Intelligent decision-making method, device and equipment for old cell environment transformation and medium

Abstract

The invention relates to an intelligent decision-making method and device for old cell environment transformation, equipment and a medium. The method comprises the following steps: acquiring basic data of an old community through multi-source data acquisition and integration, and constructing a three-dimensional visual community environment information model in combination with historical data and current situation investigation and survey data; secondly, according to a Maslow demand hierarchy theory, carrying out multi-level division on the reconstruction demands of the residents in the old community to form a demand classification result set; then, combining the cell environment information model with the demand classification result set, iteratively adjusting spatial layout parameters by applying a simulated annealing algorithm, and generating a preliminary environment transformation planning scheme; and finally, by constructing a comprehensive evaluation system including indexes such as resource consumption, ecological influence and economic cost, performing quantitative evaluation on the scheme, and determining an optimal reconstruction scheme. By adopting the method, the scientization and systematization of the old cell environment reconstruction decision are realized, and the decision efficiency and the implementation effect of the old cell environment reconstruction are effectively improved.

Description

Technical Field

[0001] The present invention belongs to the field of computer science and technology, and in particular relates to intelligent decision-making methods, devices, equipment and media for environmental renovation of old residential areas. Background Art

[0002] As urbanization progresses, older residential communities, due to their age, are plagued by aging buildings, outdated facilities, and irrational functional layouts, making them unable to meet the needs of residents and the demands of urban development. Traditional renovation methods rely heavily on manual experience and lack systematic data support and quantitative analysis, making it difficult for renovation plans to balance resident needs, environmental benefits, and economic costs. Furthermore, resident needs are multifaceted and diverse, and existing technologies struggle to effectively integrate spatial data such as topography, building distribution, and pipeline network layout with resident demand data. Summary of the Invention

[0003] Based on this, it is necessary to provide an intelligent decision-making method, device, equipment and medium for the environmental renovation of old communities, which can effectively improve the decision-making efficiency and implementation effect of the environmental renovation of old communities, in response to the above technical problems.

[0004] First, this application provides an intelligent decision-making method for environmental renovation of old residential areas, including:

[0005] Obtain basic data such as topography, building distribution, and pipeline network layout of old residential areas, combine historical information with current situation survey data, and construct a three-dimensional visual residential environment information model.

[0006] According to Maslow's hierarchy of needs theory, the needs of residents in old communities for the renovation of the community environment are divided into multiple levels, and a demand classification result set is generated.

[0007] Combining the residential area environmental information model with the demand classification result set, the simulated annealing algorithm is used to iteratively adjust the spatial layout parameters to generate a preliminary environmental transformation planning scheme.

[0008] Construct a comprehensive evaluation system to conduct quantitative assessment of environmental transformation planning schemes and obtain the optimal transformation plan.

[0009] In one embodiment, constructing a three-dimensional visualized cell environment information model includes:

[0010] Surface morphological characteristic parameters are generated based on multi-source terrain elevation data of old communities; the surface morphological characteristic parameters include slope distribution and drainage path.

[0011] Establish topological connection relationships based on the buried depth and pipe diameter data of the pipeline network layout.

[0012] The surface morphological characteristic parameters and topological connection relationships are integrated to generate a three-dimensional pipe network pressure simulation model.

[0013] The coordinate space of the building distribution is spatially aligned with the pipe network pressure simulation model to obtain a registered three-dimensional environmental data set.

[0014] A multi-layer grid model is constructed based on a three-dimensional environmental data set; the multi-layer grid model includes a surface grid layer and an underground pipe network layer.

[0015] Update building material parameters in multi-layer mesh models based on historical building renovation records.

[0016] The pressure distribution results of the pipeline network layer are corrected based on the sensor monitoring data from the current situation survey.

[0017] The implicit surface algorithm is used to fuse the building material parameters with the corrected pressure distribution results to generate a three-dimensional visual community environment information model.

[0018] In one embodiment, the needs of residents in old residential communities for renovation of their residential environments are divided into multiple levels according to Maslow's hierarchy of needs theory, and a demand classification result set is generated, including:

[0019] Obtain text data on renovation needs of residents in old residential areas.

[0020] Keyword extraction is performed based on the transformation demand text data to generate a structured demand feature vector; the demand feature vector contains quantitative indicators of multi-level needs such as physiological needs, safety needs, social needs, respect needs and self-realization needs.

[0021] The structured demand feature vector is input into the trained transformation demand classification model, and the hierarchical clustering algorithm is used to output a multi-level demand label set.

[0022] The demand label set is matched with the Maslow hierarchy mapping table to generate a demand classification result set.

[0023] In one embodiment, the demand feature vector is calculated using the following formula:

[0024]

[0025] Among them, V j Indicates the corresponding demand level c j The required feature vector, σ represents the activation function, Indicates the need level c j The weight vector transpose, A·e i Indicates the use of attention mechanism to calculate word ω i The degree of association with each demand level, assuming that the demand text is T = {ω1,ω2,..,ω n}, n represents the total number of words in the demand text T, ω irepresents the i-th word in the text, e i Represents word vector, e i =E(ω i ), E represents the pre-trained word embedding model, C represents the demand hierarchy category set, C={c1,c2,..,c m}, c j Represents a level in Maslow's hierarchy of needs, and the corresponding weight vector is α j , m represents the number of categories in Maslow’s hierarchy of needs.

