CN118015727A

CN118015727A - Vehicle fuel consumption prediction method and system based on vehicle-mounted CAN data

Info

Publication number: CN118015727A
Application number: CN202410318592.5A
Authority: CN
Inventors: 张周平; 樊华春; 王宁; 许钧奇; 聂辽栋; 尚凯; 徐佳祥; 廖程亮; 林小刚
Original assignee: Nanchang Intelligent New Energy Vehicle Research Institute
Current assignee: Nanchang Intelligent New Energy Vehicle Research Institute
Priority date: 2024-03-20
Filing date: 2024-03-20
Publication date: 2024-05-10

Abstract

The invention provides a vehicle oil consumption prediction method and a vehicle oil consumption prediction system based on vehicle CAN data, wherein the method comprises the steps of extracting characteristics of original CAN parameters; performing cluster analysis processing on characteristic variables in the driver operation and driving risk characteristic set; determining running parameters and road parameters of a vehicle and determining congestion level data of a running road section; constructing a preset fuel consumption prediction model, and dividing a processing data set into a training set and a testing set; the invention can improve the counting accuracy in complex scenes, can process counting tasks in open-set environment, and avoids predefining target categories.

Description

Vehicle fuel consumption prediction method and system based on vehicle-mounted CAN data

Technical Field

The invention belongs to the technical field of fuel consumption prediction, and particularly relates to a vehicle fuel consumption prediction method and system based on vehicle-mounted CAN data.

Background

With the rapid development of society and the continuous increase of traffic demand, the traffic of highway goods has a trend of rising year by year, which makes accurate prediction of vehicle fuel consumption particularly important. The fuel consumption of a highway freight truck is not only related to the characteristics of the vehicle itself, the driving situation, the type of cargo, the load capacity, etc., but is also affected by the actual driving environment and external conditions. In particular, in the case of traffic congestion, fuel consumption and emission are significantly increased due to frequent acceleration and deceleration and reduced average speed, and in the past, although many technologies have attempted to directly measure and predict fuel consumption using various sensors and specialized equipment, they are often limited due to high cost, implementation complexity and real-time problems, and fuel consumption estimation methods relying solely on vehicle dynamics models and neural networks are also difficult to avoid errors due to data noise, characteristics of nonlinear systems and large computational resource requirements, and prediction bias of conventional methods is also obvious in the face of complex driving environments and road conditions.

For example, chinese patent application number 202010918106.5, a method and apparatus for evaluating energy consumption of heavy diesel vehicles, determines driving mileage using ignition switch signals and engine state signals of the vehicle, and collects driving data based on a CAN bus. Features highly correlated with energy consumption are screened out by refining and extracting key features and using pearson correlation coefficients. Energy consumption assessment is then performed in combination with these features using a pre-trained neural network-based model. The defects are as follows: the pearson correlation coefficients have limitations that may ignore certain non-linear or more complex characteristic relationships. A large amount of data is required to train the neural network and there may be evaluation bias for certain specific driving scenarios or vehicle models that are not involved.

The China patent application 'method and system for estimating the fuel consumption of the transient state automobile based on the correction of the related parameters of the fuel consumption intensity' with the application number 202210310009.7 provides a method for estimating the fuel consumption based on the steady state and unsteady state running data of the vehicle. The method combines various data analysis technologies, such as principal component analysis, a cluster analysis method and polynomial fitting, and establishes a vehicle fuel consumption model. The defects are as follows: the influence of traffic flow state is not considered, for example, in the case of busy urban roads or traffic jams, the frequent start and stop of the vehicle may lead to increased fuel consumption, and these unaccounted actual driving scenario factors may lead to deviation of fuel consumption estimation in some cases.

The patent with application number CN202010234278.0 provides a prediction scheme of fuel consumption of the fuel vehicle based on multi-working condition fuel vehicle prediction method and system of Gaussian process regression, which is trained by using a Gaussian process regression model, and a data set is optimized through a sequential sampling algorithm, and the method and the system are repeatedly executed until a preset stop condition is reached, and then a prediction result is output. The defects are as follows: repeated sequential sampling may extend computation time and gaussian process based models may require significant computational resources.

Disclosure of Invention

In order to solve the technical problems, the invention provides a vehicle fuel consumption prediction method and system based on vehicle-mounted CAN data, which are used for solving the technical problems in the prior art.

In a first aspect, the present invention provides the following technical solutions, and a vehicle fuel consumption prediction method based on vehicle-mounted CAN data, including:

Acquiring original CAN parameters of a vehicle in real time, and extracting features of the original CAN parameters to obtain first input features, wherein the original CAN parameters at least comprise vehicle speed, engine speed, accelerator position, oil consumption, brake pressure, running longitude and latitude data, running mileage and engine key parameters;

constructing a driver operation and driving risk feature set, and performing cluster analysis processing on feature variables in the driver operation and driving risk feature set by adopting a K-Means unsupervised cluster analysis algorithm to obtain a second input feature;

Determining a running parameter and a road parameter of a vehicle based on the running longitude and latitude data, and determining congestion level data of a running road section based on the running parameter and the road parameter;

a preset oil consumption prediction model is built, the first input feature, the second input feature and the congestion level data of the driving road section are stored in a processing data set, and the processing data set is divided into a training set and a testing set;

And inputting the training set into the preset fuel consumption prediction model to train so as to obtain a training fuel consumption prediction model, and inputting the testing set into the training fuel consumption prediction model so as to obtain a predicted fuel consumption result.

