CN111198501A - Method for determining fuel equivalent factor by RBF neural network - Google Patents

Method for determining fuel equivalent factor by RBF neural network Download PDF

Info

Publication number
CN111198501A
CN111198501A CN202010039719.1A CN202010039719A CN111198501A CN 111198501 A CN111198501 A CN 111198501A CN 202010039719 A CN202010039719 A CN 202010039719A CN 111198501 A CN111198501 A CN 111198501A
Authority
CN
China
Prior art keywords
neural network
fuel
rbf neural
average
equivalent factor
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.)
Granted
Application number
CN202010039719.1A
Other languages
Chinese (zh)
Other versions
CN111198501B (en
Inventor
余世明
马包胜
何德峰
仇翔
宋秀兰
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Zhejiang University of Technology ZJUT
Original Assignee
Zhejiang University of Technology ZJUT
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Zhejiang University of Technology ZJUT filed Critical Zhejiang University of Technology ZJUT
Priority to CN202010039719.1A priority Critical patent/CN111198501B/en
Publication of CN111198501A publication Critical patent/CN111198501A/en
Application granted granted Critical
Publication of CN111198501B publication Critical patent/CN111198501B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
    • G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
    • G05B13/04—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators
    • G05B13/042—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators in which a parameter or coefficient is automatically adjusted to optimise the performance

Landscapes

  • Engineering & Computer Science (AREA)
  • Software Systems (AREA)
  • Artificial Intelligence (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Computation (AREA)
  • Medical Informatics (AREA)
  • Health & Medical Sciences (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Automation & Control Theory (AREA)
  • Electric Propulsion And Braking For Vehicles (AREA)
  • Combined Controls Of Internal Combustion Engines (AREA)
  • Hybrid Electric Vehicles (AREA)

Abstract

The invention discloses a method for calculating fuel equivalent factors by an RBF neural network. When the Pontryagin minimum value principle is applied to energy management of a hybrid electric vehicle, a Hamiltonian based on fuel economy can be constructed according to a whole vehicle power model, and a covariate can be regarded as a fuel equivalent factor for converting instantaneous electric quantity consumption into instantaneous fuel consumption. Aiming at the problem that the optimal fuel equivalent factor is difficult to determine so that the fuel economy of the whole vehicle reaches the global optimal value, the RBF neural network is adopted to optimize the fuel equivalent factor. And taking the average speed, the average acceleration, the average deceleration and the average impact degree as input layers of the neural network, and taking the fuel equivalent factor as output. Compared with the method for determining the fuel equivalent factor only through the type of the working condition of the vehicle, the method can obtain a better equivalent factor; compared with a method for calculating the fuel equivalent factor through an intelligent optimization algorithm, the method provided by the invention has the advantages of smaller calculated amount and stronger real-time property.

