Highway traffic pollutant emission amount multi-scenario interval prediction method

Abstract

The application discloses a kind of expressway traffic pollutant emission quantity multi-scenario interval prediction method, comprising the following steps: obtaining the traffic distribution observation data set of expressway based on entrance and exit ramp toll collection system in prediction space-time range;Generation policy control scenario mode combination based on different emission standard updating scheme;Calculate baseline pollutant emission factor, obtain the emission factor data set under each scenario;The seasonal SARIMA-SVR model is used to predict the vehicle population under different scenarios in the future, and the pollutant emission in each scenario in the future month is estimated;Evaluate the emission reduction effect of each combination of emission standard control policy.The application realizes the quantitative evaluation of multi-scenario expressway traffic pollutant emission reduction effect under the consideration of different emission standard updating control policy, and improves the seasonal prediction accuracy by using the SARIMA-SVR model, which provides technical support for reasonably formulating carbon peak policy.

Description

Technical Field

[0001] This invention belongs to the field of carbon emission prediction technology, and more specifically, relates to a multi-scenario interval prediction method for pollutant emissions from highway transportation. Background Technology

[0002] Data shows that the transportation sector accounts for 15% of my country's final carbon emissions, and due to insufficient fuel efficiency improvements to meet growing demand, carbon emissions from the transportation sector continue to rise. Specifically, road transport accounts for 74% of total carbon emissions from the transportation sector, and heavy-duty trucks account for 54% of total road transport carbon emissions. In recent years, with the surge in motor vehicle ownership and the rapid development of the express delivery industry, highway transport volume has increased exponentially, making highway carbon emissions a significant source of carbon emissions from road transport.

[0003] Therefore, it is of great significance to study the prediction of carbon emissions from highway transportation in order to reduce the negative impact of carbon emissions.

[0004] According to existing literature, most studies are based on the macro-influencing factors obtained by decomposing the IPAT model, Kaya identity, STIRPAT and its extended models. They use linear regression methods such as least squares regression, ridge regression, and quantile regression, as well as machine learning methods such as support vector machines and neural networks to construct the regression relationship between predicting influencing factors and carbon emissions, thereby predicting regional traffic carbon emissions.

[0005] In addition, some studies estimate future vehicle carbon emissions by using bottom-up carbon emission calculation methods (such as the LEAP model, COPERT, MOVES, GEI, and other emission factor models) based on predicted vehicle stock growth, vehicle technology development, vehicle mileage, fuel efficiency, and other important vehicle operating parameters. A method and device for predicting carbon emissions based on transportation (application number 202210631985.2) has been published. Based on a transportation volume dataset of different transportation carbon emission chains, modes of transportation, and activity characteristics within a specified geographical area, it calculates the total carbon emissions from transportation through a pre-set transportation carbon emission factor library, performs stabilization corrections and optimizations, and establishes an optimized comprehensive autoregressive moving average model to predict the change curve of carbon emissions. A method and system for predicting carbon dioxide emissions from road traffic have also been disclosed (application number 202110923903.7). This method predicts weighted road network spatial changes based on urban construction land and population spatial datasets, predicts the number of fuel vehicles based on vehicle survival curves and the sales penetration rate of new energy vehicles, allocates the number of fuel vehicles to spatial grids according to weighted road network density, and predicts the carbon emissions from the road traffic spatial grids in future years by establishing a functional relationship between the number of fuel vehicles and carbon dioxide emissions.

[0006] Based on domestic and international research and applications, predictions of total transportation carbon emissions relying on macroeconomic indicators such as population, GDP, energy intensity, total consumption, and total social output suffer from insufficient data accuracy and low industry segmentation. While bottom-up calculation models have attempted to improve the accuracy of carbon emission factors and influencing factors predictions, they have not fully considered the impact of different emission standards on highway fleet composition and pollutant emission factors. They also fail to adequately assess the emission reduction effects of combined scenarios of various emission standard control policies, thus affecting the accuracy of the prediction results. Summary of the Invention

[0007] To address at least one of the aforementioned technical problems, according to one aspect of the present invention, a multi-scenario interval prediction method for pollutant emissions from highway transportation is provided, comprising the following steps:

[0008] S1, determine the spatiotemporal range for predicting pollutant emissions from highways and obtain traffic distribution observation datasets for each highway;

[0009] Specifically, the spatiotemporal scope includes the geographical area where the highway is located and the range of the representative year and month and the target year and month; for the highways in the test area representing the year and month, traffic distribution observation datasets for each highway are obtained, including the length of the observation interval, the number of vehicles passing through the observation interval, and the corresponding attribute data of the vehicles, such as vehicle type, size, purpose, and emission standards;

[0010] Based on the vehicle ownership data and new vehicle registration data for representative months within the geographical spatial boundary, the survival rate of various vehicle types and their total proportion are calculated. Combined with the updated emission standard policies, the emission standard attributes of vehicles passing through the dataset are obtained. The survival rate of various vehicle types is calculated as follows:

[0011]

[0012] Among them, SVP n,i,t RP represents the number of unscrapped vehicles of type i registered in month n, at the estimated month t. n,m SR represents the number of vehicles of type i registered in month n. n,i,t Let T be the vehicle survival rate of type i vehicles registered in month n when estimating month t; n,i and k n,i These are characteristic parameters.