Compared with the prior art, the invention has the beneficial effects that: firstly, acquiring original CAN parameters of a vehicle in real time, performing feature extraction on the original CAN parameters to obtain a first input feature, then constructing a driver operation and driving risk feature set, and performing cluster analysis processing on feature variables in the driver operation and driving risk feature set by adopting a K-Means unsupervised cluster analysis algorithm to obtain a second input feature; then, determining running parameters and road parameters of the vehicle based on the running longitude and latitude data, and determining the congestion level data of the running road section based on the running parameters and the road parameters; then, a preset oil consumption prediction model is built, the first input feature, the second input feature and the congestion level data of the driving road section are stored in a processing data set, and the processing data set is divided into a training set and a testing set; finally, the training set is input into a preset fuel consumption prediction model for training to obtain a training fuel consumption prediction model, and the test set is input into the training fuel consumption prediction model to obtain a predicted fuel consumption result. The method solves the problems in data accuracy and training requirements, adopts the MI method to accurately screen key vehicle condition characteristics highly related to oil consumption, obviously reduces characteristic redundancy, avoids the problem of local characteristic limitation, utilizes real vehicle running data to construct a driver transportation behavior image as an independent characteristic, provides macroscopic and microscopic viewing angles for prediction, is beneficial to comprehensively considering actual driving environment factors, and simultaneously effectively calculates the road congestion index of a vehicle running section based on the ratio of free flow speed to CAN speed, so that the oil consumption prediction is more accurate.

Preferably, the step of obtaining the original CAN parameters of the vehicle in real time and extracting features of the original CAN parameters to obtain the first input features includes:

Acquiring original CAN parameters of a vehicle in real time, and sequentially carrying out smooth filtering, interpolation deficiency and abnormal data processing on the original CAN parameters to obtain processed CAN parameters;

calculating mutual information between the oil consumption variable and the processing CAN parameter ：

；

In the method, in the process of the invention,Parameter set composed for all processing CAN parameters,/>For parameter/>Sum parameter/>Is a joint probability distribution,/>And/>Parameters/>, respectivelySum parameter/>Is a boundary probability distribution of (1);

Based on mutual information Sorting the processing CAN parameters from large to small to obtain sorting CAN parameters, and selecting the first several parameters in the sorting CAN parameters as first input features.

Preferably, the step of constructing a driver operation and driving risk feature set, and performing cluster analysis processing on feature variables in the driver operation and driving risk feature set by adopting a K-Means unsupervised cluster analysis algorithm to obtain a second input feature comprises:

Constructing a driver operation and driving risk feature set, wherein the driver operation and driving risk feature set at least comprises operation days, total operation time, average operation time, total stay time, average stay time, total operation efficiency, total driving mileage, average driving mileage, total night driving time, night driving time proportion, night driving frequency and fatigue driving frequency;

Determining the number of data points in the driver operation and driving risk feature set and the number of clusters to be divided, selecting a group of clustered central points and minimizing the distance from each data point to the nearest central point of the cluster, and iterating continuously until the clustered center is not changed or the iteration stopping condition is met;

Determining the square sum of errors corresponding to different cluster numbers based on an elbow method, and drawing a K-SSE curve based on the square sum of errors corresponding to the different cluster numbers;

And determining an optimal K value which is rapidly slowed down in the K-SSE curve, and taking characteristic data corresponding to the optimal K value as a second input characteristic.

Preferably, the step of determining the driving parameter and the road parameter of the vehicle based on the driving longitude and latitude data includes:

Performing visualization processing on the vehicle running track on a map based on the running longitude and latitude data to obtain a running route map;

matching the geographic information system with a driving route map to determine driving parameters of vehicle driving, wherein the driving parameters at least comprise specific roads, road sections, driving time and driving distance of the vehicle driving;

And determining the road type information and the number of lanes based on the specific road on which the vehicle runs so as to obtain the road parameters of the vehicle.

Preferably, the step of determining the congestion level data of the driving road section based on the driving parameter and the road parameter includes:

determining a dynamic segment service level class table based on the driving parameters and the road parameters, and determining a road traffic free flow speed based on the dynamic segment service level class table;

Based on the road traffic free flow speed With vehicle CAN speed/>Determining road Congestion index/>：

；

Based on the road congestion indexAnd distributing the congestion level to each road section of the vehicle to obtain the congestion level data of the running road section.