Description

Method for determining fuel equivalent factor by RBF neural network
Technical Field
The invention relates to the field of automobiles, optimizes a hybrid electric vehicle energy management method based on the Pontryagin minimum value principle, and particularly relates to a method for calculating a fuel equivalent factor through a Radial Basis Function (RBF) neural network.
Background
Before a new energy automobile replaces a traditional fuel automobile, a hybrid electric automobile as a transition automobile type exists for a long time, and an energy management strategy of the hybrid electric automobile is a key for solving the problems of fuel economy, exhaust emission and the like. Aiming at the nonlinear time-varying system such as a hybrid electric vehicle, an energy management strategy can be designed by applying the Pontryagin minimum value principle. According to a dynamic model of the whole vehicle, a Hamiltonian can be constructed, wherein the covariates can be regarded as fuel equivalent factors for converting instantaneous electric quantity consumption into instantaneous fuel consumption.
The energy management method based on the Pontryagin minimum value is a global optimization algorithm, meanwhile, the energy management can be converted into an instantaneous problem through a Hamiltonian, the key is a fuel equivalent factor, and the optimal fuel equivalent factor can ensure the best fuel economy of the whole vehicle. The method of determining the fuel equivalent factor only according to the type of the working condition of the vehicle or the driving style of the driver ignores the influence of other vehicle condition information on the fuel equivalent factor, and is difficult to obtain the optimal fuel equivalent factor. The method for searching the optimal fuel equivalent factor through the intelligent optimization algorithm has large calculation amount and poor real-time performance.
Disclosure of Invention
In order to solve the technical problems in the prior art, the invention provides a method for determining a fuel equivalent factor by an RBF neural network so as to quickly obtain the nearly optimal fuel equivalent factor and improve the fuel economy of the whole vehicle.
The technical scheme for solving the problems is as follows:
1) and (5) preprocessing the working condition data (offline). Extracting a plurality of 600s working condition segments from the existing driving record working condition library, calculating the optimal power distribution of each working condition segment by a Hamiltonian, and determining the fuel equivalent factor lambda by an enumeration method. And calculating the average speed of each working condition segment
Figure BDA0002366593990000011
Average acceleration
Figure BDA0002366593990000012
Average deceleration
Figure BDA0002366593990000013
And average impact strength
Figure BDA0002366593990000014
2) The RBF neural network is trained (offline). At an average speed
Figure BDA0002366593990000015
Average acceleration
Figure BDA0002366593990000016
Average deceleration
Figure BDA0002366593990000017
And average impact strength
Figure BDA0002366593990000018
And (3) taking the optimal fuel equivalent factor lambda as an output for inputting layer parameters, and training the RBF neural network by a gradient descent method.
3) The fuel equivalence factor is calculated (on-line). Extracting the latest 120s of working conditions from the recorded running working conditions of the vehicle, and calculating the average speed
Figure BDA0002366593990000021
Average acceleration
Figure BDA0002366593990000022
Average deceleration
Figure BDA0002366593990000023
And average impact strength
Figure BDA0002366593990000024
And then calculating the fuel equivalent factor through the trained RBF neural network.
In the step 1, the impact degree is the magnitude of the acceleration change rate
Figure BDA0002366593990000025
Where v (t) is the current running speed.
In the step 1, the Hamiltonian is
Figure BDA0002366593990000026
Wherein, x (t) is a state variable of the system, which is referred to as a battery SOC state; u (T) is a control variable of the system, here motor torque Tm。
Figure BDA0002366593990000027
The instantaneous oil consumption of the engine is lambda (t) & f (x (t), u (t), and t), the equivalent oil consumption of the motor is t,
Figure BDA0002366593990000028
λ (t) is a fuel equivalent factor. When the Hamiltonian is minimal, there is an optimal power distribution of the engine and the electric machine. The fuel equivalent factor which enables the fuel economy of the whole vehicle to reach the best is the optimal equivalent factor.
In the step 2, the RBF neural network comprises an input layer, a hidden layer and an output layer. Taking Gaussian radial basis function as activation function of hidden layer
Figure BDA0002366593990000029
Wherein, ciIs the radial basis function center, σ is the radial basis function width, and x is the input layer parameter; the output layer is a linear combination of hidden layer neuron outputs with different weights
Figure BDA00023665939900000210
Wherein, ω isiY is the output value of the output layer.
In the step 2, each coefficient of the neural network is adjusted by a gradient descent method. The output mean square error is taken as an objective function, and because the single-output RBF neural network is constructed by the method, the method comprises the following steps:
Figure BDA00023665939900000211
Figure BDA00023665939900000212
where e is the output error, y is the actual value,
Figure BDA00023665939900000213
is the value output by the RBF neural network. In order to minimize the objective function, the individual parameters are adjusted in the direction of the negative gradient, there are
ωi←ωi+Δωi，ci←ci+Δci，σi←σi+Δσ
Figure BDA00023665939900000214
Figure BDA0002366593990000031
Figure BDA0002366593990000032
Where, ← is the assignment symbol, and η is the learning rate.
