Intelligent connected road transportation system energy consumption and emission detection and analysis method and electronic device

Abstract

The present application relates to the technical field of intelligent transportation, and especially relates to a kind of intelligent network connection road traffic system energy consumption emission detection analysis method and electronic equipment, the detection method includes, obtaining first macroscopic traffic data and first microscopic traffic data in traffic data stream;First macroscopic traffic data and first microscopic traffic data are respectively input into macroscopic energy consumption emission model and microscopic energy consumption emission model, and macroscopic energy consumption emission benchmark and microscopic energy consumption emission benchmark are obtained;First macroscopic traffic data and first microscopic traffic data are respectively processed, and second macroscopic traffic data and second microscopic traffic data are obtained;Second macroscopic traffic data and second microscopic traffic data are respectively input into macroscopic energy consumption emission model and microscopic energy consumption emission model, and first fusion data and second fusion data are respectively obtained;According to macroscopic energy consumption emission benchmark, microscopic energy consumption emission benchmark, first fusion data and second fusion data, the accurate detection result of traffic energy consumption emission is obtained.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent transportation, and in particular to an intelligent networked road traffic system energy consumption and emission detection and analysis method and an electronic device. BACKGROUND

[0002] Vehicle emissions, as the main source of traffic emissions, are one of the serious environmental problems currently faced by the world. With the increasing energy consumption and emissions of motor vehicles, many countries have proposed a large number of energy consumption and emission models.

[0003] Most of the existing energy consumption and emission models input traffic data into the same level of energy consumption and emission model to calculate energy consumption and emissions. Since the data of the traffic flow model is difficult to be consistent with the input data of the energy consumption and emission model, there are few studies on the fusion of traffic flow models and energy consumption and emission models of different levels. Most of the existing studies focus on using a combination of a traffic flow model and an energy consumption and emission model to estimate the energy consumption and emissions of a road network or a region, and few studies compare and analyze the calculation results of different combinations on the same road network or region. Therefore, the monitoring results obtained by the existing traffic energy consumption and emission detection method are not accurate enough. SUMMARY

[0004] The present application provides an intelligent networked road traffic system energy consumption and emission detection and analysis method to solve the technical defect that the monitoring results obtained by the existing traffic energy consumption and emission detection method are not accurate enough.

[0005] In one aspect, the present application provides an intelligent networked road traffic system energy consumption and emission detection and analysis method, comprising:

[0006] obtaining first macroscopic traffic data and first microscopic traffic data in a traffic data stream;

[0007] inputting the first macroscopic traffic data and the first microscopic traffic data into a macroscopic energy consumption and emission model and a microscopic energy consumption and emission model respectively to obtain a macroscopic energy consumption and emission benchmark and a microscopic energy consumption and emission benchmark respectively;

[0008] decomposing the first macroscopic traffic data to obtain second microscopic traffic data, and aggregating the first microscopic traffic data to obtain second macroscopic traffic data;

[0009] inputting the second macroscopic traffic data and the second microscopic traffic data into the macroscopic energy consumption and emission model and the microscopic energy consumption and emission model respectively to obtain first fusion data and second fusion data respectively;

[0010] obtaining a traffic energy consumption and emission detection result according to the macroscopic energy consumption and emission benchmark, the microscopic energy consumption and emission benchmark, the first fusion data and the second fusion data.

[0011] The intelligent network connection road traffic system energy consumption emission detection analysis method provided by the application can obtain the first macroscopic traffic data in the traffic data stream.

[0012] The second microscopic traffic data is the data of the traffic flow collected by the mobile sensor arranged on the vehicle.

[0013] The intelligent network connection road traffic system energy consumption emission detection analysis method provided by the application can obtain the first macroscopic traffic data in the traffic data stream.

[0014] The intelligent network connection road traffic system energy consumption emission detection analysis method provided by the application can obtain the first macroscopic traffic data in the traffic data stream.

[0015] The intelligent network connection road traffic system energy consumption emission detection analysis method provided by the application can obtain the first macroscopic traffic data in the traffic data stream.

[0016] The intelligent network connection road traffic system energy consumption emission detection analysis method provided by the application can obtain the first macroscopic traffic data in the traffic data stream.

[0017] The intelligent network connection road traffic system energy consumption emission detection analysis method provided by the application can obtain the first macroscopic traffic data in the traffic data stream.

[0018] The intelligent network connection road traffic system energy consumption emission detection analysis method provided by the application can obtain the first macroscopic traffic data in the traffic data stream.

[0019] The intelligent network connection road traffic system energy consumption emission detection analysis method provided by the application can obtain the first macroscopic traffic data in the traffic data stream.

[0020] The intelligent network connection road traffic system energy consumption emission detection analysis method provided by the application can obtain the first macroscopic traffic data in the traffic data stream.

[0021] The intelligent network connection road traffic system energy consumption emission detection analysis method provided by the application can obtain the first macroscopic traffic data in the traffic data stream.

[0022] The intelligent network connection road traffic system energy consumption emission detection analysis method provided by the application can obtain the first macroscopic traffic data in the traffic data stream.

