Evaluation model system and method for upgrading and reconstruction value of parking lot

Abstract

The invention relates to the technical field of data analysis, in particular to a parking lot upgrading and reconstruction value evaluation system and method, and the system comprises a data intelligent management module, an intelligent analysis and evaluation module, a decision support center, a visual interaction system, and a system management background. Compared with the prior art that a parking lot transformation scheme is screened only according to experience or a single index, subjectivity is high, evaluation is incomplete and the like, the scheme adopts a method of combining hard condition filtering with multi-target scoring and comprehensive scoring sorting, and the current parking lot condition is accurately matched, so that the accuracy of the parking lot transformation scheme is improved. According to the method, the economic benefit index, the policy matching degree and the social benefit are comprehensively considered, the Pareto optimal solution set is generated by applying the multi-objective optimization algorithm, sorting is adjusted in combination with the risk coefficient, scientific screening and efficient optimization of the transformation scheme are realized, and the method has the advantages of comprehensive evaluation, accurate decision and remarkable scheme optimization effect.

Description

Technical Field

[0001] The present invention relates to the technical field of data analysis, and particularly to an evaluation model system and method for the value of parking lot upgrade and transformation. Background Art

[0002] The screening of traditional parking lot renovation plans mainly relies on experience or a single indicator, and this method has obvious deficiencies. On the one hand, it ignores the comprehensive impacts of parking lot renovation on economy, policies, society, etc., resulting in incomplete evaluation results; on the other hand, the decision-making process is highly subjective and lacks a scientific and systematic evaluation system, making it difficult for renovation plans to achieve the optimal effect.

[0003] Specifically, the problems of the existing technology are reflected in: First, the evaluation indicators are single and cannot comprehensively reflect the comprehensive benefits of parking lot renovation; second, the decision-making process lacks objective basis and is easily affected by subjective judgments; finally, the plan optimization is insufficient and it is difficult to comprehensively compare and optimize the plan from multiple dimensions. These problems limit the scientific nature and feasibility of parking lot renovation projects and affect the efficiency and effect of urban parking management. Summary of the Invention

[0004] In order to overcome the problems proposed in the above background art, the present invention proposes an evaluation model system and method for the value of parking lot upgrade and transformation.

[0005] The technical solution of the present invention is: An evaluation model system for the value of parking lot upgrade and transformation, including:

[0006] A data intelligent management module, used for collecting, integrating, cleaning and feature extracting the data of the parking lot;

[0007] An intelligent analysis and evaluation module, used for processing and analyzing the data collected by the data intelligent management module and evaluating the value of the parking lot;

[0008] A decision support center, used for reporting output according to the data results of the intelligent analysis and evaluation module and generating an intelligent plan;

[0009] A visualization interaction system, used for three-dimensional visualization display of the output data of the decision support center;

[0010] A system management background, including a permission management function, a data tracking function and an operation audit function.

[0011] Preferably, the intelligent management module includes:

[0012] A11: A data collection module, used for connecting data sources including parking lot sensors, municipal platforms and third-party APIs;

[0013] A12: Federal learning center, used to achieve cross-parking lot data collaborative modeling;

[0014] A13: Data cleaning module, used to repair outliers and handle missing values in the obtained data;

[0015] A14: Feature engineering library, used to process the cleaned data and generate spatio-temporal features.

[0016] Preferably, when the intelligent analysis and evaluation module processes and analyzes the data collected by the data intelligent management module and evaluates the value of the parking lot, it specifically includes:

[0017] S11: Dynamic prediction, input historical traffic flow data, weather forecasts, and surrounding commercial activity schedules, process the input data, and predict the traffic flow heat map by time period;

[0018] S12: Multidimensional value calculation, input the preset renovation plan parameters, and predict the economic value, social value, and policy value of the parking lot upgrade and renovation based on the traffic flow heat map by time period and the renovation plan parameters;

[0019] S13: Scheme optimization, combine the pre-stored renovation reference schemes in the database, optimize and recommend the renovation schemes, and perform comprehensive scoring and ranking.