[0013] S2, Generate a combination of policy control scenario models based on emission standard updates; wherein, the policy control scenario includes a combination of pollutant emission factor change rates, emission standard update intervals, and the early scrapping year for vehicles with older standards;

[0014] The rate of change of pollutant emission factors is set to two scenarios: uniform change A1 and fluctuating change A2; the emission standard update interval is set to three scenarios: high frequency B1, general B2, and low frequency B3; the advance year of old standard vehicles is set to three scenarios: natural elimination C1, last-place elimination C2, and second-last-place elimination C3.

[0015] The three scenarios mentioned above are randomly combined to generate 18 policy regulation scenario combination Pr (r = 1, 2, 3, ..., 18), which correspond to combinations AaBbCc (a = 1, 2; b, c = 1, 2, 3).

[0016] S3, calculate the baseline pollutant emission factor and obtain the emission factor change dataset under different policy control scenarios;

[0017] The baseline pollutant emission factor is denoted as BEF. i,j,s,p Where i represents the type of vehicle from the pollution source, including small passenger cars, medium-sized passenger cars, large passenger cars, small trucks, medium-sized trucks, large trucks, and container trucks; j represents the fuel type of the vehicle from the pollution source, including gasoline, diesel, and hybrid electric new energy vehicles; s represents the emission standard of the vehicle from the pollution source, including the existing China I pre-emission, China I, China II, China III, China IV, China V, and China VI emission standards, and the proposed China VII, China VIII, China IX, China X, and China XI emission standards; p represents the type of pollutant, including NOx, VOCs, PM2.5, PM10, CO, and CO2.

[0018] Based on the Basic Emission Factors (BEF) measured under standard conditions using COPERT simulations, and combined with local conditions and highway road conditions, correction parameters reflecting driving conditions and environmental factors are used for localization. Simultaneously, based on the rate of change of emission factors under different future policy control scenarios, the emission factors corresponding to each emission standard in the proposed future scenarios are obtained. The calculation formula is as follows:

[0019]

[0020] Among them, EF i,j,s,p,r Under policy regulation scenario Pr, the emission factor (g*km⁻¹) of category p pollutants emitted by category i vehicles with energy source j and emission standard s; BEF i,j,s,p The corresponding basic emission factor is given by (g*km⁻¹). For the corresponding environmental correction factors, including temperature and humidity; ω i,j,s,p This is the corresponding velocity correction factor; δ i,j,s,p Other corresponding correction factors include load factor, lubricant parameters, oil quality, etc.; θ i,j,s,p,aUnder the policy regulation scenario Pr, the rate of change of emission factors corresponds to the combination of rate of change of pollutant emission factors Aa.

[0021] S4, determine the age distribution of various types of vehicles representing the month, and predict the number of vehicles under different future scenarios using the SARIMA-SVR model;

[0022] The specific process for determining the age distribution of various types of vehicles representing a month is as follows:

[0023] Based on the vehicle ownership data and new vehicle registration data for representative months, and according to the survival rate of each vehicle type, the age distribution of each vehicle type for representative months is obtained. The calculation formula is as follows:

[0024]

[0025] Among them, SVP n,i,t RP represents the number of unscrapped vehicles of type i registered in month n, at the estimated month t. n,m SR represents the number of vehicles of type i registered in month n. n,i,t Let T be the vehicle survival rate of type i vehicles registered in month n when estimating month t; n,i and k n,i For characteristic parameters;

[0026] The specific process for obtaining the vehicle ownership under different future scenarios using the SARIMA-SVR model is as follows:

[0027] Based on the monthly traffic volume data of the highway observation interval representing the month to be measured, the monthly number of vehicles with different vehicle types, fuel types, and emission standards is obtained according to the vehicle age distribution and the year of implementation of vehicle emission standards.

[0028] The monthly number of vehicles of different vehicle types and fuel types in the observation interval of the highway in the test area is subjected to ADF test and white noise test to determine whether the time series is stationary. If it is, the subsequent steps are carried out; otherwise, the time series is stationary by performing d-order differencing.