Preferably, the step of inputting the training set into the preset fuel consumption prediction model for training to obtain a training fuel consumption prediction model includes:

Inputting the training set into the preset oil consumption prediction model, and training the preset oil consumption prediction model by using a random forest algorithm;

Performing parameter iterative optimization on the preset fuel consumption prediction model through grid search and cross verification until the average absolute error of model parameters is not greater than a performance threshold value so as to obtain an optimized fuel consumption prediction model;

Calculating a predictive score for the optimized fuel consumption predictive model ：

；

In the method, in the process of the invention,For/>True value of fuel consumption,/>Output of predictive model for optimizing fuel consumptionPredicted value of fuel consumption,/>The fuel consumption sample number;

Judging the prediction score of the optimized fuel consumption prediction model Whether the predicted fuel consumption is larger than a scoring threshold value or not, if so, the predicted scoring/>, of the optimized fuel consumption prediction modelIf the predicted fuel consumption is larger than the scoring threshold, the optimized fuel consumption prediction model is used as a training fuel consumption prediction model, and if the predicted scoring/>, of the optimized fuel consumption prediction model is larger than the scoring threshold, the predicted scoring/>And if the fuel consumption is not greater than the scoring threshold value, carrying out parameter iterative optimization on the optimized fuel consumption prediction model again.

In a second aspect, the present invention provides the following technical solutions, a vehicle fuel consumption prediction system based on vehicle-mounted CAN data, where the system includes:

the first determining module is used for acquiring original CAN parameters of the vehicle in real time, extracting features of the original CAN parameters to obtain first input features, wherein the original CAN parameters at least comprise vehicle speed, engine speed, accelerator position, oil consumption, braking pressure, running longitude and latitude data, running mileage and engine key parameters;

The second determining module is used for constructing a driver operation and driving risk feature set, and performing cluster analysis processing on feature variables in the driver operation and driving risk feature set by adopting a K-Means unsupervised cluster analysis algorithm so as to obtain a second input feature;

The third determining module is used for determining running parameters and road parameters of the vehicle based on the running longitude and latitude data and determining congestion level data of a running road section based on the running parameters and the road parameters;

The construction module is used for constructing a preset oil consumption prediction model, storing the first input characteristic, the second input characteristic and the congestion level data of the driving road section into a processing data set, and dividing the processing data set into a training set and a testing set;

The prediction module is used for inputting the training set into the preset fuel consumption prediction model for training to obtain a training fuel consumption prediction model, and inputting the testing set into the training fuel consumption prediction model to obtain a predicted fuel consumption result.

Preferably, the first determining module includes:

the processing sub-module is used for acquiring original CAN parameters of the vehicle in real time, and sequentially carrying out smooth filtering, interpolation deficiency and abnormal data processing on the original CAN parameters to obtain processed CAN parameters;

a mutual information determination submodule for calculating mutual information between the oil consumption variable and the processing CAN parameter ：

；

A sequencing sub-module for based on mutual information Sorting the processing CAN parameters from large to small to obtain sorting CAN parameters, and selecting the first several parameters in the sorting CAN parameters as first input features.

In a third aspect, the present invention provides a computer, including a memory, a processor, and a computer program stored in the memory and capable of running on the processor, where the processor implements the vehicle fuel consumption prediction method based on vehicle CAN data as described above when executing the computer program.

In a fourth aspect, the present invention provides a storage medium, where a computer program is stored, where the computer program when executed by a processor implements the vehicle fuel consumption prediction method based on vehicle CAN data as described above.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are needed in the embodiments or the description of the prior art will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings can be obtained according to these drawings without inventive effort for a person skilled in the art.

Fig. 1 is a flowchart of a vehicle fuel consumption prediction method based on vehicle-mounted CAN data according to an embodiment of the present invention;

FIG. 2 is a graph of K-SSE provided in accordance with an embodiment of the present invention;

fig. 3 is a structural block diagram of a vehicle fuel consumption prediction system based on vehicle-mounted CAN data according to a second embodiment of the present invention;

fig. 4 is a schematic hardware structure of a computer according to another embodiment of the invention.

Embodiments of the present invention will be further described below with reference to the accompanying drawings.

Detailed Description

Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to like or similar elements or elements having like or similar functions throughout. The embodiments described below by referring to the drawings are exemplary and intended to illustrate embodiments of the invention and should not be construed as limiting the invention.

In the description of the embodiments of the present invention, it should be understood that the terms "length," "width," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," and the like indicate orientations or positional relationships based on the orientation or positional relationships shown in the drawings, merely to facilitate description of the embodiments of the present invention and simplify description, and do not indicate or imply that the devices or elements referred to must have a specific orientation, be configured and operated in a specific orientation, and thus should not be construed as limiting the present invention.

Furthermore, the terms "first," "second," and the like, are used for descriptive purposes only and are not to be construed as indicating or implying a relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defining "a first" or "a second" may explicitly or implicitly include one or more such feature. In the description of the embodiments of the present invention, the meaning of "plurality" is two or more, unless explicitly defined otherwise.

