Evaluation method, system and equipment for use efficiency of TOD (time of day) site and medium
By using an improved BP neural network model and K-means clustering method, combined with resident experience data, the problem of inaccurate evaluation of TOD site usage efficiency was solved, enabling more comprehensive evaluation and dynamic adjustment, and improving the accuracy and adaptability of the evaluation results.
Patent Information
- Application Number
- CN202511541067.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-10-27
- Publication Date
- 2026-02-17
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
Existing efficiency assessment methods for TOD sites fail to fully consider residents' experience, resulting in inaccurate and incomplete assessment results.
An improved BP neural network model combined with K-means clustering was adopted. The original dataset of TOD sites was obtained and normalized, standardized, decorrelated, and dimensionality reduced. The weights of the evaluation indicators were determined by using the entropy weight method, combined with residents' commuting time satisfaction and community activity participation. Dropout regularization layer and adaptive learning rate were introduced, and Softmax activation function was used. Finally, K-means clustering was performed to evaluate the utilization efficiency of TOD sites.
It enables a more accurate and comprehensive assessment of the efficiency of TOD site usage, reflecting residents' actual experience, improving the accuracy and comprehensiveness of assessment results, and supporting the dynamic adjustment of cluster boundaries to adapt to changes in urban development.
Smart Images
Figure CN121543866A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of urban planning, and in particular to a method, system, device and medium for evaluating the utilization efficiency of TOD (Transit-Oriented Development) sites. Background Technology
[0002] In urban planning, TOD (Transit-Oriented Development) is a non-car-oriented planning and design approach that maximizes the use of public transportation when planning a residential or commercial area. It aims to alleviate urban problems such as traffic congestion and environmental degradation. TOD stations serve as transit hubs within the TOD model. The utilization efficiency of TOD stations is a crucial indicator of urban development models and needs to be evaluated to update urban development strategies.
[0003] The existing TOD site utilization efficiency only considers commercial value and does not evaluate from the perspective of residents. The evaluation method has problems such as incomplete evaluation factors and inaccurate evaluation results. Summary of the Invention
[0004] This invention aims to address at least one of the technical problems existing in the prior art. To this end, this invention proposes a method for evaluating the utilization efficiency of TOD (Transit-Oriented Development) sites, which takes into account residents' experiences and feelings, and accurately and comprehensively assesses the utilization efficiency of TOD sites.
[0005] The present invention also proposes a system, device and medium having the above-mentioned method for evaluating the utilization efficiency of TOD sites.
[0006] A method for evaluating the utilization efficiency of TOD (Transit-Oriented Development) sites according to a first aspect of the present invention includes: Obtain the original datasets of several TOD sites; wherein, the original datasets include: passenger flow carrying capacity per unit area, vertical space utilization rate, accessibility of public facilities, saturation of transportation facilities, commercial space efficiency, growth rate of enterprise occupancy, residents' commuting time satisfaction, and community activity participation. The original datasets of each TOD site were normalized, standardized, decorrelated, and dimensionality reduced in sequence to obtain the evaluation index set of each TOD site. The evaluation index set of each TOD site is input into the improved BP neural network model to obtain the utilization efficiency score of each TOD site. The usage efficiency scores and evaluation index sets of each TOD site are used for K-means clustering to obtain the evaluation results of the usage efficiency of each TOD site.
[0007] According to an embodiment of the present invention, a method for evaluating the utilization efficiency of TOD sites has at least the following beneficial effects: Since the original data used to evaluate the utilization efficiency of TOD sites, i.e., the original dataset, has characteristics of unstable data sources, multiple categories, and complexity, and there are complex nonlinear relationships between the data, it is necessary to first perform normalization, standardization, decorrelation, and dimensionality reduction processing on the obtained original dataset in sequence. To adapt to the nonlinear relationships between the original data and obtain accurate evaluation results, a nonlinear BP neural network model is used for improvement, resulting in a preliminary evaluation result, i.e., a utilization efficiency score. Then, by comprehensively clustering the evaluation index set and the utilization efficiency score, the evaluation results of the utilization efficiency of each TOD site are obtained more accurately. Furthermore, the present invention obtains residents' commuting time satisfaction and community activity participation as original data, reflecting the actual usage experience and behavioral feedback of residents around the TOD sites, thus providing a more comprehensive evaluation of the utilization efficiency of each TOD site.
