CN115438570A - GA-BP neural network-based vehicle dynamic fuel consumption prediction model method - Google Patents
GA-BP neural network-based vehicle dynamic fuel consumption prediction model method Download PDFInfo
- Publication number
- CN115438570A CN115438570A CN202210921511.1A CN202210921511A CN115438570A CN 115438570 A CN115438570 A CN 115438570A CN 202210921511 A CN202210921511 A CN 202210921511A CN 115438570 A CN115438570 A CN 115438570A
- Authority
- CN
- China
- Prior art keywords
- oil consumption
- neural network
- correlation coefficient
- prediction model
- vehicle
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/10—Numerical modelling
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Evolutionary Computation (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biophysics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computer Hardware Design (AREA)
- Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Biomedical Technology (AREA)
- Geometry (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Combined Controls Of Internal Combustion Engines (AREA)
Abstract
The invention relates to the technical field of data processing, in particular to a vehicle dynamic oil consumption prediction model method based on a GA-BP neural network, which comprises the steps of mining and processing original data in the vehicle driving process to obtain factors influencing an oil consumption model; calculating the correlation degree of factors influencing the oil consumption model to obtain a correlation coefficient matrix; constructing a principal component expression by using the correlation coefficient matrix; establishing a BP neural network by using a principal component expression, and optimizing the initial weight and the threshold of the BP neural network by a genetic algorithm; the method comprises the steps of establishing an oil consumption prediction model by using the optimized BP neural network, predicting the dynamic oil consumption of the vehicle by using the oil consumption prediction model, and obtaining a prediction result.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to a GA-BP neural network-based vehicle dynamic fuel consumption prediction model method.
Background
The fuel economy of the vehicle is influenced by various factors, the influence degree of the various factors on the fuel economy of the vehicle is different, the analysis on the fuel economy influence factors is beneficial to distinguishing the primary and secondary relations among the factors, the optimization is made aiming at the main influence, and the important theoretical and practical significance is achieved for improving the vehicle fuel utilization rate.
The current oil consumption analysis and research thinking is mainly based on an engine oil consumption MAP, and the engine oil consumption MAP has a simple structure and can accurately reflect the steady-state driving oil consumption characteristics of a vehicle. However, in daily operation, the vehicle tends to be in an unstable state, and the operating state frequently changes. In this case, the steady-state model based on the fuel consumption map of the engine has a large deviation and poor practicability, and it is difficult to accurately estimate the fuel consumption of the vehicle and to accurately evaluate various novel fuel consumption control strategies.
The model established by the method of 'steady state initial estimation + transient correction' usually has higher precision, and the method is based on the initial estimation of the oil consumption by a simple and common steady state module, then establishes a transient model by real-time working parameters of a vehicle and an engine, corrects the initially estimated oil consumption error, and can improve the precision of the model. However, the influence of the current environment of the vehicle on the fuel consumption is not considered, so that the requirement on real-time performance is difficult to meet.
Disclosure of Invention
The invention aims to provide a GA-BP neural network-based vehicle dynamic fuel consumption prediction model method, and aims to solve the problem that the existing model does not consider the influence of the current environment of a vehicle on fuel consumption and is difficult to meet the requirement of real-time performance.
In order to achieve the aim, the invention provides a vehicle dynamic oil consumption prediction model method based on a GA-BP neural network, which comprises the following steps:
original data in the vehicle running process are mined and processed to obtain factors influencing the oil consumption model;
calculating the correlation degree of the factors influencing the oil consumption model to obtain a correlation coefficient matrix;
constructing a principal component expression by using the correlation coefficient matrix;
establishing a BP neural network by using the principal component expression, and optimizing the initial weight and the threshold value of the BP neural network by a genetic algorithm;
constructing a fuel consumption prediction model by using the optimized BP neural network;
and predicting the dynamic oil consumption of the vehicle through the oil consumption prediction model to obtain a prediction result.
The specific mode for mining and processing the original data in the vehicle driving process to obtain the factors influencing the oil consumption model is as follows:
collecting original data in the running process of a vehicle;
filtering the original data to obtain filtered data;
and preprocessing the filtering data by adopting a moving average data preprocessing method to obtain factors influencing the oil consumption model.
The specific way for calculating the degree of association of the factors affecting the fuel consumption model to obtain the correlation coefficient matrix is as follows:
establishing a characteristic matrix of the factors influencing the oil consumption by using the factors influencing the oil consumption model;
obtaining a correlation coefficient of the factors influencing the oil consumption by utilizing a covariance formula based on the characteristic matrix of the factors influencing the oil consumption;
and establishing a correlation coefficient matrix based on the correlation coefficients.