[0023] The high-dimensional parameter space of the first micro-traffic data is reduced to a low-dimensional parameter space by using principal component analysis, and the reduced data is clustered into different families by using an aggregation method, and the center speed of each family is taken as the average speed of the entire family.

[0024] The traffic average speed, traffic flow and traffic density are calculated according to the average speed of each family.

[0025] The application further provides an electronic device, including a memory, a processor and a computer program stored in the memory and executable on the processor, and the processor implements the intelligent networked road traffic system energy consumption emission detection and analysis method according to any one of the above when executing the program.

[0026] The application further provides a non-transitory computer-readable storage medium having a computer program stored thereon, and the computer program is executable by a processor to implement the intelligent networked road traffic system energy consumption emission detection and analysis method according to any one of the above.

[0027] The application further provides a computer program product, including a computer program, and the computer program is executable by a processor to implement the intelligent networked road traffic system energy consumption emission detection and analysis method according to any one of the above.

[0028] The intelligent networked road traffic system energy consumption emission detection and analysis method provided by the application first acquires first macro-traffic data and first micro-traffic data in a traffic data stream, then inputs the first macro-traffic data and the first micro-traffic data into a macro-energy consumption emission model and a micro-energy consumption emission model respectively to obtain a macro-energy consumption emission benchmark and a micro-energy consumption emission benchmark respectively, decomposes the first macro-traffic data to obtain second micro-traffic data, and aggregates the first micro-traffic data to obtain second macro-traffic data, inputs the second macro-traffic data and the second micro-traffic data into the macro-energy consumption emission model and the micro-energy consumption emission model respectively to obtain first fusion data and second fusion data respectively, and finally obtains a traffic energy consumption emission detection result according to the macro-energy consumption emission benchmark, the micro-energy consumption emission benchmark, the first fusion data and the second fusion data. The detection result obtained by using the detection method of the application comprehensively considers various data in traffic, couples two energy consumption emission models, and obtains an accurate detection result according to the coupling result. BRIEF DESCRIPTION OF DRAWINGS

[0029] In order to more clearly illustrate the technical solutions of the present application or the prior art, the following will briefly introduce the drawings needed in the embodiments or prior art description. Obviously, the drawings in the following description are some embodiments of the present application, and those skilled in the art can also obtain other drawings according to these drawings without creative labor.

[0030] Figure 1 Figure 1 is a schematic diagram of a method for detecting and analyzing energy consumption and emission of an intelligent networked road traffic system according to an embodiment of the present application.

[0031] Figure 2 Figure 2 is a schematic diagram of a method for detecting and analyzing energy consumption and emission of an intelligent networked road traffic system according to another embodiment of the present application.

[0032] Figure 3 Figure 3 is a schematic diagram of a three-dimensional shock wave theory according to an embodiment of the present application.

[0033] Figure 4 Figure 4 is a schematic diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0034] In order to make the objectives, technical solutions and advantages of the present application clearer, the technical solutions of the present application will be described below in detail with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments of the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the scope of the present application.

[0035] Energy consumption and greenhouse gas emission caused by road traffic is one of the problems that are widely concerned in the world at present. In order to effectively quantify the energy consumption and emission of regional traffic, macro and micro traffic flow models are generally coupled with energy consumption and emission models for calculation. However, due to the difficulty in obtaining accurate traffic data and the existence of parameter approximation error when different level models are coupled, the estimated results of energy emission are not objective and accurate enough. Therefore, in view of the energy emission estimation error problem of coupling of different granularity traffic flow models and energy consumption and emission models, the present application provides a method for detecting and analyzing energy consumption and emission of an intelligent networked road traffic system coupled with macro and micro models.

[0036] In order to solve the problems of the prior art, the present application aims at the error of energy consumption and emission estimation caused by coupling different macro and micro traffic flow models and energy consumption and emission models, and in order to solve the inconsistency between the data of the traffic flow model and the input data of the energy consumption and emission model, a process of converting macro and micro traffic data is established, and by inputting traffic data of different granularities into corresponding energy consumption and emission models, macro and micro energy consumption and emission benchmarks are obtained respectively by using micro and macro traffic data.

[0037] The intelligent connected road traffic system energy consumption and emission detection analysis method of the present application will be described below. Figures 1-2 The intelligent connected road traffic system energy consumption and emission detection analysis method of the present application will be described below.

[0038] Figure 1 The intelligent connected road traffic system energy consumption and emission detection analysis method of the present application will be described below. Figure 2 The intelligent connected road traffic system energy consumption and emission detection analysis method of the present application will be described below. Figure 1 and Figure 2 The intelligent connected road traffic system energy consumption and emission detection analysis method of the present application will be described below.

[0039] S101, acquiring first macro traffic data and first micro traffic data in a traffic data stream.

[0040] In one or more embodiments, traffic data can be basically divided into two categories: macro traffic data based on fixed sensors and micro traffic data based on mobile sensors. The first macro traffic data collected by fixed sensors (such as double-ring detectors, microwaves, radars, etc.) refers to the aggregated or decomposed data collected by fixed sensors within a certain time period, and the first macro traffic data includes average speed, flow density and traffic volume, etc.