[0026] In one embodiment, a preliminary environmental renovation plan is generated by combining a residential area environment information model with a demand classification result set and using a simulated annealing algorithm to iteratively adjust spatial layout parameters, including:

[0027] Obtain topological structure data of the community environment information model; the topological structure data includes building coordinates and vegetation coverage parameters.

[0028] The priority weights of the requirement classification result set are mapped to the topological structure data in a matrix to generate an initial set of layout parameters with constraints.

[0029] The disturbance amplitude of the initial set of layout parameters is adjusted according to the temperature attenuation coefficient of the simulated annealing algorithm to generate a neighborhood solution set including path width and facility spacing.

[0030] Calculate the dynamic energy consumption index and spatial coverage of the neighborhood solution set; the dynamic energy consumption index is obtained based on light intensity and ventilation path length.

[0031] The Pareto front screening strategy is used to perform multi-objective optimization of dynamic energy consumption indicators and spatial coverage, generating an optimization parameter sequence including facility distribution density and green belt area.

[0032] The spatial layout is rendered according to the optimized parameter sequence, and a preliminary environmental transformation planning scheme with elevation annotation and functional zoning is obtained.

[0033] In one embodiment, a comprehensive evaluation system is constructed to quantitatively evaluate environmental renovation plans and obtain the optimal renovation plan, including:

[0034] Obtain multi-dimensional indicator data for environmental transformation planning schemes; multi-dimensional indicator data include resource consumption rate, ecological impact index and economic cost parameters.

[0035] The scheme feature vector is constructed based on multi-dimensional indicator data, and the resource constraint matrix is ​​used to screen the scheme feature vector; the resource constraint matrix includes regional environmental capacity thresholds and policy compliance conditions.

[0036] The filtered scheme feature vector is input into the multi-objective optimization function, and the multi-objective optimization function is iteratively calculated through the gradient descent algorithm to maximize its value, and the comprehensive evaluation value of the corresponding scheme is obtained.

[0037] Generate ranking results based on the comprehensive evaluation values, and determine the optimal transformation plan based on the ranking results.

[0038] In one embodiment, the multi-objective optimization function is expressed using the following formula:

[0039]

[0040] Among them, F(s i ) represents the multi-objective optimization function, s i represents the i-th solution, X i Representation schemes i The multi-dimensional indicator data constitutes a vector, X i =[r i ,e i ,d i ],r i Representation schemes i Resource consumption rate, e i Representation schemes i Ecological impact index, d i Representation schemes i The economic cost parameter, ρ a , ρ b , ρ c Represent the weights of resource consumption rate, ecological impact index and economic cost parameter, respectively. a (r i ) represents the utility function of resource consumption rate, f b (e i ) represents the normalized function of the ecological impact index, f c (d i ) represents the normalized function of economic cost parameters, g i (X i ) represents scheme s i The jth resource consumption or ecological impact indicator, R j represents the corresponding regional environmental capacity threshold, h k (X i ) represents scheme s i is the kth policy compliance indicator, and σ represents the activation function.

[0041] Secondly, the present application also provides an intelligent decision-making device for environmental renovation of old residential areas, which includes:

[0042] The data modeling module is used to obtain basic data such as the topography, building distribution, and pipeline network layout of old residential communities, and to build a three-dimensional visual community environment information model by combining historical data with current situation survey data.

[0043] The demand analysis module is used to divide the renovation needs of residents in old communities for the community environment into multiple levels according to Maslow's hierarchy of needs theory, and generate a demand classification result set.

[0044] The scheme generation module is used to combine the community environmental information model with the demand classification result set, use the simulated annealing algorithm to iteratively adjust the spatial layout parameters, and generate a preliminary environmental renovation planning scheme; it is also used to build a comprehensive evaluation system to quantitatively evaluate the environmental renovation planning scheme and obtain the optimal renovation plan.

[0045] In a third aspect, the present application further provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the above method when executing the computer program.

[0046] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, which implements the above method when executed by a processor.

[0047] The above-mentioned intelligent decision-making method, device, computer equipment, and storage medium for environmental renovation of old residential communities acquire and integrate basic data such as topography, building distribution, and pipeline network layout of old residential communities through multi-source data collection and integration. Combining historical data with current situation survey data, this method constructs a three-dimensional visualized residential community environmental information model. Secondly, based on Maslow's hierarchy of needs theory, the renovation needs of residents in old residential communities are divided into multiple levels, forming a structured demand classification result set. Then, the residential community environmental information model and the demand classification result set are combined, and a simulated annealing algorithm is used to iteratively adjust spatial layout parameters to generate a preliminary environmental renovation plan. Finally, a comprehensive evaluation system is constructed that includes indicators such as resource consumption, ecological impact, and economic cost to quantitatively evaluate the plans and determine the optimal renovation plan. This method combines multi-source data processing, needs hierarchy analysis, intelligent algorithm optimization, and quantitative evaluation to achieve scientific and systematic decision-making for environmental renovation of old residential communities, effectively improving the decision-making efficiency and implementation effectiveness of environmental renovation in old residential communities. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the following briefly introduces the drawings required for use in the embodiments or related technical descriptions. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0049] Figure 1 A flowchart of an intelligent decision-making method for environmental renovation of old residential areas provided by an embodiment of the present invention;

[0050] Figure 2 This is a structural block diagram of the intelligent decision-making device for environmental renovation of old residential communities provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0051] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0052] In one embodiment, Figure 1 As shown, this application provides an intelligent decision-making method for environmental renovation of old residential areas, which may include the following steps:

[0053] Step S101: Obtain basic data such as topography, building distribution, and pipe network layout of old residential areas, combine historical data with current situation survey data, and construct a three-dimensional visual residential area environment information model.

[0054] Specifically, using geographic information systems (GIS), building information models (BIM), and drone mapping technology, we collected basic data such as topographic elevation data for older residential communities, building coordinates and structural parameters, and the depth and connectivity of underground pipeline networks. We also collected historical data such as completed residential area drawings and renovation records. Combined with on-site photos and sensor monitoring data obtained during current situation surveys, after data cleaning, coordinate unification, and format conversion, we used 3D modeling algorithms to construct a visual information model encompassing elements such as surface morphology, building entities, and underground pipeline networks.

[0055] In step S102 , the renovation needs of residents of old residential communities for the residential community environment are divided into multiple levels according to Maslow's hierarchy of needs theory, and a demand classification result set is generated.

[0056] Specifically, online questionnaires, offline interviews, and community forum data capture were used to capture the textual demands of residents of older residential communities for environmental improvements. Natural language processing techniques were used to segment the text, perform part-of-speech tagging, and extract keywords. Combined with a pre-trained word embedding model, the text was converted into structured feature vectors. Based on Maslow's hierarchy of needs, a machine learning classification algorithm was used to categorize the needs into five levels: physiological, safety, social, respect, and self-actualization. Correspondence between these needs and spatial locations and facility types was established, ultimately generating a classification result set containing need descriptions, priorities, and spatial orientations.

[0057] Step S103 , combining the community environment information model with the demand classification result set, using a simulated annealing algorithm to iteratively adjust the spatial layout parameters, and generate a preliminary environmental renovation planning scheme.

[0058] The topological structure data (such as building spacing, road orientation, and green space extent) in the three-dimensional environmental information model is mapped to the demand classification result set to form an initial solution for spatial layout parameters with demand constraints. A simulated annealing algorithm is used to dynamically adjust the temperature parameter to control the solution space search range, iteratively optimizing layout parameters such as road width, facility location, and functional zoning boundaries. During the iterative process, performance indicators such as ventilation efficiency and sunshine duration are calculated using tools such as CFD fluid simulation and lighting analysis. Multi-objective screening is performed in conjunction with Pareto frontier theory, ultimately generating a preliminary environmental renovation plan that meets functional requirements, spatial rationality, and performance indicators.

[0059] Step S104: construct a comprehensive evaluation system to quantitatively evaluate the environmental renovation plan and obtain the optimal renovation plan.

[0060] Specifically, a comprehensive evaluation system was constructed, encompassing resource consumption (building materials usage, energy consumption indicators), ecological impact (green space ratio, carbon emission reduction), economic costs (construction investment, operation and maintenance costs), construction feasibility (construction period, technical difficulty), and resident satisfaction predictions. The weights of each indicator were determined using the Analytic Hierarchy Process (AHP), and the quantitative data of the preliminary plan was converted into eigenvectors. Compliance screening was performed using a resource constraint matrix (including environmental capacity thresholds and planning regulations and standards). For the selected plans, a gradient descent algorithm was used to solve the multi-objective optimization function. Iterative calculations were performed with the goal of maximizing the comprehensive evaluation value. The optimal renovation plan, which balances economic, ecological, and social benefits, was determined based on the final score ranking.

[0061] The above-mentioned intelligent decision-making method for environmental renovation of old residential communities collects and integrates multi-source data to obtain basic data such as the old residential community's topography, building distribution, and pipeline network layout. Combining historical data with current situation survey data, it constructs a three-dimensional visual community environmental information model. Secondly, based on Maslow's hierarchy of needs theory, it divides the renovation needs of residents in old residential communities into multiple levels, forming a structured demand classification result set. Then, the community environmental information model and the demand classification result set are combined, and a simulated annealing algorithm is used to iteratively adjust spatial layout parameters to generate a preliminary environmental renovation plan. Finally, by constructing a comprehensive evaluation system that includes indicators such as resource consumption, ecological impact, and economic cost, the plans are quantitatively evaluated to determine the optimal renovation plan. This method combines multi-source data processing, needs hierarchy analysis, intelligent algorithm optimization, and quantitative evaluation to achieve scientific and systematic decision-making for environmental renovation of old residential communities, effectively improving the decision-making efficiency and implementation effectiveness of environmental renovation in old residential communities.

[0062] In one embodiment, constructing a three-dimensional visualized cell environment information model may include the following steps:

[0063] Step S201 : generating surface morphological characteristic parameters based on multi-source terrain elevation data of old residential areas; the surface morphological characteristic parameters include slope distribution and drainage path.

[0064] Step S202: establishing a topological connection relationship based on the buried depth and pipe diameter data of the pipe network layout.

[0065] Step S203: The surface morphological characteristic parameters and the topological connection relationship are integrated to generate a three-dimensional pipe network pressure simulation model.

[0066] Step S204 : spatially registering the building distribution coordinate space with the pipe network pressure simulation model to obtain a registered three-dimensional environmental data set.

[0067] Step S205 , constructing a multi-layer grid model based on the three-dimensional environmental data set; the multi-layer grid model includes a surface grid layer and an underground pipe network layer.

[0068] Step S206 : updating the building material parameters in the multi-layer grid model according to the building renovation records in the historical data.

[0069] Step S207: Correct the pressure distribution result of the pipe network layer in combination with the sensor monitoring data in the current situation investigation.

[0070] Step S208: Using an implicit surface algorithm, the building material parameters are integrated with the corrected pressure distribution results to generate a three-dimensional visual community environment information model.

[0071] Furthermore, multi-source terrain elevation data for older residential communities was collected. Data processing and analysis generated surface morphological parameters, including slope distribution and drainage paths. Simultaneously, graph theory algorithms were used to establish pipeline network topological connectivity based on pipeline network layout data such as burial depth and pipe diameter. Subsequently, these surface morphological parameters were integrated with the pipeline network topology, and a three-dimensional pipeline network pressure simulation model was constructed using fluid mechanics principles. Next, through spatial coordinate transformation and error correction, the coordinate space of the building distribution was precisely aligned with the pipeline network pressure simulation model, forming a unified three-dimensional environmental data set. Based on this data set, a spatial segmentation algorithm was used to construct a multi-layered grid model consisting of a surface grid layer and an underground pipeline network layer. Building material parameters in the model were then updated based on historical building renovation records, and the pressure distribution results for the pipeline network layer were corrected using sensor monitoring data obtained from current status surveys. Finally, an implicit surface algorithm was used to integrate the building material parameters with the corrected pressure distribution results, generating a three-dimensional visual community environmental information model with spatial topology and physical properties.

[0072] This embodiment achieves refined modeling of terrain, pipeline networks, buildings, and other elements of older residential communities through the integration of multi-source data and cross-domain technology. Compared to traditional single-data modeling approaches, this approach offers the advantage of truly reflecting the spatial structure and physical characteristics of residential communities. The three-dimensional visualization model provides an intuitive representation of the current state of the community, assisting planners in spatial analysis and decision-making. The integration of physical properties such as pipeline pressure simulation and material parameters also provides a scientific basis for infrastructure renovation and environmental performance assessment, improving the accuracy and reliability of decisions regarding the renovation of older residential communities.

[0073] In one embodiment, the needs of residents of old residential communities for renovation of the residential environment are divided into multiple levels according to Maslow's hierarchy of needs theory, and a needs classification result set is generated, which may include the following steps:

[0074] Step S301: Obtain text data on renovation needs of residents in old residential areas.

[0075] Step S302 , extracting keywords from the transformation demand text data to generate a structured demand feature vector; the demand feature vector includes quantitative indicators of multiple levels of demand, including physiological needs, safety needs, social needs, respect needs, and self-actualization needs.

[0076] Step S303: Input the structured demand feature vector into the trained transformation demand classification model, and use the hierarchical clustering algorithm to output a multi-level demand label set.

[0077] Step S304: Match the Maslow hierarchy mapping table according to the demand label set to generate a demand classification result set.

[0078] First, through multiple channels, including online questionnaires, offline home interviews, and community forums, we extensively collected text data on the environmental renovation needs of residents in older communities. This data was preprocessed using the TF-IDF algorithm combined with the Word2Vec word embedding model to extract high-frequency and representative keywords. Based on the semantic associations of these keywords across the various levels of Maslow's hierarchy of needs, we constructed a structured feature vector encompassing physiological, safety, social, respect, and self-actualization needs, thereby achieving a quantitative representation of residents' needs. Secondly, we fed these structured need feature vectors into a pre-trained deep learning classification model. Using a hierarchical clustering algorithm, we clustered similar needs and output a set of hierarchically related need labels. Finally, based on a pre-set Maslow hierarchy mapping table, we matched the need labels to their corresponding need hierarchies, generating a structured set of need classification results that clearly defined the hierarchical attribution and priority of different needs.

[0079] This example uses natural language processing and machine learning technologies to transform residents' renovation demands from unstructured text into structured data. Compared to traditional manual summarization methods, this method can quickly and accurately identify and classify massive amounts of demand information, avoiding subjective judgment bias. The quantified demand feature vectors and hierarchical classification results not only intuitively present residents' core demands but also form data linkages with subsequent renovation plan designs, making plans more aligned with actual needs and improving the pertinence and effectiveness of decisions regarding renovations of older residential communities.

[0080] In one embodiment, the demand feature vector can be calculated using the following formula:

[0081]

[0082] Among them, V j Indicates the corresponding demand level c j The required feature vector, σ represents the activation function, Indicates the need level c j The weight vector transpose, A·e i Indicates the use of attention mechanism to calculate word ω i The degree of association with each demand level, assuming that the demand text is T = {ω1,ω2,..,ω n}, n represents the total number of words in the demand text T, ω i represents the i-th word in the text, e i Represents word vector, e i =E(ω i ), E represents the pre-trained word embedding model, C represents the demand hierarchy category set, C={c1,c2,..,c m}, c j Represents a level in Maslow's hierarchy of needs, and the corresponding weight vector is α j , m represents the number of categories in Maslow’s hierarchy of needs.

[0083] This example systematically quantifies the relationship between demand text and Maslow's hierarchy of needs by integrating word embedding, an attention mechanism, and a weighted summation strategy. This method dynamically captures the contribution of words in the text to different levels of needs, avoiding misjudgments of needs caused by single keywords. The quantified feature vector accurately reflects the hierarchical distribution of residents' needs, providing standardized data input for subsequent demand classification and renovation plan design. This improves the objectivity and accuracy of demand analysis during the decision-making process, ensuring that renovation plans for older residential communities better align with residents' actual needs.

[0084] In one embodiment, combining the community environment information model with the demand classification result set, using a simulated annealing algorithm to iteratively adjust spatial layout parameters to generate a preliminary environmental transformation plan, may include the following steps:

[0085] Step S401: Acquire topological structure data of a cell environment information model; the topological structure data includes building coordinates and vegetation coverage parameters.

[0086] Step S402 : Matrix mapping is performed between the priority weights of the demand classification result set and the topology structure data to generate an initial set of layout parameters with constraint conditions.

[0087] Step S403 : adjusting the disturbance amplitude of the initial set of layout parameters according to the temperature attenuation coefficient of the simulated annealing algorithm to generate a neighborhood solution set including the path width and the facility spacing.

[0088] Specifically, after obtaining an initial set of layout parameters with constraints, the temperature attenuation coefficient is used to dynamically control the amplitude of parameter perturbations based on the principles of the simulated annealing algorithm. In the simulated annealing algorithm, temperature serves as a key control variable. Initially, higher temperatures allow for larger parameter perturbations, enabling the algorithm to conduct a broad search within the solution space and avoid falling into local optima. As iterations proceed, the temperature gradually decreases according to the preset attenuation coefficient, reducing the perturbation amplitude and allowing the algorithm to focus on a detailed search near the current optimal solution. During this process, random perturbations are applied to variables in the initial set of layout parameters, such as path width and facility spacing, to generate a series of neighborhood solution sets that satisfy the constraints.

[0089] Step S404, calculating the dynamic energy consumption index and space coverage of the neighborhood solution set; the dynamic energy consumption index is obtained based on the light intensity and the ventilation path length.

[0090] Step S405 , adopting the Pareto front screening strategy to perform multi-objective optimization on the dynamic energy consumption index and the spatial coverage rate, and generating an optimization parameter sequence including the facility distribution density and the green belt area.

[0091] Step S406: Perform spatial layout rendering according to the optimized parameter sequence to obtain a preliminary environmental renovation planning scheme with elevation annotations and functional zoning.

[0092] First, topological data, such as building coordinates and vegetation coverage, is extracted from the community environmental information model. Matrix operations are performed on the weights of the demand priorities at each level of the demand classification result set and the topological data to establish a correlation between demand priorities and spatial layout, generating an initial set of layout parameters with spatial constraints and demand guidance. Subsequently, the perturbation amplitude of the initial set is dynamically adjusted based on the temperature decay mechanism of the simulated annealing algorithm. A neighborhood solution set containing parameters such as path width and facility spacing is generated through random perturbations. Next, using illumination simulation and fluid dynamics analysis tools, the dynamic energy consumption index (based on parameters such as light intensity and ventilation path length) and spatial coverage of each solution in the neighborhood solution set are calculated. A Pareto front screening strategy is then used to optimize the dual objectives, retaining non-dominated solutions and generating an optimized sequence containing parameters such as facility distribution density and green belt area. Finally, based on the optimized parameter sequence, a preliminary environmental renovation plan with elevation annotations and functional zoning is generated using 3D rendering technology.

[0093] This example achieves scientific generation of renovation plans through the deep integration of demand and spatial data and intelligent algorithm optimization. Matrix mapping transforms abstract demands into concrete spatial constraints, ensuring that the plan meets the actual needs of residents. The global search capabilities of the simulated annealing algorithm and the Pareto frontier strategy are leveraged to balance multiple objectives, such as dynamic energy consumption and spatial coverage, improving the plan's environmental performance and spatial efficiency. Three-dimensional rendering technology makes planning results intuitive and judicious, effectively enhancing the rationality and feasibility of renovation plans for older residential areas.

[0094] In one embodiment, constructing a comprehensive evaluation system to quantitatively evaluate environmental renovation plans and obtain the optimal renovation plan may include the following steps:

[0095] Step S501: Acquire multi-dimensional indicator data of the environmental transformation planning scheme; the multi-dimensional indicator data includes resource consumption rate, ecological impact index and economic cost parameters.

[0096] Step S502 : constructing a solution feature vector based on the multi-dimensional indicator data, and screening the solution feature vector using a resource constraint matrix; the resource constraint matrix includes regional environmental capacity thresholds and policy compliance conditions.

[0097] In step S503 , the filtered solution feature vector is input into the multi-objective optimization function, and the multi-objective optimization function is iteratively calculated by the gradient descent algorithm to maximize its value, thereby obtaining a comprehensive evaluation value of the corresponding solution.

[0098] Furthermore, the selected scheme feature vectors are input into a multi-objective optimization function, which integrates multiple evaluation dimensions such as resource consumption rate, ecological impact index, and economic cost parameters, and performs a weighted summation of each indicator using pre-set weights. On this basis, a gradient descent algorithm is used for iterative calculations. This algorithm calculates the gradient of the objective function under the current parameters and adjusts the parameters along the negative direction of the gradient to gradually reduce the function value (because the goal is to maximize the comprehensive evaluation value, it is actually converted into a minimization problem by taking the negative gradient). Each iteration updates the parameters in the scheme feature vector until the multi-objective optimization function converges to its maximum value. Ultimately, a comprehensive evaluation value corresponding to each scheme is obtained, which quantitatively reflects the overall performance of the scheme under the multi-objective balance.

[0099] Step S504: generating a ranking result according to the comprehensive evaluation value, and determining the optimal transformation plan based on the ranking result.

[0100] First, collect quantitative data on environmental transformation planning schemes in multiple dimensions, such as resource consumption rate, ecological impact index, and economic cost parameters. These data are constructed into scheme feature vectors. Subsequently, the scheme feature vectors are screened through a resource constraint matrix that includes regional environmental capacity thresholds (such as carbon emission caps, water resource consumption limits) and policy compliance conditions (such as planning regulations and construction standards), and schemes that do not meet the basic conditions are eliminated. Secondly, the screened scheme feature vectors are substituted into the multi-objective optimization function, and the gradient descent algorithm is used for iterative calculation. By continuously adjusting the parameters to maximize the function value, the comprehensive evaluation value corresponding to each scheme is obtained. Finally, all schemes are ranked according to the comprehensive evaluation value, and the scheme with the highest evaluation value is selected as the optimal transformation scheme to achieve scheme decision-making under multi-objective balance.

[0101] This embodiment is data-driven and incorporates core elements such as resource consumption, ecological benefits, and economic costs into quantitative assessments to avoid subjective bias; it uses a resource constraint matrix to ensure that the plan complies with environmental carrying capacity and policy regulations; through multi-objective optimization functions and gradient descent algorithms, it automatically seeks the best solution among complex objectives and accurately screens out the optimal solution that takes into account economy, ecology, and feasibility, providing data support and scientific basis for the renovation of old communities.

[0102] In one embodiment, the multi-objective optimization function can be expressed using the following formula:

[0103]

[0104] Among them, F(s i ) represents the multi-objective optimization function, s i represents the i-th solution, X i Representation schemes i The multi-dimensional indicator data constitutes a vector, X i =[ri ,e i ,d i ],r i Representation schemes i Resource consumption rate, e i Representation schemes i Ecological impact index, d i Representation schemes i The economic cost parameter, ρ a , ρ b , ρ c Represent the weights of resource consumption rate, ecological impact index and economic cost parameter, respectively. a (r i ) represents the utility function of resource consumption rate, f b (e i ) represents the normalized function of the ecological impact index, f c (d i ) represents the normalized function of economic cost parameters, g i (X i ) represents scheme s i The jth resource consumption or ecological impact indicator, R j represents the corresponding regional environmental capacity threshold, h k (X i ) represents scheme s i is the kth policy compliance indicator, and σ represents the activation function.

[0105] Preferably, Among them, β r represents the adjustment coefficient, Indicates the ideal resource consumption rate benchmark value, r i Indicates the resource consumption rate.

[0106] Preferably, Among them, min(e) represents the minimum value of the ecological impact index among all options, and max(e) represents the maximum value of the ecological impact index among all options.

[0107] Preferably, Among them, max(d) represents the maximum economic cost among all options, and min(d) represents the minimum economic cost among all options.

[0108] This multi-objective optimization function integrates multi-dimensional evaluation indicators, environmental constraints, and policy compliance checks to construct a systematic scheme evaluation system. It integrates core objectives such as resources, ecology, and economy through weighted summation to avoid biased decision-making. It also introduces environmental capacity thresholds and policy compliance conditions as constraints to ensure scheme feasibility. It also uses functional transformation to unify indicators of different dimensions, enabling objective comparisons between schemes. The gradient descent algorithm, combined with iterative calculations of this function, efficiently identifies the optimal scheme for overall performance, providing a scientific and quantitative basis for decision-making in the renovation of older residential communities, improving resource utilization efficiency and overall project benefits.

