Traffic travel carbon emission space-time prediction method based on land function evolution modeling

Abstract

The embodiment of the invention provides a traffic travel carbon emission space-time prediction method based on land function evolution modeling, and belongs to the technical field of data processing, and the method specifically comprises the steps: 1, constructing a multi-scenario land layout scenario constraint based on historical land function change data in a target region; step 2, utilizing the target model to predict urban land function space distribution of the target area under the constraint of the multi-scenario land layout scenario; 3, according to the urban land function space distribution, predicting the travel times of the passenger vehicles in each traffic zone, and according to the travel times, calculating the traffic travel distribution; and step 4, calculating the urban traffic travel carbon emission under the constraint of the multi-scenario land layout according to the traffic travel distribution. Through the scheme disclosed by the invention, the prediction efficiency, accuracy and adaptability are improved.

Description

Technical Field

[0001] The embodiments of the present disclosure relate to the field of data processing technology, and in particular to a method for spatiotemporal prediction of carbon emissions from transportation based on land use function evolution modeling. Background Art

[0002] Currently, there are two main types of methods for predicting carbon emissions from transportation: those based on historical data regression fitting and those based on transportation system modeling. Historical data regression fitting relies on feature-level fitting of multi-source historical data. This model, which ignores the true transmission dependencies between systems, can only make short-term, small-scale predictions based on historical development trends and is unable to meet the needs of long-term planning and open systems. Transportation system modeling-based prediction methods typically focus only on the evolutionary process and high-dimensional characteristics of the transportation system itself, resulting in complex model systems, precise covariate data requirements, and high computational costs.

[0003] It can be seen that there is an urgent need for a spatiotemporal prediction method for transportation carbon emissions based on land use function evolution modeling with high prediction efficiency, accuracy and adaptability. Summary of the Invention

[0004] In view of this, the embodiments of the present disclosure provide a spatiotemporal prediction method for transportation carbon emissions based on land use function evolution modeling, which at least partially solves the problems of poor prediction efficiency, accuracy and adaptability in the existing technology.

[0005] The present disclosure provides a method for spatiotemporal prediction of carbon emissions from transportation based on land use function evolution modeling, including:

[0006] Step 1: Construct multi-scenario land use layout scenario constraints based on historical land use function change data in the target area;

[0007] Step 2: Use the target model to predict the spatial distribution of urban land functions in the target area under the constraints of multiple land use layout scenarios;

[0008] Step 3: predict the number of passenger vehicle trips in each traffic zone based on the spatial distribution of urban land functions, and calculate the traffic trip distribution accordingly;

[0009] Step 4: Calculate urban transportation carbon emissions under multi-scenario land use layout constraints based on transportation distribution.

[0010] According to a specific implementation of the embodiment of the present disclosure, step 1 specifically includes:

[0011] Step 1.1: Construct a baseline development scenario based on historical land use function change data, keep the transition probability unchanged, and predict the conversion area of each type of land use based on an unrestricted Markov chain;

[0012] Step 1.2: Construct an urbanization development scenario based on historical land use function change data, control the ratio of residential land to work land within a specific range, and predict the conversion area of each type of land use based on a restricted Markov chain;

[0013] Step 1.3: Construct an ecological protection development scenario. Based on the historical land use function change data, increase the probability of other land use types being converted into green space, square land, and public facilities land types, and predict the conversion area of each type of land based on the restricted Markov chain.

[0014] According to a specific implementation of the embodiment of the present disclosure, step 2 specifically includes:

[0015] Step 2.1: Based on the baseline development scenario, urbanization development scenario, and ecological protection development scenario, the Markov model in the target model is used to predict the land use function area and land use type conversion probability matrix of the target area within the preset future period;

[0016] In step 2.2, based on the land use functional area, land use type conversion probability matrix and historical land use function change data, the dynamic land segmentation and vector unit automaton model in the target model are used to evolve the urban land use functional space in the preset period to obtain the urban land use functional space distribution.

[0017] According to a specific implementation of the embodiment of the present disclosure, step 2.2 specifically includes:

[0018] Step 2.2.1, segment each land unit in the historical land use function change data using the minimum area boundary rectangle to obtain the block distribution;

[0019] Step 2.2.2, calculating the development probability of each land unit being converted to various land use types, where the development probability includes overall development suitability, neighborhood effects, constraint factors, and random factors;

[0020] Step 2.2.3, select the type with the highest development probability and exceeding the development threshold in each land unit for conversion;

[0021] Step 2.2.4: Repeat steps 2.2.1 to 2.2.3 until the functional area of the land is met.

[0022] According to a specific implementation of the embodiment of the present disclosure, step 3 specifically includes:

[0023] Step 3.1: Use the mixed-use development model to analyze the urban land functions and spatial attributes corresponding to the spatial distribution of urban land functions, simulate and evaluate the travel demand of each transportation zone in the target area, and extract the number of passenger vehicle trips based on this;

[0024] In step 3.2, the traffic travel distribution among each traffic zone is calculated based on the gravity model and the number of passenger vehicle trips.