In the step 3, after the training of the RBF neural network is completed, each coefficient omega is obtainedi、ci、σiIs already a definite value, at which time the speed will be averaged
Figure BDA0002366593990000033
Average acceleration
Figure BDA0002366593990000034
Average deceleration
Figure BDA0002366593990000035
And average impact strength
Figure BDA0002366593990000036
Substituting to obtain the fuel equivalent factor.
Compared with the existing method for determining the fuel equivalent factor, the method has the following advantages: the influence of each item of vehicle condition information on the fuel equivalent factor is fully considered, the mapping relation between the basic vehicle condition information and the optimal fuel equivalent factor is established through the RBF neural network, and the fuel equivalent factor close to the optimal can be quickly obtained in a real-time environment.
Drawings
Fig. 1 is a diagram of an RBF neural network architecture.
Fig. 2 is a flow chart of energy management for a PMP-based hybrid vehicle.
Detailed Description
The invention is further described below with reference to the accompanying drawings.
Fig. 1 is a block diagram of an RBF neural network. As shown in FIG. 2, the characteristic parameters are extracted from the driving record, the fuel equivalent factor is calculated by an RBF neural network, and the power distribution of the engine and the motor is calculated by a Hamilton function in combination with the required power.
The invention discloses a method for determining fuel equivalent factors by an RBF neural network, which comprises the following steps:
step 1: and (5) preprocessing the working condition data (offline). Extracting 600s driving segments from the existing working condition library, and calculating the average speed of the driving segments
Figure BDA0002366593990000037
Average acceleration
Figure BDA0002366593990000038
Average deceleration
Figure BDA0002366593990000039
And average impact strength
Figure BDA00023665939900000310
The impact degree is the magnitude of the change rate of the acceleration, and is shown in formula 1:
Figure BDA00023665939900000311
the optimal power distribution is determined by the Hamilton function, so that the fuel equivalent factor with the optimal fuel economy of the whole vehicle is the optimal fuel equivalent factor of the working condition, and the optimal fuel equivalent factor can be determined by a stepping method. The Hamiltonian is shown in equation 2:
Figure BDA00023665939900000312
wherein, x (t) is a state variable of the system, which is referred to as a battery SOC state; u (T) is a control variable of the system, here motor torque Tm。
Figure BDA0002366593990000041
The instantaneous oil consumption of the engine is lambda (t) & f (x (t), u (t), and t), the equivalent oil consumption of the motor is t,
Figure BDA0002366593990000042
λ (t) is a fuel equivalent factor. When the Hamiltonian is minimum, there is an optimal power allocation.
Step 2: the RBF neural network is trained (offline). The RBF neural network comprises an input layer, a hidden layer and an output layer, and a Gaussian radial basis function is used as an activation function of the hidden layer, as shown in formula 3:
Figure BDA0002366593990000043
wherein, ciDetermining the initial radial basis function centers of the hidden layer neurons by K-means clustering, wherein the radial basis function centers are sigma, the radial basis function widths are sigma, the input layer parameters are x, and the number of the hidden layer neurons is 7; output layer is a linear with different weights for hidden layer neuron outputA combination, as shown in formula 4:
Figure BDA0002366593990000044
the output of the gaussian radial basis function is highly correlated with the euclidean distance, and in order to avoid that high-order parameters have a decisive influence on the output of the gaussian radial basis function, normalization processing needs to be performed on the parameters of the input layer to make them in the same order of magnitude. Meanwhile, in order to reflect the difference of the influence degree of each input layer parameter on the output, the normalized parameters may be weighted, as shown in equation 5.
Figure BDA0002366593990000045
a1+a2+a3+a41 (6) wherein a1、a2、a3、a4Is the weight coefficient, and is the assignment symbol.
And adjusting each coefficient of the neural network by a gradient descent method. The output mean square error is taken as an objective function, the output layer only has one parameter and is a fuel equivalent factor, so the RBF neural network constructed by the invention is a single-output neural network. Equation 7 is the objective function, and equation 8 is the error expression.
Figure BDA0002366593990000046
Figure BDA0002366593990000047
Where e is the output error, y is the actual value,
Figure BDA0002366593990000048
is the value output by the RBF neural network. In order to minimize the objective function, the individual parameters are adjusted in the direction of the negative gradient, there are
ωi←ωi+Δωi，ci←ci+Δci，σi←σi+Δσ (9)
Figure BDA0002366593990000049
Figure BDA0002366593990000051
Figure BDA0002366593990000052
And the step of either refining the RBF neural network or refining the RBF neural network, wherein the step of either refining the RBF neural network or refining the RBF neural network is used as the key, and the step of refining the RBF neural network is used as the key.
And step 3: the equivalence factor is calculated (on-line). Extracting the latest 120s of vehicle condition information from the driving record, and calculating the average speed
Figure BDA0002366593990000053
Average acceleration
Figure BDA0002366593990000054
Average deceleration
Figure BDA0002366593990000055
And average impact strength
Figure BDA0002366593990000056
Substituting the calculated equivalent fuel factors into the trained RBF neural network to obtain the equivalent fuel factors required by the next stroke, calculating the optimal power distribution by combining the Hamiltonian, and updating the equivalent fuel factors every 3 seconds.