In the embodiments of the present invention, unless explicitly specified and limited otherwise, the terms "mounted," "connected," "secured" and the like are to be construed broadly and include, for example, either permanently connected, removably connected, or integrally formed; can be mechanically or electrically connected; can be directly connected or indirectly connected through an intermediate medium, and can be communicated with the inside of two elements or the interaction relationship of the two elements. The specific meaning of the above terms in the embodiments of the present invention will be understood by those of ordinary skill in the art according to specific circumstances.

Example 1

In a first embodiment of the present invention, as shown in fig. 1, a vehicle fuel consumption prediction method based on vehicle-mounted CAN data includes:

S1, acquiring original CAN parameters of a vehicle in real time, and extracting features of the original CAN parameters to obtain first input features, wherein the original CAN parameters at least comprise vehicle speed, engine speed, accelerator position, oil consumption, brake pressure, running longitude and latitude data, running mileage and engine key parameters;

Specifically, in step S1, the original CAN parameters may be obtained by using a CAN bus to obtain initial unprocessed data, and at the same time, the obtained original CAN parameters at least include vehicle speed, engine speed, accelerator position, fuel consumption, brake pressure, running longitude and latitude data, running mileage, and engine key parameters, where the fuel consumption parameters further include fuel consumption parameters, where the fuel consumption parameters refer to fuel consumption data monitored in real time from a previous time to a current time, and the purpose of the present invention is to predict fuel consumption at a next time or a next period.

The step S1 specifically includes:

s1, acquiring original CAN parameters of a vehicle in real time, and sequentially carrying out smooth filtering, interpolation deficiency and abnormal data processing on the original CAN parameters to obtain processed CAN parameters;

Specifically, for the original CAN parameters, abnormal conditions such as signal jump and distortion may exist, so that a preprocessing process is required to be performed on the data, for a smoothing filtering process, a low-pass filter is introduced to perform smoothing processing on the original signals, noise is filtered, for an interpolation supplementing process, for a missing point of the data, a linear interpolation method is introduced to perform data supplementation, for an abnormal data processing process, impulse noise and other obvious abnormal values in the data are detected and removed, and it is required to be noted that the frequency of data acquisition is 1Hz in the invention.

S2, calculating mutual information between the oil consumption variable and the processing CAN parameter：

；

in particular, mutual information is a method for measuring the interdependence between two random variables, and different from a correlation coefficient, the mutual information not only can capture a linear relation, but also can capture a more complex nonlinear relation.

S3, based on mutual informationSorting the processing CAN parameters from large to small to obtain sorting CAN parameters, and selecting the first several parameters in the sorting CAN parameters as first input features;

specifically, in the invention, mutual information is selected after the CAN parameters are processed to be ordered The top 10 feature serves as the first input feature.

S2, constructing a driver operation and driving risk feature set, and performing cluster analysis processing on feature variables in the driver operation and driving risk feature set by adopting a K-Means unsupervised cluster analysis algorithm to obtain a second input feature;

wherein, the step S2 includes:

S21, constructing a driver operation and driving risk feature set, wherein the driver operation and driving risk feature set at least comprises operation days, total operation time, average operation time, total stay time, average stay time, total operation efficiency, total driving mileage, average driving mileage, total night driving time, night driving time proportion, night driving frequency and fatigue driving frequency.

S22, determining the number of data points in the driver operation and driving risk feature set and the number of clusters to be divided, selecting a group of clustered central points, minimizing the distance from each data point to the nearest central point, and iterating continuously until the clustered central point is not changed or the iteration stop condition is met;

specifically, in step S22, for a data set X including n-dimensional data points and to be divided into K clusters, the euclidean distance is used between the data objects to measure dissimilarity, and the targets are clustered Can be expressed as:

；

In the method, in the process of the invention, Is the center of the kth cluster,/>Represents the/>, in the datasetPoint,/>Is the number of data points in the kth cluster;

While new cluster centers iterate Can be expressed as:

。

s23, determining the square sum of errors corresponding to different cluster numbers based on an elbow method, and drawing a K-SSE curve based on the square sum of errors corresponding to the different cluster numbers;

specifically, in this step, the number of clusters is the K value in the K-SSE curve, and the sum of squares of the errors is the SSE value in the K-SSE curve.

S24, determining an optimal K value which is rapidly slowed down in the K-SSE curve, and taking characteristic data corresponding to the optimal K value as a second input characteristic;

Specifically, after determining the K-SSE curve, a sharp slowing point is found in the K-SSE curve as an optimal K value, that is, a significant inflection point is found in the K-SSE curve, with the iteration of the K value increasing, the SSE is continuously reduced, an optimal K value for the sharp slowing of the SSE is found, and the specific K-SSE curve can be seen with reference to fig. 2, and the reduction of the SSE change occurs when k=3, so that when the driver behavior characteristic parameter is constructed to be 3, the data aggregation effect is the best, and therefore, the characteristic data corresponding to the optimal K value=3 is used as the second input characteristic.