Wherein, the specific way of constructing the principal component expression by using the correlation coefficient matrix is as follows:
constructing a characteristic equation of a correlation coefficient array by using the correlation coefficient matrix;
calculating a characteristic vector of a correlation coefficient and a characteristic value of a factor influencing oil consumption by using the characteristic equation;
constructing a feature vector equation by using the feature vector of the correlation coefficient and the feature value;
solving the contribution rate of the characteristic value by using the characteristic vector equation;
and dividing the contribution rate into three main components to obtain a main component expression.
The factors influencing the oil consumption model comprise the speed, the acceleration, the engine torque, the engine speed, the gradient of the current position and the adhesion coefficient of the road surface of the current position of the vehicle.
Wherein the three principal components include a kinetic energy factor, an acceleration factor, and a dissipation factor.
According to the vehicle dynamic oil consumption prediction model method based on the GA-BP neural network, the factors influencing the oil consumption model are obtained by mining and processing the original data in the vehicle driving process; calculating the correlation degree of the factors influencing the oil consumption model to obtain a correlation coefficient matrix; constructing a principal component expression by using the correlation coefficient matrix; establishing a BP neural network by using the principal component expression, and optimizing the initial weight and the threshold of the BP neural network by a genetic algorithm; the optimized BP neural network is used for constructing a fuel consumption prediction model, dynamic fuel consumption of a vehicle is predicted through the fuel consumption prediction model, and a prediction result is obtained.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the embodiments or the prior art descriptions will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without creative efforts.
FIG. 1 is a schematic diagram of a schematic principle of a vehicle dynamic fuel consumption prediction model method based on a GA-BP neural network provided by the invention.
FIG. 2 is a schematic diagram of a fuel consumption influence factor determination neural network neuron of a GA-BP neural network-based vehicle dynamic fuel consumption prediction model method provided by the invention.
Fig. 3 is a BP neural network structure diagram of a GA-BP neural network-based vehicle dynamic fuel consumption prediction model method provided by the present invention.
FIG. 4 is a GA-BP neural network optimization flow chart of the vehicle dynamic fuel consumption prediction model method based on the GA-BP neural network provided by the invention.
FIG. 5 is a flowchart of a method for predicting dynamic fuel consumption of a vehicle based on a GA-BP neural network provided by the invention.
Detailed Description
Reference will now be made in detail to embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to the same or similar elements or elements having the same or similar function throughout. The embodiments described below with reference to the drawings are illustrative and intended to be illustrative of the invention and are not to be construed as limiting the invention.
Referring to fig. 1 to 5, the invention provides a method for predicting dynamic fuel consumption of a vehicle based on a GA-BP neural network, comprising the following steps:
s1, mining and processing original data in a vehicle driving process to obtain factors influencing an oil consumption model;
the concrete mode is as follows:
s11, collecting original data in the driving process of a vehicle;
s12, filtering the original data to obtain filtered data;
specifically, the original data is firstly filtered to improve the data quality. Thereby eliminating the uncertain interference on the result caused by data loss, data jump and the like.
And S13, preprocessing the filtering data by adopting a moving average data preprocessing method to obtain factors influencing the oil consumption model.
Specifically, the factors influencing the fuel consumption model comprise the speed, the acceleration, the engine torque, the engine speed, the gradient of the current position and the adhesion coefficient of the road surface of the current position of the vehicle.
S2, calculating the correlation degree of the factors influencing the oil consumption model to obtain a correlation coefficient matrix;
the concrete method is as follows:
s21, establishing a characteristic matrix of the factors influencing the oil consumption by using the factors influencing the oil consumption model;
specifically, a factor matrix of the fuel consumption influence is established through preprocessed factor data of the fuel consumption influence model;
wherein: m represents the number of factors influencing oil consumption, including the speed, the acceleration, the engine torque, the engine speed, the gradient of the current position and the adhesion coefficient of the road surface of the current position of the vehicle, and n represents the number of groups of the elements influencing oil consumption in the selected experiment.
S22, obtaining a correlation coefficient of the oil consumption influencing factors by utilizing a covariance formula based on the characteristic matrix of the oil consumption influencing factors;
specifically, the correlation coefficient r of the influencing factors in each group of influencing factor data is calculated and obtained by using a covariance formula ij ；
S23, establishing a correlation coefficient matrix based on the correlation coefficients.