[0020] Preferably, when combining the pre-stored renovation reference schemes in the database, optimizing and recommending the renovation schemes, and performing comprehensive scoring and ranking, it specifically includes:

[0021] S21: Hard condition filtering, read all the renovation reference schemes, filter the renovation reference schemes according to the hard conditions of the current parking lot and the renovation plan parameters, and filter out the renovation reference schemes that do not meet the existing conditions;

[0022] S22: Multi-objective scoring, calculate the economic benefit index, policy matching degree, and social benefit for each of the remaining renovation reference schemes separately, and perform weighted integration on the calculated economic benefit index, policy matching degree, and social benefit;

[0023] S23: Comprehensive scoring and ranking, use the multi-objective optimization algorithm to generate the Pareto optimal solution set, and adjust the ranking in combination with the risk coefficient, and output the ranked schemes.

[0024] Preferably, when calculating the economic benefit index, policy matching degree, and social benefit for each of the remaining renovation reference schemes separately, the principle formula is:

[0025] Economic benefit index:

[0026]

[0027] f3 = 0.6ΔC + 0.4K;

[0028] Among them, f1 is the economic benefit index, NPV is the net present value, and the calculation formula is the sum of discounted future cash flows minus the initial investment. The industry benchmark value is the average NPV of parking lot projects of the same type; f2 is the policy matching degree, ω k is the weight of the k-th policy, Compliance k is the compliance with the k policies, m is the total number of policies to be met, f3 is the comprehensive social benefit score, ΔC is the estimated annual reduction in carbon emissions, and K is the congestion mitigation coefficient.

[0029] Preferably, when performing weighted synthesis on the calculated economic benefit index, policy matching degree, and social benefits, the principle formula is:

[0030]

[0031] Among them, Score i is the comprehensive score of the i-th plan, W j is the weight of the j-th objective. Among them, for the economic benefit index, W1 = 0.5, for the policy matching degree, W2 = 0.3, and for the social benefit, W3 = 0.2, f j max is the maximum value of the j-th objective among all plans, f j min is the minimum value of the j-th objective among all plans.

[0032] Preferably, when adjusting the ranking by combining the risk coefficient, the principle formula is:

[0033] Adj payback = Payback × (1 + risk coefficient);

[0034] Adj payback ≤ threshold;

[0035]

[0036] Among them, Adj payback is the investment payback period, Payback is the ideal investment payback period, s a represents Plan a, s b represents Plan b, f i (s a ) represents the score of Plan a on the i-th objective, f i (s b ) represents the score of Plan b on the i-th objective, f j (s a ) represents the score of Plan a on the i-th objective, f j(s b ) represents the score of Plan b on the jth objective.

[0037] Preferably, when the decision support center outputs a report and generates an intelligent plan based on the data results of the intelligent analysis and evaluation module, it specifically includes:

[0038] S31: Intelligent plan generation. After inputting the parking lot CAD drawings, the owner's budget, and policy requirements, the building structure is analyzed by AI to automatically generate 3 - 5 feasible renovation plans. Based on the space optimization algorithm, the maximum parking space capacity and investment range are calculated, and finally, a three - dimensional interactive model and a key parameter table are output;

[0039] S32: Dynamic risk scanning. After inputting the real - time updated policy and regulation library, equipment manufacturer's failure records, and market forecast data, policy keywords are captured through natural language processing. Combining the mean time between failures of equipment, the technical risk probability is calculated, and a heat map is output to intuitively label high - risk items and generate risk response suggestions;

[0040] S33: Quantitatively score the candidate plans from three dimensions: economy, compliance, and social benefits. Display the advantages and disadvantages distribution of each plan through a radar chart, and generate a list of advantages and disadvantages.

[0041] Preferably, when the visual interaction system performs three - dimensional visual display on the output data of the decision support center, it specifically includes:

[0042] S41: Data conversion and mapping. Read the parking lot CAD drawings, real - time sensor data, and renovation parameters, convert the original data into a standardized format, and map the numerical values to visual attributes. Convert the parking space size and floor height parameters into three - dimensional space coordinates through an algorithm, and output a structured data set with space labels;

[0043] S42: Three - dimensional model construction. Based on the structured data set with space labels, construct a three - dimensional grid model including parking spaces, aisles, and equipment, and output a three - dimensional scene that can be viewed in layers;

[0044] S43: Dynamic rendering optimization. Use ray - tracing technology to simulate real lighting and automatically switch the rendering mode according to the device performance;

[0045] S44: Interaction response. When the user clicks on a parking space, details can be viewed. When dragging and rotating the perspective, level - of - detail rendering is triggered. And in the AR mode, when the mobile phone scans the scene, the virtual model is superimposed in real - time, and an interactive picture of virtual - real fusion is output.

