CN108128309A - A kind of method that vehicle working condition is predicted in real time - Google Patents
A kind of method that vehicle working condition is predicted in real time Download PDFInfo
- Publication number
- CN108128309A CN108128309A CN201710783574.4A CN201710783574A CN108128309A CN 108128309 A CN108128309 A CN 108128309A CN 201710783574 A CN201710783574 A CN 201710783574A CN 108128309 A CN108128309 A CN 108128309A
- Authority
- CN
- China
- Prior art keywords
- speed
- state
- working condition
- predicted
- prediction
- 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
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W40/00—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models
- B60W40/10—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models related to vehicle motion
- B60W40/105—Speed
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W50/00—Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
- B60W2050/0001—Details of the control system
- B60W2050/0019—Control system elements or transfer functions
- B60W2050/0028—Mathematical models, e.g. for simulation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Mathematical Physics (AREA)
- Transportation (AREA)
- Mechanical Engineering (AREA)
- Electric Propulsion And Braking For Vehicles (AREA)
Abstract
The present invention relates to a kind of method that vehicle working condition is predicted in real time, this method carries out prediction of speed with the Markov model with Optimization Prediction condition, and BP neural network can compensate the error of prediction of speed, specific steps：The speed and acceleration of vehicle working condition are recorded on X-Y scheme,And by its gridding,Delete the grid of no state,And work condition state is encoded to a dimension,The state transition probability of adjacent moment is calculated using formula,Current state is determined as a numerical value,Then one group of random number is established,According to formula,It is another numerical value by the status predication of subsequent time,And so on,It searches in the table and obtains predetermined speed,Compensation speed is maintained with trained BP neural network and according to actual speed,The present invention proposes a kind of new vehicle operating mode real-time predicting method,The method is without additional hardware or sensor,It ensure that computational efficiency simultaneously,The state transition matrix established based on the vehicle working condition data recorded predicts vehicle working condition,Vehicle is not needed to additionally to increase hardware cost,And the adaptivity with different state of cyclic operation.
Description
Technical field
The present invention relates to a kind of methods that vehicle working condition is predicted in real time.
Background technology
Nowadays, mixed power electric car (HEV) is due to having the potentiality for improving automobile fuel ecomomy and reducing discharge
It is widely used, energy-optimised management in mixed power electric car (HEV) and plug-in hybrid electric automobile (PHEV)
Performance, either in terms of accuracy or in terms of computational efficiency, be all highly dependent in the prediction to Shape Of Things To Come operating mode.
The Forecasting Methodology of a variety of different vehicle working conditions has been proposed in academic circles at present, according to whether using remote sensing or vehicle
The different situations such as set sensor information, the Forecasting Methodology of these vehicle working conditions can be divided into two classes.
There is different method and algorithm using the operating mode Forecasting Methodology of remote sensing or onboard sensor information, however, using
The vehicle working condition Forecasting Methodology of remote sensing or onboard sensor information can generate higher cost, and whole vehicle control unit
(VCU) by increasingly complex and computation complexity higher.
To solve the problems, such as this, will be more applicable in without using the vehicle working condition Forecasting Methodology of remote sensing or onboard sensor information.
We propose the method for establishing Stochastic Markov chain and neural network respectively, for prediction of speed, in fact, these vehicles
The Forecasting Methodology of operating mode is the state transition matrix established based on the vehicle working condition data recorded to predict vehicle working condition
, vehicle is not needed to additionally to increase hardware cost, and the adaptivity with different state of cyclic operation.
Invention content
The present invention proposes a kind of new vehicle operating mode real-time predicting method, and the method is without additional hardware or sensing
Device, while ensure that computational efficiency, which is characterized in that it is carried out with formula (1) Markov model with Optimization Prediction condition
Prediction of speed：
P{X(tm+1)=j | X (t1)=x1,X(t2)=x2,...,X(tm)=xm}
=P { X (tm+1)=j | X (tm)=xm},j∈I
(1)
Trained BP neural network compensates the error of prediction of speed.
The method and step that vehicle operating mode is predicted in real time is as follows：
The speed and acceleration of vehicle working condition are recorded on X-Y scheme, and by its gridding.