[0041] However, the detection range of fixed sensors is limited, the deployment rate is low, and the missing detection rate is high, so the entire space-time vehicle information cannot be obtained. Microscopic traffic data based on mobile sensors refers to the complete vehicle trajectory provided by probe vehicles and intelligent connected vehicles equipped with GPS and other mobile sensors. This method of collecting first microscopic traffic data overcomes the bottleneck of limited detection range in space and time dimensions of fixed sensors, but the penetration rate of probe vehicles and intelligent connected vehicles is low, so the collected trajectory data is usually sparse and noisy, and the sampling and data processing cost is high, increasing the difficulty of data collection.

[0042] S102, input the first macroscopic traffic data and the first microscopic traffic data into a macroscopic energy consumption and emission model and a microscopic energy consumption and emission model respectively to obtain a macroscopic energy consumption and emission benchmark and a microscopic energy consumption and emission benchmark respectively.

[0043] According to the research scope and precision of energy consumption emission model, it can be divided into macro and micro energy consumption emission model. Macro energy consumption emission model studies the energy consumption and emission of motor vehicles under different road types, road sections and traffic flow conditions. This model is usually based on actual road driving data and motor vehicle performance parameters, combined with traffic flow model and emission model, etc. to simulate and analyze the road traffic conditions. Common macro energy consumption emission models mainly include Mobile Source Emission Factor Model (MOBILE), Emissions Factors (EMFAC), International Vehicle Emissions Model (IVE), Computer Programme to Calculate Emissions from Road Transport (COPERT), etc. Micro energy consumption emission model mainly studies the energy consumption and emission of a single vehicle in the specific driving process. This model is based on the dynamics model and fuel consumption model of motor vehicles, combined with driving behavior and vehicle performance parameters, etc. to simulate and analyze the vehicle driving process. The main micro energy consumption emission models currently include Comprehensive Modal Emissions Model (CMEM), Motor Vehicle Emission Simulator (MOVES), VT-Micro (Virginal Tech Microscopic), etc. Among them, MOVES model is more perfect in structure compared with other models, has strong portability, has the ability to simulate macro, micro and different levels between macro and micro emission, and considers more detailed uncertain influencing factors, and the calculation result is closer to the actual average hourly emission of urban road. The essence of the calculation of this model is to use the method of VSP to represent the driving characteristics of motor vehicles except speed, which is defined as formula (1).

[0044]

[0045] wherein v is the speed (m / s), a is the acceleration (m / s 2 ), m is the weight of the vehicle (t), and A, B, and C are all vehicle road load coefficients.

[0046] In this embodiment, the first macro traffic data and the first micro traffic data are first input into the corresponding macro energy consumption emission model and micro energy consumption emission model respectively to obtain the macro energy consumption emission benchmark and the micro energy consumption emission benchmark respectively.

[0047] S103, decompose the first macroscopic traffic data to obtain second microscopic traffic data, and aggregate the first microscopic traffic data to obtain second macroscopic traffic data.

[0048] In one or more embodiments, decomposing the first macroscopic traffic data to obtain the second microscopic traffic data comprises:

[0049] The first macroscopic traffic data is decomposed by using a microscopic vehicle trajectory reconstruction method to obtain the second microscopic traffic data.

[0050] In one or more embodiments, the microscopic vehicle trajectory reconstruction method comprises one or more of a spatiotemporal information interpolation method, a filtering method, and a multi-source data fusion method.

[0051] The spatiotemporal information interpolation method mainly relies on the spatiotemporal autocorrelation characteristics of fixed detector data. First, the weight is calculated according to the distance between data points, and then the weighted average of each data point measurement value is used to fill in each point in the spatiotemporal region. Instantaneous model, dynamic time slice model and linear model can estimate the spatiotemporal velocity distribution between fixed detectors within the observation range by using the velocity recursive method. These models use the average method in data collection and trajectory reconstruction, which has low accuracy in congestion conditions, and some higher and lower values are discarded, resulting in reconstructed trajectories that cannot reflect the different behaviors between vehicles. First, the road speed is predicted based on fixed point speed measurements, and then the time when the vehicle enters the network is given to reconstruct its driving trajectory on the road. The prior knowledge of real-time traffic information is considered to add random speed to the speed obtained by using the four-corner smoothing method to generate a comprehensive speed trajectory. The density of each point in the spatiotemporal region is interpolated according to the density of the four corners of the road spatiotemporal region, and then the trajectory of each vehicle is reconstructed by using the triangular basic graph. However, since the density is collected in an aggregated manner, if the densities of the four corners are the same, the density of each point in the spatiotemporal region will be the same, which means that the vehicle will travel at a constant speed. Using traffic data from a single loop detector, a method for estimating vehicle trajectories on highways is proposed, which uses traffic flow theory to infer local traffic conditions. However, this method is not suitable for the transition period when the traffic condition changes from non-congestion to congestion, limiting the application range of the method.