[0046] An evaluation method for the value of parking lot upgrade and transformation includes the following steps:

[0047] S51: Data intelligent processing, collecting and integrating multi-source data through a preset interface, locally training a prediction model, aggregating gradient parameters by a central server to generate a globally optimized model, and then extracting composite features from the data.

[0048] S52: Intelligent analysis and evaluation. First, use a spatio-temporal convolutional network to predict the traffic flow in different time periods in the next 72 hours, generate a heat map with a confidence interval, then calculate the economic value, policy matching degree and social benefits respectively, output a three-dimensional radar chart to quantify the evaluation results, and finally optimize and rank the solutions.

[0049] S53: Decision support. Analyze the CAD drawings of the parking lot, automatically generate 3 - 5 feasible solutions in combination with budget constraints, output a three-dimensional adjustable model and a key parameter table, and monitor policy changes and equipment failure rates in real time, generate a heat map of red, yellow and green risk levels and corresponding suggestions, and finally compare the advantages and disadvantages of the solutions through a radar chart.

[0050] S54: Visualization delivery. Convert the parking space coordinates into a WebGL three-dimensional model, dynamically bind real-time data, and support mobile phone scanning on-site to overlay the virtual model.

[0051] Advantages of the present invention:

[0052] 1. Compared with the prior art that only selects parking lot renovation solutions based on experience or a single indicator, which has the disadvantages of strong subjectivity and incomplete evaluation, this solution adopts a method of hard condition filtering combined with multi-objective scoring and comprehensive scoring ranking. By accurately matching the current parking lot conditions, comprehensively considering the economic benefit index, policy matching degree and social benefits, and using a multi-objective optimization algorithm to generate a Pareto optimal solution set, and adjusting the ranking in combination with the risk coefficient, it realizes the scientific screening and efficient optimization of renovation solutions, and has the advantages of comprehensive evaluation, accurate decision-making and remarkable solution optimization effect.

[0053] 2. Compared with the prior art that comprehensively evaluates the economic benefit index, policy matching degree and social benefits by using a simple average or manual weighting method, which has the disadvantages of inaccurate evaluation results and strong subjectivity, this solution adopts a weighted comprehensive scoring method. By assigning different weights to the economic benefit index, policy matching degree and social benefits, and using a normalization formula to calculate the comprehensive score of each solution, it effectively avoids subjectivity and one-sidedness in the evaluation process, improves the accuracy and scientificity of the evaluation results, and provides strong support for the selection of optimal parking lot upgrade and renovation solutions.

[0054] 3. Compared with the existing technologies that adopt static two-dimensional display or simple three-dimensional modeling methods, there are disadvantages such as unintuitive display effects, poor interactivity, and difficulty in truly reflecting the renovation effects of parking lots. This visualization interaction system adopts advanced technical solutions such as data conversion and mapping, three-dimensional model construction, dynamic rendering optimization, and interaction response. It can accurately convert the output data of the decision support center into a three-dimensional visualization scene, not only realizing the intuitive display of the parking lot renovation effects, but also providing a rich user experience through dynamic rendering and interaction response functions, enabling users to easily view the details of parking spaces, rotate the perspective to observe details, and achieve an interactive picture of virtual-real integration in the AR mode, significantly improving the display effects and decision-making efficiency of the parking lot renovation project. Brief Description of the Drawings

[0055] Figure 1 The structural schematic diagram of the evaluation model system for the value of the parking lot upgrade and renovation of the present invention is shown;

[0056] Figure 2 The flowchart of the evaluation model method for the value of the parking lot upgrade and renovation of the present invention is shown. Detailed Embodiments

[0057] The present invention will be further described below in conjunction with the drawings and embodiments.