[0008] According to some embodiments of the present invention, the training method for the improved BP neural network model includes: Obtain the original datasets of several TOD sample sites; Based on the original datasets of each TOD sample site, an evaluation index set for each TOD sample site is obtained; Based on the evaluation index set of each TOD sample site, the comprehensive score of each TOD sample site is obtained; The evaluation index set of each TOD sample site is used as input, and the comprehensive score of each TOD sample site is used as the supervision signal to train the BP neural network model.
[0009] According to some embodiments of the present invention, obtaining the comprehensive score of each TOD sample site based on the evaluation index set of each TOD sample site includes: For each TOD sample site, the weight of each evaluation indicator in the evaluation indicator set is determined by using the entropy weight method in combination with a preset TOD site indicator library, based on the evaluation indicator set. The overall score of the TOD sample sites is determined based on the set of evaluation indicators and their corresponding weights.
[0010] According to some embodiments of the present invention, the improved BP neural network model introduces a Dropout regularization layer in the hidden layer and sets an adaptive learning rate adjustment formula, wherein the adaptive learning rate adjustment formula is specifically as follows: in, The adjusted learning rate, The preset initial learning rate, The decay coefficient is t, and the number of training iterations is t. The improved BP neural network model uses the Softmax activation function in its output layer. The specific formula for the Softmax activation function is as follows: in, The output value of the k-th output node. is the input value of the k-th node in the output layer, where K is the total number of output nodes.
[0011] According to some embodiments of the present invention, the step of performing normalization, standardization, decorrelation, and dimensionality reduction processing on the original datasets of each TOD site in sequence to obtain the evaluation index set for each TOD site includes: Determine whether the residents' commuting time satisfaction and residents' expected commuting time in the original dataset of each TOD site are invalid or abnormal; if so, remove the data; otherwise, retain the data. The retained data were standardized, and residents' commuting time satisfaction and expected commuting time were transformed into data that conformed to a standard normal distribution. The transformed data was then added to the evaluation indicator set. Other raw data, including the passenger flow capacity per unit area, the vertical space utilization rate, the accessibility of public facilities, the saturation of transportation facilities, the commercial space efficiency, and the growth rate of enterprise occupancy, are then normalized, standardized, decorrelated, and dimensionality reduced in sequence.
[0012] According to some embodiments of the present invention, in the process of performing K-means clustering on the usage efficiency scores and evaluation index sets of each TOD site, Euclidean distance is used to calculate the similarity of each TOD site, and the Euclidean distance formula is: in, Let x be the Euclidean distance between TOD site x and TOD site y. Let i be the value of the i-th feature of TOD site x. Let be the value of the i-th feature of the TOD site y, and n be the dimension of the feature; wherein, the feature includes the usage efficiency score and the evaluation indicators in the evaluation indicator set.
[0013] According to some embodiments of the present invention, it further includes: The locations of each TOD station and each K-means cluster center are converted into spatial vector data of a geographic information system to obtain the location coordinates of each TOD station and the location coordinates of each K-means cluster center. Periodically calculate the Euclidean distance between each TOD site and each K-means cluster center; Determine the rate of change between the Euclidean distance between each TOD site and each K-means cluster center and the Euclidean distance of the previous period; where the Euclidean distance of the previous period refers to the Euclidean distance between each TOD site and each K-means cluster center in the previous period. If the rate of change exceeds a preset rate of change threshold, an alarm is issued and the preset strategy library is invoked to update the development strategy.
[0014] A TOD site utilization efficiency evaluation system according to a second aspect of the present invention is used to perform a TOD site utilization efficiency evaluation method according to any one of the first aspects, comprising: The data acquisition module is used to obtain the raw datasets from several TOD sites; The data processing module, connected to the data acquisition module, is used to receive the original datasets of TOD sites sent by the data acquisition module, and to perform normalization, standardization, decorrelation and dimensionality reduction processing on the original datasets of each TOD site in sequence to obtain the evaluation index set of each TOD site. The model evaluation module, connected to the data processing module, is used to receive the evaluation index set of each TOD site sent by the data processing module, input the evaluation index set of each TOD site into the improved BP neural network model, and obtain the utilization efficiency score of each TOD site. The classification decision module, connected to the model evaluation module, is used to receive the usage efficiency scores of each TOD site sent by the model evaluation module, perform K-means clustering on the usage efficiency scores and evaluation index set of each TOD site, and obtain the evaluation results of the usage efficiency of each TOD site.
