Optimal design method and system for bus lanes considering exhaust emissions at the road section scale

Abstract

The present invention proposes a bus lane optimization design method and system considering tail gas emissions at the road section scale, the method comprising the following steps: collecting design element information of urban bus lanes, road section traffic flow information, and road section basic information; using a virtual simulation method to obtain traffic flow operation condition data of the road section where the urban bus lane is located, and obtaining the tail gas emission equivalent of vehicles in the road section; using an encapsulated feature selection method to screen feature variables from the design element information of urban bus lanes; using an interpretable machine learning method to analyze the influence of feature variables on road section vehicle tail gas emissions, and output the influence contribution ranking of feature variables; using the objective function of minimizing the total tail gas emissions of road section vehicles, using an optimization algorithm to optimize the design scheme of bus lanes. The present invention provides an ecological design scheme for bus lanes to significantly reduce the tail gas emissions of vehicles in urban bus lane sections.

Description

Technical Field

[0001] The present invention relates to the technical field of urban public transportation exhaust emissions, and in particular to a bus lane optimization design method and system taking into account section-scale exhaust emissions. Background Art

[0002] As a traffic planning strategy that gives priority to public transportation, bus lanes ensure that buses have priority by granting them independent road rights. Setting up bus lanes can increase the speed of buses, improve the efficiency of the public transportation system, reduce traffic congestion, and provide more reliable public transportation services. However, the setting up of bus lanes will reduce the number of lanes available for social vehicles, increase the traffic volume that non-dedicated lanes need to bear, cause congestion, and then cause changes in the exhaust emissions of vehicles on the road section.

[0003] In response to this, existing studies in this field have used virtual simulation tests and theoretical analysis methods to explore the impact mechanism of setting up bus lanes on vehicle exhaust emissions on road sections. However, most studies remain at the qualitative analysis level, and few studies have proposed an optimization design method for bus lanes that considers vehicle exhaust emissions at the road section scale from the perspective of traffic environment. Therefore, effectively quantifying the impact of bus lanes on vehicle exhaust emissions on road sections and optimizing bus lane design elements are key issues that need to be addressed and have important practical significance. Summary of the invention

[0004] The purpose of the present invention is to propose a bus lane optimization design method and system taking into account the exhaust emissions at the road section scale, so as to significantly reduce the exhaust emissions of vehicles on urban bus lane sections.

[0005] To achieve the above object, the technical solution of the present invention is as follows:

[0006] In a first aspect, the present invention proposes a method for optimizing the design of bus lanes considering tail gas emissions at the road section scale, which specifically includes the following steps:

[0007] Step S1, collecting design element information of urban bus lanes, road section traffic flow information, and road section basic information;

[0008] Step S2, using the design element information of the urban bus lane, the road section traffic flow information, and the road section basic information obtained in step S1, a virtual simulation method is used to construct a vehicle driving analysis scenario that maps the actual bus lane road section, obtain the traffic flow operation condition data of the road section where the urban bus lane is located, and obtain the exhaust emission equivalent of the vehicles on the road section;

[0009] Step S3, using the design element information of the urban bus lane obtained in step S1, combined with the exhaust emission equivalent information of the road section vehicles obtained in step S2, to integrate the data, and adopt an encapsulated feature selection method to screen the feature variables;

[0010] Step S4, using the exhaust emission equivalent information of the road section vehicles obtained in step S2 and the characteristic variables obtained in step S3, an interpretable machine learning method is used to build a prediction model, and the contribution of the characteristic variables to the exhaust emission of the road section vehicles is obtained by the SHAP value method;

[0011] Step S5, using the influence contribution of the characteristic variables obtained in step S4, taking minimizing the total amount of vehicle exhaust emissions on the road section as the objective function, adopting an optimization algorithm, performing parameter optimization on the characteristic variables, and determining an optimal design scheme for the bus lane.

[0012] Preferably, traffic flow information of urban bus lane sections is obtained through on-site collection using drone aerial photography technology; design element information and basic section information of urban bus lanes are obtained through on-site collection using drone aerial photography technology, or relevant information is provided by the road design department.

