A method for predicting medium- and long-term trends of carbon emissions from highway traffic flows
By combining statistical analysis, causal relationship and system dynamics, a carbon emission prediction model for traffic flow was constructed, which solved the problem that complex relationships and long-term effects could not be accurately captured in the existing technology, and achieved more accurate carbon emission trend prediction and reduction Schedule strategy formulation.
Patent Information
- Application Number
- CN202411231945.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-04
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2044-09-04
AI Technical Summary
The existing technology cannot accurately capture complex relationships in the forecast of carbon emission trends in traffic flows, and it is difficult to reflect the impact of policy changes and technological progress on carbon emissions. The reliability of model predictions depends on data quality and cannot effectively predict long-term effects.
Using multiple regression analysis methods based on statistical analysis, causal relationship and system dynamics, combined with visualization tools and simulation and comparison simulation technology, a traffic flow carbon emission prediction model is constructed, key factors are identified, causal chains and feedback mechanisms are discussed, and model stability is verified. Sensitivity, analyze carbon emission trends under different policies.
It improves the accuracy and scope of application of carbon emission forecasts, can effectively reflect the dynamic changes in traffic flow, supports the formulation of scientific emission reduction strategies, and provides reliable decision-making basis.
Smart Images

Figure CN118735142B_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of traffic flow carbon emission prediction, and in particular relates to a method for predicting the medium- and long-term trend of highway traffic flow carbon emission. Background Art
[0002] In recent years, the global greenhouse gas emission problem has become increasingly serious. As one of the main greenhouse gases, carbon dioxide (CO2) emissions continue to grow. According to statistics from the International Energy Agency (IEA), China's CO2 emissions increased by 565 million tons in 2023 to 12.6 billion tons, an increase of 4.7%. The transportation sector is one of the main sources of global greenhouse gas and air pollutant emissions, accounting for 15% of global greenhouse gas emissions, and is showing a continuous growth trend. Among various modes of transportation, road transportation accounts for an important proportion of carbon emissions.
[0003] At present, scholars have adopted a variety of model methods in the study of carbon emission trends, including input-output models, linear programming models, MARKAL, IEPM models, scenario analysis models, integrated energy and environmental policy assessment models (IPAC), STIRPAT models, and linear regression models. However, most of the above models simplify the complexity of the real world for the convenience of calculation and processing, which may lead to the inability to accurately capture the complex relationship between economic activities and carbon emissions; many models, such as linear programming and linear regression models, are based on linear structures and are difficult to capture the nonlinear characteristics of actual systems; the accuracy of models such as MARKAL is highly dependent on the quality and completeness of input data, and insufficient or inaccurate data will directly affect the reliability of model predictions; existing models have limitations in integrating the impact of policy changes, market conditions, and technological progress on carbon emission trends, and cannot effectively reflect the complex interactions between the various influencing parts of the emission system, nor do they fully consider the dynamic changes of the carbon emission system over time, making it difficult to predict the long-term effects of these factors. Summary of the invention
[0004] The purpose of the present invention is to provide a method for predicting the medium- and long-term trend of carbon emissions from highway traffic flows, aiming to solve the technical problems existing in the prior art identified in the background technology.
[0005] The present invention is implemented as follows: a method for predicting the medium- and long-term trend of carbon emissions from highway traffic flow, the method comprising:
[0006] S1. Based on the historical data of carbon emissions from highways in the study area, statistical analysis techniques and visualization tools are used to deeply explore the temporal and spatial variation patterns and characteristics of the data, find the most significant change cycle, and determine the spatial and temporal boundaries of the model.
[0007] S2. Based on multivariate regression analysis technology, the core factors of traffic flow carbon emissions are identified, and in-depth analysis is conducted to determine the key variables that need to be studied, revealing the main factors affecting carbon emissions, and providing a basis for the construction and optimization of subsequent prediction models.
[0008] S3. Use causal analysis techniques to explore the causal chain and feedback mechanism between the factors affecting traffic flow carbon emissions, divide the system levels, explore the relationship between variables within the subsystem, and draw system flow charts and causal relationship diagrams.
[0009] S4. Based on the system dynamics modeling technology, a prediction model for traffic flow carbon emissions is constructed, variable types based on causal relationships are set, and mathematical function relationships between variables are defined. On the basis of establishing model parameters, a system dynamics model is established.
[0010] S5. Apply model verification technology to comprehensively verify the traffic flow carbon emission model, including structural verification, historical backtracking verification and sensitivity verification, so as to evaluate the stability, effectiveness and sensitivity of the model.
[0011] S6. Use simulation control technology to analyze the carbon emission trends of traffic flows under different highway carbon reduction policies. By comparing the simulation results under different policy options, effective carbon reduction strategies can be screened out, providing a quantitative basis for policy decisions.