S3, determining running parameters and road parameters of the vehicle based on the running longitude and latitude data, and determining congestion level data of a running road section based on the running parameters and the road parameters;

Specifically, the step S3 includes: s31, determining the running parameters and road parameters of the vehicle based on the running longitude and latitude data; s32, determining the congestion level data of the driving road section based on the driving parameters and the road parameters.

Wherein, the step S31 includes:

and S311, carrying out visualization processing on the vehicle running track on a map based on the running longitude and latitude data so as to obtain a running route map.

S312, matching the geographic information system with the driving route map to determine driving parameters of vehicle driving, wherein the driving parameters at least comprise specific roads, road sections, driving time and driving distance of the vehicle driving.

S313, determining road type information and the number of lanes based on the specific road on which the vehicle is running so as to obtain the road parameters of the vehicle.

Wherein, the step S32 includes:

s321, determining a dynamic segment service level class table based on the driving parameters and the road parameters, and determining a road passing free flow speed based on the dynamic segment service level class table;

specifically, the road traffic free flow speed is determined according to a dynamic segment service level class table, the dynamic segment service level class table is determined according to the running speeds of different road segments at different congestion levels, and the specific dynamic segment service level class table is as follows:

dynamic segment service level class table

Road segment class service level	Expressway	Arterial road	Secondary trunk road	Branch circuit
					1	>55Km/h	>40 Km/h	>30 Km/h	>30 Km/h
2	>40-55Km/h	>30-40Km/h	>20-30Km/h	>20-30Km/h
					3	>30-40Km/h	>20-30Km/h	>15-20Km/h	>15-20Km/h
4	>20-30Km/h	>15-20Km/h	>10-15Km/h	>10-15Km/h
					5	≤20 Km/h	≤15 Km/h	≤10 Km/h	≤10 Km/h

From the above table, service levels 1 to 5 represent higher and higher congestion levels, and the road traffic free flow speed is the road traffic speed at service level 1.

S322, based on the road traffic free flow speedWith vehicle CAN speed/>Determining road Congestion index/>：

；

S323 based on the road congestion indexDistributing congestion level for each road section of the vehicle to obtain congestion level data of the running road section;

Specifically, in this step, the congestion level data of the traveling road section is allocated according to different congestion levels of different road sections, and the different congestion levels may be classified into clear (1), slow (2), congestion (3), medium congestion (4) and severe congestion (5).

S4, constructing a preset oil consumption prediction model, storing the first input feature, the second input feature and the congestion level data of the driving road section into a processing data set, and dividing the processing data set into a training set and a testing set;

Specifically, in the step, the preset oil consumption prediction model is specifically an oil consumption prediction RF random forest model, when a feature data set is constructed, the first input feature, the second input feature and the congestion level data of the driving road section are sequentially stored in a processing data set, and the processing data set is divided into a training set and a testing set according to a ratio of 7:3.

S5, inputting the training set into the preset fuel consumption prediction model for training to obtain a training fuel consumption prediction model, and inputting the testing set into the training fuel consumption prediction model to obtain a predicted fuel consumption result;

The step S5 includes:

s51, inputting the training set into the preset oil consumption prediction model, and training the preset oil consumption prediction model by using a random forest algorithm.

S52, carrying out parameter iterative optimization on the preset fuel consumption prediction model through grid search and cross verification until the average absolute error of model parameters is not greater than a performance threshold value so as to obtain an optimized fuel consumption prediction model;

Specifically, a random forest algorithm is used for training, parameters are optimized through grid search and cross verification, so that the prediction precision of a model is improved, and a performance threshold is set in the parameter optimization process.

S53, calculating a prediction score of the optimized fuel consumption prediction model：

；

Specifically, the prediction score is the average absolute percentage error, and the larger the error is, the lower the model prediction precision is, and the smaller the error is, the higher the model prediction precision is.

S54, judging the prediction scores of the optimized fuel consumption prediction modelWhether the predicted fuel consumption is larger than a scoring threshold value or not, if so, the predicted scoring/>, of the optimized fuel consumption prediction modelIf the predicted score is larger than the score threshold, carrying out parameter iterative optimization on the optimized fuel consumption prediction model again, and if the predicted score/>, of the optimized fuel consumption prediction modelIf the fuel consumption is not greater than the scoring threshold value, the optimized fuel consumption prediction model is used as a training fuel consumption prediction model;

specifically, the score threshold here is 5%, when the score is predicted When the prediction accuracy of the model is not required due to the fact that the prediction accuracy of the model is not required when the prediction accuracy of the model is larger than the scoring threshold, the process of steps S51-S52 needs to be carried out again until the prediction accuracy of the model is required, and if the prediction score/>, of the optimized fuel consumption prediction model, is obtainedThe prediction accuracy of the model reaches the prediction requirement, so that the optimized fuel consumption prediction model can be used as a training fuel consumption prediction model, and then the test set can be input into the training fuel consumption prediction model for prediction output so as to obtain a predicted fuel consumption result.