[0013] Preferably, the design element information of the urban bus lane includes at least: the type of bus lane, the location of the bus lane, the time when the bus lane is activated, the type of stop, and the size of the parking area;

[0014] The road section traffic flow information at least includes: the traffic volume of each lane of the road section during peak hours, the vehicle type composition and proportion of each lane of the road section during peak hours, the travel speed of different vehicle types in each lane during peak hours, the headway of vehicles in each lane during peak hours, the probability of vehicles changing lanes on the left and right sides in each lane during peak hours, and the average stop time of public transport vehicles during peak hours;

[0015] The basic information of the road section includes at least: road type, road section length, number of road section lanes, and lane width.

[0016] Preferably, the specific process of step S2 is as follows:

[0017] Step S21, based on the acquired urban bus lane design element information, road section traffic flow information, and road section basic information, a virtual road section vehicle following model is built in the software using a virtual simulation method, lane change rules are formulated, and a vehicle driving analysis scenario that maps the actual bus lane section is constructed;

[0018] Step S22, configuring bus and social vehicle models based on the simulation scenario;

[0019] Step S23, based on the vehicle driving analysis scenario of the constructed bus lane section, the bus lane type, bus lane setting location, bus lane activation time, bus stop type, and parking area size are adjusted to carry out a virtual control test;

[0020] Step S24, outputting the road section vehicle operating condition data of different test scenarios; wherein the vehicle operating condition data includes: virtual simulation timestamp 、 The instantaneous velocity and acceleration at the moment;

[0021] Step S25, based on the vehicle operating condition data of the road sections in different test scenarios, using the exhaust emission measurement model, obtain the exhaust emission equivalent of the vehicles on the road sections in different test scenarios.

[0022] Preferably, the types of bus lanes include: conventional bus lanes, intermittent bus lanes, and intermittent bus priority bus lanes;

[0023] The locations of the bus lanes include: roadside bus lanes, secondary roadside bus lanes, and median bus lanes;

[0024] The activation time of the bus lanes mentioned above includes: peak-hour bus lanes and all-day bus lanes;

[0025] The types of bus stops include: straight bus stops and bay bus stops;

[0026] The exhaust emission calculation models include: MOVES model, MOBILE model, CMEM model and IVE model.

[0027] Preferably, the specific process of step S3 is as follows:

[0028] Step S31, integrating the acquired information on design elements of the urban bus lanes with the exhaust emission equivalents of vehicles on the road sections under different test scenarios acquired in step S2, to form a structured data set containing a mapping relationship between design elements and emission equivalents;

[0029] Step S32, using an adapted encapsulated feature selection method, iteratively screening feature variables of the urban bus lane design element information through variable evaluation and model performance feedback mechanism, using model performance evaluation indicators (accuracy, AUC) as evaluation indicators, and retaining the top 30% of feature variables with importance scores;

[0030] Step S33, using the mutual information method to verify the nonlinear correlation between the retained characteristic variables and the exhaust emission equivalent, determining the dynamic threshold interval through a statistical significance test (p<0.05), eliminating variables with mutual information values ​​less than the threshold, and obtaining the characteristic variables for the initial screening;

[0031] In step S34, the stepwise regression method is used to evaluate the stability and generalization ability of the initially screened feature variables in different test scenarios, and the feature sensitivity is analyzed through multi-scenario cross-validation. The feature variables with variance inflation factor (VIF>5) are eliminated to determine the final feature variables.

[0032] Preferably, the encapsulated feature selection method includes: recursive feature elimination method, simulated annealing-based feature selection method, genetic algorithm-based feature selection method, heuristic search algorithm, exhaustive search method, sequential forward selection method, and sequential backward elimination method;

[0033] Preferably, the specific process of step S4 is as follows:

[0034] Step S41, based on the feature variable set obtained in S3, combined with the exhaust emission equivalent data of vehicles on the road section under different test scenarios, an interpretable machine learning algorithm is used to load the data set and construct a prediction model of feature variables and exhaust emissions;

[0035] The machine learning algorithms include: random forest method, XGBoost algorithm, gradient boosting decision tree method;

[0036] Step S42, based on the characteristic variables and the exhaust emission prediction model constructed in step S41, the characteristic variables are encoded or standardized; the data set is divided, 70% of the data set is used for model training, and the remaining 30% of the data set is used for model testing, model parameters are set, and model training is carried out;

[0037] Step S43, using a model performance diagnosis method, through error propagation analysis, calculating the mean absolute percentage error (MAPE) between the model prediction value and the actual emission equivalent, evaluating the model performance, and ensuring the model stability and generalization ability;

[0038] Step S44, using a grid search method to tune the hyperparameters of the model, and through sensitivity analysis, evaluating the impact of different hyperparameters on model performance, and further optimizing the model;

[0039] Step S45, using the SHAP value method, calculate the Shapley value of each characteristic variable in the prediction model, obtain the contribution of different characteristic variables to vehicle exhaust emissions on the bus lane section based on the calculated Shapley value, and generate a ranking of the contribution of the characteristic variables.

[0040] Preferably, the specific process of step S5 is as follows:

[0041] Step S51, based on the influence contribution ranking information of the characteristic variables obtained in step S4, combined with the exhaust emission equivalent of vehicles on the road section under different test scenarios, an optimization algorithm is used to carry out the optimization design of the bus lane;

[0042] The optimization algorithms include: genetic algorithm, ant colony optimization algorithm, particle swarm optimization algorithm;

[0043] Step S52, setting the optimization objective function: the goal is to minimize the total amount of vehicle exhaust emissions on the road section, and the emission types include CO 2 , CO, NO X , PM2.5, using weighted summation, with weights set according to national standards (GB18352.6-2016), setting constraints, and setting optimization priorities in descending order of the contribution of characteristic variables, so as to prioritize the optimization of characteristic variables with large contribution;

[0044] The constraints include: road section form, road section traffic volume, road section speed limit, bus departure frequency, road design specifications, environmental regulations, and budget constraints;

[0045] Step S53, setting optimization algorithm parameters, using an automated parameter adjustment method to determine key parameters of the algorithm, including the number of iterations, crossover probability, and mutation probability;

[0046] The automated parameter adjustment methods include: grid search method, random search method, Bayesian optimization method;

[0047] Step S54, running the optimization algorithm, optimizing the characteristic variables and outputting the optimization results, and determining the optimal design scheme of the bus lane;

[0048] Step S55, feeding back the optimization scheme to the simulation software to verify the emission reduction effect of the optimization scheme.

[0049] In the second aspect, the present invention proposes a bus lane optimization design system that takes into account exhaust emissions at the road section scale, including a processor, a memory, and a computer program stored on the memory. When the processor executes the computer program, it specifically executes the steps in the above-mentioned bus lane optimization design method that takes into account exhaust emissions at the road section scale.

[0050] Compared with the prior art, the present invention has the following beneficial effects:

[0051] (1) By using the measured road section traffic flow information, road section basic information, and urban bus lane design element information as model input parameters, the technical solution disclosed in the present invention can be applied to actual urban bus lane road section scenarios, and the optimized design of bus lanes considering vehicle exhaust emissions at the road section scale can be achieved, filling the gap in technical solutions in this field;

[0052] (2) It is compatible with the measured data of road sections and the existing virtual testing technology, avoiding the disadvantage that pure theoretical calculations easily lead to calculation results that are too ideal compared to the actual situation;

[0053] (3) The proposed bus lane optimization design method that takes into account exhaust emissions at the road section scale can significantly reduce vehicle exhaust emissions on urban bus lane sections, making up for the deficiency that existing research only optimizes bus lanes for the operating efficiency of the bus system, but does not propose an ecological design plan for bus lanes from the perspective of the traffic environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 A flowchart of a bus lane optimization design method provided by an embodiment of the present invention;

[0055] Figure 2 This is a flow chart of an embodiment of the present invention using a virtual simulation method to obtain the equivalent of vehicle exhaust emissions on a road section;

[0056] Figure 3 It is a flowchart of constructing a vehicle exhaust emission prediction model for a bus lane section and screening characteristic variables provided by an embodiment of the present invention;

[0057] Figure 4 It is a flow chart of the optimized design of bus lane provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0058] The following is combined with Figure 1-4 , the technical solution of the present invention is specifically described.

[0059] It should be noted that the following detailed descriptions are illustrative and are intended to provide further explanation of the present application. Unless otherwise specified, all technical and scientific terms used in this specification have the same meanings as those commonly understood by those skilled in the art to which the present application belongs.

[0060] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, it indicates the presence of features, steps, operations, devices, components and / or combinations thereof.

[0061] like Figure 1 As shown, the present invention proposes a method for optimizing the design of bus lanes that takes into account exhaust emissions at the road section scale, by collecting design element information of urban bus lanes, road section traffic flow information, and road section basic information; using a virtual simulation method to obtain traffic flow operating condition data of the road section where the urban bus lanes are located, and obtaining the exhaust emission equivalent of vehicles on the road section; using an encapsulated feature selection method to screen feature variables from the design element information of urban bus lanes; using an interpretable machine learning method to analyze the impact of feature variables on exhaust emissions of road section vehicles, and outputting the impact contribution ranking of feature variables; using minimization of the total amount of exhaust emissions of road section vehicles as the objective function, using an optimization algorithm to optimize the design of bus lanes. Specifically, it includes the following steps:

[0062] Step S1, collecting design element information of urban bus lanes, road section traffic flow information, and road section basic information;

[0063] The design element information of the urban bus lane includes at least: the type of bus lane, the location of the bus lane, the time when the bus lane is activated, the type of stop, and the size of the parking area;

[0064] The road section traffic flow information at least includes: the traffic volume of each lane of the road section during peak hours, the vehicle type composition and proportion of each lane of the road section during peak hours, the travel speed of different vehicle types in each lane during peak hours, the headway of vehicles in each lane during peak hours, the probability of vehicles changing lanes on the left and right sides in each lane during peak hours, and the average stop time of public transport vehicles during peak hours;

[0065] The basic information of the road section includes at least: road form, road section length, number of lanes in the road section, and lane width;

[0066] Among them, traffic flow information of urban bus lane sections can be obtained through on-site collection methods using drone aerial photography technology; basic information of road sections and design elements of bus lanes can be obtained through on-site collection using drone aerial photography technology, or relevant information can be provided by the road design department;

[0067] Step S2, using a virtual simulation method to obtain the traffic flow operating condition data of the section where the urban bus lane is located, and obtain the exhaust emission equivalent of the vehicles on the section; the flow chart of this step is as follows Figure 2 As shown;

[0068] Step S21, based on the acquired urban bus lane design element information, road section traffic flow information, and road section basic information, a road section vehicle following model is built in the software using a virtual simulation method, lane change rules are formulated, and a vehicle driving analysis scenario that maps the actual bus lane section is constructed;

[0069] Among them, the software-based virtual simulation method can be realized independently or jointly with scenario-based urban bus lane virtual simulation software such as SUMO, VISSIM, and Carla. The modeling effectiveness of the above software has been widely verified in the field.

[0070] Step S22, configuring bus and social vehicle models based on the simulation scenario;

[0071] Step S23, based on the vehicle driving analysis scenario of the constructed bus lane section, the bus lane type, bus lane setting location, bus lane activation time, bus stop type, and parking area size are adjusted to carry out a virtual control test;

[0072] The types of bus lanes include, but are not limited to: conventional bus lanes, intermittent bus lanes, and intermittent bus priority bus lanes;

[0073] The location of the bus lane includes but is not limited to: roadside bus lane, secondary roadside bus lane, and middle road bus lane;

[0074] The activation time of the bus lanes includes but is not limited to: peak-hour bus lanes and all-day bus lanes;

[0075] The types of bus stops include but are not limited to: linear bus stops and bay bus stops;

[0076] Step S24, outputting the road section vehicle operating condition data of different test scenarios; wherein the vehicle operating condition data includes: virtual simulation timestamp 、 The instantaneous velocity and acceleration at the moment;

[0077] Step S25, based on the vehicle operating condition data of the road sections in different test scenarios, using the exhaust emission calculation model, obtaining the exhaust emission equivalent of the vehicles on the road sections in different test scenarios;

[0078] Among them, the calculation of vehicle exhaust emissions on urban bus lanes can be achieved by relying on vehicle exhaust emission prediction models such as the MOVES model, MOBILE model, CMEM model, and IVE model.

[0079] Step S3, using the encapsulated feature selection method, selects the design element information of the urban bus lane as the feature variable; the flow chart of this step is as follows Figure 3 As shown;

[0080] Step S31, integrating the acquired information on design elements of the urban bus lanes with the exhaust emission equivalents of vehicles on the road sections under different test scenarios acquired in step S2, to form a structured data set containing a mapping relationship between design elements and emission equivalents;

[0081] Step S32, using an adapted encapsulated feature selection method, iteratively screening feature variables of the urban bus lane design element information through variable evaluation and model performance feedback mechanism, using model performance evaluation indicators (accuracy, AUC) as evaluation indicators, and retaining the top 30% of feature variables with importance scores;

[0082] The encapsulated feature selection methods include: recursive feature elimination method, simulated annealing-based feature selection method, genetic algorithm-based feature selection method, heuristic search algorithm, exhaustive search method, sequential forward selection method, and sequential backward elimination method;

[0083] Step S33, using the mutual information method to verify the nonlinear correlation between the retained characteristic variables and the exhaust emission equivalent, determining the dynamic threshold interval through a statistical significance test (p<0.05), eliminating variables with mutual information values ​​less than the threshold, and obtaining the characteristic variables for the initial screening;

[0084] In step S34, the stepwise regression method is used to evaluate the stability and generalization ability of the initially screened feature variables in different test scenarios, and the feature sensitivity is analyzed through multi-scenario cross-validation. The feature variables with variance inflation factor (VIF>5) are eliminated to determine the final feature variables.

[0085] Step S4, using an interpretable machine learning method to analyze the impact of characteristic variables on vehicle exhaust emissions on the road section; the flowchart of this step is as follows Figure 3 As shown;

[0086] Step S41, based on the feature variable set obtained in S3, combined with the exhaust emission equivalent data of vehicles on the road section under different test scenarios, an interpretable machine learning algorithm is used to load the data set and construct a prediction model of feature variables and exhaust emissions;

[0087] The machine learning algorithms include: random forest method, XGBoost algorithm, gradient boosting decision tree method;

[0088] Step S42, based on the characteristic variables and the exhaust emission prediction model constructed in step S41, the characteristic variables are encoded or standardized; the data set is divided, 70% of the data set is used for model training, and the remaining 30% of the data set is used for model testing, model parameters are set, and model training is carried out;

[0089] Step S43, using a model performance diagnosis method, through error propagation analysis, calculating the mean absolute percentage error (MAPE) between the model prediction value and the actual emission equivalent, evaluating the model performance, and ensuring the model stability and generalization ability;

[0090] Step S44, using a grid search method to tune the hyperparameters of the model, and through sensitivity analysis, evaluating the impact of different hyperparameters on model performance, and further optimizing the model;

[0091] Step S45, using the SHAP value method, calculate the Shapley value of each characteristic variable in the prediction model, obtain the contribution of different characteristic variables to vehicle exhaust emissions on the bus lane section based on the calculated Shapley value, and generate a ranking of the contribution of the characteristic variables.

[0092] Step S5, taking minimizing the total amount of vehicle exhaust emissions on the road section as the objective function, an optimization algorithm is used to optimize the design of the bus lane; the flow chart of this step is as follows Figure 4 As shown;

[0093] Step S51, based on the influence contribution ranking information of the characteristic variables obtained in step S4, combined with the exhaust emission equivalent of vehicles on the road section under different test scenarios, an optimization algorithm is used to carry out the optimization design of the bus lane;

[0094] The optimization algorithms include: genetic algorithm, ant colony optimization algorithm, particle swarm optimization algorithm;

[0095] Step S52, setting the optimization objective function: the goal is to minimize the total amount of vehicle exhaust emissions on the road section, and the emission types include CO 2 , CO, NO X , PM2.5, etc., using a weighted summation method, with weights set according to national standards (GB18352.6-2016), constraints set, and the priority of optimization set in descending order of the contribution of the characteristic variables, so as to give priority to optimizing the characteristic variables with large contribution;

[0096] The constraints include: road section form, road section traffic volume, road section speed limit, bus departure frequency, road design specifications, environmental regulations, and budget constraints;

[0097] Step S53, setting optimization algorithm parameters, using an automated parameter adjustment method to determine key parameters such as the number of iterations, crossover probability, and mutation probability of the algorithm;

[0098] The automated parameter adjustment methods include: grid search method, random search method, Bayesian optimization method;

[0099] Step S54, running the optimization algorithm, optimizing the characteristic variables and outputting the optimization results, and determining the optimal design scheme of the bus lane;

[0100] Step S55, feeding back the optimization scheme to the simulation software to verify the emission reduction effect of the optimization scheme.

[0101] According to the above key information, the optimized design of bus lanes taking into account vehicle exhaust emissions at the road section scale can be achieved based on the solution provided in this embodiment.

[0102] The present invention also proposes a bus lane optimization design system that takes into account exhaust emissions at the road section scale, including a processor, a memory, and a computer program stored in the memory. When the processor executes the computer program, it specifically executes the steps in the above-mentioned bus lane optimization design method that takes into account exhaust emissions at the road section scale.

[0103] In summary, the present invention designs a bus lane optimization design method and system that considers exhaust emissions at the road section scale. By collecting the design element information of the urban bus lane, the road section traffic flow information, and the road section basic information as the model input, a virtual simulation method is used to obtain the exhaust emission data of the road section vehicles, and interpretable machine learning is used to analyze the influence of characteristic variables. With the goal of minimizing exhaust emissions, an optimization algorithm is used to improve the design of the bus lane, so as to significantly reduce the exhaust emissions of vehicles on the road sections of the urban bus lanes. The design method of the present invention fills the gap in the technical solutions in this field; avoids the drawback that pure theoretical calculations easily lead to calculation results that are too ideal compared to the actual situation; and makes up for the shortcomings of the existing research that only optimizes the bus lanes for the operating efficiency of the bus system, and does not propose an ecological design solution for the bus lanes from the perspective of the traffic environment.

[0104] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0105] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0106] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0107] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0108] The above is only a preferred embodiment of the present invention, and does not limit the present invention in other forms. Any technician familiar with the profession may use the above disclosed technical content to change or modify it into an equivalent embodiment with equivalent changes. However, any simple modification, equivalent change and modification made to the above embodiment according to the technical essence of the present invention without departing from the technical solution of the present invention still belongs to the protection scope of the technical solution of the present invention.

[0109] This patent is not limited to the above-mentioned optimal implementation mode. Anyone can derive various other forms of a road traffic node section driving adaptability evaluation method for autonomous driving under the inspiration of this patent. All equal changes and modifications made according to the scope of the patent application of the present invention should be covered by this patent.

Claims

1. A bus lane optimization design method considering tail gas emissions at the road section scale, characterized in that: The specific steps include: Step S1, collecting design element information of urban bus lanes, road section traffic flow information, and road section basic information; Step S2, using the design element information of the urban bus lane, the road section traffic flow information, and the road section basic information obtained in step S1, a virtual simulation method is used to construct a vehicle driving analysis scenario that maps the actual bus lane road section, obtain the traffic flow operation condition data of the road section where the urban bus lane is located, and obtain the exhaust emission equivalent of the vehicles on the road section; Step S3, using the design element information of the urban bus lane obtained in step S1, combined with the exhaust emission equivalent information of the road section vehicles obtained in step S2, to integrate the data, and adopt an encapsulated feature selection method to screen the feature variables; Step S4, using the exhaust emission equivalent information of the road section vehicles obtained in step S2 and the characteristic variables obtained in step S3, an interpretable machine learning method is used to build a prediction model, and the contribution of the characteristic variables to the exhaust emission of the road section vehicles is obtained by the SHAP value method; Step S5, using the influence contribution of the characteristic variables obtained in step S4, taking minimizing the total amount of vehicle exhaust emissions on the road section as the objective function, adopting an optimization algorithm, performing parameter optimization on the characteristic variables, and determining an optimal design scheme for the bus lane; The specific process of step S3 is as follows: Step S31, integrating the acquired information on design elements of urban bus lanes with the exhaust emission equivalents of vehicles on the road sections under different test scenarios acquired in step S2, to form a structured data set containing a mapping relationship between design elements and emission equivalents; Step S32, using an adapted encapsulated feature selection method, iteratively screening the feature variables of the design element information of the urban bus lane through variable evaluation and model performance feedback mechanism, using the model performance evaluation index as the evaluation index, and retaining the feature variables ranked in the top 30% of the importance score; Step S33, using the mutual information method to verify the nonlinear correlation between the retained characteristic variables and the exhaust emission equivalent, determining the dynamic threshold interval through a statistical significance test, eliminating the variables with mutual information values ​​less than the threshold, and obtaining the characteristic variables for the initial screening; Step S34, using stepwise regression method to evaluate the stability and generalization ability of the initially screened feature variables in different test scenarios, analyzing feature sensitivity through multi-scenario cross-validation, eliminating feature variables with variance inflation factors, and confirming the final feature variables; The specific process of step S4 is as follows: Step S41, based on the feature variable set obtained in S3, combined with the exhaust emission equivalent data of vehicles on the road section under different test scenarios, an interpretable machine learning algorithm is used to load the data set and construct a prediction model of feature variables and exhaust emissions; The machine learning algorithms include: random forest method, XGBoost algorithm, gradient boosting decision tree method; Step S42, based on the characteristic variables and the exhaust emission prediction model constructed in step S41, the characteristic variables are encoded or standardized; the data set is divided, 70% of the data set is used for model training, and the remaining 30% of the data set is used for model testing, model parameters are set, and model training is carried out; Step S43, using a model performance diagnosis method, through error propagation analysis, calculating the mean absolute percentage error between the model prediction value and the actual emission equivalent, evaluating the model performance, and ensuring the model stability and generalization ability; Step S44, using a grid search method to tune the hyperparameters of the model, and through sensitivity analysis, evaluating the impact of different hyperparameters on model performance, and further optimizing the model; Step S45, using the SHAP value method to calculate the Shapley value of each characteristic variable in the prediction model, and according to the calculated Shapley value, obtaining the contribution of different characteristic variables to vehicle exhaust emissions in the bus lane section, and generating a ranking of the contribution of the characteristic variables; The specific process of step S5 is as follows: Step S51, based on the influence contribution ranking information of the characteristic variables obtained in step S4, combined with the exhaust emission equivalent of vehicles on the road section under different test scenarios, an optimization algorithm is used to carry out the optimization design of the bus lane; The optimization algorithms include: genetic algorithm, ant colony optimization algorithm, particle swarm optimization algorithm; Step S52, setting the optimization objective function: the goal is to minimize the total amount of vehicle exhaust emissions on the road section, and the emission types include CO2, CO, NO X , PM2.5, using the weighted summation method, setting constraints, and setting the optimization priority in descending order of the contribution of the characteristic variables, so as to give priority to optimizing the characteristic variables with large contribution; The constraints include: road section form, road section traffic volume, road section speed limit, bus departure frequency, road design specifications, environmental regulations, and budget constraints; Step S53, setting optimization algorithm parameters, using an automated parameter adjustment method to determine key parameters of the algorithm, including the number of iterations, crossover probability, and mutation probability; The automated parameter adjustment methods include: grid search method, random search method, Bayesian optimization method; Step S54, running the optimization algorithm, optimizing the characteristic variables and outputting the optimization results, and determining the optimal design scheme of the bus lane; Step S55, feeding back the optimization scheme to the simulation software to verify the emission reduction effect of the optimization scheme.