[0012] As a further solution of the present invention, S1 specifically includes:
[0013] Collect and clean historical highway traffic flow and carbon emissions data;
[0014] Use the Z-score method to identify and remove outliers and remove outliers:
[0015] ;
[0016] in, represents the standard deviation distance of the data point from the mean, represents a data point, is the mean value of the data, is the standard deviation of the data, if , then the determination value is an outlier;
[0017] Interpolation and mean completion methods are used to handle missing data values;
[0018] Specific:
[0019] Interpolation method:
[0020] ;
[0021] in, Indicates at a point in time The observed value of Indicates at a point in time The observed value of Indicates at a point in time The observed value of Here, it represents the time index, indicating different time points in the sequence;
[0022] Mean Completion Method:
[0023] ;
[0024] in, represents the estimated value of the missing value, represents the total number of observed samples, Indicates Observations, represents different observation samples;
[0025] Analyze and visualize data based on statistical analysis tools and visualization tools to reveal the distribution characteristics, change trends and correlations of data;
[0026] Use time series analysis models, spatial autocorrelation Moran's I model in geographic information system analysis, and visualization tools to obtain the temporal and spatial development laws and trends of data;
[0027] Specific:
[0028] Time Series Analysis Model:
[0029] ;
[0030] in Representative time The last observed value, Refers to seasonal ingredients, Refers to other random components;
[0031] Moran's I Model
[0032] ;
[0033] in represents the total number of samples, Represents the variable value, is the variable mean, represents the spatial weight matrix, ;
[0034] represents the Moran's I index, which is used to measure spatial autocorrelation. represents the total number of observed samples, is the variable mean, represents the spatial weight matrix, , and They represent different observation samples, and is corresponding to and Observed value of
[0035] The main periodic components in carbon emission data were identified based on fast Fourier transform and inverse transform;
[0036] Among them, the carbon emissions data are:
[0037] ,in, ;
[0038] Here, Indicated in Carbon emission data of sample points, represents the total number of observed samples, The last index indicating the number of samples;
[0039] The mathematical expression of fast Fourier transform is:
[0040] ;
[0041] in Indicates the frequency The complex coefficients of represents an imaginary unit;
[0042] Indicated in Carbon emission data of sample points, Represents the index of the frequency in the frequency domain after FFT transformation, The value range is 0 to ;
[0043] Represents the sample point index in the time domain, ranging from 0 to , which is related to time and represents the Points, represents the total number of observed samples;
[0044] By calculation The spectrum is obtained, expressed as amplitude spectrum and phase spectrum, specifically:
[0045] Amplitude spectrum:
[0046] ;
[0047] Phase spectrum:
[0048] ;
[0049] in, represents the frequency index, Indicates the frequency The phase at
[0050] and Representing the real part and the imaginary part respectively, by analyzing the frequency spectrum, finding the frequency corresponding to the peak in the amplitude spectrum, and identifying the main periodic components in the carbon emission data;
[0051] Convert the frequency domain signal back to the time domain using an inverse fast Fourier transform:
[0052] ;
[0053] in, Indicates at time The reconstructed data at Represents the sample point index in the time domain, ranging from 0 to , which is related to time and represents the number of points; Represents the total number of observed samples and the size of the entire data set; Indicates the index of the frequency in the frequency domain after FFT transformation, indicates the index of the frequency in the frequency domain after FFT transformation, The value range is 0 to ; represents an imaginary unit;
[0054] As a further solution of the present invention, S2 specifically includes:
[0055] Based on regression analysis, the variables that affect carbon emissions are determined, and a multivariate linear regression analysis model is established to conduct regression analysis:
[0056] ;
[0057] in Refers to the response variable of the model, specifically carbon emissions, Refers to the model independent variables, specifically traffic flow, vehicle type, and fuel use, is the intercept term, is the regression coefficient, is the error term;
[0058] Variable screening is performed based on the stepwise regression method, specifically:
[0059] Perform significance test on variables:
[0060] ;
[0061] in express The standard error of The corresponding model independent variables, It represents a statistic, which is a numerical value and is used to determine whether the independent variable has a significant impact on the response variable;
[0062] Test the significance of variables by standardizing regression coefficients:
[0063] ;
[0064] in is the independent variable The standard deviation of is the response variable The standard deviation of represents the standardized regression coefficient, The corresponding model independent variables, represents the unstandardized regression coefficient;
[0065] Based on the above results, the key variables affecting carbon emissions are determined and the determination coefficient is used As a reference, measure the fitting effect, mean square error Evaluate the model optimization effect and optimize the prediction model structure;
[0066] in, Represents the sample point index in the time domain, The corresponding model independent variables;
[0067] As a further solution of the present invention, S3 specifically includes:
[0068] Collect and organize traffic flow carbon emission data, and apply Granger causality test method to determine the causal relationship and feedback mechanism between various factors. Specifically:
[0069] ;
[0070] in is the explained variable, is the regression coefficient, is the error term;
[0071] is the explained variable, indicating the of carbon emissions, Indicates at time of carbon emissions, Indicates at time The independent variable reflects the impact of past factors on current carbon emissions;
[0072] is the regression coefficient, is the error term, represents the number of lags, which refers to the number of past data points considered in the regression model;
[0073] Divide the system level and decompose the traffic flow carbon emission system into different subsystems, including the socio-economic subsystem, the highway traffic flow subsystem, and the energy consumption and emission subsystem. Within each subsystem, explore the relationship between variables, analyze the interaction and impact, and the degree of contribution to the overall system;
[0074] Use the data and analysis results to draw system flow charts and cause-and-effect diagrams;
[0075] As a further solution of the present invention, S4 specifically includes:
[0076] Based on the causal relationship, the types of variables in the traffic flow carbon emission system are set, including identifying state variables and rate variables;
[0077] Analyze the mathematical function relationship between various variables in the traffic flow carbon emission system, including data analysis and model fitting. Through the analysis of historical data and model fitting, quantify the causal relationship between variables and establish a mathematical function model. Specifically:
[0078] Quantify the mathematical relationship between variables based on regression analysis and time series analysis methods:
[0079] Regression Analysis:
[0080] ;
[0081] Time Series Analysis:
[0082] ;
[0083] in, Representative time The last observed value, is a constant term, is the autoregressive term coefficient, is the moving average coefficient, is the error term;
[0084] The system dynamics model simulates the dynamic evolution of the system and captures the relationship between internal variables. Specifically:
[0085] Determine the model state equations and control equations;
[0086] Model state equation:
[0087] ;
[0088] in, Represents the time index, which is the same as the previous Consistency;
[0089] Indicates at time The state variable represents the current state of the system. Indicates at time Add time increment , used to describe the state of the system at the next time step, Indicates at time The state variables are The difference is that the former reflects the future state, represents the time step;
[0090] Governing equations:
[0091] ;
[0092] in state variables, represents the control variable, Refers to the mathematical functional relationship between variables;
[0093] As a further solution of the present invention, S5 specifically includes:
[0094] Verify the traffic flow carbon emission model, including structural test, historical backtest and sensitivity test;
[0095] Check whether the model is reasonable based on its structure, including the logical consistency between variables and the validity of mathematical expression;
[0096] Perform model backtesting based on historical data to evaluate the model's fit and prediction accuracy in different time periods;
[0097] The sensitivity test of the model is performed by simulating parameter changes to evaluate the response of the model to parameter changes. The response of the model to parameter changes is evaluated. The parameter change rate is:
[0098] ;
[0099] in are model parameters, is the parameter variation, is the change in the response variable.
[0100] As a further solution of the present invention, S6 specifically includes:
[0101] Based on the established system dynamics model, the carbon emission trends of highways under different scenarios are simulated to evaluate the impact of policy implementation and industry development;
[0102] Based on the comparison of simulation results of different policy options, their carbon reduction effects and cost-effectiveness are evaluated.
[0103] The beneficial effects of the present invention are:
[0104] 1. Using statistical analysis techniques and visualization tools can accurately understand the temporal and spatial distribution characteristics of data, providing a scientific basis for the definition of the model. In-depth exploration of data characteristics helps determine the most significant change cycles and spatial and temporal boundaries, enhancing the accuracy and applicability of model predictions.
[0105] 2. Accurately identify and analyze key factors affecting carbon emissions through regression analysis technology. Provide specific variables and their relationship basis for the construction and optimization of the prediction model, ensuring that the model can effectively reflect the actual situation.
[0106] 3. Causal analysis technology reveals the deep connections between variables and the feedback mechanism within the system. It helps build more complex models that are closer to the actual working mode of the system, and improves the model's adaptability to future changes and prediction accuracy.
[0107] 4. The system dynamics model can comprehensively consider the influence of multiple factors and reflect the complex relationship between the dynamic changes of traffic flow and carbon emissions. It provides a powerful tool for in-depth understanding of the mechanism and future trend of traffic flow carbon emissions, and supports the formulation of more effective emission reduction strategies.
[0108] 5. The application of model verification technology improves the stability, effectiveness and sensitivity of the model. Through comprehensive verification of the model, the credibility of the prediction results is ensured, providing a reliable scientific basis for decision makers.
[0109] 6. Simulation control Simulation technology allows different policy options to be tested in a virtual environment, which can effectively compare the impacts of different carbon reduction policies, screen out the optimal strategy, provide quantitative analysis for the formulation of carbon reduction policies, and improve the scientificity and effectiveness of policy making. BRIEF DESCRIPTION OF THE DRAWINGS
[0110] Figure 1 A flowchart of a method for predicting the medium- and long-term trend of carbon emissions from highway traffic flow provided by an embodiment of the present invention;
[0111] Figure 2 A cause-and-effect relationship loop diagram of various variables in the highway traffic flow carbon emission system provided by the present invention;
[0112] Figure 3 The present invention provides a carbon emission stock flow diagram for highway traffic flow. DETAILED DESCRIPTION
[0113] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.
[0114] It is understood that the terms "first", "second", etc. used in this application may be used herein to describe various elements, but unless otherwise specified, these elements are not limited by these terms. These terms are only used to distinguish a first element from another element. For example, without departing from the scope of this application, a first xx script may be referred to as a second xx script, and similarly, a second xx script may be referred to as a first xx script.
[0115] In the analysis of future regional carbon emission trends, the complexity of the highway traffic flow carbon emission system stems from the dynamic changes of multiple factors, including the socio-economic subsystem, the highway traffic flow subsystem, and the energy and emission subsystem. The system dynamics method has obvious advantages in analyzing such complex problems. Therefore, the present invention focuses on the carbon emission problem of highway traffic flow, aiming to use the system dynamics method to comprehensively consider multiple subsystems and factors of the highway traffic flow carbon emission system, such as GDP growth, changes in highway passenger and freight transportation demand, new energy vehicle policies and market penetration, and fuel vehicle technology development, to construct a highway traffic flow carbon emission system evolution model to predict the medium- and long-term trends of carbon emissions. Through the multi-source data calibration model of each module in the model, the baseline value and correlation function are explored to explore the impact of different control measures on carbon emissions, and provide a scientific basis for the low-carbon sustainable development policy of highways.
[0116] like Figure 1 As shown, a method for predicting the medium- and long-term trend of carbon emissions from highway traffic flow is provided, the method comprising:
[0117] S1. Obtain historical data on carbon emissions from highways in the study area, and use statistical analysis techniques and visualization tools to analyze the temporal and spatial variation patterns and characteristics to derive the spatial and temporal boundaries of the model;
[0118] S2. Based on regression analysis technology, identify the factors affecting traffic flow carbon emissions and obtain research variables and factors affecting carbon emissions;
[0119] S3. Analyze the causal chain and feedback mechanism between the factors affecting traffic flow carbon emissions, divide the system level, and analyze the relationship between variables within the subsystem;
[0120] S4. Based on the system dynamics modeling technology, a prediction model for traffic flow carbon emissions is constructed, variable types based on causal relationships are set, mathematical function relationships between variables are defined, and a system dynamics model is established based on the established model parameters;
[0121] S5. Verify the traffic flow carbon emission model, including structural test, historical backtest and sensitivity test;
[0122] S6. Use simulation control technology to analyze the carbon emission trends of traffic flows under different highway carbon reduction policies, and obtain the final carbon reduction strategy by comparing the simulation results under different policy options.
[0123] In S1, based on the historical data of carbon emissions from highways in the study area, statistical analysis techniques and visualization tools are used to deeply explore the temporal and spatial variation patterns and characteristics of the data, identify the most significant change cycles, and determine the spatial and temporal boundaries of the model.
[0124] First, collect historical data, including highway traffic flow and carbon emissions data, perform necessary cleaning and preprocessing on the collected data, and use the Z-score method to identify and remove outliers. ,in represents a data point, is the mean value of the data, is the standard deviation of the data, if , then the determination value is an outlier. Use interpolation method , and mean completion method , handle missing data values and ensure that all data timestamps and units are consistent.
[0125] Based on the historical data of highway carbon emissions, this paper uses statistical techniques such as descriptive statistical analysis, time series analysis and periodic analysis to preliminarily reveal the basic characteristics and changing patterns of the data. Descriptive statistical analysis helps to understand the distribution of carbon emission data, and time series analysis shows the changing trend of carbon emissions over time. Using the time series analysis model ,in Representative time The last observed value, Refers to seasonal ingredients, refers to other random components; and Moran's I model of spatial autocorrelation in GIS analysis ,in represents the total number of samples, Represent variable values, and use data visualization tools to analyze data histograms, density maps, box plots and other images through drawing to obtain the temporal and spatial development patterns and trends of the data.
[0126] Periodicity analysis uses fast Fourier transform (FFT) and inverse transform to identify the main periodic components in carbon emission data. The mathematical expression of FFT is: ,in Indicates the frequency The complex coefficients of Represents an imaginary unit, obtained by calculation The spectrum can be expressed as the amplitude spectrum and phase spectrum in and Represent the real part and the imaginary part respectively. By analyzing the spectrum, we can find the frequencies corresponding to the peaks in the amplitude spectrum. The periodic components corresponding to these frequencies are the main periodic components in the data, thus identifying the main periodic components in the carbon emissions data. The frequency domain signal is converted back to the time domain through the inverse FFT (ifffT) Through the above steps, the periodic patterns in carbon emission data can be revealed, such as daily, weekly, and annual periodic components. Based on these periodic components, further predictions or other analyses can be performed.
[0127] According to the data analysis results, the spatial boundary of the model is determined to be the entire Guangdong Province, and the time boundary is from 2016 to 2035. It is found that the main change cycle of carbon emissions from Guangdong Province's highways is one year, and carbon emissions are higher in spring and autumn, and relatively lower in winter and summer. These findings provide a scientific basis for the formulation of carbon reduction policies in the future, which will help achieve the goal of low-carbon sustainable development in Guangdong Province.
[0128] In S2, based on regression analysis technology, the core factors of traffic flow carbon emissions are identified and studied, and in-depth analysis is performed to determine the key variables that need to be solved, revealing the main factors affecting carbon emissions, and providing a basis for the construction and optimization of subsequent prediction models.
[0129] This paper uses regression analysis technology to identify and study the core factors that affect carbon emissions from highway traffic flow in Guangdong Province. First, a large amount of data related to traffic flow was collected, including variables such as vehicle type, traffic flow, fuel type, and weather conditions. Through multiple regression analysis, the degree of influence of these variables on carbon emissions was deeply explored, and several key variables were preliminarily identified. These key variables lay the foundation for the subsequent construction of a more accurate carbon emission prediction model.
[0130] A multivariate linear regression analysis model was established, and using the prepared data, traffic flow, carbon emissions, etc. were used as independent variables to conduct regression analysis with other factors that may affect carbon emissions. ,in Refers to the response variable of the model (specifically carbon emissions), refers to the model's independent variables (specifically traffic flow, vehicle type, fuel use, etc.), is the intercept term, is the regression coefficient, is the error term)
[0131] Stepwise regression, forward selection, backward elimination or stepwise regression method are used to select significant variables, in which significance test of variables is used ,in express The standard error of the regression coefficient is: , is the independent variable The standard deviation of is the response variable The standard deviation of the variables was used to test the significance of the variables.
[0132] Based on the above results, the key variables that have the most significant impact on carbon emissions are determined, and the prediction model structure is optimized, including the addition or removal of variables, the adjustment of model parameters, etc. , mean square error (MSE): Evaluate the model optimization effect, ensure the accuracy and reliability of the model, and conduct in-depth analysis of the degree and direction of the impact of the screened variables on carbon emissions.
[0133] The results show that vehicle type and traffic flow are the main factors affecting carbon emissions, among which heavy vehicles and peak traffic flow contribute most significantly to carbon emissions. At the same time, fuel type and weather conditions also have a significant impact on carbon emissions, especially in extreme weather conditions, carbon emissions have obvious fluctuations. These findings provide specific goals and directions for carbon emission control.
[0134] Finally, through the regression analysis results, the key variables that need to be addressed are identified. The study found that optimizing traffic flow management, promoting clean energy vehicles, and improving fuel quality are effective emission reduction strategies. In addition, it is preliminarily found that carbon emissions can be further reduced by adjusting peak traffic flow and improving traffic management under extreme weather conditions. These conclusions provide a scientific basis for the construction and optimization of subsequent prediction models.
[0135] like Figure 2As shown in Figure 3, in S3, the causal analysis technology is used to explore the causal chain and feedback mechanism between the factors affecting traffic flow carbon emissions, divide the system level, analyze the relationship between the internal variables of the subsystem, and draw the system flow chart and causal relationship diagram.
[0136] Based on the identification of core factors, the present invention adopts Granger causality test, where is the explained variable, is the regression coefficient, The basic idea is that if the lag term of X still helps to explain the change of the current value of Y while controlling the lag term (past value) of Y, then X is considered to have a causal effect on Y. Deeply explore the causal chain and feedback mechanism between the influencing factors of traffic flow carbon emissions. By constructing a structural equation model, the complex interactive relationship between the factors is revealed, and the status and role of different variables in the carbon emission system are clarified. This method helps to understand the internal mechanism of the carbon emission system and provides theoretical support for the accuracy of the model. Explore the causal chain and feedback mechanism between the influencing factors of traffic flow carbon emissions, and use the Vensim tool to construct a variable causal relationship loop diagram.
[0137] Secondly, the relationship between system-level and subsystem-internal variables was further analyzed using causal diagrams. By drawing causal diagrams, the study found that factors such as traffic flow, vehicle type, fuel use, and weather conditions interact with each other in the carbon emission system, forming complex causal chains and feedback loops. For example, an increase in traffic flow will lead to an increase in carbon emissions, and changes in carbon emissions will in turn affect traffic flow and vehicle type selection. Through this analysis, the study revealed the interdependence of the variables in the system.
[0138] Finally, through causal analysis, the system hierarchy and the relationship between variables within the subsystem were established. This structured analysis method enables us to understand the dynamic behavior of the system more clearly and provides a clear framework for the subsequent construction of the prediction model. The research results show that only by comprehensively considering the interaction of various variables can the changing trend of traffic flow carbon emissions be accurately predicted.
[0139] In S4, based on the system dynamics modeling technology, a prediction model for traffic flow carbon emissions is constructed, variable types based on causal relationships are set, and mathematical function relationships between variables are defined. On the basis of establishing model parameters, a system dynamics model is established.
[0140] Based on the previous research, this paper uses system dynamics modeling technology to construct a prediction model for carbon emissions from highway traffic flow in Guangdong Province. First, based on the results of causal relationship analysis, the variable types in the model are set, and the mathematical function relationship between the variables is defined. Based on the system causal relationship, the mathematical function relationship between the variables in the traffic flow carbon emission system is mined, including data analysis and model fitting. Through the analysis of historical data and model fitting, regression analysis and time series analysis methods are used to quantify the mathematical relationship between variables. , Time Series Analysis (ARIMA Model) , Representative time The last observed value, is a constant term, is the autoregressive term coefficient, is the moving average coefficient, As the error term, a complete mathematical function model is established.
[0141] Based on the system dynamics model, the Vensim tool is used to visualize these relationships and simulate the dynamic changes of the carbon emission system, providing a scientific basis for carbon emission prediction. In this model, state variables (Stock Variables): variables that describe the state or characteristics of the system, such as Indicates time Traffic flow at a given moment, Flow Variables: Variables that describe the rate of change of state variables, such as Indicates time The carbon emission rate at the time, constructing the state equation , the governing equation ,in represents the state variable, represents the control variable, Refers to the mathematical functional relationship between variables.
[0142] Secondly, on the basis of establishing the model parameters, a complete system dynamics model was established. The model parameters were determined by historical data and relevant research results to ensure the accuracy and reliability of the model. The model includes multiple subsystems, such as traffic flow subsystem, vehicle type subsystem, fuel use subsystem and weather condition subsystem. Through the interaction of these subsystems, the model can dynamically simulate the changing process of traffic flow carbon emissions.
[0143] Finally, through model simulation and simulation, the study predicted the carbon emission trends under different scenarios. The results show that without any intervention measures, carbon emissions from highways in Guangdong Province will continue to grow. However, carbon emissions can be significantly reduced by promoting clean energy vehicles, optimizing traffic flow management, and improving fuel quality. These simulation results provide a scientific basis for policymakers to help them choose the best emission reduction strategy and promote the goal of low-carbon sustainable development.
[0144] In S5, model verification technology is applied to comprehensively verify the traffic flow carbon emission model, including structural verification, historical backtracking verification and sensitivity verification, so as to evaluate the stability, effectiveness and sensitivity of the model.
[0145] The structural test uses parameter estimation techniques using the method of least squares (OLS): Or the Maximum Likelihood Estimation (MLE): Evaluate the suitability of the model structure.
[0146] By applying the model to historical data for historical backtesting, the model's prediction results are compared with the actual observed values by backtracking the historical data. The research results show that the model can reproduce the trend of historical carbon emissions well, indicating that the model has a high degree of fit on historical data. At the same time, historical backtesting also helps us discover possible deviations in the model and make corresponding adjustments and optimizations to the model.
[0147] The sensitivity test is carried out by evaluating the sensitivity of the model to the changes of different variables. , are model parameters, is the parameter variation, The response variable is the change in the amount of change. Adjust the key parameters in the model, observe the changes in the model output, and determine the sensitivity of the model to different variables. The sensitivity test results show that traffic flow and vehicle type are the most sensitive variables to carbon emissions, which provides policymakers with specific regulatory directions.
[0148] Combined with Figure 3 ,In S6, the simulation control simulation technology is used to ,analyze the carbon emission trends of traffic flows under different ,highway carbon reduction policies. By comparing the simulation results under ,different policy schemes, effective carbon reduction strategies are ,screened out, providing a quantitative basis for policy ,decision-making.
[0149] Based on the construction and verification of the prediction model, the study used simulation control simulation technology to analyze the carbon emission trends of traffic flow under different highway carbon reduction policies. First, by setting different policy scenarios, such as restricting the passage of high-emission vehicles, promoting clean energy vehicles, and optimizing traffic flow management, the impact of these policy measures on carbon emissions was simulated. The simulation results show that there are significant differences in the reduction effects of different policy combinations on carbon emissions.
[0150] Secondly, by comparing the simulation results under different policy options, the most effective carbon reduction strategy can be screened out. For example, the study found that the combination of restricting the passage of high-emission vehicles and promoting clean energy vehicles can significantly reduce carbon emissions. At the same time, optimizing traffic flow management, such as improving traffic signal control and improving road traffic efficiency, also has a significant effect on carbon emissions. These findings provide policymakers with specific quantitative evidence to help them choose the best emission reduction strategy.
[0151] Finally, the emission reduction effects and implementation difficulties of different policy measures are summarized, and a variety of feasible emission reduction plans are proposed. By comprehensively considering the emission reduction effect, implementation cost and social acceptance, it is recommended to adopt comprehensive emission reduction measures, combining technical means and policies and regulations to jointly promote the reduction of carbon emissions. These suggestions provide a scientific basis and policy reference for achieving the goal of low-carbon sustainable development in Guangdong Province, and help decision makers formulate more effective and feasible emission reduction strategies.
[0152] It should be understood that, although each step in the flow chart of each embodiment of the present invention is shown in sequence according to the indication of the arrow, these steps are not necessarily performed in sequence according to the order indicated by the arrow. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be performed in other orders. Moreover, at least a portion of the steps in each embodiment may include a plurality of sub-steps or a plurality of stages, and these sub-steps or stages are not necessarily performed at the same time, but can be performed at different times, and the execution order of these sub-steps or stages is not necessarily performed in sequence, but can be performed in turn or alternately with at least a portion of other steps or sub-steps or stages of other steps.
[0153] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.
[0154] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.
Claims
1. A method for predicting the medium- and long-term trend of carbon emissions from highway traffic flow, characterized in that: The method comprises: S1. Obtain historical data on carbon emissions from highways in the study area, and use statistical analysis techniques and visualization tools to analyze the temporal and spatial variation patterns and characteristics to derive the spatial and temporal boundaries of the model; S2. Based on regression analysis technology, identify the factors affecting traffic flow carbon emissions and obtain research variables and factors affecting carbon emissions; S3. Analyze the causal chain and feedback mechanism between the factors affecting traffic flow carbon emissions, divide the system level, and analyze the relationship between variables within the subsystem; S4. Based on the system dynamics modeling technology, a prediction model for traffic flow carbon emissions is constructed, variable types based on causal relationships are set, mathematical function relationships between variables are defined, and a system dynamics model is established based on the established model parameters; S5. Verify the traffic flow carbon emission model, including structural test, historical backtest and sensitivity test; S6. Use simulation comparison technology to analyze the carbon emission trend of traffic flow under different highway carbon reduction policies, and obtain the final carbon reduction strategy by comparing the simulation results under different policy options; Among them, S1 specifically includes: Collect and clean historical highway traffic flow and carbon emissions data; Use the Z-score method to identify and remove outliers and remove outliers: ; in, represents the standard deviation distance of the data point from the mean, represents a data point, is the mean value of the data, is the standard deviation of the data, if , then the determination value is an outlier; Interpolation and mean completion methods are used to handle missing data values; Specific: Interpolation method: ; in, Indicates at a point in time The observed value of Indicates at a point in time The observed value of Indicates at a point in time The observed value of Here, it represents the time index, indicating different time points in the sequence; Mean Completion Method: ; in, represents the estimated value of the missing value, represents the total number of observed samples, Indicates Observations, represents different observation samples; Analyze and visualize data based on statistical analysis tools and visualization tools to reveal the distribution characteristics, change trends and correlations of data; Use time series analysis models, spatial autocorrelation Moran's I model in geographic information system analysis, and visualization tools to obtain the temporal and spatial development laws and trends of data; Specific: Time Series Analysis Model: ; in Representative time The last observed value, Refers to seasonal ingredients, Other random components; Moran's I Model ; in represents the Moran's I index, which is used to measure spatial autocorrelation. represents the total number of observed samples, Represents the variable value, is the variable mean, Represents the spatial weight matrix , and They represent different observation samples, and is corresponding to and Observed value of The main periodic components in carbon emission data were identified based on fast Fourier transform and inverse transform; Among them, the carbon emissions data are: ,in, ; Here, Indicated in Carbon emission data of sample points, represents the total number of observed samples, The last index indicating the number of samples; The mathematical expression of fast Fourier transform is: ; in Indicates the frequency The complex coefficients of represents an imaginary unit; Indicated in Carbon emission data of sample points, Represents the index of the frequency in the frequency domain after FFT transformation, The value range is 0 to ; Represents the sample point index in the time domain, ranging from 0 to , which is related to time and represents the Points, represents the total number of observed samples; Calculated Get the spectrum, expressed as amplitude spectrum and phase spectrum, specifically: Amplitude spectrum: ; Phase spectrum: ; in, represents the frequency index, Indicates the frequency The phase at and Representing the real part and the imaginary part respectively, by analyzing the frequency spectrum, finding the frequency corresponding to the peak in the amplitude spectrum, and identifying the main periodic components in the carbon emission data; Convert the frequency domain signal back to the time domain using an inverse fast Fourier transform: ; in, Indicates at time The reconstructed data at Represents the sample point index in the time domain, ranging from 0 to , which is related to time and represents the number of points; Represents the total number of observed samples and the size of the entire data set; Indicates the index of the frequency in the frequency domain after FFT transformation, indicates the index of the frequency in the frequency domain after FFT transformation, The value range is 0 to ; represents an imaginary unit; Analyze periodic patterns in carbon emissions data; S2 specifically includes: Based on regression analysis, the variables that affect carbon emissions are determined, and a multivariate linear regression analysis model is established to conduct regression analysis: ; in Refers to the response variable of the model, specifically carbon emissions, Refers to the model independent variables, specifically traffic flow, vehicle type, and fuel use, The intercept term, is the regression coefficient, Error term; Variable screening is performed based on the stepwise regression method, specifically: Perform significance test on variables: ; in express The standard error of The corresponding model independent variables, It represents a statistic, which is a numerical value and is used to determine whether the independent variable has a significant impact on the response variable; Test the significance of variables by standardizing regression coefficients: ; in is the independent variable The standard deviation of Response variable The standard deviation of represents the standardized regression coefficient, The corresponding model independent variables, represents the unstandardized regression coefficient; Based on the above results, the key variables affecting carbon emissions are determined and the determination coefficient is used As a reference, measure the fitting effect, mean square error Evaluate the model optimization effect and optimize the prediction model structure; in, Represents the sample point index in the time domain, The corresponding model independent variables; The S3 specifically includes: Collect and organize traffic flow carbon emission data, and apply Granger causality test method to determine the causal relationship and feedback mechanism between various factors. Specifically: ; in is the explained variable, is the regression coefficient, Error term; is the explained variable, indicating the of carbon emissions, Indicates at time of carbon emissions, Indicates at time The independent variable reflects the impact of past factors on current carbon emissions; is the regression coefficient, is the error term, represents the number of lags, which refers to the number of past data points considered in the regression model; Divide the system level and decompose the traffic flow carbon emission system into different subsystems, including the socio-economic subsystem, the highway traffic flow subsystem, and the energy consumption and emission subsystem. Within each subsystem, explore the relationship between variables, analyze the interaction and impact, and the degree of contribution to the overall system; Use the data and analysis results to draw system flow charts and cause-and-effect diagrams; The S4 specifically includes: Based on the causal relationship, the types of variables in the traffic flow carbon emission system are set, including identifying state variables and rate variables; Analyze the mathematical function relationship between various variables in the traffic flow carbon emission system, including data analysis and model fitting. Through the analysis of historical data and model fitting, quantify the causal relationship between variables and establish a mathematical function model. Specifically: Quantify the mathematical relationship between variables based on regression analysis and time series analysis methods: Regression Analysis: ; Time Series Analysis: ; in, Representative time The last observed value, is a constant term, is the autoregressive term coefficient, is the moving average coefficient, is the error term; The system dynamics model simulates the dynamic evolution of the system and captures the relationship between internal variables. Specifically: Determine the model state equations and control equations; Model state equation: ; in, Represents the time index, which is the same as the previous Consistency; Indicates at time The state variable represents the current state of the system. Indicates at time Add time increment , used to describe the state of the system at the next time step, Indicates at time The state variables are The difference is that the former reflects the future state, represents the time step; Governing equations: ; in represents the state variable, represents the control variable, Refers to the mathematical functional relationship between variables; The S5 specifically includes: Verify the traffic flow carbon emission model, including structural test, historical backtest and sensitivity test; Check whether the model is reasonable based on its structure, including the logical consistency between variables and the validity of mathematical expression; Perform model backtesting based on historical data to evaluate the model's fit and prediction accuracy in different time periods; The sensitivity test of the model is performed by simulating parameter changes to evaluate the response of the model to parameter changes. The response of the model to parameter changes is evaluated. The parameter change rate is: ; in are model parameters, is the parameter change, is the change in the response variable; The S6 specifically includes: Based on the established system dynamics model, the carbon emission trends of highways under different scenarios are simulated to evaluate the impact of policy implementation and industry development; Based on the comparison of simulation results of different policy options, their carbon reduction effects and cost-effectiveness are evaluated.