According to the vehicle fuel consumption prediction method based on the vehicle-mounted CAN data, firstly, original CAN parameters of a vehicle are obtained in real time, feature extraction is carried out on the original CAN parameters to obtain first input features, then a driver operation and driving risk feature set is constructed, and a K-Means unsupervised cluster analysis algorithm is adopted to carry out cluster analysis processing on feature variables in the driver operation and driving risk feature set to obtain second input features; then, determining running parameters and road parameters of the vehicle based on the running longitude and latitude data, and determining the congestion level data of the running road section based on the running parameters and the road parameters; then, a preset oil consumption prediction model is built, the first input feature, the second input feature and the congestion level data of the driving road section are stored in a processing data set, and the processing data set is divided into a training set and a testing set; finally, the training set is input into a preset fuel consumption prediction model for training to obtain a training fuel consumption prediction model, and the test set is input into the training fuel consumption prediction model to obtain a predicted fuel consumption result. The method solves the problems in data accuracy and training requirements, adopts the MI method to accurately screen key vehicle condition characteristics highly related to oil consumption, obviously reduces characteristic redundancy, avoids the problem of local characteristic limitation, utilizes real vehicle running data to construct a driver transportation behavior image as an independent characteristic, provides macroscopic and microscopic viewing angles for prediction, is beneficial to comprehensively considering actual driving environment factors, and simultaneously effectively calculates the road congestion index of a vehicle running section based on the ratio of free flow speed to CAN speed, so that the oil consumption prediction is more accurate.

Example two

As shown in fig. 3, in a second embodiment of the present invention, there is provided a vehicle fuel consumption prediction system based on vehicle-mounted CAN data, the system including:

the first determining module 1 is configured to obtain an original CAN parameter of a vehicle in real time, and perform feature extraction on the original CAN parameter to obtain a first input feature, where the original CAN parameter at least includes a vehicle speed, an engine speed, an accelerator position, oil consumption, a brake pressure, running longitude and latitude data, a running mileage, and an engine key parameter;

the second determining module 2 is used for constructing a driver operation and driving risk feature set, and performing cluster analysis processing on feature variables in the driver operation and driving risk feature set by adopting a K-Means unsupervised cluster analysis algorithm to obtain a second input feature;

A third determining module 3, configured to determine a driving parameter and a road parameter of a vehicle based on the driving longitude and latitude data, and determine driving road section congestion level data based on the driving parameter and the road parameter;

the construction module 4 is used for constructing a preset oil consumption prediction model, storing the first input feature, the second input feature and the congestion level data of the driving road section into a processing data set, and dividing the processing data set into a training set and a testing set;

And the prediction module 5 is used for inputting the training set into the preset fuel consumption prediction model for training to obtain a training fuel consumption prediction model, and inputting the testing set into the training fuel consumption prediction model to obtain a predicted fuel consumption result.

The first determining module 1 includes:

；

The second determining module 2 includes:

The risk feature set determining submodule is used for constructing a driver operation and driving risk feature set, and the driver operation and driving risk feature set at least comprises operation days, total operation duration, average operation duration, total stay duration, average stay duration, total operation efficiency, total driving mileage, average driving mileage, total night driving duration, night driving duration proportion, night driving frequency and fatigue driving frequency;

the clustering sub-module is used for determining the number of data points in the driver operation and driving risk feature set and the number of clusters to be divided, selecting a group of clustered central points and minimizing the distance from each data point to the nearest central point of the cluster, and iterating continuously until the clustered center is not changed or the iteration stopping condition is met;

the curve determining submodule is used for determining the square sum of errors corresponding to the different cluster numbers based on an elbow method and drawing a K-SSE curve based on the square sum of errors corresponding to the different cluster numbers;

and the K value determining submodule is used for determining an optimal K value which is rapidly slowed down in the K-SSE curve, and taking characteristic data corresponding to the optimal K value as a second input characteristic.

The third determining module 3 includes:

The route map determining submodule is used for carrying out visual processing on the running track of the vehicle on a map based on the running longitude and latitude data so as to obtain a running route map;

The system comprises a driving parameter determining submodule, a driving parameter determining submodule and a driving parameter determining module, wherein the driving parameter determining submodule is used for matching a geographic information system with a driving route map to determine driving parameters of vehicle driving, and the driving parameters at least comprise specific roads, road sections, driving time and driving distance of the vehicle driving;

And the road parameter determination sub-module is used for determining road type information and the number of lanes based on the specific road on which the vehicle runs so as to obtain the road parameter of the vehicle.

The third determining module 3 further comprises:

the free flow speed determining submodule is used for determining a dynamic segment service level grade table based on the running parameter and the road parameter and determining road passing free flow speed based on the dynamic segment service level grade table;

a road congestion index determination submodule for determining the free flow speed based on the road traffic With vehicle CAN speed/>Determining road Congestion index/>：

；

A congestion level determination sub-module for determining a congestion level based on the road congestion indexAnd distributing the congestion level to each road section of the vehicle to obtain the congestion level data of the running road section.