[0015] An electronic device according to a third aspect of the present invention includes: Memory, used to store programs; A processor for executing a program stored in the memory, wherein when the processor executes the program stored in the memory, the processor is configured to perform the method as described in any one of the first aspects.
[0016] According to a fourth aspect of the present invention, a storage medium stores computer-executable instructions for performing the method as described in any one of the first aspects.
[0017] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the description, claims, and drawings. Attached Figure Description
[0018] The accompanying drawings are provided to further understand the technical solutions of the present invention and constitute a part of the specification. They are used together with the embodiments of the present invention to explain the technical solutions of the present invention, and do not constitute a limitation on the technical solutions of the present invention.
[0019] Figure 1 This is a flowchart of a method for evaluating the efficiency of TOD site usage, provided in an embodiment of the present invention. Detailed Implementation
[0020] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.
[0021] It should be understood that in the description of the embodiments of the present invention, "multiple" (or "amounts") means two or more, "greater than," "less than," and "exceeding" are understood to exclude the stated number, while "above," "below," and "within" are understood to include the stated number. If "first," "second," etc., are used in the description, they are only for the purpose of distinguishing technical features and should not be construed as indicating or implying relative importance, or implicitly indicating the number of indicated technical features, or implicitly indicating the order of the indicated technical features.
[0022] like Figure 1 As shown, this embodiment of the invention provides a method for evaluating the utilization efficiency of TOD (Transit-Oriented Development) sites, including: Step S100: Obtain the original datasets of several TOD sites; wherein, the original datasets include: passenger flow carrying capacity per unit area, vertical space utilization rate, accessibility of public facilities, saturation of transportation facilities, commercial space efficiency, growth rate of enterprise occupancy, residents' commuting time satisfaction, and community activity participation. Step S200: Normalize, standardize, decorrelate, and reduce the dimensionality of the original datasets of each TOD site in sequence to obtain the evaluation index set of each TOD site; Step S300: Input the evaluation index set of each TOD site into the improved BP neural network model to obtain the utilization efficiency score of each TOD site. Step S400: Perform K-means clustering on the usage efficiency scores and evaluation index sets of each TOD site to obtain the evaluation results of the usage efficiency of each TOD site.
[0023] Because the raw data used to assess the efficiency of TOD sites is characterized by unstable data sources, multiple categories, and complexity, and contains complex nonlinear relationships, the raw dataset must first be normalized, standardized, decorrelated, and dimensionality reduced sequentially. To adapt to the nonlinear relationships between the raw data and obtain accurate assessment results, a nonlinear BP neural network model is used for improvement, yielding a preliminary assessment result, namely the efficiency score. Then, the assessment index set and the efficiency score are combined for clustering to more accurately derive the efficiency assessment results of each TOD site. Furthermore, this invention uses residents' commuting time satisfaction and community activity participation as raw data to reflect the actual usage experience and behavioral feedback of residents around the TOD sites, thus providing a more comprehensive assessment of the efficiency of each TOD site.
[0024] In one embodiment, K-means clustering uses the elbow method to determine the K value, K=3, and divides the evaluation results of the utilization efficiency of each TOD site into three categories: inefficient utilization, moderate utilization, and inefficient utilization.
[0025] In one embodiment, the training method for the improved BP neural network model includes: Obtain the original datasets of several TOD sample sites; Based on the original datasets of each TOD sample site, an evaluation index set for each TOD sample site is obtained; Based on the evaluation index set of each TOD sample site, the comprehensive score of each TOD sample site is obtained; The evaluation index set of each TOD sample site is used as input, and the comprehensive score of each TOD sample site is used as the supervision signal to train the BP neural network model.
[0026] In one embodiment, the comprehensive score of each TOD sample site is obtained based on the evaluation index set of each TOD sample site, including: For each TOD sample site, based on the evaluation indicator set, the weight of each evaluation indicator in the evaluation indicator set is determined using the entropy weight method combined with a pre-set TOD site indicator library. The specific calculation formula is as follows: , , in, Let be the weight of the j-th evaluation indicator in the evaluation indicator set for TOD sample sites. Let be the information entropy of the j-th evaluation indicator, n be the number of evaluation indicators for the TOD sample sites, and m be the number of indicator values for the corresponding evaluation indicator in the TOD site indicator database. Let be the value of the j-th indicator of the i-th evaluation indicator; The overall score of the TOD sample sites is determined based on the set of evaluation indicators and their corresponding weights.