Claims (5)

1. A method for determining a fuel equivalent factor by an RBF neural network is characterized by comprising the following steps:
in an off-line environment, effective driving segments are selected from an existing working condition library, the optimal fuel equivalent factors of the driving segments are calculated, and an RBF neural network is constructed to fit the mapping relation between the basic information of the vehicle condition and the optimal fuel equivalent factors;
and after the RBF neural network training is finished, extracting corresponding characteristic parameters from the driving record to calculate the fuel equivalent factor in a real-time online environment.
2. A method for determining fuel equivalence factors by an RBF neural network as claimed in claim 1, wherein, under off-line conditions, effective 600s operating condition segments are selected from an existing operating condition library and their average speeds are calculated respectively
Figure FDA0002366593980000016
Average acceleration
Figure FDA0002366593980000017
Average deceleration
Figure FDA0002366593980000018
And average impact strength
Figure FDA0002366593980000019
The degree of impact is defined as
Figure FDA0002366593980000011
According to the extreme value principle of Pontryagin, a Hamiltonian is constructed according to a mathematical model of the hybrid electric vehicle
Figure FDA0002366593980000012
When the Hamiltonian has the minimum value, the whole vehicle has the optimal power distribution, when the fuel equivalent factor lambda takes the optimal value, the whole vehicle has the optimal fuel economy, and the optimal equivalent factors of all working condition segments are determined by a stepping method.
3. An RBF neural network as claimed in claim 1 for determining fuel equivalence factorsMethod, characterized by constructing RBF neural network fitting basic vehicle condition information (average speed)
Figure FDA00023665939800000110
Average acceleration
Figure FDA00023665939800000111
Average deceleration
Figure FDA00023665939800000112
Average degree of impact
Figure FDA00023665939800000113
) And an optimal equivalence factor λ; the RBF neural network is divided into three layers, including an input layer, a hidden layer and an output layer, wherein the hidden layer takes a Gaussian radial basis function as an activation function:
Figure FDA0002366593980000013
wherein, ciIs the radial basis function center, σ is the radial basis function width, and x is the input layer parameter; the output layer is a linear combination of hidden layer neuron outputs with different weights
Figure FDA0002366593980000014
Wherein, ω isiY is the output value of the output layer.
4. A method for determining fuel equivalence factors by an RBF neural network as claimed in claim 1, wherein the parameters of the input layer are normalized and then weighted
Figure FDA0002366593980000015
a1+a2+a3+a4＝1
Wherein, a1、a2、a3、a4Is the weight coefficient, and is the assignment symbol. Training RBF neural network by gradient descent method to minimize output mean square error as objective function
ωi←ωi+Δωi,ci←ci+Δci,σi←σi+Δσ
Figure FDA0002366593980000021
Figure FDA0002366593980000022
Figure FDA0002366593980000023
Where, ← is the assignment symbol, and η is the learning rate.
5. A method for determining fuel equivalence factors by an RBF neural network as claimed in claim 1, wherein after training of the RBF neural network is completed, basic vehicle condition information including average speed is extracted from the last 120s driving record under real-time on-line condition
Figure FDA0002366593980000024
Average acceleration
Figure FDA0002366593980000025
Average deceleration
Figure FDA0002366593980000026
Average degree of impact
Figure FDA0002366593980000027
Substituting into RBF neural network to calculate fuel equivalent factor every 3 secondsAnd updating the primary fuel equivalent factor.
CN202010039719.1A 2020-01-14 2020-01-14 Method for determining fuel equivalent factor by RBF neural network Active CN111198501B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010039719.1A CN111198501B (en) 2020-01-14 2020-01-14 Method for determining fuel equivalent factor by RBF neural network