[0025] According to a specific implementation of the embodiment of the present disclosure, step 3.1 specifically includes:

[0026] Step 3.1.1: Combined with the road network, the target area is divided into multiple traffic zones according to the functional spatial distribution of urban land;

[0027] Step 3.1.2, calculating the impact factor of each transportation zone, wherein the impact factor includes land use, transportation accessibility, and socioeconomic factors;

[0028] Step 3.1.3, based on the impact factors, calculate the travel demand of each transportation zone using the travel rates in the ITE travel generation manual;

[0029] Step 3.1.4, decompose travel demand into internal trips and external trips and calculate the total external trips accordingly;

[0030] Step 3.1.5: Decompose the total number of external trips into walking trips, bicycle trips, public transportation trips, and passenger vehicle trips, and calculate the probabilities of walking, cycling, and public transportation in external trips using a pre-set formula;

[0031] In step 3.1.6, the probabilities of walking, cycling, and public transportation are subtracted from the total probability to obtain the trip probability of passenger vehicles in external trips. The number of passenger vehicle trips in each traffic zone is calculated based on the total number of external trips and the trip probability of passenger vehicles.

[0032] According to a specific implementation of the embodiment of the present disclosure, step 3.2 specifically includes:

[0033] Step 3.2.1: Calculate the spatial interaction intensity between every two traffic areas of all traffic areas based on the number of passenger vehicle trips. The expression of the spatial interaction intensity is:

[0034]

[0035] Among them, T ij represents the number of trips from traffic area i to traffic area j, T ji represents the number of trips from traffic area j to traffic area i, T cari and T carj denote the number of passenger vehicle trips in traffic area i and traffic area j respectively;

[0036] Step 3.2.2, assume that the Euclidean distance between traffic area i and traffic area j is Dij , the spatial interaction strength is C ij , for the spatial interaction intensity C ij and Euclidean distance D ij Perform logarithmic transformation and perform linear fitting on the transformed data, and calculate the distance decay index β based on this, where the linear fitting expression is:

[0037] log(C ij )=α-βlog(D ij )

[0038] Among them, D ij represents the Euclidean distance between regions i and j, C ij represents the spatial interaction intensity, α is a constant term, and β represents the distance decay exponent;

[0039] Step 3.2.3, based on the distance decay index, use the gravity model formula to distribute the number of passenger vehicle trips in each traffic zone to different traffic zones, and obtain the traffic trip distribution between each traffic zone.

[0040]

[0041] According to a specific implementation of the embodiment of the present disclosure, step 4 specifically includes:

[0042] Based on the distribution of traffic trips among different traffic zones, the vehicle emission calculation model is used to calculate the predicted carbon emissions of traffic trips in the target area based on the travel distance and number of trips of passenger vehicles.

[0043]

[0044] Among them, E co2 Indicates the carbon dioxide emissions of passenger vehicles, OD ij represents the travel distribution from traffic area i to traffic area j, R gasoline It represents the proportion of gasoline vehicles, EI represents the energy intensity of fuel, and EF represents the carbon dioxide emission factor.

[0045] The spatiotemporal prediction scheme for transportation carbon emissions based on land use function evolution modeling in the embodiment of the present disclosure includes: step 1, constructing multi-scenario land use layout scenario constraints based on historical land use function change data in the target area; step 2, using the target model to predict the spatial distribution of urban land use functions in the target area under the multi-scenario land use layout scenario constraints; step 3, predicting the number of passenger vehicle trips in each traffic zone based on the spatial distribution of urban land use functions, and calculating the transportation travel distribution accordingly; step 4, calculating the urban transportation carbon emissions under the multi-scenario land use layout constraints based on the transportation travel distribution.

[0046] The beneficial effects of the embodiments of the present disclosure are as follows: through the scheme of the present disclosure, a transportation carbon emission prediction framework based on the evolution of land use functions is constructed, the future land use pattern is predicted through vector land use function evolution simulation, and a chain prediction model of land use function-transportation-carbon emissions is established to achieve reliable prediction of future urban transportation carbon emissions; a quantitative model of the relationship between land use functions and carbon emissions under multi-scenario constraints is established, and socio-economic and policy factors are modeled as land function layout constraints to predict transportation carbon emissions under different scenarios, providing a low-cost, highly reliable decision support tool for urban management; the mixed-use development model is connected in series with the gravity model to solve the difficulty of traditional methods in collecting a large amount of micro-traffic monitoring data; the restrictive Markov chain under multi-scenario constraints is introduced to achieve parameterized expression of different policy scenarios, thereby improving prediction efficiency, accuracy and adaptability. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0048] Figure 1 A schematic flow chart of a method for spatiotemporal prediction of carbon emissions from transportation based on land use function evolution modeling provided by an embodiment of the present disclosure;

[0049] Figure 2 A schematic diagram of a specific implementation process of a method for spatiotemporal prediction of carbon emissions from transportation based on land use function evolution modeling provided by an embodiment of the present disclosure;

[0050] Figure 3 A multi-source geographical environment element feature distribution map provided in an embodiment of the present disclosure;

[0051] Figure 4 This is a diagram showing the predicted spatial distribution of urban land functions under different scenarios provided by the embodiments of the present disclosure;

[0052] Figure 5 This is a graph of the predicted results of total carbon emissions from transportation under different scenarios provided by the embodiments of the present disclosure. DETAILED DESCRIPTION

[0053] The embodiments of the present disclosure are described in detail below with reference to the accompanying drawings.