[0058] Please refer to Figure 1 , the present invention provides an embodiment: an evaluation model system for the value of the parking lot upgrade and renovation, including:

[0059] A data intelligent management module for collecting, integrating, cleaning, and extracting features of the data of the parking lot;

[0060] An intelligent analysis and evaluation module for processing and analyzing the data collected by the data intelligent management module and evaluating the value of the parking lot;

[0061] A decision support center for reporting output according to the data results of the intelligent analysis and evaluation module and generating an intelligent plan;

[0062] A visualization interaction system for three-dimensional visualization display of the output data of the decision support center;

[0063] A system management background including a permission management function, a data tracking function, and an operation audit function.

[0064] As described above, the present invention integrates functional modules such as data intelligent management, intelligent analysis and evaluation, decision support, visual interaction, and system management background, achieving comprehensive and efficient data processing and accurate value evaluation. It not only improves the scientific and intelligent level of decision-making, but also enhances the intuitiveness of information understanding through three-dimensional visualization. At the same time, it ensures the security and traceability of system operation, providing strong technical support and management guarantee for the upgrade and transformation of parking lots.

[0065] Preferably, the intelligent management module includes:

[0066] A11: Data acquisition module, used to connect to data sources including parking lot sensors, municipal platforms, and third-party APIs;

[0067] A12: Federated learning center, used to realize cross-parking lot data collaborative modeling;

[0068] A13: Data cleaning module, used to repair outliers and process missing values in the obtained data;

[0069] A14: Feature engineering library, used to process the data after cleaning and generate spatio-temporal features.

[0070] As described above, the present invention effectively connects to multiple data sources by integrating functions such as data acquisition, federated learning, data cleaning, and feature engineering, realizes cross-parking lot data collaborative modeling and efficient integration, and at the same time ensures data quality. High-quality data with spatio-temporal characteristics is generated through fine feature engineering, laying a solid foundation for subsequent parking lot value evaluation and decision support.

[0071] Preferably, when the intelligent analysis and evaluation module processes and analyzes the data collected by the data intelligent management module and evaluates the value of the parking lot, it specifically includes:

[0072] S11: Dynamic prediction, input historical traffic flow data, weather forecasts, and surrounding commercial activity schedules, process the input data, and predict the traffic flow heat map by time period;

[0073] S12: Multidimensional value calculation, input the preset transformation plan parameters, and predict the economic value, social value, and policy value of the parking lot upgrade and transformation according to the traffic flow heat map by time period and the transformation plan parameters;

[0074] S13: Plan optimization, combine the pre-stored transformation reference plans in the database, optimize and recommend the transformation plans, and perform comprehensive scoring and ranking.

[0075] As described above, the present invention dynamically predicts the traffic heat map by time period, comprehensively evaluates the economic, social and policy values of the parking lot upgrade and transformation by combining multi-dimensional value calculation, and optimizes and recommends the plan based on the transformation reference plan in the database, realizing the accurate evaluation of the parking lot value and the efficient optimization of the transformation plan, providing a scientific basis and an optimal plan for decision-makers.

[0076] Preferably, when combining the pre-stored transformation reference plans in the database, optimizing and recommending the transformation plan, and performing comprehensive scoring and ranking, it specifically includes:

[0077] S21: Hard condition filtering. Read all the transformation reference plans, filter the transformation reference plans according to the hard conditions of the current parking lot and the transformation plan parameters, and filter out the transformation reference plans that do not meet the existing conditions.

[0078] S22: Multi-objective scoring. Calculate the economic benefit index, policy matching degree and social benefit separately for all the remaining transformation reference plans, and perform weighted integration on the calculated economic benefit index, policy matching degree and social benefit.

[0079] S23: Comprehensive scoring and ranking. Use the multi-objective optimization algorithm to generate the Pareto optimal solution set, and adjust the ranking in combination with the risk coefficient, and output the sorted plan.

[0080] Preferably, when calculating the economic benefit index, policy matching degree and social benefit separately for all the remaining transformation reference plans, the principle formula is:

[0081] Economic benefit index:

[0082]

[0083] f3 = 0.6ΔC + 0.4K;

[0084] Among them, f1 is the economic benefit index, NPV is the net present value, and the calculation formula is the sum of the discounted future cash flows minus the initial investment. The industry benchmark value is the average NPV of the same type of parking lot project; f2 is the policy matching degree, ω k is the weight of the kth policy, Compliance k is the compliance with the kth policy, m is the total number of policies to be met, f3 is the comprehensive social benefit score, ΔC is the estimated annual reduction in carbon emissions, and K is the congestion mitigation coefficient.