[0052] Among them, the filtering method is a space-time traffic state estimation method that can consider the propagation of traffic congestion, which uses a nonlinear low-pass filter to estimate the traffic state between fixed detectors. The prior art proposes an adaptive smoothing method that uses a nonlinear space-time low-pass filter for fixed detector data, so that in congested traffic, disturbances move upstream at a nearly constant speed, while in free traffic, they propagate downstream. By improving the adaptive smoothing method, it can fuse data from various traffic detectors to obtain reliable traffic state estimates. Based on the random macroscopic traffic flow model and the extended Kalman filter, a general adaptive method for highway section or road network traffic state is proposed, so as to reconstruct the vehicle trajectory on the highway. A trajectory estimation algorithm based on the extended car-following model is established to estimate the undetected part of each trajectory, and on this basis, a trajectory fusion algorithm based on the particle filter is proposed to minimize the error of the estimated trajectory. The prior art also proposes a hybrid method to reconstruct the complete vehicle trajectory at signalized intersections, which is based on the variational network and Kalman filter, and reproduces the stochastic characteristics of the reconstructed queue boundary curve. The missing trajectory between probe vehicles is reconstructed using a particle filter-based method. Using the particle filter method, five space-time trajectory correction factors are considered, and based on the data collected by automatic vehicle identification and traditional detectors, the vehicle trajectory in a large-scale network is reconstructed. Through the particle filter, different data sources are integrated to estimate the vehicle trajectory on the main road of the signalized intersection.

[0053] Among them, multi-source data fusion aims to combine the data of fixed detectors and mobile detectors for processing and analysis, making up for the low accuracy of fixed detector data and the sparsity of mobile sensor data, so as to obtain more comprehensive and reliable traffic data. Based on the three-dimensional shock wave theory and under the guidance of variational theory, a new data fusion framework is proposed to reconstruct the vehicle trajectory on urban main roads. Also based on variational theory, the floating car and signal timing data are used as the fusion object to reconstruct the vehicle trajectory. However, the model based on variational theory ignores the stochastic wave characteristics of speed. The prior art proposes a two-stage method to reconstruct multi-vehicle trajectories, first calculating the speed of each cell through filtering smoothing, and then using bicubic interpolation to calculate the displacement at each time step. By utilizing the mixed characteristics of fixed position detection data and space-time single mobile sensor data, a macro-micro integrated framework for vehicle trajectory reconstruction on highways is developed, which provides a critical speed baseline for trajectory reconstruction through data fusion, car-following and inverse car-following models are used to generate candidate trajectories, and the optimal trajectory with the minimum speed baseline difference is determined according to dynamic programming.

[0054] Specifically, in terms of trajectory reconstruction, due to the low deployment rate and long aggregation time of fixed detectors on highways, most studies use the aggregated values of microscopic vehicle trajectories in a short time (usually 30s) as virtual detector data for trajectory reconstruction, rather than using real fixed detector data for trajectory reconstruction; the trajectory reconstruction methods of most studies are often suitable for single vehicle trajectory reconstruction or trajectory reconstruction under free flow conditions, and cannot reflect the stop-and-go behavior of vehicles under congested flow conditions; some trajectory reconstruction methods are based on the estimation of travel time, without considering the kinematic characteristics of vehicles and the authenticity of trajectories, resulting in abnormal vehicle speed and acceleration, which cannot be applied to energy consumption and emission calculation. Therefore, it is very meaningful to reconstruct vehicle trajectories under various traffic conditions using real fixed detector data as input for energy consumption and emission models.

[0055] In one or more embodiments, the microscopic vehicle trajectory reconstruction method comprises:

[0056] The road space-time region is divided into a cell grid, the traffic speed evolution of the road space-time network is estimated by fusing fixed sensor data and mobile sensor data using a non-parametric kernel smoothing method, and a space-time variation network is established according to the cell speed. Under the constraint of sparse probe vehicle trajectories, the cumulative number of vehicles in each space-time variation network is calculated based on the three-dimensional shock wave theory, and the complete travel trajectory of the vehicle is reconstructed by the shortest path method. The specific content is as follows:

[0057] The non-parametric kernel smoothing method first needs to estimate the macroscopic traffic flow fundamental diagram. The macroscopic traffic flow fundamental diagram is a model that describes the relationship between road traffic density and traffic flow, and can intuitively represent the characteristics of road traffic flow. The most common traffic flow macroscopic fundamental diagram is a triangular fundamental diagram with three parameters: a fixed free-flow speed (v free ), a maximum flow or capacity (q max ), and a jam density (k jam ). This fundamental diagram uses a steady flow-density relationship to describe the state of traffic flow, and the traffic density k is in a constant linear relationship with the corresponding flow q and speed v. The key parameters v free , q max , k jam , etc. in the fundamental diagram, and their calculation formulas are as follows:

[0058]

[0059]

[0060] k jam =max(max(K i,j ),150)⑷

[0061] wherein, Ki,j These represent the speed, traffic flow, and density counted by coil detector j at time i, respectively.