[0054] The following describes the embodiments of the present disclosure through specific examples, and those skilled in the art can easily understand other advantages and effects of the present disclosure from the contents disclosed in this specification. Obviously, the described embodiments are only a part of the embodiments of the present disclosure, rather than all of the embodiments. The present disclosure can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present disclosure. It should be noted that, in the absence of conflict, the following embodiments and features in the embodiments can be combined with each other. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present disclosure.

[0055] It should be noted that various aspects of the embodiments within the scope of the appended claims are described below. It should be apparent that the aspects described herein can be embodied in a wide variety of forms, and any specific structure and / or function described herein is merely illustrative. Based on this disclosure, it should be understood by those skilled in the art that an aspect described herein can be implemented independently of any other aspect, and two or more of these aspects can be combined in various ways. For example, any number of aspects described herein can be used to implement an apparatus and / or practice a method. In addition, other structures and / or functionalities other than one or more of the aspects described herein can be used to implement this apparatus and / or practice this method.

[0056] It should also be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present disclosure. The illustrations only show components related to the present disclosure and are not drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component can be changed at will, and the component layout type may also be more complicated.

[0057] Additionally, in the following description, specific details are provided to provide a thorough understanding of the examples. However, one skilled in the art will appreciate that the aspects described can be practiced without these specific details.

[0058] Current methods for predicting carbon emissions from transportation fall into two main categories: those based on historical data regression fitting and those based on transportation system modeling. Historical data regression fitting relies on feature-level fitting of multi-source historical data. This model, which ignores the true transmission dependencies between systems, can only make short-term, small-scale predictions based on historical development trends and is unable to meet the needs of long-term planning and open systems. Transportation system modeling-based methods typically focus solely on the evolution and high-dimensional characteristics of the transportation system itself, resulting in complex model systems, precise covariate data requirements, and high computational costs.

[0059] Real-world spatial decision-making involves a multi-faceted, multi-faceted system. Evolutionary predictions and analysis based solely on a single spatiotemporal system can easily yield spurious or even contradictory planning results. However, current methods for predicting carbon emissions from transportation primarily assess the effects of future transportation carbon reductions based on adjustments to transportation structure, neglecting the a priori guidance of policies and socioeconomic benefits. This makes it difficult to effectively integrate carbon emission predictions with current policy and socioeconomic development trends for decision-making analysis. Empirical research and theoretical mechanisms of historical urbanization processes demonstrate that changes in urban land use significantly influence transportation patterns and their carbon emissions. Urban land use is the primary decision-making element in urban planning and construction. Any urban planning activity, based on changes in land use patterns, influences urban development, including transportation systems and carbon emissions. Qualitative research indicates that appropriate land use mix and a rational land use pattern distribution can effectively reduce light-duty vehicle travel and mileage, increase the proportion of walking and public transportation trips, and thus reduce total transportation carbon emissions. Therefore, changes in urban land use functional layout will directly impact the future pattern and total amount of transportation carbon emissions. However, under the current planning-led land use layout development scenario, there is still a research gap in how urban transportation carbon emissions change quantitatively.

[0060] The existing prediction models based on regression fitting rely on complete covariate data, which makes it difficult to adapt to the spatiotemporal dynamic changes in variable relationships and is only applicable to short-term stable scenarios; the models based on system evolution focus too much on micro-traffic characteristics, and the models are complex and computationally expensive.

[0061] Existing methods only focus on the evolution of the transportation system itself, ignore the impact of land use changes on transportation travel behavior, and ignore changes in land use distribution and socio-economic systems, resulting in "scenario bias" and distorted prediction results.

[0062] Therefore, there is an urgent need for a new method that can predict the future development trend of urban transportation carbon emissions under different land use layout scenarios from the perspective of urban land use functions, and provide decision support for policymakers and urban managers.

[0063] The embodiments of the present disclosure provide a spatiotemporal prediction method for transportation carbon emissions based on land use function evolution modeling, which can be applied to the transportation carbon emissions prediction process in urban management scenarios.