[0029] Based on the Akaike Information Criterion (AIC), the optimal parameters of the SARIMA(p,d,q)×(P,D,Q)s model are determined using grid search, where p, d, and q are the number of non-seasonal autoregressive terms, the number of moving average lag terms, and the difference order, respectively; P, D, and Q are the number of seasonal autoregressive terms, the number of moving average lag terms, and the difference order, respectively; and s is the seasonal period of the time series.

[0030] The SARIMA model is trained using stationary time series training samples, and the prediction function of the SARIMA model is used for prediction to obtain the fitted time series sequence. and the corresponding residual sequence R i,j,t ; By sliding the window to R i,j,t The sequence is reconstructed into a sequence of order u, and the residual sequence is trained and predicted using an SVR model to obtain the residual fitted sequence. The SARIMA-SVR model predicts the monthly number of vehicles traveling under different vehicle types and fuel types. for and sum.

[0031] Based on the exponential smoothing method, the number of new vehicle registrations in the target month is predicted. Combined with the emission standard update frequency Bb and the early scrapping scheme Cc for vehicles with older emission standards under the policy control scenario Pr, the vehicle age and emission standard distribution for each target month are obtained. The monthly number of vehicles is then calculated. Based on emission standards, the monthly number of vehicles passing through the target month under scenario Pr, categorized by vehicle type, fuel type, and emission standard, is obtained.

[0032] Step 5: Estimate the pollutant emissions for each scenario in the coming months and assess the emission reduction effects of the emission standard adjustment policies adopted for each scenario.

[0033] Based on the monthly average vehicle mileage of the highway observation interval in the target area for the representative month, the monthly average vehicle mileage for the target month is predicted using the quadratic exponential smoothing method.

[0034] The combined Monthly pollutant emissions from highway transportation under various scenarios for the target month. and emission factor EF i,j,s,p,r The formula for calculating the estimated emissions of Class P pollutants from vehicles in month t under scenario Pr is as follows:

[0035]

[0036] By analyzing the predicted emission trends of pollutant type p under various scenarios in the target month, the emission reduction effect of the adopted emission standard adjustment policy is evaluated.

[0037] According to another aspect of the present invention, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps in the multi-scenario interval prediction method for highway transportation pollutant emissions of the present invention.

[0038] According to another aspect of the present invention, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of the multi-scenario interval prediction method for highway transportation pollutant emissions of the present invention.

[0039] Compared with the prior art, the present invention has at least the following beneficial effects:

[0040] This invention, based on historical highway traffic data, improves the prediction accuracy and generalization ability of seasonal models through the SARIMA-SVR model. At the same time, it takes into account the impact of policy combinations on future pollutant emissions from different emission standard update intensities, update intervals, and early scrapping policies. This provides technical support for understanding future pollutant emission trends on highways and for rationally formulating carbon peaking policies. Attached Figure Description

[0041] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings of the embodiments will be briefly described below. Obviously, the drawings described below only relate to some embodiments of the present invention and are not intended to limit the present invention.

[0042] Figure 1 This is a flowchart illustrating the prediction process for carbon emissions from highway transportation according to the present invention. Detailed Implementation

[0043] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention.

[0044] Unless otherwise defined, the technical or scientific terms used herein shall have the ordinary meaning as understood by one of ordinary skill in the art to which this invention pertains.

[0045] Example 1:

[0046] The present invention provides a multi-scenario interval prediction method for pollutant emissions from highway transportation, such as... Figure 1 As shown, the specific steps of the multi-scenario interval prediction method for pollutant emissions from highway transportation include 1-5:

[0047] S1, determine the spatiotemporal range for predicting pollutant emissions from highways and obtain traffic distribution observation datasets for each highway;

[0048] Specifically, cities and city clusters in the sense of administrative divisions are used as geographical boundaries, and continuous year and month intervals are used as representative year-month intervals and target year-month intervals, respectively.

[0049] Then, for the highways in the test area representing the year and month, traffic distribution observation datasets are obtained for each highway, including the length of the observation interval, the number of vehicles passing through the observation interval, and the corresponding attribute data of the vehicles, such as vehicle type, size, purpose, and emission standards.

[0050] Based on the vehicle ownership data and new vehicle registration data for representative months within the geographical spatial boundary, the survival rate of various vehicle types is calculated. Combined with the updated emission standard policies, the emission standard attributes of vehicles passing through the dataset are obtained. The survival rate of various vehicle types is calculated as follows:

[0051]

[0052] Among them, SVP n,i,t RP represents the number of unscrapped vehicles of type i registered in month n, at the estimated month t. n,m SR represents the number of vehicles of type i registered in month n. n,i,t Let T be the vehicle survival rate of type i vehicles registered in month n when estimating month t; n,i and k n,i These are characteristic parameters.