The grid of no state is deleted, and work condition state is encoded to a dimension.
The state transition probability of adjacent moment is calculated using formula (2)：
Pij=Nij/Ni (2)
In formula：
Nij=from state i to the conversion times of state j；
Ni=from state i to stateful total conversion times；
Pij=from state i to the transition probability of state j.
Current state is determined as k0, then establish one group of random number { r1,r2,r3,...}。
It is k by the status predication of subsequent time according to formula (3)1,
It is similar, according to ki, NextState is predicted as ki+1。
Predicted state is determined in original mesh, then searches in the table and obtains predetermined speed.
Train BP neural network using the actual speed of input and predetermined speed error of output, then reuse by
Trained BP neural network simultaneously maintains compensation speed according to actual speed.
Final predetermined speed subtracts equal to the speed predicted by optimal Markov model and uses trained BP nerve nets
The compensation speed of network.
Typical Cities in China working condition measurement is selected to verify the accuracy of proposed operating mode Forecasting Methodology, and using formula
(4) in root-mean-square error (RMSE) come assess operating mode prediction accuracy：
Description of the drawings
Fig. 1 is the grid schematic diagram of work condition state in the working condition measurement of Typical Cities in China.
Fig. 2 is the BP neural network schematic diagram by actual speed and the training of predetermined speed error.
Fig. 3 is the prediction result schematic diagram using the Typical Cities in China working condition measurement of original Markov model.
Fig. 4 is the prediction result schematic diagram using the Typical Cities in China working condition measurement of optimal Markov model.
Fig. 5 is the prediction result of the Typical Cities in China working condition measurement using optimal Markov model and BP neural network
Schematic diagram.
Fig. 6 is the holistic approach schematic diagram of patent of the present invention.
Fig. 7 is the Energy Management System schematic diagram of hybrid vehicle.
Specific embodiment
Describe the specific embodiment of the present invention in detail below in conjunction with technical solution and attached drawing.
Technical scheme of the present invention is as shown in Figure 6.
The present invention proposes a kind of new vehicle operating mode real-time predicting method, and the method is without additional hardware or sensing
Device, while ensure that computational efficiency, which is characterized in that it is carried out with formula (1) Markov model with Optimization Prediction condition
Prediction of speed：
P{X(tm+1)=j | X (t1)=x1,X(t2)=x2,...,X(tm)=xm}
=P { X (tm+1)=j | X (tm)=xm},j∈I
(1)
Trained BP neural network compensates the error of prediction of speed.
The speed and acceleration of vehicle working condition are recorded on X-Y scheme, and by its gridding, Fig. 1 is Typical Cities in China
The grid schematic diagram of work condition state in working condition measurement.
The grid of no state is deleted, and work condition state is encoded to a dimension.
The state transition probability of adjacent moment is calculated using formula (2)：
Pij=Nij/Ni (2)
In formula：
Nij=from state i to the conversion times of state j；
Ni=from state i to stateful total conversion times；
Pij=from state i to the transition probability of state j.
Current state is determined as k0, then establish one group of random number { r1,r2,r3,...}。
It is k by the status predication of subsequent time according to formula (3)1,
It is similar, according to ki, NextState is predicted as ki+1。
Predicted state is determined in original mesh, then searches in the table and obtains predetermined speed.
Train BP neural network using the actual speed of input and predetermined speed error of output, then reuse by
Trained BP neural network simultaneously maintains compensation speed according to actual speed, shown in Fig. 2 to be missed by actual speed and predetermined speed
The BP neural network of difference training.
Final predetermined speed subtracts equal to the speed predicted by optimal Markov model and uses trained BP nerve nets
The compensation speed of network.
Typical Cities in China working condition measurement is selected to verify the accuracy of proposed operating mode Forecasting Methodology, and using formula
(4) in root-mean-square error (RMSE) come assess operating mode prediction accuracy：
In order to prove the validity of proposed operating mode Forecasting Methodology, we compared using original Markov model
Operating mode Forecasting Methodology, original Markov forecast techniques condition are intended to find the corresponding state of maximum probability converted with state,
As shown in Figure 3, the RMSE of predetermined speed is 1.1298 meter per seconds to prediction result.