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010039719.1A CN111198501B (en) 2020-01-14 2020-01-14 Method for determining fuel equivalent factor by RBF neural network

Publications (2)

Publication Number Publication Date
CN111198501A true CN111198501A (en) 2020-05-26
CN111198501B CN111198501B (en) 2022-05-13

Family

ID=70746496

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010039719.1A Active CN111198501B (en) 2020-01-14 2020-01-14 Method for determining fuel equivalent factor by RBF neural network

Country Status (1)

Country Link
CN (1) CN111198501B (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111923897A (en) * 2020-09-29 2020-11-13 北京航空航天大学 Intelligent energy management method for plug-in hybrid electric vehicle
CN112193232A (en) * 2020-09-23 2021-01-08 江苏大学 Self-adaptive energy management system and method for hybrid electric vehicle
CN116118709A (en) * 2023-03-14 2023-05-16 合众新能源汽车股份有限公司 Energy management method and system for hybrid electric vehicle

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103914985A (en) * 2014-04-25 2014-07-09 大连理工大学 Method for predicting future speed trajectory of hybrid power bus
CN105527110A (en) * 2015-12-03 2016-04-27 东软集团股份有限公司 Evaluation method and device of automotive fuel economy
US20170146362A1 (en) * 2015-11-19 2017-05-25 GM Global Technology Operations LLC Method and apparatus for fuel consumption prediction and cost estimation via crowd-sensing in vehicle navigation system
CN107688343A (en) * 2017-08-01 2018-02-13 北京理工大学 A kind of energy control method of motor vehicle driven by mixed power
CN109131350A (en) * 2018-08-23 2019-01-04 北京理工大学 A kind of hybrid vehicle energy management method and system
CN109143867A (en) * 2018-09-26 2019-01-04 上海海事大学 A kind of energy management method of the hybrid power ship based on ANN Control
CN110194172A (en) * 2019-06-28 2019-09-03 重庆大学 Based on enhanced neural network plug-in hybrid passenger car energy management method

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103914985A (en) * 2014-04-25 2014-07-09 大连理工大学 Method for predicting future speed trajectory of hybrid power bus
US20170146362A1 (en) * 2015-11-19 2017-05-25 GM Global Technology Operations LLC Method and apparatus for fuel consumption prediction and cost estimation via crowd-sensing in vehicle navigation system
CN105527110A (en) * 2015-12-03 2016-04-27 东软集团股份有限公司 Evaluation method and device of automotive fuel economy
CN107688343A (en) * 2017-08-01 2018-02-13 北京理工大学 A kind of energy control method of motor vehicle driven by mixed power
CN109131350A (en) * 2018-08-23 2019-01-04 北京理工大学 A kind of hybrid vehicle energy management method and system
CN109143867A (en) * 2018-09-26 2019-01-04 上海海事大学 A kind of energy management method of the hybrid power ship based on ANN Control
CN110194172A (en) * 2019-06-28 2019-09-03 重庆大学 Based on enhanced neural network plug-in hybrid passenger car energy management method