[0064] See also Figure 1 , which is a flow chart of a method for spatiotemporal prediction of carbon emissions from transportation based on land use function evolution modeling provided by an embodiment of the present disclosure. Figure 1 and Figure 2 As shown, the method mainly includes the following steps:

[0065] Step 1: Construct multi-scenario land use layout scenario constraints based on historical land use function change data in the target area;

[0066] In specific implementation, multi-scenario land use layout scenario constraints can be constructed based on the perspective of transportation carbon emissions. The specific process is as follows:

[0067] 1.1 Constructing a baseline development scenario

[0068] Based on the historical land use function change data, the transfer probability remains unchanged and the conversion area of each type of land is predicted based on the unrestricted Markov chain;

[0069] 1.2 Constructing urbanization development scenarios

[0070] Based on historical land use function change data, the ratio of residential land to work land is controlled within a specific range, and the conversion area of each type of land use is predicted based on a restricted Markov chain.

[0071] 1.3 Constructing an ecological protection and development scenario

[0072] Based on the historical land use function change data, the probability of other land use types being converted into green space, square land, public facilities land, etc. is increased, and the conversion area of each type of land use is predicted based on the restrictive Markov chain.

[0073] Step 2: Use the target model to predict the spatial distribution of urban land functions in the target area under the constraints of multiple land use layout scenarios;

[0074] In specific implementation, the target model Markov-DLPS-VCA includes a Markov model and a dynamic land segmentation and vector unit automaton model. The Markov-DLPS-VCA model is used to predict the spatial distribution of urban land functions under the constraints of multiple land use layout scenarios. The specific process is as follows:

[0075] 2.1 Prediction of functional land area based on Markov model

[0076] Markov chain is a prediction and optimization control method that assumes that state transitions depend on historical states. In geographic simulation, Markov chain can not only reveal the transition probability between different land types, but also predict the proportion of various land uses in the future. Therefore, it is widely used in the prediction process of geographic research. The Markov chain prediction process is shown in formulas (1) and (2):

[0077] S(t+1)=P i,j S(t) (1)

[0078]

[0079] In the formula, S(t) and S(t+1) represent the state of the system at time t and t+1 respectively, P i,j It is a matrix representing the probability of land use type conversion, that is, the development probability.

[0080] 2.2 Spatial evolution of urban land use functions based on the DLPS-VCA model

[0081] The Dynamic Land Segmentation and Vector Cell Automata model (DLPS-VCA model) aims to consider dynamic land segmentation during urban land use change by integrating the random forest algorithm (RFA) and vector cell automation technology, while simulating urban expansion and land use change at the land cell scale. The DLPS-VCA model can more accurately predict and simulate urban land use changes. The implementation of the model includes the following three main steps:

[0082] (1) Land segmentation: First, each vector land unit is segmented by the minimum area bounding rectangle (MABR) to obtain a reasonable block distribution along the urban road. This process ensures that the direction of the land segmentation is consistent with the original land block, making the simulation results more consistent with the actual urban layout. Then, the average area of each land type block (μ i ) and standard deviation (σ i ), and the plots P with an area larger than (μi+2σi) i,j Then, by iterating the vertices of the polygon, we can get the plot P i,j Make sure the MABR is oriented in the same direction as the original plot. Make a perpendicular bisector of the long side of the MABR and divide the plot P i,j Divided into two new land parcels P1 i,j and P2 i,j Finally, repeat the above steps until the area of all land parcels is smaller than the current (μi+2σi).

[0083] (2) Creation of auxiliary spatial variables and introduction of RFA model: After the land parcels are subdivided, they are treated as basic units and simulated based on the cellular automaton model. Previous studies have shown that the development probability P of each unit is composed of four factors, including the overall development suitability Pg, the domain effect Ω, the constraint factor Pc, and the random factor RA. By calculating the probability of each land unit being converted to various land use types, the unit with the highest conversion probability and exceeding the development threshold is selected for conversion, as shown in formula (3):

[0084]

[0085] Where, is the probability that plot i changes to land use type k at time t, Indicates the overall development suitability, is the neighborhood effect, is a constraint factor, and RA is a random factor. The calculation process of each factor is described in detail below: ① Formula (4) is the calculation formula for overall development suitability, where I(·) is the indicator function of the decision tree set, M is the total number of decision trees, x is a high-dimensional vector consisting of auxiliary spatial variables in the plot, and h n (x) The predicted type of x by the nth decision tree, that is, the land use change type result of each decision tree for the i-th non-construction land transformation; ② Formula (5) is the neighborhood effect calculation formula, where e is the exponential constant, d ij is the center distance between plot i and plot j, S i and S j Represent the area of plot i and plot j respectively, S max and S min Represent the maximum and minimum values of the plot area in the study area respectively; ③ Formula (6) is the calculation formula for the constraint factor of plot i, U i To determine the development suitability of the land parcel, in this study, water bodies (including rivers, lakes, and sea areas) and roads are designated as restricted development areas. This step is performed until the land use area predicted in step 2.1 is met.