[0053] In one embodiment, a certain urban agglomeration is used as the observation space, with the representative month-year interval from January 2004 to December 2020, and the target month-year interval from January 2021 to December 2030. Based on the number of large trucks in operation and the number of new vehicle registrations in the representative years, combined with the emission standard update policies of the predicted urban agglomeration, the emission standard distribution of the observed vehicles is obtained. Taking large trucks passing through the observation area as an example, the distribution of the proportion of each emission standard for large trucks is shown in Table 1.

[0054] Table 1 Distribution of Emission Standards for Large Trucks

[0055] years Country 0 Country 1 National 2 National 3 National 4 National 5 National 6 2004 0.2797 0.7203 0.0000 0.0000 0.0000 0.0000 0.0000 2005 0.2442 0.6293 0.1265 0.0000 0.0000 0.0000 0.0000 2006 0.2262 0.5850 0.1888 0.0000 0.0000 0.0000 0.0000 2007 0.2007 0.5230 0.2763 0.0000 0.0000 0.0000 0.0000 2008 0.1748 0.4625 0.3627 0.0000 0.0000 0.0000 0.0000 2009 0.1552 0.4220 0.3345 0.0883 0.0000 0.0000 0.0000 2010 0.1049 0.2985 0.2415 0.3552 0.0000 0.0000 0.0000 2011 0.0632 0.1933 0.1623 0.5812 0.0000 0.0000 0.0000 2012 0.0422 0.1435 0.1282 0.6861 0.0000 0.0000 0.0000 2013 0.0284 0.1130 0.1113 0.7473 0.0000 0.0000 0.0000 2014 0.0162 0.0800 0.0909 0.6293 0.1836 0.0000 0.0000 2015 0.0081 0.0542 0.0754 0.5498 0.3126 0.0000 0.0000 2016 0.0033 0.0331 0.0608 0.4819 0.4209 0.0000 0.0000 2017 0.0012 0.0202 0.0538 0.4830 0.4419 0.0000 0.0000 2018 0.0002 0.0087 0.0371 0.3990 0.3719 0.1830 0.0000 2019 0.0000 0.0031 0.0240 0.3288 0.3169 0.3272 0.0000 2020 0.0000 0.0008 0.0138 0.2616 0.2667 0.3669 0.0902

[0056] S2, Generate a combination of policy control scenarios based on emission standard updates. These policy control scenarios include combinations of pollutant emission factor change rates, emission standard update intervals, and the early scrapping year for vehicles meeting older emission standards.

[0057] The rate of change of pollutant emission factors is set to two scenarios: uniform change A1 and fluctuating change A2; the emission standard update interval is set to three scenarios: high frequency B1, general B2, and low frequency B3; the advance year of old standard vehicles is set to three scenarios: natural elimination C1, last-place elimination C2, and second-last-place elimination C3.

[0058] The above three scenarios are randomly combined to generate 18 policy regulation scenario model combinations Pr (r=1,2,3,...,18), which correspond to combinations AaBbCc (a=1,2;b,c=1,2,3);

[0059] In this embodiment, the 18 policy regulation scenario combinations are shown in Table 2.

