classification. Through multiple iterative optimizations, the prediction accuracy and robustness of the model are improved.

[0089] On this basis, the social force model, multi-agent system and particle filter algorithm are integrated to construct a pedestrian pathfinding gray box model. The social force model is used to simulate the interaction between pedestrians and the influence of the environment on pedestrians, the multi-agent system is used to simulate the collective behavior and path selection strategy of pedestrians, and the particle filter algorithm is used to track and predict pedestrian trajectories in real time.

[0090] In the construction of the pathfinding graybox model, each agent is first assigned an initial position and velocity to simulate the initial state of the pedestrian group. These initial states reflect the position and direction of movement of pedestrians at a specific point in time. Next, a set of particles is initialized for each agent, which represent possible trajectory states. The particle filter algorithm uses these particles to represent the possible position and movement path of pedestrians in future time steps.

[0091] After initialization, the trained GCN model is used to predict the next target node of the pedestrian path, providing each agent with a clear short-term running goal and providing direction for subsequent motion calculations. The motion of each agent at the current time step is then calculated based on the social force model. The social force model calculates the acceleration and velocity changes of pedestrians by considering the interaction forces between pedestrians, the attraction between pedestrians and targets, and the repulsion between pedestrians and obstacles, thereby updating the position of pedestrians.

[0092] On this basis, the particle filter algorithm is applied to predict and track the trajectory of the intelligent agent. The particle filter uses the social force model to predict the movement path of each particle, and adjusts the weight of each particle based on the actual observation data. By adjusting the weights, the particle filter can improve the estimation accuracy of the pedestrian's true trajectory. The resampling step reduces impossible trajectories by selecting particles with high weights, thereby enhancing the reliability of trajectory prediction.

[0093] Based on the results of the particle filter algorithm, the state of the agent is updated. This process ensures that the trajectory of each agent not only accurately reflects its own movement rules, but also is consistent with the actual situation when interacting with the environment and other pedestrians. Through this dynamic adjustment, the movement trajectory of each agent becomes more accurate and reliable.

[0094] Finally, by continuously iterating the above steps, the trajectory of the agent is gradually optimized and accurately predicted. At each time step, the trajectory of the agent is adjusted and optimized according to the new observation data and the prediction results of the particle filter algorithm until the prediction of all time steps is completed and the final pedestrian pathfinding gray box model is obtained.

[0095] The step 1.3 is specifically as follows:

[0096] Deploy drones in typical cold-region walking spaces, and plan the drone routes according to the terrain and layout of the target area to ensure coverage of the entire walking space. Drones should conduct multiple low-altitude flights in various time periods (including 08:00-09:00, 09:00-10:00, 10:00-11:00, 11:00-12:00, 12:00-13:00, 13:00-14:00, 14:00-15:00, 15:00-16:00, 16:00-17:00, 17:00-18:00) to collect multi-angle image data, and use high-resolution camera equipment for low-altitude photography to ensure the clarity and coverage of the image data and avoid occlusion and ghosting. The route can be planned using Altizure software. After planning the measurement range, height, heading overlap, lateral overlap, and camera tilt angle, the software will automatically calculate the shooting time.

[0097] Next, the multi-angle image data taken by the drone is preprocessed, including noise removal and color correction, and then imported into the ContextCapture Master software. The coordinates of the same-name points of the drone-collected images are used for image matching of multiple data sources, and the correction equation is established by setting the connection point and control point coordinates to improve the image matching accuracy. The modeling process will automatically match the best images at different angles at each location, apply the SMF algorithm to match the feature points, and convert the multi-angle image data into point cloud data. After the dense point cloud data obtained by aerial triangulation is segmented and processed, different levels of TIN triangulated mesh generation are performed to convert the point cloud data into an irregular triangular mesh, simplify the triangular mesh model, and then automatically map the texture to generate a high-precision three-dimensional information model of the cold block, ensuring the accuracy and details of the model, which can clearly display the terrain, buildings, roads and other information of the walking space.

[0098] On this basis, multiple typical frames are selected from the image data taken by the drone, which should be able to represent the distribution of pedestrians in the walking space in different time periods. Then, the position of each pedestrian is located in the selected typical frames, and these position data are recorded in the high-precision 3D information model.