[0027] It is easy to understand that the TOD sample sites are not the same as the TOD sites whose utilization efficiency needs to be evaluated. The TOD sites whose utilization efficiency needs to be evaluated (i.e., the input of this method) are equivalent to unknown sites. The evaluation index set of each TOD sample site is used as input, and the comprehensive score of each TOD sample site is used as a supervision signal. This allows the BP neural network model to better fit the nonlinear relationship between the evaluation indexes, rather than simply making a direct evaluation based on the comprehensive score. This is because the comprehensive score is actually obtained by weighted summation with preset weights. The comprehensive score cannot accurately reflect the nonlinear relationship between the evaluation indexes. Therefore, it is necessary to use the BP neural network model for fitting.
[0028] It should be noted that the TOD site indicator library stores evaluation indicators (equivalent to labels) and indicator values (equivalent to data), with each evaluation indicator corresponding to at least one indicator value; this invention only uses the entropy weight method to determine indicator weights. The entropy weight method is a completely data-driven objective weighting method that avoids the influence of subjective judgment on weights in the analytic hierarchy process, making weight determination more objective and accurate.
[0029] In one embodiment, the improved BP neural network model introduces a Dropout regularization layer in the hidden layer and sets an adaptive learning rate adjustment formula, which is as follows: in, The adjusted learning rate, The preset initial learning rate, The decay coefficient is t, and the number of training iterations is t. The improved BP neural network model uses the Softmax activation function in its output layer. The specific formula for the Softmax activation function is as follows: in, The output value of the k-th output node. is the input value of the k-th node in the output layer, where K is the total number of output nodes.
[0030] The Dropout regularization layer randomly discards 50% of neurons during training, forcing the BP neural network model to learn more robust features, while the adaptive learning rate ensures smooth convergence in the parameter space, avoiding training oscillations caused by Dropout. The combination of these two features improves the model's generalization ability on high-dimensional sparse TOD data by 40% (comparative experimental data).
[0031] In one embodiment, in step S100, the method for obtaining the passenger flow carrying capacity per unit area includes: sending a command to the infrared sensor and the turnstile at the entrance of the TOD station to obtain data; receiving the card swiping data sent by the infrared sensor and the turnstile at the entrance of the TOD station respectively; and determining the passenger flow carrying capacity per unit area based on the card swiping data and the surrounding building area of the TOD station. Specifically, the passenger flow carrying capacity per unit area = card swiping data / surrounding building area of the TOD station. The methods for obtaining vertical space utilization rate include: obtaining the usable area of each floor of the surrounding buildings of the TOD site and the commercial distribution area of each floor; obtaining the vertical space utilization rate based on the usable area of each floor and the commercial distribution area of each floor. Specifically, vertical space utilization rate = average of space utilization rate of each floor, and space utilization rate of each floor = commercial distribution area of each floor / usable area of each floor. Methods for obtaining accessibility to public facilities include: obtaining residents' mobile phone signaling data and road network data; obtaining residents' walking time to the nearest public facility based on the mobile phone signaling data and road network data; summarizing and processing residents' walking time to the nearest public facility to obtain accessibility to public facilities; Methods for obtaining the saturation of transportation facilities include: obtaining the passenger flow of public transportation inside and around the TOD site; and obtaining the saturation of transportation facilities based on the passenger flow and the corresponding approved passenger capacity of public transportation. Specifically, the saturation of transportation facilities = passenger flow / approved passenger capacity of public transportation. Methods for obtaining commercial space efficiency include: obtaining the monthly turnover and commercial area of shopping malls around the TOD station; and obtaining commercial space efficiency based on the monthly turnover and commercial area. Specifically, commercial space efficiency = monthly turnover / commercial area. The methods for obtaining the enterprise entry growth rate include: obtaining the number of newly added enterprises in the current year and the number of newly added enterprises in the previous year; and obtaining the enterprise entry growth rate based on the number of newly added enterprises in the current year and the number of newly added enterprises in the previous year. Specifically, the enterprise entry growth rate = number of newly added enterprises in the current year / number of newly added enterprises in the previous year. The methods for obtaining residents' commuting time satisfaction include: obtaining residents' actual commuting time and residents' expected commuting time; comparing residents' actual commuting time and residents' expected commuting time, and using a 5-point Likert scale to quantify the comparison results to obtain residents' commuting time satisfaction. The methods for obtaining community activity participation include: obtaining the number of people who checked in at community activities around TOD sites; calculating the proportion of the number of people who checked in at community activities around TOD sites to the total population of the community, and the result is the community activity participation rate; It should be noted that the area surrounding a TOD station is the one-kilometer radius of the TOD station's influence.
[0032] In one embodiment, in step S200, the original datasets of each TOD site are sequentially normalized, standardized, decorrelated, and dimensionality reduced to obtain the evaluation index set for each TOD site, including: Determine whether the residents' commuting time satisfaction and residents' expected commuting time in the original dataset of each TOD site are invalid or abnormal; if so, remove the data; otherwise, retain the data. The retained data were Z-score standardized to transform residents’ commuting time satisfaction and residents’ expected commuting time into data that conforms to a standard normal distribution. The transformed data was then added to the evaluation indicator set. The other raw data are then normalized, standardized, decorrelated, and dimensionality reduced in sequence.
[0033] It should be noted that, since this invention needs to consider residents' experience, and the acquisition of residents' actual commuting time and expected commuting time depends on residents' subjectivity, it is characterized by instability and a large amount of invalid data, so data cleaning is required; while other raw data, such as commercial space efficiency, have centralized and stable data sources, so there is no need to remove abnormal data.
[0034] In one embodiment, in step S400, during the K-means clustering process of the usage efficiency scores and evaluation index sets of each TOD site, Euclidean distance is used to calculate the similarity of each TOD site. The Euclidean distance formula is as follows: in, Let x be the Euclidean distance between TOD site x and TOD site y. Let i be the value of the i-th feature of TOD site x. Let be the value of the i-th feature of TOD site y, and n be the dimension of the feature; where the feature includes the efficiency score and the evaluation indicators in the evaluation indicator set.
[0035] In one embodiment, the evaluation results obtained after performing step S400 are shown in the following table: The mean refers to the average value of the corresponding indicator in the indicator library. For example, the mean in "commercial space efficiency ≥ 1.2 times the mean value" is the average value of commercial space efficiency in the indicator library.
[0036] It is easy to understand that the evaluation metrics in the evaluation metric set are actually obtained by processing the original dataset, and they all belong to the data types listed in the original dataset.
[0037] In one embodiment, the method further includes: The locations of each TOD station and each K-means cluster center are converted into spatial vector data of a geographic information system to obtain the location coordinates of each TOD station and the location coordinates of each K-means cluster center. Periodically calculate the Euclidean distance between each TOD site and each K-means cluster center; Determine the rate of change between the Euclidean distance between each TOD site and each K-means cluster center and the Euclidean distance of the previous period; where the Euclidean distance of the previous period refers to the Euclidean distance between each TOD site and each K-means cluster center in the previous period. If the rate of change exceeds the preset rate of change threshold, an alarm will be issued and the preset strategy library will be invoked to update the development strategy.
[0038] This embodiment can dynamically adapt the cluster boundaries of TOD sites based on the evaluation results. When the distribution of indicators of TOD sites in a certain area changes with urban development, the Euclidean distance can automatically adjust the cluster boundaries. For example, as a certain area becomes commercially prosperous, the Euclidean distance between the original "medium utilization" sites and the "high-efficiency utilization" center shortens. The system can issue an early warning and suggest preventive upgrade strategies to avoid the lag caused by traditional static clustering.
[0039] This invention also provides a system for evaluating the utilization efficiency of TOD (Transit-Oriented Development) sites, comprising: The data acquisition module is used to obtain the raw datasets from several TOD sites; The data processing module, connected to the data acquisition module, is used to receive the raw datasets of TOD sites sent by the data acquisition module, and to perform normalization, standardization, decorrelation and dimensionality reduction on the raw datasets of each TOD site in sequence to obtain the evaluation index set of each TOD site. The model evaluation module, connected to the data processing module, is used to receive the evaluation index set of each TOD site sent by the data processing module, input the evaluation index set of each TOD site into the improved BP neural network model, and obtain the utilization efficiency score of each TOD site. The classification decision module, connected to the model evaluation module, receives the usage efficiency scores of each TOD site from the model evaluation module, performs K-means clustering on the usage efficiency scores and evaluation index set of each TOD site, and obtains the evaluation results of the usage efficiency of each TOD site.
[0040] This invention also provides an electronic device, which includes, but is not limited to: Memory, used to store programs; The processor is used to execute programs stored in memory. When the processor executes programs stored in memory, it is used to perform one of the TOD site utilization efficiency evaluation methods described above.
[0041] The processor and memory can be connected via a bus or other means.
[0042] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs, such as the method described in the embodiments of the present invention. The processor implements the above method by running the non-transitory software program and instructions stored in the memory.
[0043] The memory may include a program storage area and a data storage area, wherein the program storage area may store the operating system and application programs required for at least one function; the data storage area may store data for executing the methods described above. Furthermore, the memory may include high-speed random access memory and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, the memory may optionally include memory remotely located relative to the processor, which can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.
[0044] The non-transitory software program and instructions required to implement the above terminal selection method are stored in memory and are executed by one or more processors.
[0045] This invention also provides a storage medium storing computer-executable instructions for performing the above-described methods.
[0046] In one embodiment, the storage medium stores computer-executable instructions that are executed by one or more control processors.
[0047] The embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.
[0048] It will be understood by those skilled in the art that all or some of the steps and systems in the methods disclosed above can be implemented as software, firmware, hardware, and suitable combinations thereof. Some or all of the physical components can be implemented as software executed by a processor, such as a central processing unit, digital signal processor, or microprocessor, or as hardware, or as an integrated circuit, such as an application-specific integrated circuit. Such software can be distributed on a computer-readable medium, which can include computer storage media (or non-transitory media) and communication media (or transient media). As is known to those skilled in the art, the term computer storage media includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data). Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disc (DVD) or other optical disc storage, magnetic cartridges, magnetic tape, disk storage or other magnetic storage devices, or any other medium that can be used to store desired information and is accessible to a computer. Furthermore, as is known to those skilled in the art, communication media typically include computer-readable instructions, data structures, program modules, or other data in modulated data signals such as carrier waves or other transmission mechanisms, and may include any information delivery medium.
[0049] This document describes embodiments of the invention, including preferred embodiments known to the inventors for carrying out the invention. Variations of these embodiments will become apparent to those skilled in the art upon reading the foregoing description. The inventors encourage those skilled in the art to adopt such variations as appropriate, and the inventors intend to practice embodiments of the invention in ways other than those specifically described herein. Therefore, the scope of the invention includes all modifications and equivalents of the subject matter set forth in the appended claims, as permitted by applicable law. Furthermore, the scope of the invention covers any combination of the foregoing elements in all possible variations thereof, unless otherwise indicated herein or otherwise clearly contradicted by the context.
Claims
1. A method for evaluating TOD site usage efficiency, characterized in that, include: Obtain the original datasets of several TOD sites; wherein, the original datasets include: passenger flow carrying capacity per unit area, vertical space utilization rate, accessibility of public facilities, saturation of transportation facilities, commercial space efficiency, growth rate of enterprise occupancy, residents' commuting time satisfaction, and community activity participation. The original datasets of each TOD site were normalized, standardized, decorrelated, and dimensionality reduced in sequence to obtain the evaluation index set of each TOD site. The evaluation index set of each TOD site is input into the improved BP neural network model to obtain the utilization efficiency score of each TOD site. The usage efficiency scores and evaluation index sets of each TOD site are used for K-means clustering to obtain the evaluation results of the usage efficiency of each TOD site.
2. The method of claim 1, wherein, Improved training methods for BP neural network models include: Obtain the original datasets of several TOD sample sites; Based on the original datasets of each TOD sample site, an evaluation index set for each TOD sample site is obtained; Based on the evaluation index set of each TOD sample site, the comprehensive score of each TOD sample site is obtained; The evaluation index set of each TOD sample site is used as input, and the comprehensive score of each TOD sample site is used as the supervision signal to train the BP neural network model.
3. The method of claim 2, wherein, The comprehensive score for each TOD sample site, obtained based on the evaluation index set of each TOD sample site, includes: For each TOD sample site, the weight of each evaluation indicator in the evaluation indicator set is determined by using the entropy weight method in combination with a preset TOD site indicator library, based on the evaluation indicator set. The overall score of the TOD sample sites is determined based on the set of evaluation indicators and their corresponding weights.
4. The method of claim 1, wherein, The improved BP neural network model introduces a Dropout regularization layer in the hidden layer and sets an adaptive learning rate adjustment formula, which is specifically as follows: wherein, is the adjusted learning rate, is the preset initial learning rate, is the decay coefficient, and t is the number of training iterations. The improved BP neural network model uses the Softmax activation function in its output layer. The specific formula for the Softmax activation function is as follows: in, The output value of the k-th output node. is the input value of the k-th node in the output layer, where K is the total number of output nodes.
5. The method for evaluating the utilization efficiency of a TOD site according to claim 1, characterized in that, The original datasets for each TOD site are sequentially normalized, standardized, decorrelated, and dimensionality reduced to obtain the evaluation index set for each TOD site, including: Determine whether the residents' commuting time satisfaction and residents' expected commuting time in the original dataset of each TOD site are invalid or abnormal; if so, remove the data; otherwise, retain the data. The retained data were standardized, and residents' commuting time satisfaction and expected commuting time were transformed into data that conformed to a standard normal distribution. The transformed data was then added to the evaluation indicator set. Other raw data, including the passenger flow capacity per unit area, the vertical space utilization rate, the accessibility of public facilities, the saturation of transportation facilities, the commercial space efficiency, and the growth rate of enterprise occupancy, are then normalized, standardized, decorrelated, and dimensionality reduced in sequence.
6. The method for evaluating the utilization efficiency of a TOD site according to claim 1, characterized in that, In the process of performing K-means clustering on the usage efficiency scores and evaluation index sets of each TOD site, Euclidean distance is used to calculate the similarity of each TOD site. The Euclidean distance formula is as follows: in, Let x be the Euclidean distance between TOD site x and TOD site y. Let i be the value of the i-th feature of TOD site x. Let be the value of the i-th feature of the TOD site y, and n be the dimension of the feature; wherein, the feature includes the usage efficiency score and the evaluation indicators in the evaluation indicator set.
7. The method for evaluating the utilization efficiency of TOD sites according to claim 1, characterized in that, Also includes: The locations of each TOD station and each K-means cluster center are converted into spatial vector data of a geographic information system to obtain the location coordinates of each TOD station and the location coordinates of each K-means cluster center. Periodically calculate the Euclidean distance between each TOD site and each K-means cluster center; Determine the rate of change between the Euclidean distance between each TOD site and each K-means cluster center and the Euclidean distance of the previous period; where the Euclidean distance of the previous period refers to the Euclidean distance between each TOD site and each K-means cluster center in the previous period. If the rate of change exceeds a preset rate of change threshold, an alarm is issued and the preset strategy library is invoked to update the development strategy.
8. A system for evaluating the utilization efficiency of TOD (Transit-Oriented Development) sites, characterized in that, include: The data acquisition module is used to obtain the raw datasets from several TOD sites; The data processing module, connected to the data acquisition module, is used to receive the original datasets of TOD sites sent by the data acquisition module, and to perform normalization, standardization, decorrelation and dimensionality reduction on the original datasets of each TOD site in sequence to obtain the evaluation index set of each TOD site. The model evaluation module, connected to the data processing module, is used to receive the evaluation index set of each TOD site sent by the data processing module, input the evaluation index set of each TOD site into the improved BP neural network model, and obtain the utilization efficiency score of each TOD site. The classification decision module, connected to the model evaluation module, is used to receive the usage efficiency scores of each TOD site sent by the model evaluation module, perform K-means clustering on the usage efficiency scores and evaluation index set of each TOD site, and obtain the evaluation results of the usage efficiency of each TOD site.
9. An electronic device, characterized in that, include: Memory, used to store programs; A processor for executing a program stored in the memory, wherein when the processor executes the program stored in the memory, the processor is configured to perform the method as described in any one of claims 1 to 7.
10. A storage medium, characterized in that, The device stores computer-executable instructions for performing the method as described in any one of claims 1 to 7.