[0086] (3) DLPS-VCA model validation: This study uses the traditional cell-to-cell method to evaluate the simulation accuracy of the DLPS-VCA model by converting the actual and simulated urban land use data into raster data and calculating the overall accuracy (OA) and Kappa coefficient. At the same time, this study uses the Figure-of-Merit (FoM) to evaluate the accuracy of the simulation results. This indicator represents the proportion of consistency between the observed changes and the predicted changes. The formulas are shown in (7) to (9).

[0087]

[0088] Where A represents the error when land use change actually occurs but the simulation result does not change; B represents the error when land use change actually occurs but the simulation result changes correctly; C represents the error when land use change actually occurs but the simulation result changes in fact; D represents the error when land use change does not actually occur but occurs in the simulation.

[0089] Step 3: predict the number of passenger vehicle trips in each traffic zone based on the spatial distribution of urban land functions, and calculate the traffic trip distribution accordingly;

[0090] In specific implementation, the specific process of estimating the spatial distribution of traffic volume and traffic trip distribution based on the predicted spatial distribution of urban land functions is as follows:

[0091] 3.1 Calculation of spatial distribution of traffic volume

[0092] The Mixed-Use Development Model (MXD) is a land-use-transportation model used to estimate travel demand. By analyzing the functional and spatial attributes of urban land, the model can simulate and assess travel demand for each transportation zone within the study area, including total trips and trips by different modes. The main steps are as follows:

[0093] (1) Research unit division. Divide the traffic zones according to the distribution and pattern of different urban land use functions. Based on the existing traffic network, fully consider the road network constraints to ensure that each zone can accurately reflect the traffic behavior characteristics within it.

[0094] (2) Calculation of vehicle trip times

[0095] The Institute of Transportation Engineers (ITE) model based on land use was used to estimate the total number of trips using the trip rates from the ITE trip generation manual.

[0096] After calculating the total number of trips, vehicle trip types were categorized into three types based on purpose: home-to-work trips, home-to-other trips, and non-home trips. Based on the NCHRP report, the total number of trips was broken down into the total number of trips by purpose. The estimated vehicle trips were then broken down into internal and external vehicle trips. The log-odds ratio was indexed to obtain the ratio of internal trips, which was then converted to a probability. Finally, the total number of internal trips was subtracted from the total number of trips estimated using the ITE Trip Generation Manual to obtain the total number of external trips.

[0097] The estimated external travel trips are decomposed into walking trips, bicycle trips, public transportation trips and light passenger vehicle trips, and the probabilities of walking, cycling and public transportation in external trips are calculated through the equations in the model.

[0098] The probability of a light-duty vehicle trip in external travel is obtained by subtracting the probabilities of walking, cycling, and public transportation from the total probability, and the total external travel volume is multiplied by this probability to obtain the total amount of external travel generated by light-duty vehicles.

[0099] T totalj =∑ i f i (X ij ) (10)

[0100] lo INj =β IN +Σ i αi ln(Variable) (11)

[0101] p INj = exp(lo INj ) / (1 + exp(lo INj )) (12)

[0102] T EAj = T totalj × (1 - p INj ) (13)

[0103] lo walkj = β walk + Σ k α k ln(Variable) (14)

[0104] p walkj = exp(lo walkj ) / (1 + exp(lo walkj )) (15)

[0105] lo bikej = β bike + Σ k α m ln(Variable) (16)

[0106] p bikej = exp(lo bikej ) / (1 + exp(lo bikej )) (17)

[0107] lo ptj = β pt + ∑ m α n ln(Variable) (18)

[0108] p ptj = exp(lo ptj ) / (1 + exp(lo ptj )) (19)

[0109] p carj = 1 - p walkj - p bikej - p ptj (20)

[0110] T carj = T EAj × p carj (21)

[0111] In the formula, T totalj is the total vehicle trips in zone j, Xij is the area of urban land function i in region j; function f i (X ij ) Calculate the total number of trips for urban land use type i starting from traffic zone j. The specific calculation formula is introduced in the ITE model. This process estimates the trip volume of each land use type by the degree of influence of different land use function types on the trip volume; INj represents the internal trip occurrence ratio from traffic zone j; p INj represents the internal trip probability from traffic zone j; T EAj represents the total external travel volume from traffic zone j; Variable is the influencing factor related variable; β IN , α i It is a correlation coefficient based on existing research. The estimation of this coefficient in existing research is mainly obtained by regression of a large amount of real survey data. Since the survey data comes from cities in different regions and at different levels of development around the world, the reliability of this model is widely recognized and widely used in cities around the world. walkj 、lo bikej 、lo ptj represents the occurrence ratio of walking trips, bicycle trips and public transportation trips from the external trip zone j; p walkj 、p bikej 、p ptj represents the probability of walking trips, bicycle trips, and public transportation trips from the external trips of transportation community j; Variable is the influencing factor related variable; β walk , β bike , β pt , α k , α m , α n The calculation method is the same as above; p carj represents the occurrence ratio of light passenger vehicles in external travel; T carj represents the number of light passenger vehicle trips.