[0099] After that, use the public pedestrian detection dataset or custom dataset to train the YOLO algorithm so that it can accurately detect pedestrian targets in the image. Apply the trained YOLO model to continuous drone image frames to automatically detect pedestrian targets in the image and record the position and time information of each pedestrian. Initialize the CSRT (Discriminative Correlation Filter with Channel and SpatialReliability) tracker at each detected pedestrian target position, apply the CSRT algorithm to track each detected pedestrian in continuous drone image frames, and record the position change information of each pedestrian in each time frame.

[0100] Finally, the manually annotated pedestrian positions in typical frames and the pedestrian position change information obtained through the CSRT algorithm are integrated and fused into a high-precision three-dimensional information model. Based on the integrated pedestrian position data, the trajectory reconstruction algorithm is used to reconstruct the pedestrian's motion trajectory with high precision to ensure that the reconstructed trajectory can accurately reflect the pedestrian's movement path and behavior pattern in the cold-land walking space. Through the GIS platform or three-dimensional visualization tools, the reconstructed pedestrian trajectory is visualized in the high-precision three-dimensional information model to provide intuitive pedestrian behavior analysis results.

[0101] Step 2: Constructing thermal environment data of cold-region walking space and pedestrian thermal sensation mapping;

[0102] The step 2 is specifically as follows:

[0103] Step 2.1: Pedestrian thermal sensation acquisition in typical cold-weather walking spaces based on ecological instantaneous assessment;

[0104] Step 2.2: Construction of pedestrian thermal sensation model under the winter and summer heat environment of cold-region walking space;

[0105] Step 2.3: Calibration of pedestrian thermal sensation model in winter and summer in cold-region walking space based on reinforcement learning.

[0106] The step 2.1 is specifically as follows:

[0107] Step 2.1.1: Based on the mobile devices (such as smart phones) carried by pedestrians, the geographical locations of pedestrians are recorded in real time to obtain the locations of pedestrians in cold-weather walking spaces;

[0108] Step 2.1.2: Real-time thermal sensation collection of pedestrians in cold-weather walking spaces under changing pedestrian positions;

[0109] Step 2.1.3: Deploy temperature, humidity, wind speed and radiation heat sensors in a grid-like manner in typical cold-region walking spaces to collect real-time thermal environment data of typical cold-region walking spaces.

[0110] The step 2.1.2 is specifically as follows:

[0111] Based on wearable devices with sensors, including smart wearable ear clip sensors, smart wearable wrist sensors and smart wearable finger sensors, real-time skin temperature, heart rate and EDA skin electrical signal data are collected under the changes of pedestrian position in cold walking spaces, and these data are transmitted to the central data processing server via wireless communication (such as Bluetooth or Wi-Fi).

[0112] At the same time, during pedestrians' walking, we regularly collect pedestrians' subjective evaluation data on their current thermal sensations through questionnaire surveys, including subjective indicators such as thermal acceptability, thermal comfort, cold and warm sensations, and thermal pleasure. These subjective evaluation data will be matched with physiological indicator data in time and space to form a complete data set.

[0113] Then, big data analysis and machine learning techniques are used to construct a mapping relationship between pedestrian physiological index data and thermal sensation in cold-region walking spaces. First, the collected data is cleaned and normalized to eliminate noise and outliers. Then, key features such as skin temperature change rate, heart rate fluctuation pattern, EDA signal change, etc. are extracted as input variables of the model. Next, a neural network model is used to train the mapping model to predict pedestrian thermal sensation. After model verification and optimization, real-time thermal sensation prediction of pedestrian position changes in cold-region walking spaces is achieved.

[0114] Step 3: Generate and design the layout of cold-region walking space driven by dynamic thermal comfort data;

[0115] The step 3 is specifically as follows:

[0116] Step 3.1: Obtaining the layout plan of cold-region walking space driven by generation rules;

[0117] Step 3.2: Optimal design of cold-region walking space layout driven by dynamic thermal comfort data;

[0118] Step 3.3: Decision support for cold-region walking space layout design based on human-computer interaction.

[0119] The step 3.1 is specifically as follows:

[0120] First, a large amount of cold-region block layout model data is obtained and voxelized, that is, the three-dimensional model is divided into a series of regular small cubic units. Then, a three-dimensional convolutional neural network is used to process the voxelized data to extract the layout characteristics and building characteristics of the cold-region block.

[0121] Based on the extracted cold-region block scale features, the predefined cold-region pedestrian space is divided into three-dimensional units of appropriate size. The size of these three-dimensional units should be set according to the actual design conditions and space requirements to ensure that each unit can accurately reflect the details of the pedestrian space. Then, according to the design conditions of the cold-region pedestrian space and the extracted layout features, constraints are set on the three-dimensional matrix. These constraints include geometric and topological requirements, such as building scale restrictions, road width requirements, and the proximity of blocks to roads.

[0122] Under the constraints, the three-dimensional units are allocated based on the multi-agent system. The multi-agent system simulates the combination mode of the three-dimensional units. Each agent represents a three-dimensional unit and uses specific rules and decision-making mechanisms to optimize the combination and layout. The agents interact with each other based on the design conditions and constraints, and gradually adjust the position and distribution of each three-dimensional unit through an iterative process to generate a pedestrian space layout plan that meets the design goals.

[0123] The step 3.2 is specifically as follows:

[0124] Step 3.2.1: Construct the mathematical model and parameterized model for cold-region walking space layout optimization based on the design objectives and constraints;

[0125] Step 3.2.2: Obtaining dynamic thermal comfort of pedestrians during travel periods under cold-region pedestrian space layout;

[0126] Step 3.2.3: Based on the dynamic thermal comfort objectives of pedestrians, a multi-objective optimization algorithm is applied to optimize the layout of cold-region walking space and obtain the Parento solution.

[0127] The step 3.2.2 is specifically as follows:

[0128] Step 3.2.2.1: Construct a typical thermal environment prediction model for each travel period in winter and summer under the cold-region pedestrian space layout based on a convolutional neural network;

[0129] Step 3.2.2.2: Obtain the typical thermal environment prediction images for each travel period in winter and summer through the typical thermal environment prediction model for each travel period in winter and summer constructed in step 3.2.1, and obtain the pedestrian routes for each travel period under the current cold-region pedestrian space layout through the cold-region pedestrian space path-finding agent model constructed in step 1, so as to realize the prediction of the thermal environment of pedestrians under the typical routes of cold-region pedestrian space during winter and summer travel periods;

[0130] Step 3.2.2.3: Based on the thermal environment of pedestrians on typical routes during winter and summer travel periods in the cold-region walking space obtained in step 3.2.2, the thermal environment data of the cold-region walking space and the pedestrian thermal sensation mapping model constructed in step 2 are used to obtain the thermal sensation of pedestrians during typical route travel periods under the layout of the cold-region walking space.

[0131] The step 3.2.2.1 is specifically as follows:

[0132] The microclimate of 1000 typical cold-region walking spaces was simulated by ENVI-MET software, and the thermal environment images of each travel period in winter and summer under the layout of cold-region walking spaces were obtained. The images were preprocessed and converted from RGB color space to HSV (hue, saturation, brightness) color space to better separate color information and brightness information. Then, the color histogram of the image was calculated, and the color distribution of the three HSV channels was counted respectively to obtain the color histogram of each channel, which was normalized to ensure that the value range was between 0 and 1. In order to capture the color distribution law, the image was divided into grids or blocks, the color histogram of each block was calculated, and these histograms were combined into a feature vector. This method can reflect the spatial distribution of colors in the image. By extracting the color histogram of each grid block and combining them, a complete color distribution feature vector was generated. After the feature extraction was completed, the extracted color distribution feature vectors were clustered using the K-means clustering algorithm to obtain the thermal environment clustering images of these 1000 typical cold-region walking spaces in each travel period in winter and summer. Finally, the convolutional neural network is used to construct the mapping relationship between the cold-region walking space layout and the clustering images of each travel period in winter and summer.

[0133] The step 3.3 is specifically as follows:

[0134] First, 200 designers were invited to evaluate and score the morphology of 1,000 cold-region walking space schemes, and the corresponding dynamic thermal comfort data of the schemes were obtained. Next, the collected data was cleaned to remove noise and outliers. The dynamic thermal comfort data and morphology evaluation data were integrated into a comprehensive data set to be input into the random forest model.

[0135] Then, key features are extracted from the comprehensive data set, and statistical analysis and correlation analysis methods are used to select features that have a greater impact on layout decisions to improve the model's training efficiency and prediction accuracy. The processed data is trained using the random forest algorithm to build a preliminary layout decision model. The performance of the model is evaluated using the cross-validation method to ensure that the model has good generalization ability and prediction accuracy.

[0136] Next, VR equipment and environmental control equipment are used to provide users with a simulated walking experience of the scheme. High-performance VR headsets and controllers are selected to ensure that users can obtain an immersive walking experience. At the same time, environmental control equipment such as air conditioners and humidifiers are configured to control the ambient temperature, humidity and wind speed to simulate real climate conditions and environmental changes. And use 3D modeling software to build a virtual scene of the cold-region walking space layout plan, including elements such as buildings, roads, and green spaces. After that, invite users to wear VR equipment for a virtual walking experience. During this process, according to the dynamic thermal comfort data of pedestrians in the scheme, set the environmental parameters in the virtual scene and controllable environment to simulate the real thermal comfort experience. During and after the experience, record their evaluation of the scheme, including feedback on comfort, aesthetics, and functionality, and organize it into a structured evaluation data set.

[0137] Finally, the collected user evaluation data is used to optimize and train the machine learning model, adjust the parameters and optimization strategy of the layout decision model, and gradually improve the decision accuracy and matching degree of the model. Based on the optimized cold-region walking space decision model, the optimal cold-region walking space layout design scheme is obtained, providing designers with efficient and comprehensive decision support.

[0138] The present invention also proposes an electronic device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of a method for generating a design method for a cold-region urban walking space layout based on dynamic thermal comfort prediction are implemented.

[0139] The present invention also proposes a computer-readable storage medium for storing computer instructions, which, when executed by a processor, implement the steps of a method for generating a design for a cold-region urban pedestrian space layout based on dynamic thermal comfort prediction.

[0140] The memory in the embodiments of the present application may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of RAM are available, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), and direct RAM bus RAM (DRRAM). It should be noted that the memory of the method described in the present invention is intended to include, but is not limited to, these and any other suitable types of memory.

[0141] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions may be transmitted from a website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (digital subscriber line, DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, server or data center. The computer-readable storage medium may be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more available media integrated. The available medium may be a magnetic medium (eg, a floppy disk, a hard disk, a magnetic tape), an optical medium (eg, a high-density digital video disc (DVD)), or a semiconductor medium (eg, a solid state disc (SSD)).

[0142] In the implementation process, each step of the above method can be completed by an integrated logic circuit of hardware in a processor or an instruction in the form of software. The steps of the method disclosed in conjunction with the embodiment of the present application can be directly embodied as a hardware processor for execution, or a combination of hardware and software modules in a processor for execution. The software module can be located in a storage medium mature in the art such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory or an electrically erasable programmable memory, a register, etc. The storage medium is located in a memory, and the processor reads the information in the memory and completes the steps of the above method in conjunction with its hardware. To avoid repetition, it is not described in detail here.

[0143] It should be noted that the processor in the embodiment of the present application can be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method embodiment can be completed by an integrated logic circuit of hardware in the processor or an instruction in the form of software. The above processor can be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The methods, steps and logic block diagrams disclosed in the embodiments of the present application can be implemented or executed. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the method disclosed in the embodiment of the present application can be directly embodied as a hardware decoding processor to perform, or the hardware and software modules in the decoding processor can be combined and performed. The software module can be located in a mature storage medium in the field such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory or an electrically erasable programmable memory, a register, etc. The storage medium is located in a memory, and the processor reads the information in the memory and completes the steps of the above method in combination with its hardware.

[0144] The above is a detailed introduction to the cold-region urban pedestrian space layout generation and design method based on dynamic thermal comfort prediction proposed in the present invention. This article uses specific examples to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea; at the same time, for general technical personnel in this field, according to the idea of ​​the present invention, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

Claims

1. A method for generating and designing the layout of walking space in cold regions based on dynamic thermal comfort prediction, characterized in that: The method comprises the following steps: S1. Construction of a pathfinding agent model for cold-region walking space; The step S1 is specifically as follows: S1.1: Pedestrian group trajectory data collection at different travel times in typical cold-region walking spaces based on IoT; S1.2: Construction of gray box model for path finding in cold-region walking space at different travel periods; S1.3: Extract accurate pedestrian trajectories at different travel times in typical cold-region walking spaces based on drone image data; S1.4: Calibration of the cold-region walking space pathfinding agent model based on typical spatial IoT perception data; S2, construction of thermal environment data of cold-region walking space and pedestrian dynamic thermal comfort mapping; The step S2 is specifically as follows: S2.1: Pedestrian thermal sensation acquisition in typical cold-weather walking spaces based on ecological instantaneous assessment; S2.2: Construction of thermal sensation model of pedestrians in cold-region walking spaces under hot winter and summer conditions; S2.3: Calibration of thermal sensation model of pedestrians in winter and summer in cold-region walking space based on reinforcement learning; S3, Generative design of cold-region walking space layout driven by dynamic thermal comfort data; The step S3 is specifically as follows: S3.1: Obtaining the layout plan of cold-region walking space driven by generation rules; S3.2: Optimal design of cold-region walking space layout driven by dynamic thermal comfort data; S3.3: Decision support for cold-region walking space layout design based on human-computer interaction.

2. The method according to claim 1, characterized in that The step S2.1 is specifically as follows: S2.1.1: Pedestrian position acquisition in cold-weather walking spaces; S2.1.2: Real-time thermal sensation collection of pedestrians in cold-weather walking spaces under changing pedestrian positions; S2.1.3: Real-time collection of thermal environment data of typical cold-region walking spaces.

3. The method according to claim 2, characterized in that In step S2.1.2, the real-time skin temperature, heart rate and EDA skin electrical signal data of pedestrians in cold-region walking spaces are collected based on wearable devices as the pedestrians' positions change in the cold-region walking spaces, and the subjective evaluation data of pedestrians in cold-region walking spaces as the pedestrians' positions change are obtained. Based on this, a mapping relationship between the physiological indicator data and thermal sensation of pedestrians in cold-region walking spaces is constructed, thereby obtaining the real-time thermal sensation information of pedestrians in cold-region walking spaces as the pedestrians' positions change in the cold-region walking spaces.

4. The method according to claim 1, characterized in that: In step S3.1, the cold-region block layout model data is obtained and voxelized, and the layout and building features are extracted using a three-dimensional convolutional neural network. Based on the scale characteristics of the cold-region block, the predefined cold-region pedestrian space is divided into three-dimensional units. Based on the design conditions of the cold-region pedestrian space and the extracted features, constraints are set on the three-dimensional matrix so that its generation scheme meets the geometric and topological requirements. Under the constraints, the three-dimensional units are allocated based on the multi-agent system to realize the generation of the cold-region pedestrian space layout, thereby obtaining the optimized prototype of the cold-region pedestrian space layout.

5. The method according to claim 1, characterized in that The step S3.2 is specifically as follows: S3.2.1: Construction of optimization model for cold-region walking space layout; S3.2.2: Dynamic thermal comfort of pedestrians during travel periods under cold-region pedestrian space layout; S3.2.3: Optimization of cold-region walking space layout under the guidance of dynamic thermal comfort.

6. The method according to claim 5, characterized in that The step S3.2.2 is specifically as follows: S3.2.2.1: Construction of typical thermal environment prediction model for each travel period in winter and summer under the spatial layout of cold-region pedestrian space; S3.2.2.2: Thermal environment acquisition of pedestrians along typical routes during winter and summer travel periods in cold-region pedestrian spaces; S3.2.2.3: Pedestrian thermal sensation acquisition during typical travel periods on a cold-region pedestrian space layout.

7. The method according to claim 6, characterized in that In step S3.2.2.1, the thermal environment images of each travel period in winter and summer under the cold-region walking space layout are obtained, and the thermal environment images of each travel period are clustered respectively, and then a mapping relationship between the cold-region walking space layout and the clustered images of each travel period in winter and summer is constructed.

8. The method according to claim 1, characterized in that In step S3.3, after obtaining the optimal layout plan of cold-region walking space under the guidance of dynamic thermal comfort, the random forest model is used to construct a cold-region walking space layout decision model based on the dynamic thermal comfort data of the cold-region walking space layout plan and the designer's morphological evaluation data of the layout plan, and a preliminary decision plan is obtained; VR equipment and environmental control equipment are used to provide users with a simulation of the walking experience of the plan, obtain their evaluation of the plan, and use it as feedback to the decision model, and further adjust the cold-region walking space layout decision model, thereby providing designers with efficient and comprehensive decision support.

9. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 8 are implemented.

10. A computer-readable storage medium for storing computer instructions, characterized in that: When the computer instructions are executed by a processor, the steps of the method according to any one of claims 1 to 8 are implemented.

Images