The prediction module 5 includes:

The training sub-module is used for inputting the training set into the preset oil consumption prediction model and training the preset oil consumption prediction model by using a random forest algorithm;

the iteration sub-module is used for carrying out parameter iteration optimization on the preset fuel consumption prediction model through grid search and cross verification until the average absolute error of model parameters is not greater than a performance threshold value so as to obtain an optimized fuel consumption prediction model;

A score determination submodule for calculating a prediction score of the optimized fuel consumption prediction model ：

；

a judging sub-module for judging the predictive score of the optimized fuel consumption predictive model Whether the predicted fuel consumption is larger than a scoring threshold value or not, if so, the predicted scoring/>, of the optimized fuel consumption prediction modelIf the predicted fuel consumption is larger than the scoring threshold, the optimized fuel consumption prediction model is used as a training fuel consumption prediction model, and if the predicted scoring/>, of the optimized fuel consumption prediction model is larger than the scoring threshold, the predicted scoring/>And if the fuel consumption is not greater than the scoring threshold value, carrying out parameter iterative optimization on the optimized fuel consumption prediction model again.

In other embodiments of the present invention, a computer is provided in the embodiments of the present invention, including a memory 102, a processor 101, and a computer program stored in the memory 102 and executable on the processor 101, where the processor 101 implements the vehicle fuel consumption prediction method based on vehicle CAN data as described above when executing the computer program.

In particular, the processor 101 may include a Central Processing Unit (CPU), or an Application SPECIFIC INTEGRATED Circuit (ASIC), or may be configured as one or more integrated circuits that implement embodiments of the present invention.

Memory 102 may include, among other things, mass storage for data or instructions. By way of example, and not limitation, memory 102 may comprise a hard disk drive (HARD DISK DRIVE, abbreviated HDD), a floppy disk drive, a Solid state drive (Solid STATE DRIVE, abbreviated SSD), flash memory, an optical disk, a magneto-optical disk, a magnetic tape, or a universal serial bus (Universal Serial Bus, abbreviated USB) drive, or a combination of two or more of these. Memory 102 may include removable or non-removable (or fixed) media, where appropriate. The memory 102 may be internal or external to the data processing apparatus, where appropriate. In a particular embodiment, the memory 102 is a Non-Volatile (Non-Volatile) memory. In particular embodiments, memory 102 includes Read-Only Memory (ROM) and random access Memory (Random Access Memory, RAM). Where appropriate, the ROM may be a mask-programmed ROM, a programmable ROM (Programmable Read-Only Memory, abbreviated PROM), an erasable PROM (Erasable Programmable Read-Only Memory, abbreviated EPROM), an electrically erasable PROM (ELECTRICALLY ERASABLE PROGRAMMABLE READ-Only Memory, abbreviated EEPROM), an electrically rewritable ROM (ELECTRICALLY ALTERABLE READ-Only Memory, abbreviated EAROM), or a FLASH Memory (FLASH), or a combination of two or more of these. The RAM may be a Static Random-Access Memory (SRAM) or a dynamic Random-Access Memory (Dynamic Random Access Memory DRAM), where the DRAM may be a fast page mode dynamic Random-Access Memory (Fast Page Mode Dynamic Random Access Memory, FPMDRAM), an extended data output dynamic Random-Access Memory (Extended Date Out Dynamic Random Access Memory, EDODRAM), a synchronous dynamic Random-Access Memory (Synchronous Dynamic Random-Access Memory, SDRAM), or the like, as appropriate.

Memory 102 may be used to store or cache various data files that need to be processed and/or communicated, as well as possible computer program instructions for execution by processor 101.

The processor 101 reads and executes the computer program instructions stored in the memory 102 to implement the vehicle fuel consumption prediction method based on the vehicle CAN data.

In some of these embodiments, the computer may also include a communication interface 103 and a bus 100. As shown in fig. 4, the processor 101, the memory 102, and the communication interface 103 are connected to each other by the bus 100 and perform communication with each other.

The communication interface 103 is used to implement communications between modules, devices, units, and/or units in embodiments of the invention. The communication interface 103 may also enable communication with other components such as: and the external equipment, the image/data acquisition equipment, the database, the external storage, the image/data processing workstation and the like are used for data communication.