S3, constructing a principal component expression by using the correlation coefficient matrix;
the concrete mode is as follows:
s31, constructing a characteristic equation of a correlation coefficient array by using the correlation coefficient matrix;
specifically, the correlation coefficient r of the influencing factors in each group of influencing factor data is calculated and obtained by using a covariance formula ij (ii) a Calculating a correlation coefficient array R of all oil consumption factors; constructing a characteristic equation of the correlation coefficient array through the correlation coefficient array R of the oil consumption factors;
through calculation, a correlation coefficient array R of all factors influencing the oil consumption can be obtained;
constructing a characteristic equation of the correlation coefficient array through the correlation coefficient array R of the oil consumption factors;
|R-μE|＝0
s32, calculating a characteristic vector of a correlation coefficient and a characteristic value of a factor influencing oil consumption by using the characteristic equation;
specifically, E represents an m × m unit matrix, and a feature vector q affecting the oil consumption factor is obtained through calculation i . Obtaining the characteristic value mu of m oil consumption factors i ，(i＝1,2,......,m)。
S33, constructing a characteristic vector equation by using the characteristic vector of the correlation coefficient and the characteristic value;
s34, solving the contribution rate of the characteristic value by using the characteristic vector equation;
specifically, the characteristic value and the characteristic vector of the factors influencing the oil consumption are calculated by the characteristic equation of the factors influencing the oil consumption, and the contribution rate W of each factor influencing the oil consumption is calculated respectively i 。
Respectively calculating the contribution rate W of each oil consumption influence factor i ；
S35, dividing the contribution rate into three main components to obtain a main component expression.
Specifically, the three principal components include a kinetic energy factor, an acceleration factor, and a dissipation factor.
From the eigenvector equation q i Calculating the principal component score of each influencing oil consumption factor;
Z i the ith influencing factor principal component score is expressed.
Constructing a principal component expression by the score matrix of the principal component; dividing the vehicle speed and the acceleration to be called acceleration factors of a fuel consumption prediction model, wherein the adhesion coefficient of the road gradient and the ground is called consumption factors of fuel consumption prediction, and the engine speed and the engine torque of the vehicle are called kinetic energy factors of fuel consumption prediction;
constructing a principal component expression by the score matrix of the principal component; through the contrast of the contribution rate of the oil consumption factors, dividing the vehicle speed and the acceleration to be called acceleration factors of an oil consumption prediction model, the adhesion coefficient of the road gradient and the ground is called consumption factors of oil consumption prediction, and the engine speed and the engine torque of the vehicle are called kinetic energy factors of the oil consumption prediction;
Y i ＝X*Z i
in the formula: y is i Is the ith main component, X is a vector formed by various factors influencing the oil consumption, and Z i Is the score vector of the ith principal component.
S4, establishing a BP neural network by using the principal component expression, and optimizing the initial weight and the threshold of the BP neural network through a genetic algorithm;
specifically, the main three main components are used as BP neural network input neurons; and optimizing the initial weight and the threshold value of the BP neural network by utilizing the global search function of the genetic algorithm.
The BP neural network has a strong data fitting prediction capability, but the determination of the initial weights and thresholds has a large impact on the performance of the neural network. The genetic algorithm optimizes the BP neural network, namely the initial weight and the threshold value of the BP neural network are optimized, so that the BP neural network can complete the training of the model more scientifically and quickly.
The elements of the genetic algorithm optimization BP neural network comprise population initialization, fitness function selection, selection operation, cross operation and mutation operation. In the process of searching the optimal initial weight and the threshold value by the genetic algorithm, the proper population scale and the cross variation probability are determined, so that the neural network training can be more quickly approached to the ultimate optimal solution, and the local optimal solution and the algorithm non-convergence are avoided.
S5, constructing a fuel consumption prediction model by using the optimized BP neural network;
(1) BP neural network structure
The BP neural network has the characteristics of strong data fitting prediction capability, self-adaption and the like, and the whole training process comprises two processes of data forward transmission and error feedback correction. When the BP neural network is trained, data are transmitted to the hidden layer through the input layer according to weights, and are finally applied to the output layer after nonlinear processing. Therefore, a nonlinear BP neural network is selected to construct a nonlinear oil consumption prediction problem.