[0062] Typically, a spatiotemporal velocity map virtually divides a road spatiotemporal network into different cells, and then estimates the traffic velocity of all cells using various methods. Compared to traditional rectangular parallelogram cells, using parallelogram cells to segment the spatiotemporal network further considers backwaves, which can reduce errors in estimating microscopic vehicle trajectories and travel times under traffic congestion conditions. Using the road segment congestion wave velocity estimated from the macroscopic traffic flow fundamental map and custom time and spatial intervals to segment the spatiotemporal network, it can be divided into multiple parallelogram cells, with the velocity at the cell center used as the average velocity of the entire cell. When estimating cell velocities using multi-source data fusion, two auxiliary velocity surfaces are established to capture traffic interference: one for free-flow conditions and one for congested conditions. For each data source, the two auxiliary velocity surfaces are estimated as follows:

[0063]

[0064]

[0065] Among them, V s Let S be the velocity technical value of data point s, where S is the velocity count set within the range A(t,x), t is time (seconds), and x is position (meters). and It is given by the following formula:

[0066]

[0067]

[0068] Where S represents the reference point, t S x represents the time corresponding to the reference point. S c represents the position corresponding to the reference point. free c represents an auxiliary velocity surface estimate for a cell. con This represents another auxiliary velocity surface estimate for a cell. It is a kernel function that counts V for each velocity. s Assign a weight, defined as follows:

[0069]

[0070] Where τ and σ are the temporal smoothing width and spatial smoothing width, respectively.

[0071] After the non-parametric kernel smoothing method is used to complete the road cell speed estimation, the road speed in a local range can be obtained. However, due to the influence of the surrounding traffic state on the vehicle during the driving process, the vehicle speed changes in a large range, so the driving state of the vehicle also switches back and forth between smooth driving and stop-and-go. The trajectory data of the probe vehicle and the CAV equipped with GPS can provide the position, speed and acceleration of the vehicle during the driving process. Therefore, in order to more accurately capture the driving state change caused by the change of the vehicle speed, the trajectory information of the probe vehicle can be used as a reference basis for reconstructing the trajectory.

[0072] With reference to the trajectory information of the sparse probe vehicle, the cumulative vehicle number concept of the three-dimensional shock wave theory is used to reconstruct the trajectory of other vehicles in the space-time network, Figure 3 The three-dimensional shock wave theory diagram provided by the embodiment of the application is shown as Figure 3 , Figure 3 The trajectory of the reference probe vehicle, the forward wave and the backward wave are marked in the diagram respectively, and the black dots on the time axis are the entry and exit time of the vehicle.

[0073] On the basis of the three-dimensional shock wave theory, the space-time network is divided according to the wave speed of the forward and backward waves. Since the forward arc slope and the backward arc slope of the space-time network are deterministic, the dynamic change of the traffic state is not fully considered. Therefore, based on the estimated evolution of the road space-time speed, the free flow speed of the small grid point is set to the average speed of the cell, and the basic parameters of the initial variational network are determined, including the slope of the backward wave, the time, the space step:

[0074]

[0075] t step =1s⑾

[0076]

[0077] Among them, w is the backward arc slope, u is the forward arc slope determined by the average speed of the cell where the point (t, x) is located, t step is the predetermined time step, and s step is the space step.

[0078] After the network division is completed, the cumulative vehicle number of the first column node in the variational network is set to 1. Considering that the probe vehicle trajectory and the space-time network are difficult to completely match, the probe vehicle trajectory needs to be gridded, and the cumulative vehicle number of the grid point covered by the probe vehicle trajectory is determined according to the order in which the probe vehicle enters the space-time network, as a constraint for calculating the cumulative vehicle number of other unknown nodes in the grid. For the cumulative vehicle number of each grid point on the upper and lower boundaries of the initial variational network, it is necessary to determine when the vehicle enters or exits the network according to other external conditions. Under the constraints of the probe vehicle trajectory and the space-time network boundary, the cumulative vehicle number of each unknown grid point in the space-time variational network is calculated. The nodes with the same cumulative vehicle number in the variational network indicate the area covered by the same vehicle trajectory, and the nodes need to be connected as the vehicle trajectory. Since the traffic flow cannot be reversed in space or time, it can only be searched from the node with a smaller cumulative vehicle number to the node with a larger or equal cumulative vehicle number. For a loop-free variational network, the shortest path algorithm is used for searching.

[0079] In one or more embodiments, the first micro-traffic data is aggregated to obtain the second macro-traffic data, including:

[0080] The first micro-traffic data is aggregated to obtain the second macro-traffic data using a traffic state estimation method.

[0081] In the aspect of traffic state estimation, since it is difficult to obtain full-sampling vehicle trajectory data, most of the current research still uses fixed detector data and a small amount of probe vehicle data, and there is little research based on full-sampling multi-vehicle data; most traffic state estimations strictly depend on traffic flow models, which limits the applicability of the algorithm, and the traffic flow model is often abstract and difficult to adapt to the complex situation of actual traffic; some researches calculate the average traffic speed by mean interpolation from the observed speed at the detector, without considering the phenomenon of vehicle stop-start caused by road congestion, which makes the vehicle speed change range large. Therefore, it is necessary to estimate the traffic speed from the perspective of the large vehicle speed change range.