[0085] As described above, compared with the prior art, the prior art only selects the parking lot renovation plan based on experience or a single index, which has the disadvantages of strong subjectivity and incomplete evaluation. This solution adopts a method of combining hard condition filtering with multi-objective scoring and comprehensive scoring ranking. By accurately matching the current parking lot conditions, comprehensively considering the economic benefit index, policy matching degree, and social benefits, and using a multi-objective optimization algorithm to generate a Pareto optimal solution set, and adjusting the ranking in combination with the risk coefficient, the scientific screening and efficient optimization of the renovation plan are realized, which has the advantages of comprehensive evaluation, accurate decision-making, and remarkable plan optimization effect.

[0086] Preferably, when performing weighted synthesis on the calculated economic benefit index, policy matching degree, and social benefits, the principle formula is:

[0087]

[0088] Among them, Score i is the comprehensive score of the i-th plan, and W j is the weight of the j-th objective. Among them, for the economic benefit index, W1 = 0.5, for the policy matching degree, W2 = 0.3, and for the social benefit, W3 = 0.2. f j max is the maximum value of the j-th objective among all plans, and f j min is the minimum value of the j-th objective among all plans.

[0089] As described above, compared with the prior art, the prior art comprehensively evaluates the economic benefit index, policy matching degree, and social benefits by using a simple average or manual weighting method, which has the disadvantages of inaccurate evaluation results and strong subjectivity. This solution adopts a weighted comprehensive scoring method. By assigning different weights to the economic benefit index, policy matching degree, and social benefits, and using the normalization formula to calculate the comprehensive score of each plan, it effectively avoids subjectivity and one-sidedness in the evaluation process, improves the accuracy and scientific nature of the evaluation results, and provides strong support for the optimization of the parking lot upgrade renovation plan.

[0090] Preferably, when adjusting the ranking in combination with the risk coefficient, the principle formula is:

[0091] Adj payback = Payback × (1 + risk coefficient);

[0092] Adj payback ≤ threshold;

[0093]

[0094] Among them, Adj payback is the payback period, Payback is the ideal payback period, sa Represents solution a, s b Represents solution b, f i (s a ) represents the score of solution a on the i-th objective, f i (s b ) represents the score of solution b on the i-th objective, f j (s a ) represents the score of solution a on the i-th objective, f j (s b ) represents the score of solution b on the j-th objective.

[0095] As described above, the limitation of the present invention that risk factors are not considered when sorting solutions relative to the prior art may lead to a relatively high risk for the preferred solution in actual implementation. This solution adopts a method of adjusting the sorting by combining the risk coefficient. By calculating the score differences of different solutions on multiple objectives and comprehensively considering the risk coefficient, a more comprehensive and objective sorting of the solutions is carried out, effectively reducing the potential risks in the implementation of the preferred solution, improving the robustness and feasibility of solution selection, and providing a strong guarantee for the successful implementation of the parking lot upgrade and transformation project.

[0096] Preferably, when the decision support center outputs a report and generates an intelligent solution according to the data results of the intelligent analysis and evaluation module, it specifically includes:

[0097] S31: Intelligent solution generation. After inputting the parking lot CAD drawing, the owner's budget and policy requirements, the building structure is parsed by AI to automatically generate 3 - 5 feasible transformation solutions, and the maximum parking space capacity and investment range are calculated based on the space optimization algorithm. Finally, a three-dimensional interactive model and a key parameter table are output;

[0098] S32: Dynamic risk scanning. After inputting the real-time updated policy and regulation library, equipment manufacturer failure records and market prediction data, policy keywords are captured through natural language processing, and the technical risk probability is calculated in combination with the average trouble-free time of the equipment. A heat map is output to intuitively mark high-risk items and generate risk response suggestions;

[0099] S33: Quantitatively score the candidate solutions from three dimensions of economy, compliance and social benefits, display the advantages and disadvantages distribution of each solution through a radar chart, and generate a list of advantages and disadvantages.