(2) The determination of the number of the hidden layer neurons and the number of the single layer neurons has a remarkable influence on the performance of the whole network, the determination method is not unique, and the number of the hidden layer neurons is determined preliminarily. Can be properly adjusted in the actual training process. The number of hidden layer neurons is continually modified to achieve optimality.
In the formula: n is the number of neurons in the hidden layer, m is the number of neurons in the input layer, N is the number of neurons in the output layer, and t is an integer between 0 and 10.
(3) According to the BP neural network structure, the vehicle speed and the acceleration extracted from the factors mainly influencing the oil consumption are called acceleration factors of an oil consumption prediction model, the adhesion coefficient of the road gradient and the ground is called consumption factors of oil consumption prediction, and the engine rotating speed and the engine torque of a vehicle are called kinetic energy factors of the oil consumption prediction. The three main components are used as BP neural network input neurons.
And S6, predicting the dynamic oil consumption of the vehicle through the oil consumption prediction model to obtain a prediction result.
According to the method, firstly, the oil consumption influence data are preprocessed by adopting a moving average data preprocessing method, so that uncertain interference on results caused by data loss, data jumping and the like is eliminated. And then, through the correlation analysis of the factors influencing the oil consumption, finding out the correlation coefficient of the factors influencing the oil consumption, and obtaining a characteristic value, a principal component contribution rate and a score matrix so as to obtain a principal component expression, wherein the speed and the acceleration of the vehicle are divided to be called as an acceleration factor of an oil consumption prediction model, the adhesion coefficient of the road gradient and the ground is called as a consumption factor of the oil consumption prediction, and the engine speed and the engine torque of the vehicle are called as an oil consumption prediction kinetic energy factor. The GA-BP network is adopted for fitting, and the input layer, the hidden layer and the output layer can simplify the BP neural network structure, and meanwhile, the elements of the BP neural network optimized by the genetic algorithm comprise population initialization, fitness function selection, selection operation, cross operation and variation operation. The influence of various factors on the vehicle is focused on the running state of the vehicle, the influence factors of the fuel consumption model are comprehensively considered according to the running condition, the external environment and the driving characteristics of the vehicle, and a proper fuel consumption analysis and prediction model is established, so that the method has important theoretical and practical significance for the development and evaluation of the subsequent energy management strategy.
Although the above embodiments are only examples of the method for predicting fuel consumption of a vehicle based on a GA-BP neural network of the present invention, it is understood that the scope of the present invention is not limited thereto, and those skilled in the art can understand that all or part of the processes of the above embodiments can be implemented and equivalents thereof can be made according to the claims of the present invention.
Claims (6)
1. A GA-BP neural network-based vehicle dynamic fuel consumption prediction model method is characterized by comprising the following steps:
original data in the vehicle running process are mined and processed to obtain factors influencing the oil consumption model;
calculating the correlation degree of the factors influencing the oil consumption model to obtain a correlation coefficient matrix;
constructing a principal component expression by using the correlation coefficient matrix;
establishing a BP neural network by using the principal component expression, and optimizing the initial weight and the threshold value of the BP neural network by a genetic algorithm;
constructing a fuel consumption prediction model by using the optimized BP neural network;
and predicting the dynamic oil consumption of the vehicle through the oil consumption prediction model to obtain a prediction result.
2. A GA-BP neural network-based vehicle dynamic fuel consumption prediction model method as claimed in claim 1,
the specific mode for mining and processing the original data in the vehicle driving process to obtain the factors influencing the oil consumption model is as follows:
collecting original data in the running process of a vehicle;
filtering the original data to obtain filtered data;
and preprocessing the filtering data by adopting a moving average data preprocessing method to obtain factors influencing the oil consumption model.
3. A GA-BP neural network-based vehicle dynamic fuel consumption prediction model method according to claim 2,
the specific way for calculating the degree of correlation of the factors affecting the fuel consumption model to obtain the correlation coefficient matrix is as follows:
establishing a characteristic matrix of the factors influencing the oil consumption by using the factors influencing the oil consumption model;
obtaining a correlation coefficient of the factors influencing the oil consumption by utilizing a covariance formula based on the characteristic matrix of the factors influencing the oil consumption;
and establishing a correlation coefficient matrix based on the correlation coefficients.
4. A GA-BP neural network-based vehicle dynamic fuel consumption prediction model method as claimed in claim 3,
the specific way of constructing the principal component expression by using the correlation coefficient matrix is as follows:
constructing a characteristic equation of a correlation coefficient array by using the correlation coefficient matrix;
calculating a characteristic vector of a correlation coefficient and a characteristic value of a factor influencing oil consumption by using the characteristic equation;
constructing a feature vector equation by using the feature vector of the correlation coefficient and the feature value;
solving the contribution rate of the characteristic value by using the characteristic vector equation;
and dividing the contribution rate into three main components to obtain a main component expression.