[0082] Generally, the second macro-traffic data includes one or more of traffic average speed, traffic flow, and traffic density.

[0083] In one or more embodiments, the first micro-traffic data is aggregated to obtain the second macro-traffic data using a traffic state estimation method, including:

[0084] The high-dimensional parameter space of the first micro-traffic data is reduced to a low-dimensional parameter space using principal component analysis, and then the reduced data is clustered into different families using an aggregation method, and the center speed of each family is taken as the average speed of the entire family;

[0085] The traffic average speed, traffic flow and traffic density are calculated according to the average speed of each family.

[0086] Traffic state estimation refers to the process of inferring macroscopic traffic state variables such as flow, density, and speed on a road segment using partially observed traffic data. Through these variables, traffic planners can perceive congestion levels, identify traffic demand, and achieve traffic control, management, and optimization. Generally, estimation methods can be classified according to input data and physical or statistical assumptions. In terms of physical or statistical assumptions, traffic state estimation can be roughly divided into model-driven and data-driven methods.

[0087] Among them, the model-driven traffic state estimation method is mainly based on traffic flow theory, control theory and mathematical modeling method, establishes traffic flow model, and estimates the traffic state of the whole road network. This method mainly combines traffic flow basic graph and partial differential equation with other technologies (such as Kalman filter) to estimate traffic state through Lighthill-Whitham-Richards (LWR) model and its extended model, Green-Shields model, Aw-RascleZhang (ARZ) model and its extended model, etc. macroscopic traffic flow model. A framework based on the LWR traffic flow model is proposed in the prior art to estimate traffic state from Euler (loop) and Lagrange (probe car) data. Based on the METANET model, the Kalman filter principle is combined, and a data assimilation method is proposed in the prior art, which can estimate the density, speed, critical density, etc. of mixed traffic flow from intelligent networked vehicle data. A traffic state estimation method based on PVSME data is proposed in the prior art, including a dynamic traffic flow model, an observation model and an extended Kalman filter.

[0088] The advantage of model-driven methods is that they can reflect basic traffic principles and provide insights for interpreting the estimation process. However, model-based methods are highly dependent on various theoretical assumptions of traffic physics, which may be too abstract and biased to model noisy and random real-world data. Therefore, model-based methods often require reliable prior information to guide estimation.

[0089] Among them, the data-driven traffic state estimation method uses historical traffic data to establish a prediction model through data analysis and mining to estimate future traffic state. This method can be divided into two categories: statistical methods and machine learning methods.

[0090] Among them, the statistical method uses the average value, trend and other statistical characteristics of historical traffic data to predict the future traffic state, mainly including moving average, autoregressive moving average, multivariate linear regression prediction and other data analysis methods. Although this method has small calculation amount and high efficiency, it belongs to linear analysis and cannot accurately reflect the nonlinear characteristics of the real traffic system, so it cannot obtain satisfactory prediction results.

[0091] The machine learning method mainly uses support vector machine, convolutional neural network and non-parametric regression model. These methods use machine learning algorithms to train historical traffic data to establish a prediction model, which can better handle complex traffic flow states, but require a large amount of data for training and have high computational complexity. The prior art proposes an offline and online Bayesian traffic state estimation method to estimate traffic density, which assumes that there is a CAV (Connected Autonomous Vehicle) measurement value, uses the Bayesian paradigm to express the prior information, and thus deduces the probability distribution of the traffic density of different road segments in the traffic network, and verifies the effectiveness of the algorithm in the SUMO simulation scenario and the HighD real data set. The prior art also proposes a purely data-driven and traffic flow model-free traffic state estimation method, which estimates the traffic state as a space-time matrix difference problem, applies tensor structure and low-rank assumption to estimate the space-time average traffic speed from sparse probe vehicle trajectories. The prior art also proposes a method for real-time estimation of road traffic density, which is completely based on V2V communication and estimates in real time according to the data of vehicles moving in the same direction and future traffic. Since this method only needs to record the number of times a vehicle interacts with other vehicles, it allows vehicles to change their driving speed in any way. The prior art also proposes a traffic state estimation method that uses oncoming vehicles and stationary observers to calculate the traffic flow of a saturated intersection controlled by traffic signals. The prior art also proposes a method for establishing a Bayesian model based on historical traffic and event data, which models the probability dependency structure of the congestion cause of a specific road segment and outputs variables representing traffic performance, such as traffic flow, speed and density. The prior art also proposes a deep convolutional neural network based on motion waves to estimate high-resolution traffic speed dynamics from sparse probe vehicle trajectories. By introducing the motion wave theory, the propagation characteristics of the forward wave and the backward wave are explicitly considered, and physical constraints are imposed on the learning model of the neural network, which improves the robustness of existing deep learning-based estimation methods.

[0092] Based on the above analysis of the traffic state estimation method, it can be seen that most of the existing traffic state estimation methods are based on fixed detector data and probe vehicle data with partial penetration rate, and either need to rely on well-defined traffic flow models or need a large amount of driving data for statistical analysis or learning training. Few studies are based on full-sampling vehicle trajectory data for macroscopic traffic state estimation.

[0093] S104, input the second macroscopic traffic data and the second microscopic traffic data into a macroscopic energy consumption and emission model and a microscopic energy consumption and emission model respectively to obtain first fusion data and second fusion data respectively.

[0094] In terms of energy consumption and emission estimation, most of the existing researches input traffic data into the same level energy consumption and emission model to calculate energy consumption and emission, and since the data of traffic flow model is difficult to be consistent with the input data of energy consumption and emission model, there are few researches on coupling traffic flow model and energy consumption and emission model of different levels, and most of the previous researches focus on using one traffic flow model and one energy consumption and emission model to estimate the energy consumption and emission of a road network or a region, and there are few researches on comparative analysis of the calculation results of the same road network or region by different combinations.

[0095] S105, obtaining a detection result of traffic energy consumption and emission according to the macroscopic energy consumption and emission benchmark, the microscopic energy consumption and emission benchmark, the first fusion data and the second fusion data.

[0096] Since the microscopic traffic data has higher vehicle driving detail representation, and the microscopic energy consumption and emission model considers more factors affecting vehicle fuel consumption and emission, the coupling result of the microscopic trajectory data and the microscopic energy consumption and emission model has relatively high accuracy in the experiment, so the coupling result is taken as the benchmark data of the experiment to represent the real fuel consumption and emission result of the research area.

[0097] Specifically, the detection result can be an emission result, which can be fuel consumption and pollutant emission (CO2, CO, NOx) under each combination of traffic model and energy consumption and emission model.

[0098] Figure 4 An example of an entity structure diagram of an electronic device is shown in FIG. 1. Figure 4As shown, the electronic device can include a processor 310, a communications interface 320, a memory 330, and a communications bus 340, wherein the processor 310, the communications interface 320, and the memory 330 complete mutual communication through the communications bus 340. The processor 310 can call the logical instructions in the memory 330 to execute the intelligent networked road traffic system energy consumption emission detection and analysis method, which includes: acquiring first macroscopic traffic data and first microscopic traffic data in a traffic data stream; inputting the first macroscopic traffic data and the first microscopic traffic data into a macroscopic energy consumption emission model and a microscopic energy consumption emission model respectively to obtain a macroscopic energy consumption emission benchmark and a microscopic energy consumption emission benchmark respectively; decomposing the first macroscopic traffic data to obtain second microscopic traffic data, and aggregating the first microscopic traffic data to obtain second macroscopic traffic data;

[0099] Inputting the second macroscopic traffic data and the second microscopic traffic data into the macroscopic energy consumption emission model and the microscopic energy consumption emission model respectively to obtain first fusion data and second fusion data respectively; obtaining a traffic energy consumption emission detection result according to the macroscopic energy consumption emission benchmark, the microscopic energy consumption emission benchmark, the first fusion data, and the second fusion data.

[0100] In addition, the logical instructions in the memory 330 described above can be implemented in the form of a software functional unit and sold or used as an independent product, and can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application essentially or the part that contributes to the prior art or part of the technical solutions can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a plurality of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the method described in various embodiments of the present application. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various program code storage media.

[0101] In another aspect, the present application also provides a computer program product, which comprises a computer program, the computer program being stored in a non-transitory computer-readable storage medium, and the computer program being executable by a processor to enable a computer to perform the intelligent connected road traffic system energy consumption emission detection and analysis method provided by the above method, the method comprising: obtaining first macroscopic traffic data and first microscopic traffic data in a traffic data stream; inputting the first macroscopic traffic data and the first microscopic traffic data into a macroscopic energy consumption emission model and a microscopic energy consumption emission model respectively to obtain a macroscopic energy consumption emission benchmark and a microscopic energy consumption emission benchmark respectively; decomposing the first macroscopic traffic data to obtain second microscopic traffic data, and aggregating the first microscopic traffic data to obtain second macroscopic traffic data; inputting the second macroscopic traffic data and the second microscopic traffic data into the macroscopic energy consumption emission model and the microscopic energy consumption emission model respectively to obtain first fusion data and second fusion data respectively; and obtaining a detection result of traffic energy consumption emission according to the macroscopic energy consumption emission benchmark, the microscopic energy consumption emission benchmark, the first fusion data and the second fusion data.

[0102] In another aspect, the present application also provides a non-transitory computer-readable storage medium, which stores a computer program, and the computer program is executable by a processor to implement the intelligent connected road traffic system energy consumption emission detection and analysis method provided by the above method, the method comprising: obtaining first macroscopic traffic data and first microscopic traffic data in a traffic data stream; inputting the first macroscopic traffic data and the first microscopic traffic data into a macroscopic energy consumption emission model and a microscopic energy consumption emission model respectively to obtain a macroscopic energy consumption emission benchmark and a microscopic energy consumption emission benchmark respectively; decomposing the first macroscopic traffic data to obtain second microscopic traffic data, and aggregating the first microscopic traffic data to obtain second macroscopic traffic data; inputting the second macroscopic traffic data and the second microscopic traffic data into the macroscopic energy consumption emission model and the microscopic energy consumption emission model respectively to obtain first fusion data and second fusion data respectively; and obtaining a detection result of traffic energy consumption emission according to the macroscopic energy consumption emission benchmark, the microscopic energy consumption emission benchmark, the first fusion data and the second fusion data.