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
FENG TIANHENG等: "A Supervisory Control Strategy for Plug-In Hybrid Electric Vehicles Based on Energy Demand Prediction and Route Preview", 《 IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY》 *
NINGYUAN GUO等: "A Novel Velocity Forecast Method for Improving Predictive Energy Management of Plug-In Hybrid Electric Vehicles", 《2017 IEEE VEHICLE POWER AND PROPULSION CONFERENCE (VPPC)》 *
武振: "基于神经网络和PMP的混合动力汽车能量管理策略研究", 《中国优秀博硕士学位论文全文数据库(硕士) 工程科技Ⅱ辑》 *
王庆年等: "插电式混合动力汽车动力传动系参数优化", 《吉林大学学报(工学版)》 *

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112193232A (en) * 2020-09-23 2021-01-08 江苏大学 Self-adaptive energy management system and method for hybrid electric vehicle
CN112193232B (en) * 2020-09-23 2021-10-12 江苏大学 Self-adaptive energy management system and method for hybrid electric vehicle
CN111923897A (en) * 2020-09-29 2020-11-13 北京航空航天大学 Intelligent energy management method for plug-in hybrid electric vehicle
CN116118709A (en) * 2023-03-14 2023-05-16 合众新能源汽车股份有限公司 Energy management method and system for hybrid electric vehicle
CN116118709B (en) * 2023-03-14 2024-01-16 合众新能源汽车股份有限公司 Energy management method and system for hybrid electric vehicle

Also Published As

Publication number Publication date
CN111198501B (en) 2022-05-13

Similar Documents

Publication Publication Date Title
CN111198501B (en) Method for determining fuel equivalent factor by RBF neural network
CN107688343B (en) Energy control method of hybrid power vehicle
CN111038485B (en) Hybrid electric vehicle control method and system based on driving style recognition
WO2021103625A1 (en) Short-term vehicle speed condition real-time prediction method based on interaction between vehicle ahead and current vehicle
CN113153521B (en) System and method for predicting vehicle engine torque by using artificial neural network
CN110775065B (en) Hybrid electric vehicle battery life prediction method based on working condition recognition
Wu et al. Fuzzy energy management strategy for a hybrid electric vehicle based on driving cycle recognition
CN110962837B (en) Plug-in hybrid electric vehicle energy management method considering driving style
CN112668799A (en) Intelligent energy management method and storage medium for PHEV (Power electric vehicle) based on big driving data
CN111047085B (en) Hybrid vehicle working condition prediction method based on meta-learning
CN111775925B (en) Working mode decision method and device for power split hybrid electric vehicle
CN110929920A (en) Hybrid power train energy management method based on working condition identification
CN112046489B (en) Driving style identification algorithm based on factor analysis and machine learning
CN112026744B (en) Series-parallel hybrid power system energy management method based on DQN variants
CN115492928B (en) Economical efficiency, dynamic performance and safety comprehensive optimal gear shifting rule optimization method
CN112298155B (en) Method for predicting energy consumption of hybrid power truck based on variable time domain model
CN111923897A (en) Intelligent energy management method for plug-in hybrid electric vehicle
Montazeri-Gh et al. Driving condition recognition for genetic-fuzzy HEV control
CN117465301A (en) Fuel cell automobile real-time energy management method based on data driving
CN110077389B (en) Energy management method for plug-in hybrid electric vehicle
CN114969982A (en) Fuel cell automobile deep reinforcement learning energy management method based on strategy migration
CN107527093B (en) Wind turbine generator running state diagnosis method and device
Nguyen et al. Optimal Energy Management Strategy based on Driving Pattern Recognition for a Dual-Motor Dual-Source Electric Vehicle
CN113997925B (en) Energy management method for plug-in hybrid power system
Ji et al. Driving Condition Recognition Combined With Stochastic Prediction and Machine Learning and Its Application in Energy Management of Medium Fuel Cell Trucks

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
GR01 Patent grant
GR01 Patent grant