[0109] In one embodiment, Figure 2 As shown, the present application also provides an intelligent decision-making device for environmental renovation of old residential areas, which may include:

[0110] The data modeling module 601 is used to obtain basic data such as the topography, building distribution, and pipeline network layout of old residential areas, and to build a three-dimensional visual residential environment information model by combining historical data with current situation survey data.

[0111] The demand analysis module 602 is used to divide the renovation needs of residents in old residential areas into multiple levels according to Maslow's hierarchy of needs theory for the residential area environment, and generate a demand classification result set.

[0112] The scheme generation module 603 is used to combine the community environmental information model with the demand classification result set, use the simulated annealing algorithm to iteratively adjust the spatial layout parameters, and generate a preliminary environmental renovation planning scheme; it is also used to build a comprehensive evaluation system to quantitatively evaluate the environmental renovation planning scheme and obtain the optimal renovation scheme.

[0113] The aforementioned intelligent decision-making device for environmental renovation of old residential communities uses a data modeling module that collects basic data such as topography, building distribution, and pipeline network layout, combines historical data with current situation survey information, and uses geographic information systems (GIS) and three-dimensional modeling technology to construct a visual model of the community's environmental information. The demand analysis module, based on Maslow's hierarchy of needs theory, performs semantic analysis and feature extraction on residents' renovation demand texts, using machine learning algorithms to divide the demand hierarchy and form a structured demand classification result set. The solution generation module integrates the environmental information model with the demand classification results, optimizes spatial layout parameters using a simulated annealing algorithm, generates preliminary renovation plans, and constructs a comprehensive evaluation system that includes indicators such as resource consumption, ecological impact, and economic cost, selecting the optimal plan through quantitative evaluation. This device combines multi-source data processing, demand hierarchy analysis, intelligent algorithm optimization, and quantitative evaluation to achieve scientific and systematic decision-making for environmental renovation of old residential communities, effectively improving the decision-making efficiency and implementation effectiveness of environmental renovation in old residential communities.

[0114] It should be understood that, although the various steps in the flowcharts involved in the various embodiments described above are displayed in sequence according to the instructions of the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and these steps can be executed in other orders. Moreover, at least a portion of the steps in the flowcharts involved in the various embodiments described above can include multiple steps or multiple stages, and these steps or stages are not necessarily executed and completed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a portion of steps or stages in other steps.

[0115] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of the intelligent decision-making method, device, equipment and medium for environmental renovation of old communities as described above are implemented.

[0116] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0117] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to the partial description of the method embodiments. The device embodiments described above are merely illustrative, wherein the components described as separate parts may or may not be physically separated, and the parts displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the disclosed solution. A person of ordinary skill in the art can understand and implement it without expending creative work.

[0118] The above-described embodiments merely represent several implementation methods of the embodiments of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the patent application. It should be noted that a person skilled in the art may make various modifications and improvements without departing from the concept of the embodiments of the present application, and these modifications and improvements fall within the scope of protection of the embodiments of the present application.

Claims

1. An intelligent decision-making method for environmental renovation of old residential areas, characterized by: The method comprises: Obtain basic data on topography, building distribution, pipe network layout, etc. of old residential areas, combine historical data with current situation survey data, and build a three-dimensional visual residential environment information model; According to Maslow's hierarchy of needs theory, the needs of residents in old residential areas for renovation of the residential environment are divided into multiple levels, and a demand classification result set is generated; Combining the residential area environment information model with the demand classification result set, using a simulated annealing algorithm to iteratively adjust spatial layout parameters to generate a preliminary environmental renovation planning scheme; Construct a comprehensive evaluation system to quantitatively evaluate the environmental renovation plan and obtain the optimal renovation plan.

2. The method according to claim 1, characterized in that The constructing of a three-dimensional visualized cell environment information model includes: Generating surface morphological characteristic parameters based on the multi-source terrain elevation data of the old residential area; the surface morphological characteristic parameters include slope distribution and drainage path; Establishing a topological connection relationship based on the buried depth and pipe diameter data of the pipe network layout; Fusion of the surface morphological characteristic parameters and the topological connection relationship to generate a three-dimensional pipe network pressure simulation model; Performing spatial registration on the coordinate space of the building distribution and the pipe network pressure simulation model to obtain a registered three-dimensional environmental data set; Constructing a multi-layer grid model based on the three-dimensional environmental data set; the multi-layer grid model includes a surface grid layer and an underground pipe network layer; updating building material parameters in the multi-layer grid model according to the building renovation records in the historical data; Correct the pressure distribution results of the pipe network layer based on the sensor monitoring data in the current situation survey; The building material parameters are fused with the corrected pressure distribution results using an implicit surface algorithm to generate a three-dimensional visual community environment information model.