[0060] Scenario Mode Corresponding combinations Rate of change combination A Standard update interval B Early scrapping year C <![CDATA[P1]]> <![CDATA[A1B1C1]]> <![CDATA[Uniform change A1]]> <![CDATA[2(B1)]]> <![CDATA[0(C1)]]> <![CDATA[P2]]> <![CDATA[A1B1C2]]> <![CDATA[Uniform change A1]]> <![CDATA[2(B1)]]> <![CDATA[1(C2)]]> <![CDATA[P3]]> <![CDATA[A1B1C3]]> <![CDATA[Uniform change A1]]> <![CDATA[2(B1)]]> <![CDATA[2(C3)]]> <![CDATA[P4]]> <![CDATA[A2B1C1]]> <![CDATA[Fluctuating change A2]]> <![CDATA[2(B1)]]> <![CDATA[0(C1)]]> <![CDATA[P5]]> <![CDATA[A2B1C2]]> <![CDATA[Fluctuating change A2]]> <![CDATA[2(B1)]]> <![CDATA[1(C2)]]> <![CDATA[P6]]> <![CDATA[A2B1C3]]> <![CDATA[Fluctuating change A2]]> <![CDATA[2(B1)]]> <![CDATA[2(C3)]]> <![CDATA[P7]]> <![CDATA[A1B2C1]]> <![CDATA[Uniform change A1]]> <![CDATA[3(B2)]]> <![CDATA[0(C1)]]> <![CDATA[P8]]> <![CDATA[A1B2C2]]> <![CDATA[Uniform change A1]]> <![CDATA[3(B2)]]> <![CDATA[1(C2)]]> <![CDATA[P9]]> <![CDATA[A1B2C3]]> <![CDATA[Uniform change A1]]> <![CDATA[3(B2)]]> <![CDATA[2(C3)]]> <![CDATA[P 10 ]]> <![CDATA[A2B2C1]]> <![CDATA[Fluctuating change A2]]> <![CDATA[3(B2)]]> <![CDATA[0(C1)]]> <![CDATA[P 11 ]]> <![CDATA[A2B2C2]]> <![CDATA[Fluctuating change A2]]> <![CDATA[3(B2)]]> <![CDATA[1(C2)]]> <![CDATA[P 12 ]]> <![CDATA[A2B2C3]]> <![CDATA[Fluctuating change A2]]> <![CDATA[3(B2)]]> <![CDATA[2(C3)]]> <![CDATA[P 13 ]]> <![CDATA[A1B3C1]]> <![CDATA[Uniform change A1]]> <![CDATA[4(B3)]]> <![CDATA[0(C1)]]> <![CDATA[P 14 ]]> <![CDATA[A1B3C2]]> <![CDATA[Uniform change A1]]> <![CDATA[4(B3)]]> <![CDATA[1(C2)]]> <![CDATA[P 15 ]]> <![CDATA[A1B3C3]]> <![CDATA[Uniform change A1]]> <![CDATA[4(B3)]]> <![CDATA[2(C3)]]> <![CDATA[P 16 ]]> <![CDATA[A2B3C1]]> <![CDATA[Fluctuating change A2]]> <![CDATA[4(B3)]]> <![CDATA[0(C1)]]> <![CDATA[P 17 ]]> <![CDATA[A2B3C2]]> <![CDATA[Fluctuating change A2]]> <![CDATA[4(B3)]]> <![CDATA[1(C2)]]> <![CDATA[P 18 ]]> <![CDATA[A2B3C3]]> <![CDATA[Fluctuating change A2]]> <![CDATA[4(B3)]]> <![CDATA[2(C3)]]>

[0061] S3, calculate the baseline pollutant emission factor and obtain the emission factor change dataset under different policy control scenarios;

[0062] The baseline pollutant emission factor is denoted as BEF. i,j,s,p Where i represents the type of vehicle from the pollution source, including small passenger cars, medium-sized passenger cars, large passenger cars, small trucks, medium-sized trucks, large trucks, and container trucks; j represents the fuel type of the vehicle from the pollution source, including gasoline, diesel, and hybrid electric new energy vehicles; s represents the emission standard of the vehicle from the pollution source, including the existing China I pre-emission, China I, China II, China III, China IV, China V, and China VI emission standards, and the proposed China VII, China VIII, China IX, China X, and China XI emission standards; p represents the type of pollutant, including NOx, VOCs, PM2.5, PM10, CO, and CO2.

[0063] Taking into account local conditions and highway road conditions, correction parameters reflecting driving conditions and environmental factors are used for localization. Simultaneously, based on the rate of change of emission factors under different future policy control scenarios, the emission factors corresponding to each emission standard in the proposed future scenarios are obtained. The calculation formula is as follows:

[0064]

[0065] Among them, EF i,j,s,p,r Under policy regulation scenario Pr, the emission factor (g*km⁻¹) of category p pollutants emitted by category i vehicles with energy source j and emission standard s; BEF i,j,s,p The corresponding basic emission factor is given by (g*km⁻¹). For the corresponding environmental correction factors, including temperature and humidity; ω i,j,s,p This is the corresponding velocity correction factor; δ i,j,s,p Other corresponding correction factors include load factor, lubricant parameters, oil quality, etc.; θ i,j,s,p,a Under the policy regulation scenario Pr, the change rate of the pollutant emission factors corresponds to the change rate of the emission factors in combination Aa.

[0066] In this embodiment, the COPERT carbon emission factor library was selected to simulate the basic emission factors (BEF) measured under standard conditions. The emissions of various pollutants generated by a large truck using diesel fuel traveling 1000km were used as the baseline emission factors. The baseline emission factors corresponding to each emission standard are shown in Table 3.

[0067] Table 3. Benchmark NOx Emission Factors for Large Trucks

[0068] Emission standards In front of the country Kunichi Second grade National III National IV National V emission standard National VI <![CDATA[NO X [t]]]> 0.01240 0.00872 0.00924 0.00742 0.00525 0.00236 0.00016

[0069] Based on the average speed and load of large trucks on highways, localized simulation environmental correction factors, speed correction factors and other correction coefficients are used. Based on the Class A scenario generated in step 2, the factor change rate of future emission standards is preset, as shown in Table 4.