In the working condition measurement of Typical Cities in China, the prediction result such as figure of the Markov model of optimum prediction condition are used
Shown in 4, the RMSE of predetermined speed is 0.7592 meter per second.
Predict that we increase is provided by BP neural network based on the above-mentioned vehicle working condition using optimal Markov model
The compensation speed error of maintenance, as shown in Figure 5, the RMSE of predetermined speed is 0.6700 meter per second to final prediction result.
In contrast, the vehicle working condition Forecasting Methodology proposed has more accuracy and validity, due to Markov chain
And BP neural network is offline foundation, but operating mode prediction is on-line implement, therefore Vehicle Controller accepting computation is answered
Miscellaneous degree, in addition, this optimal vehicle working condition Forecasting Methodology and without using telemetering or onboard sensor information, from these in terms of
It sees, the vehicle working condition Forecasting Methodology proposed has preferable feasibility.
Operating mode prediction algorithm provides necessary information input for the energy management strategies predicted based on operating mode, in Fig. 7
For Energy Management System in shown hybrid vehicle, operating mode prediction module provides the speed in following certain time period
Prediction, the speed based on the prediction can be converted to the power demand of vehicle, and the energy management strategies based on operating mode prediction are such as
Global motion planning can just optimize point of the hybrid power system for engine and driving motor torque in the predicted time section
With the selection with best gear, road test experiment shows more rule-based than existing based on the energy management strategies that operating mode is predicted
Energy management strategies can improve 10~15% or so on the rate of economizing gasoline of hybrid power system.
The present invention proposes a kind of new vehicle operating mode prediction side based on optimal Markov model and BP neural network
Method, and the feasibility with good accuracy and implementation, this method can fundamentally change the optimum control plan of dynamical system
Slightly, especially for mixed power electric car and pure electric automobile, the vehicle working condition Forecasting Methodology proposed can basis
The vehicle working condition of mixed power electric car is predicted to formulate energy management strategies, it is energy saving so as to help to realize.
Claims (1)
1. a kind of method that vehicle working condition is predicted in real time, which is characterized in that can with formula (1) Ma Er with Optimization Prediction condition
Husband's model carries out prediction of speed：
P{X(tm+1)=j | X (t1)=x1,X(t2)=x2,...,X(tm)=xm}
=P { X (tm+1)=j | X (tm)=xm},j∈I (1)
Trained BP neural network compensates the error of prediction of speed.
The method and step that vehicle operating mode is predicted in real time is as follows：
1) speed and acceleration of vehicle working condition are recorded on X-Y scheme, and by its gridding.
2) grid of no state is deleted, and work condition state is encoded to a dimension.
3) state transition probability of adjacent moment is calculated using formula (2)：
Pij=Nij/Ni (2)
In formula：
Nij=from state i to the conversion times of state j；
Ni=from state i to stateful total conversion times；
Pij=from state i to the transition probability of state j.
4) current state is determined as k0, then establish one group of random number { r1,r2,r3... }.
5) it is k by the status predication of subsequent time according to formula (3)1,
It is similar, according to ki, NextState is predicted as ki+1。
6) predicted state is determined in original mesh, then searches in the table and obtains predetermined speed.
7) BP neural network is trained using the actual speed of input and predetermined speed error of output, then reused by instruction
Experienced BP neural network simultaneously maintains compensation speed according to actual speed.
8) final predetermined speed subtracts equal to the speed predicted by optimal Markov model and uses trained BP neural network
Compensation speed.