Bus 100 includes hardware, software, or both, coupling components of a computer device to each other. Bus 100 includes, but is not limited to, at least one of: data Bus (Data Bus), address Bus (Address Bus), control Bus (Control Bus), expansion Bus (Expansion Bus), local Bus (Local Bus). By way of example, and not limitation, bus 100 may comprise a graphics acceleration interface (ACCELERATED GRAPHICS Port, abbreviated as AGP) or other graphics Bus, an enhanced industry standard architecture (Extended Industry Standard Architecture, abbreviated as EISA) Bus, a Front Side Bus (Front Side Bus, abbreviated as FSB), a HyperTransport (abbreviated as HT) interconnect, an industry standard architecture (Industry Standard Architecture, abbreviated as ISA) Bus, a wireless bandwidth (InfiniBand) interconnect, a Low Pin Count (LPC) Bus, a memory Bus, a micro channel architecture (Micro Channel Architecture, abbreviated as MCA) Bus, a peripheral component interconnect (PERIPHERAL COMPONENT INTERCONNECT, abbreviated as PCI) Bus, a PCI-Express (PCI-X) Bus, a serial advanced technology attachment (SERIAL ADVANCED Technology Attachment, abbreviated as SATA) Bus, a video electronics standards Association local (Video Electronics Standards Association Local Bus, abbreviated as VLB) Bus, or other suitable Bus, or a combination of two or more of these. Bus 100 may include one or more buses, where appropriate. Although embodiments of the invention have been described and illustrated with respect to a particular bus, the invention contemplates any suitable bus or interconnect.

The computer CAN execute the vehicle fuel consumption prediction method based on the vehicle CAN data based on the obtained vehicle fuel consumption prediction system based on the vehicle CAN data, thereby realizing the prediction of the vehicle fuel consumption.

In still other embodiments of the present invention, in combination with the above-described vehicle fuel consumption prediction method based on vehicle CAN data, embodiments of the present invention provide a technical solution, a storage medium storing a computer program, where the computer program when executed by a processor implements the above-described vehicle fuel consumption prediction method based on vehicle CAN data.

Those of skill in the art will appreciate that the logic and/or steps represented in the flow diagrams or otherwise described herein, e.g., a ordered listing of executable instructions for implementing logical functions, can be embodied in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, processor-containing system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions. For the purposes of this description, a "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device.

More specific examples (a non-exhaustive list) of the readable medium would include the following: an electrical connection (electronic device) having one or more wires, a portable computer diskette (magnetic device), a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). In addition, the computer readable medium may even be paper or other suitable medium on which the program is printed, as the program may be electronically captured, via, for instance, optical scanning of the paper or other medium, then compiled, interpreted or otherwise processed in a suitable manner, if necessary, and then stored in a computer memory.

It is to be understood that portions of the present invention may be implemented in hardware, software, firmware, or a combination thereof. In the above-described embodiments, the various steps or methods may be implemented in software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, may be implemented using any one or combination of the following techniques, as is well known in the art: discrete logic circuits having logic gates for implementing logic functions on data signals, application specific integrated circuits having suitable combinational logic gates, programmable Gate Arrays (PGAs), field Programmable Gate Arrays (FPGAs), and the like.

The technical features of the above-described embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above-described embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.

The above examples illustrate only a few embodiments of the invention, which are described in detail and are not to be construed as limiting the scope of the invention. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the invention, which are all within the scope of the invention. Accordingly, the scope of protection of the present invention is to be determined by the appended claims.

Claims

1. The vehicle fuel consumption prediction method based on the vehicle-mounted CAN data is characterized by comprising the following steps of:

2. The vehicle fuel consumption prediction method based on vehicle-mounted CAN data according to claim 1, wherein the step of acquiring original CAN parameters of a vehicle in real time and extracting features of the original CAN parameters to obtain first input features comprises:

；

In the method, in the process of the invention,Parameter set composed for all processing CAN parameters,/>For parameter/>Sum parameter/>Is used to determine the joint probability distribution of (1),And/>Parameters/>, respectivelySum parameter/>Is a boundary probability distribution of (1);

3. The vehicle fuel consumption prediction method based on vehicle-mounted CAN data according to claim 1, wherein the step of constructing a driver operation and driving risk feature set and performing cluster analysis processing on feature variables in the driver operation and driving risk feature set by using a K-Means unsupervised cluster analysis algorithm to obtain a second input feature comprises:

4. The vehicle fuel consumption prediction method based on vehicle-mounted CAN data according to claim 1, wherein the step of determining a driving parameter and a road parameter of a vehicle based on the driving longitude and latitude data includes:

5. The vehicle fuel consumption prediction method based on-vehicle CAN data according to claim 1, wherein the step of determining the travel section congestion level data based on the travel parameter and the road parameter includes:

；

6. The vehicle fuel consumption prediction method based on vehicle-mounted CAN data according to claim 1, wherein the step of inputting the training set into the preset fuel consumption prediction model for training to obtain a training fuel consumption prediction model comprises:

；

7. A vehicle fuel consumption prediction system based on vehicle-mounted CAN data, the system comprising:

8. The vehicle fuel consumption prediction system based on-board CAN data of claim 7, wherein the first determination module comprises:

；

9. A computer 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 vehicle fuel consumption prediction method based on-board CAN data according to any one of claims 1 to 6 when executing the computer program.

10. A storage medium having stored thereon a computer program which, when executed by a processor, implements the vehicle fuel consumption prediction method based on-board CAN data according to any one of claims 1 to 6.