5. A GA-BP neural network-based vehicle dynamic fuel consumption prediction model method according to claim 4,
the factors influencing the oil consumption model comprise the speed, the acceleration, the engine torque, the engine rotating speed, the gradient of the current position and the adhesion coefficient of the road surface of the current position of the vehicle.
6. A GA-BP neural network-based vehicle dynamic fuel consumption prediction model method as claimed in claim 5,
the three principal components include a kinetic energy factor, an acceleration factor, and a dissipation factor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210921511.1A CN115438570A (en) | 2022-08-02 | 2022-08-02 | GA-BP neural network-based vehicle dynamic fuel consumption prediction model method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210921511.1A CN115438570A (en) | 2022-08-02 | 2022-08-02 | GA-BP neural network-based vehicle dynamic fuel consumption prediction model method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115438570A true CN115438570A (en) | 2022-12-06 |
Family
ID=84243498
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210921511.1A Pending CN115438570A (en) | 2022-08-02 | 2022-08-02 | GA-BP neural network-based vehicle dynamic fuel consumption prediction model method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115438570A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117235508A (en) * | 2023-11-15 | 2023-12-15 | 天津市普迅电力信息技术有限公司 | Vehicle fuel consumption prediction method and system based on big data |
-
2022
- 2022-08-02 CN CN202210921511.1A patent/CN115438570A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117235508A (en) * | 2023-11-15 | 2023-12-15 | 天津市普迅电力信息技术有限公司 | Vehicle fuel consumption prediction method and system based on big data |
| CN117235508B (en) * | 2023-11-15 | 2024-01-30 | 天津市普迅电力信息技术有限公司 | Vehicle fuel consumption prediction method and system based on big data |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110594317B (en) | Starting control strategy based on double-clutch type automatic transmission | |
| CN114217524A (en) | Power grid real-time self-adaptive decision-making method based on deep reinforcement learning | |
| CN111047085A (en) | Hybrid vehicle working condition prediction method based on meta-learning | |
| CN109931943B (en) | Unmanned ship global path planning method and electronic equipment | |
| CN115438570A (en) | GA-BP neural network-based vehicle dynamic fuel consumption prediction model method | |
| CN114199248B (en) | AUV co-location method for optimizing ANFIS based on mixed element heuristic algorithm | |
| CN112149883A (en) | Photovoltaic power prediction method based on FWA-BP neural network | |
| CN114912697B (en) | PSO algorithm-based boiler slagging degree prediction method and related device | |
| CN112987572B (en) | Particle swarm optimization method of adaptive ball bar system based on priori knowledge | |
| CN115390452B (en) | LQR transverse controller parameter online self-adaption method and system | |
| CN114444737B (en) | Pavement maintenance intelligent planning method based on transfer learning | |
| CN115689001A (en) | Short-term load prediction method based on pattern matching | |
| Gladwin et al. | A controlled migration genetic algorithm operator for hardware-in-the-loop experimentation | |
| CN113890633B (en) | Underwater acoustic communication system self-adaptive selection method based on deep neural network | |
| CN108256190A (en) | Multiple target Aircraft Steering Engine optimum design method | |
| CN114219274A (en) | Workshop scheduling method adapting to machine state based on deep reinforcement learning | |
| CN113537620A (en) | Vehicle speed prediction method based on Markov model optimization and working condition recognition | |
| CN112615364A (en) | Novel wide-area intelligent cooperative control method for power grid stability control device | |
| CN117901724B (en) | Control method, system and equipment for thermal management system of pure electric vehicle | |
| Zhang et al. | An on-line adaptive hybrid PID autopilot of ship heading control using auto-tuning BP & RBF neurons | |
| CN110351241A (en) | A kind of industrial network DDoS intruding detection system classification method based on GWA optimization | |
| CN113915250B (en) | Intelligent control system and control method for wet clutch based on state identification | |
| CN115855226B (en) | Multi-AUV cooperative underwater data acquisition method based on DQN and matrix completion | |
| CN113419426B (en) | AUV communication delay active compensation method based on data driving state predictor | |
| CN114326706B (en) | Path planning method based on intelligent agricultural machinery and used for rapid deviation correction |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |