Urban heat island effect regulation and control method based on particle swarm optimization

Abstract

The invention discloses an urban heat island effect regulation and control method based on particle swarm optimization, and the method comprises the steps: constructing a simulation system which couples an urban physical microclimate model and a human behavior dynamics model, carrying out the iterative optimization through employing a particle swarm optimization algorithm, and enabling each particle to represent a to-be-evaluated urban heat island regulation and control scheme, in the iteration process, a comprehensive fitness value is calculated for each regulation and control scheme through the coupling simulation system so as to guide the state updating of the particle swarm until an optimal scheme is obtained; according to the method, the urban heat island regulation and control scheme which is excellent in physical cooling, human welfare and social vitality can be found, and the scientificity and reality effectiveness of the scheme are improved.

Description

Technical Field

[0001] This invention relates to the field of urban environment simulation and intelligent optimization technology, and in particular to a method for controlling the urban heat island effect based on particle swarm optimization. Background Technology

[0002] The urban heat island (UHI) effect, referring to the phenomenon where urban areas experience significantly higher temperatures than their surrounding rural areas, has become an increasingly serious environmental and social problem in the global urbanization process. This effect not only leads to a sharp increase in urban energy consumption and air pollutant concentrations during summer, but also poses a direct threat to the health, well-being, and quality of life of residents. To address this challenge, academia and engineering have developed a series of assessment and optimization techniques centered on numerical simulation. Early research mainly focused on using computational fluid dynamics (CFD) models, such as ENVI-met and Fluent, or mesoscale meteorological models (such as WRF), to perform high-precision simulations of the microclimate environment of specific urban areas. These models can finely depict the impact mechanisms of physical elements such as building layout, underlying surface materials, and vegetation configuration on the urban thermal environment, providing a scientific physical basis for urban planning and design. However, with the improvement of computing power, some researchers have proposed combining simulation models with optimization algorithms, thus forming an integrated "simulation-optimization" framework. Within this framework, heuristic algorithms such as genetic algorithms (GA) and particle swarm optimization (PSO) are typically used to automatically search for combinations of control schemes that maximize the cooling effect within a vast space of design parameters (such as green space ratio, building albedo, and water area). This also marks a shift in urban heat island control research from a passive "evaluation and analysis" stage to an active "intelligent optimization" design stage.

[0003] However, existing urban heat island control methods based on the "simulation-optimization" integrated framework, despite significant progress at the physical level, generally suffer from a deep-seated and fundamental technical flaw: the model construction and evaluation system severely neglects the dynamic presence and adaptive behavior of the core actors in the urban environment—human beings. Specifically, existing technologies typically treat the city as a static physical container lacking agency. First, at the model input end, anthropogenic heat emissions are often simplified spatially and temporally to static or preset values ​​based on statistical yearbooks, failing to reflect the dynamic changes in residents' actual activity trajectories and aggregation patterns caused by environmental changes, thus distorting the heat source input. Second, existing technologies completely ignore human adaptive behavioral feedback to the thermal environment. In the real world, residents actively avoid overheated areas and seek cool, comfortable paths or public spaces; this spontaneous self-protective behavior reshapes urban space usage patterns and vitality distribution. Existing models fail to capture this dynamic, therefore their evaluation results of control schemes may differ significantly from actual application effects. Finally, on the optimization target side, evaluation systems are often limited to single physical indicators, such as the reduction in regional average air temperature or surface temperature. This evaluation dimension cannot measure the degree to which the control measures improve human "thermal comfort," let alone assess whether they have achieved "comfort equity" among different groups, or whether they have enhanced the socio-economic vitality of specific public spaces by guiding pedestrian flow. At the same time, these problems limit the scientific validity and effectiveness of current urban tropical effect control measures in the real world. Summary of the Invention

[0004] The purpose of this section is to outline some aspects of embodiments of the present invention and to briefly describe some preferred embodiments. Simplifications or omissions may be made in this section, as well as in the abstract and title of this application, to avoid obscuring the purpose of these documents; however, such simplifications or omissions should not be construed as limiting the scope of the invention.

[0005] In view of the aforementioned existing problems, this invention is proposed. Therefore, this invention provides a particle swarm optimization-based method for controlling the urban heat island effect, to address the problems mentioned in the background section.

[0006] To address the aforementioned technical problems, this invention provides the following technical solution: a method for controlling the urban heat island effect based on particle swarm optimization, comprising:

[0007] Construct a coupled simulation system of urban physical microclimate model and human behavior dynamics model;

[0008] Particle swarm optimization algorithm is used for iterative optimization, where each particle represents an urban heat island control scheme to be evaluated.

[0009] During the iterative optimization process, the coupled simulation system calculates a comprehensive fitness value for each urban heat island control scheme represented by each particle and guides the state update of the particle swarm until the optimal control scheme is obtained.

[0010] As a preferred embodiment of the urban heat island effect control method based on particle swarm optimization described in this invention, the comprehensive fitness value includes the following sequentially executed sub-steps:

[0011] The physical microclimate model is updated based on the urban heat island control scheme represented by the particles to generate a dynamic urban microclimate field.

[0012] Based on the dynamic urban microclimate field, the human behavior dynamics model is driven to simulate the adaptive behavioral response of human groups to the urban microclimate field.

[0013] The comprehensive fitness value is calculated based on the simulation results of the urban microclimate field and the adaptive behavioral response.

[0014] As a preferred embodiment of the urban heat island effect control method based on particle swarm optimization described in this invention, the human group in the human behavior dynamics model is represented by multiple agents. The behavioral decision of the agents is determined by comparing the environmental thermal comfort assessed by the general thermal climate index with a preset comfort reference point, and by assigning different weights to situations above and below the reference point according to asymmetric decision rules.

[0015] As a preferred embodiment of the urban heat island effect control method based on particle swarm optimization described in this invention, the metabolic heat generated by the intelligent agent during the simulation process is accumulated, and the accumulated result is input as an anthropogenic heat source term into the energy balance calculation of the corresponding spatial location of the physical microclimate model.

[0016] As a preferred embodiment of the urban heat island effect control method based on particle swarm optimization described in this invention, the comprehensive fitness value is generated through a comprehensive evaluation of the following indicators:

[0017] The physical cooling index of the control scheme in terms of its effectiveness in reducing urban surface or air temperature.

[0018] A group comfort index representing the degree of thermal comfort experienced by a human group in a simulation;

[0019] Spatial vitality index of the degree to which public spaces are effectively utilized.

[0020] As a preferred embodiment of the urban heat island effect control method based on particle swarm optimization described in this invention, the calculation of the group comfort index includes not only the evaluation of the overall average duration of thermal discomfort experienced by all agents, but also the evaluation of the differences in the distribution of thermal discomfort duration among different agents.

[0021] As a preferred embodiment of the urban heat island effect control method based on particle swarm optimization described in this invention, the calculation of the spatial vitality index includes: evaluating the spatiotemporal distribution balance of the activity density of agents in various public spaces driven by agent behavior decisions, and the path selection cost of agents from their location to a preset comfort area.

[0022] As a preferred embodiment of the urban heat island effect control method based on particle swarm optimization described in this invention, the operation of the physical microclimate model is based on solving the surface energy balance equation. The calculation of the balance equation includes the solar shading effect and sky visibility factor determined by the three-dimensional shape of the building, and calculates the sensible heat and latent heat fluxes through a mechanism model based on aerodynamic impedance and surface impedance.

[0023] As a preferred embodiment of the urban heat island effect control method based on particle swarm optimization described in this invention, the method further includes constructing and using a surrogate model to replace the sub-step of calculating the comprehensive fitness value in some iterations, and selectively executing the sub-step based on the prediction uncertainty of the surrogate model.

[0024] Compared with existing technologies, the beneficial effects of the invention are:

[0025] 1. This invention constructs a simulation system that couples a physical microclimate model with a human behavior dynamics model. It integrates human adaptive behaviors to the thermal environment (such as spatial avoidance and gathering) and the resulting dynamic human thermal feedback into the optimization process. This overcomes the fundamental defect of existing technologies that treat humans as a static background. As a result, the evaluation of the control scheme is no longer based on the isolated physical cooling effect, but on a dynamic balance result of two-way interaction between humans and the environment that is closer to the real world. This improves the reliability and prediction accuracy of the optimal scheme in practical applications.

[0026] 2. By incorporating the three dimensions of physical cooling efficiency, human thermal comfort, and urban public space vitality into the comprehensive fitness function of the optimization algorithm, the optimization process not only pursues temperature reduction but also aims to improve residents' experience, fairly allocate comfort resources, and effectively utilize urban space. This promotes social equity and urban vitality while optimizing the physical environment.

[0027] 3. The asymmetric decision-making rule, which reflects the psychology of "loss aversion," is adopted in the human behavior dynamics model, which realistically portrays human decision-making preferences when facing uncomfortable environments, thereby improving the realism of the micro-behavioral simulation. At the same time, by introducing a proxy model, it can intelligently replace some of the computationally expensive simulation evaluations without sacrificing the global optimization capability, shortening the time required to obtain the optimal solution and enhancing its practicality and operability in dealing with complex urban problems. Attached Figure Description

[0028] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein:

[0029] Figure 1 This is a flowchart illustrating the overall process of a particle swarm optimization-based method for controlling the urban heat island effect according to an embodiment of the present invention. Detailed Implementation

[0030] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0031] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0032] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0033] This invention is described in detail with reference to the schematic diagrams. When detailing the embodiments of this invention, for ease of explanation, the cross-sectional views illustrating the device structure may be partially enlarged, not adhering to the usual scale. Furthermore, the schematic diagrams are merely examples and should not be construed as limiting the scope of protection of this invention. In actual fabrication, the three-dimensional spatial dimensions of length, width, and depth should be included.

[0034] Furthermore, in the description of this invention, it should be noted that the terms "upper," "lower," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. These terms are used solely for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. In addition, the terms "first," "second," or "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0035] Unless otherwise explicitly specified and limited, the terms "installation," "connection," and "joining" in this invention should be interpreted broadly. For example, they can refer to fixed connections, detachable connections, or integral connections; similarly, they can refer to mechanical connections, electrical connections, or direct connections, or indirect connections through an intermediate medium, or internal connections between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0036] Example 1

[0037] Reference Figure 1 This is the first embodiment of the present invention, which provides a method for controlling the urban heat island effect based on particle swarm optimization, including:

[0038] S1. Construct a coupled simulation system of urban physical microclimate model and human behavior dynamics model;

[0039] It should be noted that the purpose of constructing the coupled simulation system is to create a high-fidelity virtual city environment capable of simulating two-way dynamic interaction between humans and the environment, thereby overcoming the shortcomings of traditional methods that treat the city as a static physical container. The construction of this coupled system includes the establishment of the following two core models and the implementation of their coupling mechanism:

[0040] Furthermore, an urban physical microclimate model is constructed. This model is a numerical model based on physical mechanisms, used to accurately calculate the distribution of microclimate environmental fields (such as air temperature, humidity, wind speed, mean radiation temperature, etc.) in urban space under specific urban morphology and meteorological conditions.

[0041] Preferably, the model is based on solving the surface energy balance equation, which describes the balance between various energy fluxes received and emitted by the surface and is the physical basis of microclimate simulation. Specifically, the calculation of the balance equation includes the solar shading effect and the sky view factor (SVF) determined by the three-dimensional shape of the building.

[0042] Specifically, the surface energy balance equation can be expressed as:

[0043] Rn =H+LE+G

[0044] Among them, R n It is expressed as the net surface radiation flux, which is determined by the difference between the received shortwave radiation and the longwave and shortwave radiation. Its calculation is directly related to the aforementioned solar shading effect and sky visibility factor; H is the sensible heat flux, LE is the latent heat flux; G is the soil / surface heat flux, which represents the heat entering or leaving the ground; by solving this surface energy balance equation, the surface temperature can be obtained.

[0045] Specifically, the urban physical microclimate model receives three-dimensional geometric data of the city (including buildings, terrain, vegetation, etc.) and calculates the solar shading of each surface grid cell at any time and its openness to the sky (i.e., sky visibility factor) based on the solar position algorithm.

[0046] It should be noted that the amount of shortwave radiation received and longwave radiation emitted by the Earth's surface is determined by the amount of solar shading and the sky visibility factor, which are key inputs for energy balance calculations.

[0047] Furthermore, the urban physical microclimate model also calculates sensible and latent heat fluxes through a mechanism model based on aerodynamic impedance and surface impedance. That is, the model calculates the energy exchanged between the surface and the atmosphere through convection and evapotranspiration based on the physical properties of different underlying surfaces (such as concrete, asphalt, grassland, and water bodies) (such as albedo, emissivity, roughness, soil moisture, etc.) and near-surface temperature and wind speed gradients, using the mechanism formula of the Moning-Obukhov similarity theory. This ensures that the model can accurately reflect the real impact of different urban design schemes (such as increasing green space and using highly reflective materials) on energy distribution.

[0048] Furthermore, a human behavior dynamics model is constructed, which adopts an agent-based modeling paradigm to simulate the adaptive behavioral response of human groups to the thermal environment in urban spaces.

[0049] Specifically, in this model, the human population is represented by multiple agents, each agent representing an individual or a class of individuals with autonomous decision-making capabilities, and the individual is given attributes such as initial location, destination, activity plan, and metabolic rate.

[0050] Specifically, the agent's behavioral decision-making is achieved by comparing the environmental thermal comfort assessed by the Universal Thermal Climate Index (UTCI) with a preset comfort reference point, and by determining the agent's behavioral choices based on an asymmetric decision-making rule that assigns different weights to responses to situations above and below the reference point.

[0051] Specifically, at each simulation time step, the human behavior dynamics model first uses the environmental parameters (temperature, humidity, wind speed, mean radiant temperature, etc.) output by the physical microclimate model to calculate the perceived UTCI for each agent's location, and then compares the UTCI with a comfort reference point representing a "no heat stress" state (e.g., UTCI = 26°C).

[0052] Preferably, in determining the agent's behavioral choices, the present invention adopts an asymmetric decision rule that embodies the "loss aversion" principle in behavioral economics: that is, when the UTCI is significantly higher than the reference point, the agent will have a strong aversion intention, and the probability of it choosing to find a shady place, enter an air-conditioned building, or choose a cooler path will increase sharply; while when the UTCI is lower than or slightly higher than the reference point, its tendency to change its current behavior is relatively weak.

[0053] Specifically, asymmetric decision rules can be implemented using a probabilistic decision function, where the probability of the agent choosing the "seeking a comfortable environment" behavior can be defined as:

[0054]

[0055] Where P(SC|UTCI) is the conditional probability, representing the probability that an agent will choose to perform the behavior of "seeking a comfortable environment" (e.g., moving to the shade, entering an air-conditioned building) given that the comprehensive thermal stress index of the current environment is UTCI. This probability value is between 0 and 1; UTCI ref The UTCI (Universal Temperature Compensation Index) benchmark or boundary of the comfort zone for human comfort is defined as a preset comfort reference point (e.g., set to 24°C). UTCI is related to UTCI. ref The difference quantifies the degree and direction of the current environment's deviation from the comfort state; a positive difference indicates a hotter environment, and a negative difference indicates a colder environment. k represents the decision sensitivity coefficient. Unlike traditional models, in this invention, k is not a fixed constant, but a conditional parameter that dynamically changes according to the thermal stress state: when the environment is hotter, i.e., UTCI > UTCI... ref Let k take a large positive value, denoted as k hot When the environment is relatively cold, i.e., UTCI <UTCI ref Let k take a positive value with a small absolute value, denoted as k cold For example, in a typical summer outdoor scene simulation, k can be set. hot =0.8, k cold =0.2, to reflect that the tendency to avoid overheated environments is much stronger than the reaction to slightly cold environments, but this does not constitute a limitation of the present invention. The present invention initially sets k hot >k coldThis setting makes the output probability curve in UTCI ref The slopes on both sides are different, due to k hot The probability curve is steeper on the hot side, meaning that if the UTCI slightly exceeds the comfort reference point, the probability of seeking comfort will rise sharply; while on the cold side, the probability curve is flatter, meaning that a greater deviation is needed to trigger the same intensity of behavioral change intention. This accurately portrays the "loss aversion" psychology of humans in thermal comfort decision-making, that is, the degree of aversion to "overheating" is much greater than the discomfort of "slight cold", thereby improving the microscopic realism of human behavior simulation.

[0056] Furthermore, the two models constructed above are coupled, that is, the dynamic microclimate field (especially the UTCI distribution map) generated by the physical microclimate model is used as environmental information, and this environmental information is input into the human behavior dynamics model at each time step. At the same time, the output of the human behavior dynamics model is fed back to the physical microclimate model in real time.

[0057] Furthermore, the metabolic heat generated by the intelligent agent during the simulation process is accumulated, and the accumulated result is used as an artificial heat source term to input into the energy balance calculation of the corresponding spatial location in the physical microclimate model;

[0058] Specifically, within each simulation time step, the metabolic heat (W / m³) generated by all agents based on their current activity state (e.g., sitting, walking) is calculated. 2 The heat flux of an agent is assigned to the surface grid cell of the physical microclimate model in which it resides. If multiple agents are located in the same grid cell, their metabolic heat is accumulated within that cell. The accumulated anthropogenic heat flux is directly added to the surface energy balance equation and participates in the calculation of the physical microclimate field in the next time step.

[0059] S2. Particle swarm optimization algorithm is used for iterative optimization, where each particle represents an urban heat island control scheme to be evaluated.

[0060] It should be noted that after a high-fidelity coupled simulation system is built, the question that needs to be considered is how to efficiently find the optimal control scheme in a decision space consisting of countless possible combinations of urban design parameters.

[0061] Preferably, the present invention employs a particle swarm optimization algorithm to intelligently search the decision space to find the optimal control scheme. Each particle represents a complete and executable urban heat island control scheme. The position of the particle is defined as an N-dimensional vector, where each dimension of the vector corresponds to an adjustable urban design parameter. These parameters may include, but are not limited to: the green space layout of the plot (such as green space ratio, vegetation type, and the configuration ratio of trees and shrubs), the material properties of the underlying surface (such as the albedo and emissivity of building roofs and roads), and the water body configuration (such as the water body area, shape, and location). The particle swarm optimization process is to search for the optimal position vector in the N-dimensional decision space composed of all combinations of urban design parameters.

[0062] In addition, each particle has a velocity vector, which determines the direction and distance of the particle's movement in the decision space in the next iteration;

[0063] Furthermore, before the iteration begins (t=0), an initial population containing M particles is randomly generated in the N-dimensional decision space, and the initial position and initial velocity of each particle are randomly assigned within its allowed range.

[0064] Furthermore, when entering an iterative loop, perform the following operations in each iteration:

[0065] For each particle in the particle swarm, the control scheme represented by its current position is evaluated (i.e., the quality of the scheme). The evaluation method is to call the constructed coupled simulation system and calculate a comprehensive fitness value for the scheme. This fitness value is the sole basis for guiding the evolution of the particle swarm.

[0066] Record the position of the best fitness value experienced by each particle, and call it the individual optimal position. At the same time, through the information sharing mechanism of the particle swarm, record the position of the global best fitness value of all particles, and call it the global optimal position. After each scheme evaluation, update the individual optimal position and the global optimal position.

[0067] Specifically, in each iteration, when the overall fitness value of a particle is calculated, it is compared with the particle's own historical best fitness value. If the overall fitness value is better (i.e., larger than the historical best fitness value), the particle's individual best position is updated to the current particle position, which is the candidate control scheme. After the individual best positions of all particles have been updated, the individual best positions of all particles are traversed again to find the particle with the highest overall fitness value and update its position to the new global best position.

[0068] According to the classic particle swarm optimization rules, the velocity and position of each particle are updated in the next iteration:

[0069] V n (t+1)=ω·V n (t)+c1r1·(pbest n -X n (t))+c2r2·(gbest-X n (t))

[0070] X n (t+1)=X n (t)+V n (t+1)

[0071] Where t represents the current iteration number, X n (t) represents the position vector of particle n at the t-th iteration, V n (t) represents the velocity vector of particle n at the t-th iteration; ω represents the inertia weight, used to adjust the influence of the previous velocity on the current velocity. This value is usually set to a relatively large initial value and dynamically decreased with the number of iterations. The advantage of this setting is that it helps to perform global exploration in the early stages of optimization; pbest n pbest represents the position of the best fitness value experienced by particle n; gbest represents the position of the global best fitness value of the entire particle swarm; c1 is the cognitive learning factor used to adjust the particle's own optimal position pbest. n The step size is c1; c2 is the social learning factor, used to adjust the step size of the global optimal position gbest of the particle; r1 and r2 are both independently generated random numbers in the interval [0, 1] to avoid the particle swarm getting trapped in local optima too early and to increase the diversity of the search.

[0072] Furthermore, since the computational cost of the comprehensive fitness evaluation process based on the coupled simulation system is extremely high, it is not feasible to perform a complete simulation on all particles in each iteration directly in practical applications. Therefore, this invention introduces an optimization acceleration mechanism, which involves constructing and using a surrogate model to replace some of the sub-steps in calculating the comprehensive fitness value in the iterations, and selectively executing the sub-steps for calculating the comprehensive fitness value based on the prediction uncertainty of the surrogate model itself, as follows:

[0073] By constructing a computationally inexpensive surrogate model in parallel, such as Gaussian process regression (GPR), radial basis function network (RBF), or deep neural network (DNN), the task of the surrogate model is to learn the complex mapping relationship between the control scheme (particle position) and the overall fitness value. Its training data comes from the (particle position, true fitness value) data pairs obtained after each simulation.

[0074] In the iterative process of the particle swarm optimization algorithm, for a particle to be evaluated, a surrogate model is used to quickly predict its overall fitness value and the uncertainty of this prediction (e.g., for the GPR model, the variance of the predicted value). Then, based on a predefined infill criterion, such as expected improvement or upper confidence, the infill is applied. The upper bound of confidence (UCB) criterion is used to determine whether a coupled simulation system needs to be directly invoked for real simulation. For example, when using the upper bound of confidence (UCB) criterion, for each particle in the particle swarm, its UCB value is calculated as: UCB(i) = μ(i) + k·σ(i), where μ(i) is the fitness value predicted by the surrogate model, σ(i) is the standard deviation of the prediction (uncertainty), and κ is the coefficient for balancing exploration and exploitation. In each iteration, the particle with the highest UCB value is selected, and a real simulation is performed on it using the coupled simulation system. The obtained (particle position, real fitness value) data is used to update the surrogate model. For other particles, the μ(i) predicted by the surrogate model is directly used as their comprehensive fitness value to participate in the state update of the particle swarm.

[0075] In addition, when a particle is located in a region where the surrogate model has high prediction uncertainty (i.e., the surrogate model is "not confident" in its performance, which usually refers to a region in the particle swarm that has not been fully explored), or when the overall fitness predicted by the surrogate model is very high, the coupled simulation system will be used to perform real simulation directly, and the new data points will be used to update and optimize the surrogate model. For other particles, the prediction value of the surrogate model will be used directly as its overall fitness value.

[0076] It should be noted that by using the proxy model, the algorithm's global optimization capability can be guaranteed, while the majority of computing resources are concentrated on the candidate solutions that are most likely to produce breakthroughs. This reduces unnecessary simulations, shortens the time required to obtain the optimal solution, and enhances the practicality and operability of this invention in dealing with complex urban problems.

[0077] S3. During the iterative optimization process, a comprehensive fitness value is calculated for each urban heat island control scheme represented by each particle through a coupled simulation system, and the state update of the particle swarm is guided until the optimal control scheme is obtained.

[0078] Furthermore, the formula for calculating the overall fitness value is as follows:

[0079] Fitness (S) i )=ω cool ·I cooling (S i )+ω comfort ·I comfort (Si )+ω vitality ·I vitality (S i )

[0080] Among them, Fitness (S i S represents the candidate control scheme S corresponding to the i-th particle. i The overall performance score, i.e., the overall fitness value; I cooling (S i The physical cooling index () is a normalized value; the higher the value, the better the physical cooling effect. The calculation result comes from the output of the physical microclimate model in the coupled simulation system and can be weighted by one or more of the following sub-indices: the reduction in regional average or maximum air / surface temperature, and the percentage reduction in area of ​​the general thermal climate index (e.g., UTCI > 32℃). comfort (S i (S) is used as a group comfort index to evaluate candidate control schemes S. i The higher the value, the better and fairer the overall thermal comfort experience of the human group. The calculation results are derived from the simulation output of the human behavior dynamics model and include: overall comfort, which is the reciprocal of the overall average duration of thermal discomfort experienced by all agents, and comfort fairness, which is the Gini coefficient of the distribution of thermal discomfort duration among different agents. This comfort fairness ensures that the calculation results do not sacrifice the extreme discomfort of a few individuals for the improvement of the average. vitality (S i ) is represented as a space vitality index, used to measure candidate control schemes S i The higher the value of the positive impact on the socio-economic functions of urban public spaces, the more balanced and efficient the urban space can be utilized. The calculation results are derived from the simulation output of the human behavior dynamics model and include: the spatial utilization balance, which is the information entropy of the distribution of human flow density formed by the activities of intelligent agents in each public space. The higher the information entropy value, the more balanced the distribution and the stronger the vitality; and comfort accessibility, which is the reciprocal of the average path cost for an intelligent agent to reach the nearest "heat refuge" from an uncomfortable area. The lower the path cost, the higher the accessibility of the comfortable environment; ω cool ω comfort ω vitality These are weighting coefficients for physical cooling indicators, group comfort indicators, and spatial vitality indicators. These coefficients are set by urban planners, decision-makers, or stakeholders based on specific planning objectives. For example, in a central business district, more emphasis is placed on physical cooling indicators to achieve the best cooling effect; while in a historical and cultural district or community, more emphasis is placed on group comfort indicators and spatial vitality indicators to prioritize residents' comfort and community vitality in order to meet the diverse development needs of different urban functional areas.

[0081] Furthermore, the physical cooling index is expressed as:

[0082]

[0083] Among them, UTCI base UTCI is the spatially average universal thermal climate index for the baseline scenario, calculated through simulation under conditions where no control measures were implemented (i.e., the original urban layout before optimization). i This represents the spatial average general thermal climate index of the entire urban simulation area during the specified simulation period after the i-th control scheme is implemented.

[0084] Furthermore, the group comfort index is expressed as:

[0085]

[0086] Among them, T uncomf,avg The average duration of thermal discomfort refers to the average cumulative duration during which all agents are exposed to thermally uncomfortable environments (e.g., defined as areas with a UTCI exceeding 28°C) within the simulation period; Gini(T uncomf ) represents the Gini coefficient of thermal discomfort duration. The Gini coefficient is a classic indicator for measuring the degree of inequality in distribution. It takes a value between 0 and 1 and is used to measure the differences in the distribution of thermal discomfort duration among different agents. α and β are both weighting coefficients of the group comfort index, which are set by decision-makers according to planning objectives. For example, if more attention is paid to the experience of vulnerable groups, the weight of β can be increased.

[0087] Furthermore, the spatial vitality index is expressed as:

[0088]

[0089] Among them, C path,avg The average comfortable path cost refers to the average path cost for all agents to move from their current (potentially uncomfortable) location to the nearest preset "heat refuge" (such as a shady spot, building entrance, or other comfortable area). Furthermore, this path cost can be a comprehensive metric, such as the product of path length and the average UTCI along the path. A lower cost means that people can reach the cooler area more easily and comfortably. γ and δ are both weighting coefficients for the spatial vitality index, with the same initial value and a range of (0,1). H(D) represents the Shannon entropy of spatial utilization equilibrium, used to measure the degree of equilibrium in the distribution of pedestrian flow driven by agent behavior across different public spaces. Its extended formula is as follows:

[0090]

[0091] Where D is the probability set of the human flow density distribution, D = {p1, p2, ..., p...} j};p j This refers to the proportion of agents that remain in the j-th common space during the simulation period out of the total number of agents; m represents the total number of common spaces.

[0092] It should be noted that the greater the Shannon entropy, the more evenly people are distributed in each space, avoiding situations where some spaces become deserted due to overheating while others become overcrowded due to comfort, thus representing a stronger overall space vitality.

[0093] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code. The solutions in the embodiments of this application can be implemented using various computer languages, such as the object-oriented programming language Java and the interpreted scripting language JavaScript.

[0094] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0095] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0096] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0097] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.

[0098] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A method for controlling the urban heat island effect based on particle swarm optimization, characterized in that, include: Construct a coupled simulation system of urban physical microclimate model and human behavior dynamics model; Particle swarm optimization algorithm is used for iterative optimization, where each particle represents an urban heat island control scheme to be evaluated. During the iterative optimization process, the coupled simulation system calculates a comprehensive fitness value for each urban heat island control scheme represented by each particle and guides the state update of the particle swarm until the optimal control scheme is obtained.

2. The urban heat island effect control method based on particle swarm optimization as described in claim 1, characterized in that, The overall fitness value includes the following sub-steps executed in sequence: The physical microclimate model is updated based on the urban heat island control scheme represented by the particles to generate a dynamic urban microclimate field. Based on the dynamic urban microclimate field, the human behavior dynamics model is driven to simulate the adaptive behavioral response of human groups to the urban microclimate field. The comprehensive fitness value is calculated based on the simulation results of the urban microclimate field and the adaptive behavioral response.

3. The urban heat island effect control method based on particle swarm optimization as described in claim 2, characterized in that, The human group in the human behavior dynamics model is represented by multiple agents. The agents make behavioral decisions by comparing the environmental thermal comfort assessed by the general thermal climate index with a preset comfort reference point, and by determining the agent's behavior selection based on an asymmetric decision rule that assigns different weights to situations above and below the reference point.

4. The urban heat island effect control method based on particle swarm optimization as described in claim 3, characterized in that, The metabolic heat generated by the intelligent agent during the simulation is accumulated, and the accumulated result is used as an artificial heat source term to calculate the energy balance of the corresponding spatial location in the physical microclimate model.

5. The urban heat island effect control method based on particle swarm optimization as described in claim 1 or 2, characterized in that, The overall fitness value is generated through a comprehensive evaluation of the following indicators: The physical cooling index of the control scheme in terms of its effectiveness in reducing urban surface or air temperature. A group comfort index representing the degree of thermal comfort experienced by a human group in a simulation; Spatial vitality index of the degree to which public spaces are effectively utilized.

6. The urban heat island effect control method based on particle swarm optimization as described in claim 5, characterized in that, The calculation of the group comfort index includes not only the assessment of the overall average duration of thermal discomfort experienced by all agents, but also the assessment of the differences in the distribution of the duration of thermal discomfort among different agents.

7. The urban heat island effect control method based on particle swarm optimization as described in claim 5, characterized in that, The calculation of the spatial vitality index includes: assessing the spatiotemporal distribution balance of the activity density of the agent in each public space driven by the agent's behavioral decisions, as well as the path selection cost of the agent from its location to the preset comfort zone.

8. The urban heat island effect control method based on particle swarm optimization as described in claim 1, characterized in that, The physical microclimate model operates based on solving the surface energy balance equation. The calculation of the balance equation includes the solar shading effect and sky visibility factor determined by the three-dimensional shape of the building, and calculates the sensible and latent heat fluxes through a mechanism model based on aerodynamic impedance and surface impedance.

9. The urban heat island effect control method based on particle swarm optimization as described in claim 1, characterized in that, The method further replaces the sub-step of calculating the comprehensive fitness value in some iterations by constructing and using a surrogate model, and selectively executes the sub-step based on the prediction uncertainty of the surrogate model.

Images