[0100] As described above, the present invention realizes the efficient customization and risk warning of the parking lot renovation plan through functions such as intelligent generation of integrated solutions, dynamic risk scanning, and quantitative scoring. Based on the parking lot CAD drawings, the owner's budget, and policy requirements, the solution can automatically generate 3-5 feasible renovation plans, and output a three-dimensional interactive model and a key parameter table, greatly improving the efficiency and personalization level of plan generation. At the same time, through the real-time updated policy and regulation library, equipment manufacturer failure records, and market prediction data, the solution can dynamically scan potential risks, intuitively mark high-risk items, and generate risk response suggestions, effectively reducing the uncertainty in project implementation. In addition, the solution also quantitatively scores the candidate plans from three dimensions: economy, compliance, and social benefits, and intuitively displays the advantages and disadvantages of each plan through a radar chart, providing a comprehensive and scientific decision-making basis for decision-makers, and ensuring the smooth implementation and high-benefit return of the parking lot upgrade and renovation project.

[0101] Preferably, when the visualization interaction system performs three-dimensional visualization display on the output data of the decision support center, it specifically includes:

[0102] S41: Data conversion and mapping, read the parking lot CAD drawings, real-time sensor data, and renovation parameters, convert the original data into a standardized format, and map the numerical values to visual attributes. Convert the parking space size and floor height parameters into three-dimensional space coordinates through an algorithm, and output a structured data set with spatial labels;

[0103] S42: Three-dimensional model construction, based on the structured data set with spatial labels, construct a three-dimensional grid model including parking spaces, aisles, and equipment, and output a three-dimensional scene that can be viewed in layers;

[0104] S43: Dynamic rendering optimization, use ray tracing technology to simulate real lighting, and automatically switch the rendering mode according to the device performance;

[0105] S44: Interaction response, when the user clicks on a parking space, the details can be viewed. When dragging and rotating the perspective, it triggers the rendering of different levels of detail. And in the AR mode, the mobile phone scans the scene and overlays the virtual model in real time, outputting an interactive picture that combines the virtual and the real.

[0106] As described above, compared with the prior art that uses static two-dimensional display or simple three-dimensional modeling, the present invention has disadvantages such as non-intuitive display effects, poor interactivity, and difficulty in truly reflecting the renovation effects of parking lots. This visualization interaction system adopts advanced technical solutions such as data conversion and mapping, three-dimensional model construction, dynamic rendering optimization, and interaction response, and can accurately convert the output data of the decision support center into a three-dimensional visualization scene. It not only realizes the intuitive display of the parking lot renovation effects, but also provides a rich user experience through dynamic rendering and interaction response functions, enabling users to easily view the details of parking spaces, rotate the perspective to observe details, and achieve an interactive picture of virtual-real fusion in the AR mode, significantly improving the display effects and decision-making efficiency of parking lot renovation projects.

[0107] Please refer to Figure 2 , the present invention provides an embodiment: a method for evaluating the value of parking lot upgrade and renovation, including the following steps:

[0108] S51: Intelligent data processing, collect and integrate multi-source data through a preset interface, locally train a prediction model, aggregate gradient parameters by a central server to generate a globally optimized model, and then extract composite features from the data;

[0109] S52: Intelligent analysis and evaluation, first use a spatio-temporal convolutional network to predict the traffic flow in different time periods in the next seventy-two hours to generate a heat map with a confidence interval, then calculate the economic value, policy matching degree, and social benefits respectively, output a three-dimensional radar chart to quantify the evaluation results, and finally optimize and sort the solutions;

[0110] S53: Decision support, analyze the CAD drawings of the parking lot, automatically generate 3-5 feasible solutions in combination with budget constraints, output a three-dimensional adjustable model and a key parameter table, and monitor policy changes and equipment failure rates in real time to generate a heat map of red, yellow, and green risks and corresponding suggestions. Finally, compare the advantages and disadvantages of the solutions through a radar chart;

[0111] S54: Visualization delivery, convert the parking space coordinates into a WebGL three-dimensional model, dynamically bind real-time data, and support mobile phone scanning of the site to overlay a virtual model.