3. The method according to claim 1, characterized in that According to Maslow's hierarchy of needs theory, the needs of residents in old residential areas for renovation of the residential environment are divided into multiple levels, generating a demand classification result set, including: Obtaining text data on renovation needs of residents in the old residential area; Extract keywords from the transformation demand text data to generate a structured demand feature vector; the demand feature vector includes quantitative indicators of multiple levels of needs, including physiological needs, safety needs, social needs, respect needs, and self-actualization needs; Inputting the structured demand feature vector into the trained transformation demand classification model, and using a hierarchical clustering algorithm to output a multi-level demand label set; The demand label set is matched with the Maslow hierarchy mapping table to generate a demand classification result set.

4. The method according to claim 3, characterized in that The demand feature vector is calculated using the following formula: Among them, V j Indicates the corresponding demand level c j The required feature vector, σ represents the activation function, Indicates the need level c j The weight vector transpose, A·e i Indicates the use of attention mechanism to calculate word ω i The degree of association with each demand level, assuming that the demand text is T = {ω1,ω2,..,ω n }, n represents the total number of words in the demand text T, ω i represents the i-th word in the text, e i Represents word vector, e i =E(ω i ), E represents the pre-trained word embedding model, C represents the demand hierarchy category set, C={c1,c2,..,c m }, c j Represents a level in Maslow's hierarchy of needs, and the corresponding weight vector is α j , m represents the number of categories in Maslow’s hierarchy of needs.

5. The method according to claim 1, wherein The method combines the cell environment information model with the demand classification result set, iteratively adjusts the spatial layout parameters using a simulated annealing algorithm, and generates a preliminary environmental transformation planning scheme, including: Acquiring topological structure data of the cell environment information model; the topological structure data includes building coordinates and vegetation coverage parameters; Performing matrix mapping on the priority weights of the demand classification result set and the topology structure data to generate an initial set of layout parameters with constraint conditions; adjusting the disturbance amplitude of the initial set of layout parameters according to the temperature attenuation coefficient of the simulated annealing algorithm to generate a neighborhood solution set including path width and facility spacing; Calculating a dynamic energy consumption index and a spatial coverage rate of the neighborhood solution set; the dynamic energy consumption index is obtained based on light intensity and ventilation path length; A Pareto frontier screening strategy is used to perform multi-objective optimization on the dynamic energy consumption index and spatial coverage, generating an optimization parameter sequence including facility distribution density and green belt area; The spatial layout is rendered according to the optimized parameter sequence to obtain a preliminary environmental renovation planning scheme with elevation annotations and functional zoning.

6. The method according to claim 1, wherein The comprehensive evaluation system is constructed to quantitatively evaluate the environmental renovation plan to obtain the optimal renovation plan, including: Obtaining multi-dimensional indicator data of the environmental transformation planning scheme; the multi-dimensional indicator data includes resource consumption rate, ecological impact index and economic cost parameters; Constructing a solution feature vector based on the multi-dimensional indicator data, and screening the solution feature vector using a resource constraint matrix; the resource constraint matrix includes regional environmental capacity thresholds and policy compliance conditions; Inputting the filtered solution feature vector into a multi-objective optimization function, iteratively calculating the multi-objective optimization function through a gradient descent algorithm to maximize its value, and obtaining a comprehensive evaluation value of the corresponding solution; A ranking result is generated according to the comprehensive evaluation value, and an optimal transformation plan is determined based on the ranking result.

7. The method according to claim 6, characterized in that The multi-objective optimization function is expressed using the following formula: Among them, F(s i ) represents the multi-objective optimization function, s i represents the i-th solution, X i Representation schemes i The multi-dimensional indicator data constitutes a vector, X i =[r i ,e i ,d i ],r i Representation schemes i Resource consumption rate, e i Representation schemes i Ecological impact index, d i Representation schemes i The economic cost parameter, ρ a , ρ b , ρ c Represent the weights of resource consumption rate, ecological impact index and economic cost parameter, respectively. a (r i ) represents the utility function of resource consumption rate, f b (e i ) represents the normalized function of the ecological impact index, f c (d i ) represents the normalized function of economic cost parameters, g i (X i ) represents scheme s i The jth resource consumption or ecological impact indicator, R j represents the corresponding regional environmental capacity threshold, h k (X i ) represents scheme s i is the kth policy compliance indicator, and σ represents the activation function.

8. Intelligent decision-making device for environmental renovation of old residential areas, characterized by: The device comprises: The data modeling module is used to obtain basic data such as the topography, building distribution, and pipe network layout of old residential areas, and to build a three-dimensional visual residential environment information model by combining historical data with current situation survey data; The demand analysis module is used to divide the needs of residents in old residential areas for community environment renovation into multiple levels according to Maslow's hierarchy of needs theory, and generate a demand classification result set; The scheme generation module is used to combine the community environment information model with the demand classification result set, use the simulated annealing algorithm to iteratively adjust the spatial layout parameters, and generate a preliminary environmental renovation planning scheme; it is also used to construct a comprehensive evaluation system to quantitatively evaluate the environmental renovation planning scheme and obtain the optimal renovation scheme.

9. A computer 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 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images