[0070] Table 4 NOx emission factors of large trucks under different emission factor change rate scenarios

[0071] Emission standards National Standard 7 National Eight Guojiu National Ten National Day <![CDATA[Uniform change A1]]> 0.00020 0.00013 0.00008 0.00005 0.00003 <![CDATA[Fluctuating change A2]]> 0.00034 0.00027 0.00019 0.00009 0.00001

[0072] S4, using the SARIMA-SVR model to predict the number of vehicles in different future scenarios;

[0073] Specifically, the monthly number of vehicles of different vehicle types and fuel types in the observation interval of the highway in the test area is subjected to ADF test and white noise test to determine whether the time series is stationary. If it is, the subsequent steps are carried out; otherwise, the time series is stationary by performing d-order difference.

[0074] Based on the Akaike Information Criterion (AIC), the optimal parameters of the SARIMA(p,d,q)×(P,D,Q)s model are determined using grid search, where p, d, and q are the number of non-seasonal autoregressive terms, the number of moving average lag terms, and the difference order, respectively; P, D, and Q are the number of seasonal autoregressive terms, the number of moving average lag terms, and the difference order, respectively; and s is the seasonal period of the time series.

[0075] The SARIMA model is trained using stationary time series training samples, and the prediction function of the SARIMA model is used for prediction to obtain the fitted time series sequence. and the corresponding residual sequence R i,j,t ; By sliding the window to R i,j,t The sequence is reconstructed into a sequence of order u, and the residual sequence is trained and predicted using an SVR model to obtain the residual fitted sequence. The SARIMA-SVR model predicts the monthly number of vehicles traveling under different vehicle types and fuel types. for and sum.

[0076] Based on the exponential smoothing method, the number of new vehicle registrations in the target month is predicted. Combined with the emission standard update frequency Bb and the early scrapping scheme Cc for vehicles with older emission standards under the policy control scenario Pr, the vehicle age and emission standard distribution for each target month are obtained. The monthly number of vehicles is then calculated. Based on emission standards, the monthly number of vehicles passing through the target month under scenario Pr, categorized by vehicle type, fuel type, and emission standard, is obtained.

[0077] In this embodiment, data from January 2004 to December 2018 were used as the model training set, and data from January 2019 to December 2020 were used as the model test set. After first-order trend differencing and seasonal differencing, the ADF test and white noise test results showed that the test statistics were all less than the threshold, satisfying the condition p < 0.05, indicating that the original time series was a stationary series and not a white noise series, and the differencing order d was determined to be 1. Based on the drawn autocorrelation and partial correlation plots of the original series, the preset number of autoregressive terms p could be 1 or 2, and the preset number of moving average lag terms could be 1 or 2. In addition, the data clearly showed a 12-month seasonal cycle, and s was set to 12. According to the AIC criterion, grid search was used to determine that the optimal parameter combination for the SARIMA(p,d,q)×(P,D,Q)s model was SARIMA(2,1,2)×(1,1,2)12.

[0078] The SARIMA model was trained using the original time series data to predict the fitted sequence and corresponding residual sequence from January 2019 to December 2020. The residual sequence was segmented using a 24-month sliding window, and then trained and fitted using the SVR model to obtain the residual fitted sequence. The SARIMA predicted fitted sequence and the SVR residual fitted sequence were added to obtain the final SARIMA-SVR prediction value, yielding the monthly predicted number of vehicles of different vehicle types and fuel types from January 2021 to December 2030. Table 5 compares the mean percentage error (MAPE) and symmetric mean absolute percentage error (SMAPE) of the SARIMA single model and the SARIMA-SVR ensemble model. Table 5 shows that the SARIMA-SVR model significantly reduces the prediction error and effectively improves the prediction accuracy.

[0079] Table 5 Comparison of prediction accuracy between single SARIMA model and SAIRMA-SVR model

[0080] Model SARIMA SARIMA-SVR MAPE 7.06％ 1.27％ SMAPE 6.54％ 1.21％

[0081] Based on the exponential smoothing method, the number of new vehicle registrations in the target month is predicted. Combined with the emission standard update frequency Bb and the early scrapping scheme Cc for vehicles with old emission standards under the policy control scenario Pr, the vehicle age and emission standard distribution ratio for each target month are obtained. The monthly number of vehicles passing through from January 2021 to December 2030 predicted by the SARIMA-SVR model is allocated according to the emission standard distribution ratio to obtain the monthly number of vehicles passing through the target month under different vehicle types, fuel types, and emission standards under scenario Pr.

[0082] S5 estimates pollutant emissions for future months under each scenario and assesses the emission reduction effects of the emission standard adjustment policies adopted under each scenario.

[0083] Specifically, based on the monthly average vehicle mileage of the highway observation interval in the target area for the representative month, the monthly average vehicle mileage for the target month is predicted using the quadratic exponential smoothing method.

[0084] The combined Monthly pollutant emissions from highway transportation under various scenarios for the target month. and emission factor EF i,j,s,p,r The formula for calculating the estimated emissions of Class P pollutants from vehicles in month t under scenario Pr is as follows:

[0085]

[0086] By analyzing the predicted emission trends of pollutant type p under various scenarios in the target month, the emission reduction effect of the adopted emission standard adjustment policy is evaluated.