[0112] As described above, the present invention realizes the efficient integration of multi-source data and the optimization of the prediction model through intelligent data processing, accurately predicts the traffic flow and quantitatively evaluates the renovation value in combination with intelligent analysis and evaluation, further automatically generates feasible solutions and monitors risks in real time through decision support, and finally realizes the dynamic binding of the three-dimensional model and real-time data through visualization delivery, providing an intuitive and comprehensive evaluation and decision support system. This solution not only improves the accuracy and efficiency of the evaluation, but also enhances the scientificity and operability of decision-making, providing strong technical support for parking lot upgrade and renovation projects.

[0113] The embodiments of the present invention have been described in detail above with reference to the accompanying drawings. However, the present invention is not limited to the above embodiments, and various changes can be made without departing from the spirit of the present invention within the scope of knowledge possessed by those skilled in the art.

Claims

1. An evaluation model system for the value of parking lot upgrading and transformation, characterized in that: It includes: A data intelligent management module, which is used to collect, integrate, clean and extract features from the data of the parking lot; An intelligent analysis and evaluation module, which is used to process and analyze the data collected by the data intelligent management module and evaluate the value of the parking lot; A decision support center, which is used to output reports according to the data results of the intelligent analysis and evaluation module and generate intelligent solutions; A visualization interaction system, which is used to display the output data of the decision support center in 3D visualization; A system management background, including permission management function, data tracking function and operation audit function.

2. The evaluation model system for the value of parking lot upgrade and transformation according to claim 1, characterized in that: The intelligent management module includes: A11: A data collection module, which is used to connect to data sources including parking lot sensors, municipal platforms and third-party APIs; A12: A federated learning center, which is used to realize cross-parking lot data collaborative modeling; A13: A data cleaning module, which is used to repair outliers and process missing values in the obtained data; A14: A feature engineering library, which is used to process the cleaned data and generate spatio-temporal features.

3. The evaluation model system for the value of parking lot upgrade and transformation according to claim 2, characterized in that: When the intelligent analysis and evaluation module processes and analyzes the data collected by the data intelligent management module and evaluates the value of the parking lot, it specifically includes: S11: Dynamic prediction, input historical traffic flow data, weather forecasts and surrounding commercial activity schedules, process the input data, and predict the traffic flow heat map by time period; S12: Multidimensional value calculation, input the preset renovation plan parameters, and predict the economic value, social value and policy value of the parking lot upgrade and renovation according to the traffic flow heat map by time period and the renovation plan parameters; S13: Scheme optimization, combine the pre-stored renovation reference schemes in the database, optimize and recommend the renovation schemes, and perform comprehensive scoring and ranking.

4. The evaluation model system for the value of parking lot upgrade and transformation according to claim 3, characterized in that: When combining the pre-stored renovation reference schemes in the database, optimizing and recommending the renovation schemes, and performing comprehensive scoring and ranking, it specifically includes: S21: Hard condition filtering, read all the renovation reference schemes, filter the renovation reference schemes according to the hard conditions of the current parking lot and the renovation plan parameters, and filter out the renovation reference schemes that do not meet the existing conditions; S22: Multi-objective scoring, calculate the economic benefit index, policy matching degree and social benefits of the remaining renovation reference schemes separately, and perform weighted integration on the calculated economic benefit index, policy matching degree and social benefits; S23: Comprehensive scoring and ranking, use the multi-objective optimization algorithm to generate the Pareto optimal solution set, and adjust the ranking in combination with the risk coefficient, and output the ranked schemes.

5. The evaluation model system for the value of parking lot upgrade and transformation according to claim 4, characterized in that: When calculating the economic benefit index, policy matching degree and social benefits of the remaining renovation reference schemes separately, the principle formula is: Economic benefit index: f3 = 0.6ΔC + 0.4K; Among them, f1 is the economic benefit index, NPV is the net present value, and the calculation formula is the sum of discounted future cash flows minus the initial investment. The industry benchmark value is the average NPV of the same type of parking lot project; f2 is the policy matching degree, ω k is the weight of the k-th policy, Compliance k is the compliance degree with respect to the k policies, m is the total number of policies to be satisfied, f3 is the comprehensive social benefit score, ΔC is the estimated annual reduction in carbon emissions, and K is the congestion mitigation coefficient.