9) Typical Cities in China working condition measurement is selected to verify the accuracy of proposed operating mode Forecasting Methodology, and using formula
(4) in root-mean-square error (RMSE) come assess operating mode prediction accuracy：
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710783574.4A CN108128309A (en) | 2017-09-01 | 2017-09-01 | A kind of method that vehicle working condition is predicted in real time |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710783574.4A CN108128309A (en) | 2017-09-01 | 2017-09-01 | A kind of method that vehicle working condition is predicted in real time |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108128309A true CN108128309A (en) | 2018-06-08 |
Family
ID=62388466
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710783574.4A Pending CN108128309A (en) | 2017-09-01 | 2017-09-01 | A kind of method that vehicle working condition is predicted in real time |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108128309A (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109159785A (en) * | 2018-07-19 | 2019-01-08 | 重庆科技学院 | A kind of automobile running working condition prediction technique based on Markov chain and neural network |
| CN109767619A (en) * | 2018-12-29 | 2019-05-17 | 江苏大学 | A kind of intelligent network connection pure electric automobile driving cycle prediction technique |
| CN110046719A (en) * | 2019-03-20 | 2019-07-23 | 北京物资学院 | A kind of bicycle method for diagnosing status and device |
| CN110979342A (en) * | 2019-12-30 | 2020-04-10 | 吉林大学 | Working condition information acquisition method for vehicle global energy management control |
| CN111831972A (en) * | 2020-07-16 | 2020-10-27 | 宁波工程学院 | Hybrid vehicle working condition prediction method and system based on road condition change |
| CN112441014A (en) * | 2020-12-02 | 2021-03-05 | 安徽华菱汽车有限公司 | Electric vehicle speed limit control method, device and medium |
| CN113076697A (en) * | 2021-04-20 | 2021-07-06 | 潍柴动力股份有限公司 | Typical driving condition construction method, related device and computer storage medium |
| CN114248781A (en) * | 2020-09-21 | 2022-03-29 | 比亚迪股份有限公司 | Vehicle working condition prediction method and device and vehicle |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102006054425A1 (en) * | 2005-11-22 | 2007-05-31 | Continental Teves Ag & Co. Ohg | Method for determination of value of model parameter of reference vehicle model, involves determination of statistical value of model parameter whereby artificial neural network is adapted with learning procedure |
| CN103034170A (en) * | 2012-11-27 | 2013-04-10 | 华中科技大学 | Numerical control machine tool machining performance prediction method based on intervals |
| CN103246943A (en) * | 2013-05-31 | 2013-08-14 | 吉林大学 | Vehicle operating condition multi-scale predicting method based on Markov chain |
| CN104616498A (en) * | 2015-02-02 | 2015-05-13 | 同济大学 | Markov chain and neural network based traffic congestion state combined prediction method |
| CN106427589A (en) * | 2016-10-17 | 2017-02-22 | 江苏大学 | Electric car driving range estimation method based on prediction of working condition and fuzzy energy consumption |
-
2017
- 2017-09-01 CN CN201710783574.4A patent/CN108128309A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102006054425A1 (en) * | 2005-11-22 | 2007-05-31 | Continental Teves Ag & Co. Ohg | Method for determination of value of model parameter of reference vehicle model, involves determination of statistical value of model parameter whereby artificial neural network is adapted with learning procedure |
| CN103034170A (en) * | 2012-11-27 | 2013-04-10 | 华中科技大学 | Numerical control machine tool machining performance prediction method based on intervals |
| CN103246943A (en) * | 2013-05-31 | 2013-08-14 | 吉林大学 | Vehicle operating condition multi-scale predicting method based on Markov chain |
| CN104616498A (en) * | 2015-02-02 | 2015-05-13 | 同济大学 | Markov chain and neural network based traffic congestion state combined prediction method |
| CN106427589A (en) * | 2016-10-17 | 2017-02-22 | 江苏大学 | Electric car driving range estimation method based on prediction of working condition and fuzzy energy consumption |
Non-Patent Citations (1)
| Title |
|---|
| 曹磊，陈长文: "基于马尔可夫链的汽车行驶工况预测", 《内燃机与动力装置》 * |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109159785A (en) * | 2018-07-19 | 2019-01-08 | 重庆科技学院 | A kind of automobile running working condition prediction technique based on Markov chain and neural network |
| CN109159785B (en) * | 2018-07-19 | 2020-05-01 | 重庆科技学院 | Automobile driving condition prediction method based on Markov chain and neural network |