[0112] 3.2 Urban traffic flow intensity

[0113] The gravity model-based travel allocation method is primarily used to calculate the distribution of trips between regions. The gravity model assumes that travel demand is proportional to the attractiveness of the departure and arrival regions, and inversely proportional to some power of the travel distance. By applying the gravity model, we can more accurately estimate the distribution of travel demand between regions, providing basic data for further calculations of transportation carbon emissions. This process can be broken down into the following three steps:

[0114] (1) Calculation of spatial interactions between regions

[0115] Spatial interaction refers to the interaction strength between two regions. Usually, the spatial interaction between regions can be estimated by actual taxi trajectory data. The specific method is to use taxi trajectory data to analyze the travel frequency between different regions and quantify the interaction strength between regions. Suppose there are two regions i and j, and their interaction strength C ij The calculation formula is as follows:

[0116]

[0117] Where, T ij represents the number of trips from area i to area j, T ji represents the number of trips from area j to area i, T cari and T carj denote the total number of trips in region i and region j, respectively. This step calculates the interaction strength between regions and provides basic data for subsequent gravity model calculations.

[0118] (2) Calculation of distance attenuation index

[0119] The distance decay index is a measure of how quickly the spatial interaction intensity changes over geographic distance. Specifically, the distance decay index can be calculated by linearly fitting the spatial interaction intensity and distance between different regions in the taxi trajectory data. Assume that the Euclidean distance between two regions i and j is D ij , the spatial interaction strength is C ij , then the distance decay exponent β can be calculated by the following steps: First, ij and D ij Perform logarithmic transformation and perform linear fitting on the transformed data. The fitting equation is as follows:

[0120] log(C ij )=α-βlog(D ij )

[0121] Where D ij is the Euclidean distance between regions i and j; C ij is the spatial interaction strength; α is a constant term; and β is the distance decay exponent. The β value obtained by fitting can be used to calculate the gravity model.

[0122] (3) Trip frequency distribution

[0123] Using the gravity model formula, the number of trips is allocated to each area to obtain the traffic distribution between each traffic zone. The specific formula is as follows:

[0124]

[0125] Where, OD ijis the number of trips from area i to area j; T cari is the total number of trips in area i; D ij is the Euclidean distance from the center of region i to the center of region j; β is the distance decay exponent.

[0126] Step 4: Calculate urban transportation carbon emissions under multi-scenario land use layout constraints based on transportation distribution.

[0127] In specific implementation, the specific process of calculating urban transportation carbon emissions under multi-scenario land use layout constraints based on transportation distribution is as follows:

[0128] Based on the estimated spatial distribution of traffic volume by urban land use function, a vehicle emission calculation model is used to calculate carbon dioxide emissions based on the distance traveled and the number of trips by light-duty passenger vehicles. The total carbon dioxide emissions calculation formula for light-duty passenger vehicles is as follows:

[0129]

[0130] Where, E co2 is the carbon dioxide emissions of light passenger vehicles; OD ij is the number of trips from area i to area j; D ij is the Euclidean distance from region i to region j; R gasoline is the proportion of gasoline vehicles; EI is the energy intensity of fuel; EF is the carbon dioxide emission factor.

[0131] The spatiotemporal prediction method for transportation carbon emissions based on land use function evolution modeling provided in this embodiment overcomes the problem of existing transportation carbon emissions prediction models ignoring the true evolutionary relationships of real multi-factor spatial systems by establishing a simulated evolutionary model for transportation carbon emissions under land use layout scenario constraints. By modeling basic system factors such as socio-economic and policy as scenarios constrained by urban land use function layout, a systematic prediction of the entire chain (policy guidance → socio-economic system → land system → transportation system → transportation carbon emissions) is achieved, addressing the technical bottleneck of traditional methods that cannot capture the complex interactions of multiple systems in long-term scenario predictions, and meeting the demand for low-cost, highly reliable, and decision-making-indicative transportation carbon emissions predictions. A hierarchical technology integration solution is adopted in the computing architecture, connecting the mixed-use development model with the gravity model in series, solving the difficulty of traditional methods in collecting large amounts of micro-traffic monitoring data. The introduction of a restrictive Markov chain under multi-scenario constraints enables parameterized expression of different policy scenarios, improving prediction efficiency, accuracy, and adaptability.

[0132] The method of the present disclosure will be further described below with reference to a specific example. Taking area A as the research area, the method proposed in the present invention is applied to predict the carbon emissions of transportation in 2025 and 2030 under different land use layout scenarios. The specific implementation steps are as follows:

[0133] (1) Constructing multi-scenario land use layout scenario constraints based on the perspective of carbon emissions from transportation. Based on the land use function change data of site A from 2018 to 2023, three development scenarios are constructed: ① Baseline development scenario: the transfer probability remains unchanged from the historical level, and the conversion area of each type of land use is predicted based on the unrestricted Markov chain; ② Urbanization development scenario: the area ratio of residential land to work land (including commercial land and industrial land) is controlled at about 1:1, and the conversion area of each type of land use is predicted based on the restricted Markov chain; ③ Ecological protection development scenario: the probability of converting other land use types to green space, square land, public facilities land, etc. is increased to 120% of the historical conversion probability average, and the conversion area of each type of land use is predicted based on the restricted Markov chain.

[0134] (2) Obtain the characteristics of basic geographical environment elements. Calculate 14 geographical environment element characteristics, including distance to the district center, distance to railway, distance to highway, distance to main road, digital elevation model (DEM), slope, hospital density, bus station density, restaurant density, entertainment facility density, park density, supermarket density, shopping mall density and factory density. Some of the characteristics are distributed as follows: Figure 3 As shown, (a) is the distance to the main road, (b) is the distance to the highway, (c) is the OSM road network density, (d) is the elevation, (e) is the slope, (f) is the subway station density, (g) is the bus station density, (h) is the shopping density, (i) is the restaurant density, (j) is the park density, (k) is the hospital density, (l) is the factory density, and (m) is the shopping mall density. These features serve as the input of auxiliary spatial variables in the DLPS-VCA model.

[0135] (3) Based on the Markov-DLPS-VCA model, the functional pattern of urban land use under the constraints of multiple scenarios of land use layout is predicted. First, the Markov model is used to predict the area of various types of land use in 2025 and 2030, and the land use type conversion probability matrix is calculated. Then, the spatial distribution of land use is predicted by the DLPS-VCA model: ① Each vector land unit is divided by the minimum area boundary rectangle to obtain a reasonable block distribution; ② The probability of each land unit being converted to various land use types is calculated, including overall development suitability, neighborhood effect, constraint factor and random factor; ③ The type with the highest conversion probability and exceeding the development threshold is selected for conversion, and this process is repeated until the predicted land use function area is met. The prediction results of the functional pattern of urban land use under each scenario are as follows: Figure 4As shown in the figure, (a) is the baseline development scenario in 2025, (b) is the baseline development scenario in 2030, (c) is the urbanization development scenario in 2025, (d) is the urbanization development scenario in 2030, (e) is the ecological protection development scenario in 2025, and (f) is the ecological protection development scenario in 2030.

[0136] (4) Estimate the intensity of traffic flow based on the predicted functional pattern of urban land use. Use the mixed-use development (MXD) model to estimate the travel demand of each traffic zone: ① Divide the traffic zones according to the functional distribution of different urban land uses; ② Use the Institute of Transportation Engineers (ITE) model to estimate the total number of trips; ③ Decompose the total trip volume into internal and external trips; ④ Further decompose the external trip volume into walking, bicycle, public transportation, and light passenger vehicle trips. ⑤ Based on the spatial distribution of traffic trips, apply the gravity model to calculate the distribution of trips between regions and estimate the intensity of urban traffic flow.

[0137] (5) Based on the intensity of traffic flow, calculate the carbon emissions of urban transportation under the constraints of land use layout in different scenarios. Substitute the driving distance, number of trips, energy intensity and emission factor of light passenger vehicles into the formula to calculate carbon dioxide emissions. Referring to the energy consumption parameters in relevant research literature, the energy intensity of gasoline vehicles is 0.08kgce / km and the CO2 emission factor is 2.08kgCO2 / kgce. The spatial distribution prediction results of the total carbon emissions of transportation under different scenarios are as follows: Figure 5 As shown in the figure, (a) is the baseline development scenario in 2025, (b) is the baseline development scenario in 2030, (c) is the urbanization development scenario in 2025, (d) is the urbanization development scenario in 2030, (e) is the ecological protection development scenario in 2025, and (f) is the ecological protection development scenario in 2030. As can be seen from the figure, carbon emissions from transportation are more distributed in the central and eastern parts of the urban area of Area A, while areas with lower carbon emissions from transportation are concentrated in the southern and northwestern parts of Area A.

[0138] It should be understood that various parts of the present disclosure can be implemented in hardware, software, firmware, or a combination thereof.

[0139] The above description is merely a specific embodiment of the present disclosure, but the scope of protection of the present disclosure is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this disclosure should be included in the scope of protection of the present disclosure. Therefore, the scope of protection of the present disclosure should be based on the scope of protection of the claims.

Claims

1. A spatiotemporal prediction method for transportation carbon emissions based on land use function evolution modeling, characterized by: include: Step 1: Construct multi-scenario land use layout scenario constraints based on historical land use function change data in the target area; Step 2: Use the target model to predict the spatial distribution of urban land functions in the target area under the constraints of multiple land use layout scenarios; Step 3: predict the number of passenger vehicle trips in each traffic zone based on the spatial distribution of urban land functions, and calculate the traffic trip distribution accordingly; Step 4: Calculate urban transportation carbon emissions under multi-scenario land use layout constraints based on transportation distribution.