[0103] The device embodiments described above are merely illustrative, wherein the units described as separate components can or can not be physically separate, and the components displayed as units can or can not be physical units, i.e., they can be located in one place, or distributed on multiple network units. Part or all of the modules can be selected to achieve the purpose of the present embodiment scheme according to actual needs. Those skilled in the art can understand and implement it without creative labor.

[0104] Those skilled in the art can clearly understand the technical solutions of the various embodiments from the above description of the embodiments, and the various embodiments can be implemented by means of software with the necessary general hardware platforms, and of course, can also be implemented by hardware. Based on such understanding, the above technical solutions, essentially or in other words, the part of the prior art that makes a contribution, can be embodied in the form of a software product, which can be stored in a computer readable storage medium, such as a ROM / RAM, a magnetic disk, an optical disk, and the like, and includes a number of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0105] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, rather than limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for some technical features therein; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for detecting and analyzing energy consumption and emissions of intelligent connected road transportation system, characterized in that, The method comprises the following steps: acquiring first macroscopic traffic data and first microscopic traffic data in a traffic data stream; inputting the first macroscopic traffic data and the first microscopic traffic data into a macroscopic energy consumption and emission model and a microscopic energy consumption and emission model respectively to obtain a macroscopic energy consumption and emission benchmark and a microscopic energy consumption and emission benchmark respectively; decomposing the first macroscopic traffic data to obtain second microscopic traffic data, and aggregating the first microscopic traffic data to obtain second macroscopic traffic data; inputting the second macroscopic traffic data and the second microscopic traffic data into the macroscopic energy consumption and emission model and the microscopic energy consumption and emission model respectively to obtain first fusion data and second fusion data respectively; obtaining a detection result of traffic energy consumption and emission according to the macroscopic energy consumption and emission benchmark, the microscopic energy consumption and emission benchmark, the first fusion data and the second fusion data. 2.The intelligent networked roadway transportation system energy consumption emission detection and analysis method of claim 1, wherein, The first macroscopic traffic data in the traffic data stream is data in a traffic flow collected by a fixed sensor arranged in space; The second microscopic traffic data is data in a traffic flow collected by a mobile sensor arranged on a vehicle. 3.The intelligent networked roadway transportation system energy consumption emission detection and analysis method of claim 1, wherein, The decomposition of the first macroscopic traffic data to obtain the second microscopic traffic data comprises: decomposing the first macroscopic traffic data by using a microscopic vehicle trajectory reconstruction method to obtain the second microscopic traffic data. 4.The intelligent networked roadway transportation system energy consumption emission detection and analysis method of claim 3, wherein, The microscopic vehicle trajectory reconstruction method comprises one or more of a spatiotemporal information interpolation method, a filtering method and a multi-source data fusion method. 5.The intelligent networked roadway transportation system energy consumption emission detection and analysis method of claim 3, wherein, The microscopic vehicle trajectory reconstruction method comprises: performing cell grid decomposition on a road spatiotemporal region, fusing fixed sensor data and mobile sensor data, estimating traffic speed evolution of a road spatiotemporal network by using a non-parametric kernel smoothing method, and establishing a spatiotemporal variation network according to cell speed; under the constraint of sparse probe vehicle trajectory, calculating cumulative vehicle number of each spatiotemporal variation network based on three-dimensional shock wave theory, and reconstructing complete driving trajectory of a vehicle by using a shortest path method. 6.The intelligent networked roadway transportation system energy consumption emission detection and analysis method of claim 1, wherein, The aggregation of the first microscopic traffic data to obtain the second macroscopic traffic data comprises: aggregating the first microscopic traffic data by using a traffic state estimation method to obtain the second macroscopic traffic data. 7.The intelligent networked roadway transportation system energy consumption emission detection and analysis method of claim 6, wherein, The second macroscopic traffic data comprises traffic average speed, traffic flow and traffic density. The aggregation of the first microscopic traffic data by using the traffic state estimation method to obtain the second macroscopic traffic data comprises: reducing dimensionality of a high-dimensional parameter space of the first microscopic traffic data to a low-dimensional parameter space by using a principal component analysis method, clustering the data after dimensionality reduction into different families by using an aggregation method, and taking center speed of each family as average speed of the entire family; calculating the traffic average speed, the traffic flow and the traffic density according to the average speed of each family.

8. An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The processor implements the intelligent connected road traffic system energy consumption and emission detection and analysis method according to any one of claims 1 to 7 when executing the program. 9.A non-transitory computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program implements the intelligent connected road traffic system energy consumption and emission detection and analysis method according to any one of claims 1 to 7 when executed by the processor.

Images