[0087] In this embodiment, PM2.5 emissions from large diesel trucks are used as an example. The predicted monthly number of vehicles passing through each scenario is calculated from January 2021 to December 2030. The prediction results show that, in terms of PM2.5 emission reduction, the frequency of emission standard updates (Scenario B) has the most significant effect on pollutant emissions, and this effect gradually strengthens over time. High-frequency emission standard updates, along with fluctuating updates involving technological innovation and stagnation, and policies such as scrapping older standard vehicles two years earlier (Scenarios A2B1C3), have the best emission reduction effects.

[0088] The beneficial effects of the multi-scenario interval prediction method for highway transportation pollutant emissions of the present invention are as follows: Based on historical highway traffic data, the present invention improves the prediction accuracy and generalization ability of seasonal models through SARIMA-SVR model prediction; at the same time, it takes into account the impact of different emission standard update efforts, update intervals and early scrapping policies on future pollutant emissions policy combinations, providing technical support for understanding the future pollutant emission trends of highways and rationally formulating carbon peaking policies.

[0089] Example 2:

[0090] The computer-readable storage medium of this embodiment stores a computer program that, when executed by a processor, implements the steps in the multi-scenario interval prediction method for highway transportation pollutant emissions of Embodiment 1.

[0091] The computer-readable storage medium in this embodiment can be an internal storage unit of the terminal, such as the terminal's hard disk or memory; the computer-readable storage medium in this embodiment can also be an external storage device of the terminal, such as a plug-in hard disk, smart memory card, secure digital card, flash memory card, etc. equipped on the terminal; furthermore, the computer-readable storage medium can include both the terminal's internal storage unit and external storage devices.

[0092] The computer-readable storage medium of this embodiment is used to store computer programs and other programs and data required by the terminal. The computer-readable storage medium can also be used to temporarily store data that has been output or will be output.

[0093] Example 3:

[0094] The computer device of this embodiment includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the steps in the multi-scenario interval prediction method for highway transportation pollutant emissions of Embodiment 1.

[0095] In this embodiment, the processor can be a central processing unit, or other general-purpose processors, digital signal processors, application-specific integrated circuits, off-the-shelf programmable gate arrays or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor, etc. The memory can include read-only memory and random access memory, and provides instructions and data to the processor. A portion of the memory can also include non-volatile random access memory. For example, the memory can also store device type information.

[0096] Those skilled in the art will understand that the content disclosed in the embodiments can be provided as a method, system, or computer program product. Therefore, this solution can take the form of a hardware embodiment, a software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this solution can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage and optical storage) containing computer-usable program code.

[0097] This solution is described with reference to flowchart illustrations and / or block diagrams of methods and computer program products according to embodiments of this solution. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing device to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing device, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0098] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0099] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0100] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0101] The examples described herein are merely preferred embodiments of the invention and are not intended to limit the concept and scope of the invention. Any modifications and improvements made by those skilled in the art to the technical solutions of the invention without departing from the design concept of the invention should fall within the protection scope of the invention.

Claims

1. A method for predicting pollutant emissions from highway transportation across multiple scenarios, characterized in that, Includes the following steps: S1. Determine the spatiotemporal range for predicting highway pollutant emissions and obtain traffic distribution observation datasets for highways based on the toll collection system of entrance and exit ramps. S2. Generate a combination of policy control scenario models based on updated emission standards; S3. Calculate the baseline pollutant emission factor and obtain the emission factor change dataset under different policy control scenarios; S4. Predict the number of vehicles in different future scenarios using the SARIMA-SVR model; S5. Estimate pollutant emissions for each scenario in the coming months and assess the emission reduction effects of the emission standard adjustment policies adopted for each scenario. Step S1 is as follows: The spatiotemporal range includes the geographical area where the highway is located and the range of the representative year and month and the target year and month; for the highway in the test area representing the year and month, obtain the traffic distribution observation dataset of the highway based on the entrance and exit ramp toll system, including the length of the observation interval, the number of vehicles passing through the observation interval, and the corresponding attribute data of the passing vehicles; Based on the vehicle ownership data and new vehicle registration data for representative months within the geographical spatial boundary, the survival rate of various vehicle types and their total proportion are calculated. Combined with the updated emission standard policies, the emission standard attributes of vehicles passing through the dataset are obtained. The survival rate of various vehicle types is calculated as follows: (1) in, This represents the number of unscrapped vehicles of type i registered in month n, at the estimated month t. The number of type i vehicles registered in month n; This is the vehicle survival rate of type i vehicles registered in month n when estimating month t; and These are characteristic parameters.