2. The method according to claim 1, characterized in that The step 1 specifically includes: Step 1.1: Construct a baseline development scenario based on historical land use function change data, keep the transition probability unchanged, and predict the conversion area of each type of land use based on an unrestricted Markov chain; Step 1.2: Construct an urbanization development scenario based on historical land use function change data, control the ratio of residential land to work land within a specific range, and predict the conversion area of each type of land use based on a restricted Markov chain; Step 1.3: Construct an ecological protection development scenario. Based on the historical land use function change data, increase the probability of other land use types being converted into green space, square land, and public facilities land types, and predict the conversion area of each type of land based on the restricted Markov chain.

3. The method according to claim 2, characterized in that The step 2 specifically includes: Step 2.1: Based on the baseline development scenario, urbanization development scenario, and ecological protection development scenario, the Markov model in the target model is used to predict the land use function area and land use type conversion probability matrix of the target area within the preset future period; In step 2.2, based on the land use functional area, land use type conversion probability matrix and historical land use function change data, the dynamic land segmentation and vector unit automaton model in the target model are used to evolve the urban land use functional space in the preset period to obtain the urban land use functional space distribution.

4. The method according to claim 3, characterized in that The step 2.2 specifically includes: Step 2.2.1, segment each land unit in the historical land use function change data using the minimum area boundary rectangle to obtain the block distribution; Step 2.2.2, calculating the development probability of each land unit being converted to various land use types, where the development probability includes overall development suitability, neighborhood effects, constraint factors, and random factors; Step 2.2.3, select the type with the highest development probability and exceeding the development threshold in each land unit for conversion; Step 2.2.4: Repeat steps 2.2.1 to 2.2.3 until the functional area of the land is met.

5. The method according to claim 4, characterized in that The step 3 specifically includes: Step 3.1: Use the mixed-use development model to analyze the urban land functions and spatial attributes corresponding to the spatial distribution of urban land functions, simulate and evaluate the travel demand of each transportation zone in the target area, and extract the number of passenger vehicle trips based on this; In step 3.2, the traffic travel distribution among each traffic zone is calculated based on the gravity model and the number of passenger vehicle trips.

6. The method according to claim 5, characterized in that The step 3.1 specifically includes: Step 3.1.1: Combined with the road network, the target area is divided into multiple traffic zones according to the functional spatial distribution of urban land; Step 3.1.2, calculating the impact factor of each transportation zone, wherein the impact factor includes land use, transportation accessibility, and socioeconomic factors; Step 3.1.3, based on the impact factors, calculate the travel demand of each transportation zone using the travel rates in the ITE travel generation manual; Step 3.1.4, decompose travel demand into internal trips and external trips and calculate the total external trips accordingly; Step 3.1.5: Decompose the total number of external trips into walking trips, bicycle trips, public transportation trips, and passenger vehicle trips, and calculate the probabilities of walking, cycling, and public transportation in external trips using a pre-set formula; In step 3.1.6, the probabilities of walking, cycling, and public transportation are subtracted from the total probability to obtain the trip probability of passenger vehicles in external trips. The number of passenger vehicle trips in each traffic zone is calculated based on the total number of external trips and the trip probability of passenger vehicles.

7. The method according to claim 6, characterized in that The step 3.2 specifically includes: Step 3.2.1: Calculate the spatial interaction intensity between every two traffic areas of all traffic areas based on the number of passenger vehicle trips. The expression of the spatial interaction intensity is: Among them, T ij represents the number of trips from traffic area i to traffic area j, T ji represents the number of trips from traffic area j to traffic area i, T cari and T carj denote the number of passenger vehicle trips in traffic area i and traffic area j respectively; Step 3.2.2, assume that the Euclidean distance between traffic area i and traffic area j is D ij , the spatial interaction strength is C ij , for the spatial interaction intensity C ij and Euclidean distance D ij Perform logarithmic transformation and perform linear fitting on the transformed data, and calculate the distance decay index β based on this, where the linear fitting expression is: log(C ij )=α-βlog(D ij ) Among them, D ij represents the Euclidean distance between regions i and j, C ij represents the spatial interaction intensity, α is a constant term, and β represents the distance decay exponent; Step 3.2.3, based on the distance decay index, use the gravity model formula to distribute the number of passenger vehicle trips in each traffic zone to different traffic zones, and obtain the traffic trip distribution between each traffic zone.

8. The method according to claim 7, characterized in that The step 4 specifically includes: Based on the distribution of traffic trips among different traffic zones, the vehicle emission calculation model is used to calculate the predicted carbon emissions of traffic trips in the target area based on the travel distance and number of trips of passenger vehicles. Among them, E co2 Indicates the carbon dioxide emissions of passenger vehicles, OD ij represents the travel distribution from traffic area i to traffic area j, R gasoline It represents the proportion of gasoline vehicles, EI represents the energy intensity of fuel, and EF represents the carbon dioxide emission factor.

Images