2. The bus lane optimization design method considering tail gas emissions at the road section scale according to claim 1 is characterized in that: Traffic flow information on urban bus lane sections is obtained through on-site collection using drone aerial photography technology; design element information and basic section information of urban bus lanes is obtained through on-site collection using drone aerial photography technology, or relevant information is provided by the road design department.

3. The bus lane optimization design method considering tail gas emissions at the road section scale according to claim 1 is characterized in that: The design element information of the urban bus lane includes at least: the type of bus lane, the location of the bus lane, the time when the bus lane is activated, the type of stop, and the size of the parking area; The road section traffic flow information at least includes: the traffic volume of each lane of the road section during peak hours, the vehicle type composition and proportion of each lane of the road section during peak hours, the travel speed of different vehicle types in each lane during peak hours, the headway of vehicles in each lane during peak hours, the probability of vehicles changing lanes on the left and right sides in each lane during peak hours, and the average stop time of public transport vehicles during peak hours; The basic information of the road section includes at least: road type, road section length, number of road section lanes, and lane width.

4. The bus lane optimization design method considering tail gas emissions at the road section scale according to claim 1 is characterized in that: The specific process of step S2 is as follows: Step S21, based on the acquired urban bus lane design element information, road section traffic flow information, and road section basic information, a virtual road section vehicle following model is built in the software using a virtual simulation method, lane change rules are formulated, and a vehicle driving analysis scenario that maps the actual bus lane section is constructed; Step S22, configuring bus and social vehicle models based on the simulation scenario; Step S23, based on the vehicle driving analysis scenario of the constructed bus lane section, the bus lane type, bus lane setting location, bus lane activation time, bus stop type, and parking area size are adjusted to carry out a virtual control test; Step S24, outputting the vehicle operating condition data of the road sections in different test scenarios; wherein the vehicle operating condition data includes: virtual simulation timestamp, instantaneous speed and instantaneous acceleration at the time; Step S25, based on the vehicle operating condition data of the road sections in different test scenarios, using the exhaust emission measurement model, obtain the exhaust emission equivalent of the vehicles on the road sections in different test scenarios.

5. The bus lane optimization design method considering tail gas emissions at the road section scale according to claim 4 is characterized in that: The types of bus lanes include: conventional bus lanes, intermittent bus lanes, and intermittent bus priority bus lanes; The locations of the bus lanes include: roadside bus lanes, secondary roadside bus lanes, and median bus lanes; The activation time of the bus lanes mentioned above includes: peak-hour bus lanes and all-day bus lanes; The types of bus stops include: straight bus stops and bay bus stops; The exhaust emission calculation models include: MOVES model, MOBILE model, CMEM model and IVE model.

6. The bus lane optimization design method considering tail gas emissions at the road section scale according to claim 1 is characterized in that: The encapsulated feature selection method includes: recursive feature elimination method, simulated annealing-based feature selection method, genetic algorithm-based feature selection method, heuristic search algorithm, exhaustive search method, sequential forward selection method, and sequential backward elimination method.

7. A bus lane optimization design system considering tail gas emissions at the road section scale, characterized in that: The method comprises a processor, a memory and a computer program stored in the memory. When the processor executes the computer program, the method specifically executes the steps in the bus lane optimization design method considering the exhaust emission at the road section scale as described in any one of claims 1 to 6.

Images