| CN109767619A (en) * | 2018-12-29 | 2019-05-17 | 江苏大学 | A kind of intelligent network connection pure electric automobile driving cycle prediction technique |
| CN109767619B (en) * | 2018-12-29 | 2021-05-25 | 江苏大学 | Intelligent networking pure electric vehicle running condition prediction method |
| CN110046719A (en) * | 2019-03-20 | 2019-07-23 | 北京物资学院 | A kind of bicycle method for diagnosing status and device |
| CN110979342B (en) * | 2019-12-30 | 2021-02-05 | 吉林大学 | Working condition information acquisition method for vehicle global energy management control |
| CN110979342A (en) * | 2019-12-30 | 2020-04-10 | 吉林大学 | Working condition information acquisition method for vehicle global energy management control |
| CN111831972A (en) * | 2020-07-16 | 2020-10-27 | 宁波工程学院 | Hybrid vehicle working condition prediction method and system based on road condition change |
| CN114248781A (en) * | 2020-09-21 | 2022-03-29 | 比亚迪股份有限公司 | Vehicle working condition prediction method and device and vehicle |
| CN114248781B (en) * | 2020-09-21 | 2024-04-16 | 比亚迪股份有限公司 | Vehicle working condition prediction method and device and vehicle |
| CN112441014A (en) * | 2020-12-02 | 2021-03-05 | 安徽华菱汽车有限公司 | Electric vehicle speed limit control method, device and medium |
| CN112441014B (en) * | 2020-12-02 | 2022-03-08 | 安徽华菱汽车有限公司 | Electric vehicle speed limit control method, device and medium |
| CN113076697A (en) * | 2021-04-20 | 2021-07-06 | 潍柴动力股份有限公司 | Typical driving condition construction method, related device and computer storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108128309A (en) | A kind of method that vehicle working condition is predicted in real time | |
| Panday et al. | A review of optimal energy management strategies for hybrid electric vehicle | |
| Zhang et al. | Real-time energy management strategy based on velocity forecasts using V2V and V2I communications | |
| Padmarajan et al. | Blended rule-based energy management for PHEV: System structure and strategy | |
| Liu et al. | Hybrid-trip-model-based energy management of a PHEV with computation-optimized dynamic programming | |
| Yu et al. | Model predictive control for hybrid electric vehicle platooning using slope information | |
| Li et al. | Intelligent environment-friendly vehicles: Concept and case studies | |
| Chen et al. | Energy management for a power-split plug-in hybrid electric vehicle based on dynamic programming and neural networks | |
| Liu et al. | Rule-corrected energy management strategy for hybrid electric vehicles based on operation-mode prediction | |
| Liu et al. | An on-line energy management strategy based on trip condition prediction for commuter plug-in hybrid electric vehicles | |
| CN110667434A (en) | Working condition-adaptive pure electric vehicle driving mileage estimation method and system | |
| Ganji et al. | A study on look-ahead control and energy management strategies in hybrid electric vehicles | |
| Sun et al. | An adaptive ECMS based on traffic information for plug-in hybrid electric buses | |
| Gautam et al. | Hybrid sources powered electric vehicle configuration and integrated optimal power management strategy | |
| He et al. | Global Optimal Energy Management Strategy Research for a Plug‐In Series‐Parallel Hybrid Electric Bus by Using Dynamic Programming | |
| Baker et al. | Investigation of vehicle speed prediction from neural network fit of real world driving data for improved engine on/off control of the EcoCAR3 hybrid Camaro | |
| Yu et al. | Model predictive control for connected hybrid electric vehicles | |
| Li et al. | Investigating long-term vehicle speed prediction based on GA-BP algorithms and the road-traffic environment | |
| Gan et al. | Intelligent learning algorithm and intelligent transportation-based energy management strategies for hybrid electric vehicles: A review | |
| Zhao et al. | Stochastic model predictive control for dual-motor battery electric bus based on signed Markov chain Monte Carlo method | |
| Kabir | Vehicle speed prediction based on road status using machine learning | |
| CN107464059A (en) | A kind of public transport company based on historical information automates control method of arranging an order according to class and grade | |
| He et al. | Deep reinforcement learning based energy management strategies for electrified vehicles: Recent advances and perspectives | |
| Liu et al. | A short-term traffic flow forecasting method and its applications | |
| Grossard et al. | An optimal energy-based approach for driving guidance of full electric vehicles |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20180608 |
|
| WD01 | Invention patent application deemed withdrawn after publication |