2. The method according to claim 1, characterized in that, Step S2 is as follows: The policy regulation scenario includes the combination of pollutant emission factor change rate, emission standard update interval, and the early scrapping year of vehicles with old standards; The rate of change of pollutant emission factors is set to two scenarios: uniform change A1 and fluctuating change A2; the emission standard update interval is set to three scenarios: high frequency B1, general B2, and low frequency B3; the advance year of old standard vehicles is set to three scenarios: natural elimination C1, last-place elimination C2, and second-last-place elimination C3. The three scenarios mentioned above are randomly combined to generate 18 policy regulation scenario combination patterns Pr,r=1,2,3,...,18, which correspond to combinations AaBbCc,a=1,2;b,c=1,2,3 respectively.

3. The method according to claim 2, characterized in that, Step S3 is as follows: The baseline pollutant emission factor is expressed as Where i represents the type of vehicle from the pollution source, including small passenger cars, medium-sized passenger cars, large passenger cars, small trucks, medium-sized trucks, large trucks, and container trucks; j represents the fuel type of the vehicle from the pollution source, including gasoline, diesel, and hybrid electric new energy vehicles; s represents the emission standard of the vehicle from the pollution source, including the existing China I pre-emission, China I, China II, China III, China IV, China V, and China VI emission standards, and the proposed China VII, China VIII, China IX, China X, and China XI emission standards; p represents the type of pollutant. Based on the basic emission factors measured under standard conditions using COPERT simulations, and combined with local conditions and highway road conditions, localization was performed using correction parameters reflecting driving conditions and environmental factors. Simultaneously, based on the rate of change of emission factors under different future policy control scenarios, the emission factors corresponding to each emission standard in the proposed future scenarios were obtained. The calculation formula is as follows: （2） In the formula, Under policy regulation scenario Pr, the emission factor (g*km-1) of category p pollutants emitted by category i vehicles with energy j and emission standards s; The corresponding basic emission factor is given by (g*km⁻¹). These are the corresponding environmental correction factors, including temperature and humidity; This is the corresponding speed correction factor; Other corresponding correction factors include load factor, lubricating oil parameters, oil quality, etc. Under the policy regulation scenario Pr, the change rate of the pollutant emission factors corresponds to the change rate of the emission factors in combination Aa.

4. The method according to claim 3, characterized in that, Step S4 is as follows: Perform ADF test and white noise test on the monthly number of different vehicle types and fuel types in the observation interval of the highway in the test area of ​​the time series to determine whether the time series is stationary. If it is, proceed to the next step; otherwise, perform d-order differencing on the time series to make it stationary. Based on the Akaike Information Criterion (AIC), the optimal parameters of the SARIMA(p,d,q)×(P,D,Q)s model are determined using grid search, where p, d, and q are the number of non-seasonal autoregressive terms, the number of moving average lag terms, and the difference order, respectively; P, D, and Q are the number of seasonal autoregressive terms, the number of moving average lag terms, and the difference order, respectively; and s is the seasonal period of the time series. The SARIMA model is trained using stationary time series training samples, and the prediction function of the SARIMA model is used for prediction to obtain the fitted time series sequence. and the corresponding residual sequence ; By sliding the window The sequence is reconstructed into a sequence of order u, and the residual sequence is trained and predicted using an SVR model to obtain the residual fitted sequence. The SARIMA-SVR model predicts the monthly number of vehicles traveling under different vehicle types and fuel types. for and sum; Based on the exponential smoothing method, the number of new vehicle registrations in the target month is predicted. Combined with the emission standard update frequency Bb and the early scrapping scheme Cc for vehicles with older emission standards under the policy control scenario Pr, the vehicle age and emission standard distribution for each target month are obtained. The monthly number of vehicles is then calculated. Based on emission standards, the monthly number of vehicles passing through the target month under scenario Pr, categorized by vehicle type, fuel type, and emission standard, is obtained. .

5. The method according to claim 4, characterized in that, Step S5 specifically involves: based on the monthly average vehicle mileage of the highway observation interval representing the month to be measured, using the quadratic exponential smoothing method to predict the monthly average vehicle mileage of the target month. ; The combined Monthly pollutant emissions from highway transportation under various scenarios for the target month. and emission factors The formula for calculating the estimated emissions of Class P pollutants from vehicles in month t under scenario Pr is as follows: （3） By analyzing the predicted emission trends of pollutant type p under various scenarios in the target month, the emission reduction effect of the adopted emission standard adjustment policy is evaluated.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by the processor, it implements the steps in the multi-scenario interval prediction method for highway transportation pollutant emissions as described in any one of claims 1 to 5.

7. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps in the multi-scenario interval prediction method for highway transportation pollutant emissions as described in any one of claims 1 to 5.

Images