6. The evaluation model system for the value of parking lot upgrade and transformation according to claim 5, characterized in that: When performing weighted integration on the calculated economic benefit index, policy matching degree and social benefits, the principle formula is: Among them, Score i is the comprehensive score of the i-th solution, and W j is the weight of the j-th objective. Among them, for the economic benefit index W1 = 0.5, for the policy matching degree W2 = 0.3, and for the social benefit W3 = 0.2, is the maximum value of the j-th objective among all solutions, is the minimum value of the j-th objective among all solutions.

7. The evaluation model system for the value of parking lot upgrade and transformation according to claim 6, characterized in that: When adjusting the ranking in combination with the risk coefficient, the principle formula is: Adj payback = Payback × (1 + risk coefficient); Adj payback ≤ threshold; Among them, Adj payback is the payback period, Payback is the ideal payback period, s a represents Plan a, s b represents Plan b, f i (s a ) represents the score of Plan a on the i-th objective, f i (s b ) represents the score of Plan b on the i-th objective, f j (s a ) represents the score of Plan a on the i-th objective, f j (s b ) represents the score of Plan b on the j-th objective.

8. The evaluation model system for the value of parking lot upgrade and transformation according to claim 7, characterized in that: When the decision support center outputs reports according to the data results of the intelligent analysis and evaluation module and generates intelligent solutions, it specifically includes: S31: Intelligent solution generation. After inputting the parking lot CAD drawings, the owner's budget, and policy requirements, the building structure is analyzed by AI to automatically generate 3 - 5 feasible renovation solutions. Based on the space optimization algorithm, the maximum parking space capacity and investment range are calculated, and finally, a 3D interactive model and a key parameter table are output; S32: Dynamic risk scanning. After inputting the real - time updated policy and regulation library, equipment manufacturer's fault records, and market prediction data, policy keywords are captured through natural language processing, and the technical risk probability is calculated in combination with the mean time between failures of the equipment. A heat map is output, high - risk items are intuitively marked, and risk response suggestions are generated; S33: Quantitatively score the candidate solutions from three major dimensions: economy, compliance, and social benefits. The advantages and disadvantages of each solution are displayed through a radar chart, and a list of advantages and disadvantages is generated.

9. The evaluation model system for the value of parking lot upgrade and transformation according to claim 8, wherein: When the visual interaction system performs 3D visualization of the output data of the decision - making support center, it specifically includes: S41: Data conversion and mapping. Read the parking lot CAD drawings, real - time sensor data, and renovation parameters, convert the original data into a standardized format, and map the numerical values to visual attributes. Through the algorithm, the parking space size and floor height parameters are converted into 3D space coordinates, and a structured data set with space labels is output; S42: 3D model construction. Based on the structured data set with space labels, a 3D grid model including parking spaces, aisles, and equipment is constructed, and a three - dimensional scene that can be viewed in layers is output; S43: Dynamic rendering optimization. Use ray - tracing technology to simulate real lighting and automatically switch the rendering mode according to the equipment performance; S44: Interaction response. When the user clicks on a parking space, details can be viewed. When dragging and rotating the perspective, level - of - detail rendering is triggered. In the AR mode, the mobile phone scans the scene and the virtual model is overlaid in real - time, and an interactive picture of the integration of the virtual and the real is output.

10. A method for evaluating the value of a parking lot upgrade and transformation model system, characterized in that: It includes the following steps: S51: Intelligent data processing. Through a preset interface, multi - source data is collected and integrated, and a prediction model is locally trained. The central server aggregates gradient parameters to generate a global optimization model, and then composite features are extracted from the data; S52: Intelligent analysis and evaluation. First, use a spatio - temporal convolutional network to predict the traffic flow in different time periods in the next 72 hours, generate a heat map with a confidence interval, then calculate the economic value, policy matching degree, and social benefits respectively, output a 3D radar chart to quantify the evaluation results, and finally optimize and sort the solutions; S53: Decision - making support. Analyze the parking lot CAD drawings, automatically generate 3 - 5 feasible solutions in combination with budget constraints, output a 3D adjustable model and a key parameter table, and monitor policy changes and equipment failure rates in real - time, generate a red - yellow - green three - color risk heat map and response suggestions, and finally compare the advantages and disadvantages of the solutions through a radar chart; S54: Visualization delivery. Convert the parking space coordinates into a WebGL 3D model, dynamically bind real - time data, and support the mobile phone to scan the scene and overlay the virtual model.

Images