scholarly journals Powertrain Control for Hybrid-Electric Vehicles Using Supervised Machine Learning

Vehicles ◽  
2020 ◽  
Vol 2 (2) ◽  
pp. 267-286 ◽  
Author(s):  
Craig K. D. Harold ◽  
Suraj Prakash ◽  
Theo Hofman
Keyword(s):  
Machine Learning ◽  
Dynamic Programming ◽  
Electric Vehicles ◽  
Control Strategy ◽  
Hybrid Electric Vehicles ◽  
Supervised Machine Learning ◽  
Test Case ◽  
Drive Cycle ◽  
Powertrain Control ◽  
Hybrid Electric

This paper presents a novel framework to enable automatic re-training of the supervisory powertrain control strategy for hybrid electric vehicles using supervised machine learning. The aim of re-training is to customize the control strategy to a user-specific driving behavior without human intervention. The framework is designed to update the control strategy at the end of a driving task. A combination of dynamic programming and supervised machine learning is used to train the control strategy. The trained control strategy denoted as SML is compared to an online-implementable strategy based on the combination of the optimal operation line and Pontryagin’s minimum principle denoted as OOL-PMP, on the basis of fuel consumption. SML consistently performed better than OOL-PMP, evaluated over five standard drive cycles. The EUDC performance was almost identical while on FTP75 the OOL-PMP consumed 14.7% more fuel than SML. Moreover, the deviation from the global benchmark obtained from dynamic programming was between 1.8% and 5.4% for SML and between 5.8% and 16.8% for OOL-PMP. Furthermore, a test-case was conducted to emulate a real-world driving scenario wherein a trained controller is exposed to a new drive cycle. It is found that the performance on the new drive cycle deviates significantly from the optimal policy; however, this performance gap is bridged with a single re-training episode for the respective test-case.

Depleting Mode Control Strategies for Plug-in Hybrid Electric Vehicles

2011 ◽  
Vol 130-134 ◽  
pp. 2211-2215
Author(s):  
Bing Zhan Zhang ◽  
Han Zhao ◽  
An Dong Yin
Keyword(s):  
Electric Vehicles ◽  
Fuel Economy ◽  
Control Strategy ◽  
Hybrid Electric Vehicles ◽  
Control Strategies ◽  
Great Influence ◽  
Torque Control ◽  
Drive Cycle ◽  
Vehicle Simulation ◽  
Hybrid Electric

Control strategy is the most important issue in the Plug-in Hybrid electric vehicles (PHEV) design, which has two modes: charge depleting mode (CD) and charge sustaining mode (CS). The different control strategies in depleting mode will have a great influence on PHEV dynamic performance and fuel economy. The engine optimal torque control strategy was proposed in the paper. The vehicle simulation model in Powertrain Systems Analysis Toolkit (PSAT) was adopted to evaluate the proposed control strategy. The aggressive highway drive cycle Artemis_hwy and a random drive cycle generated by Markov Process were used. The simulation results indicate the proposed control strategy has great improvement in fuel economy.

Modeling and Control Strategy of Series Hybrid Electric Vehicles

2014 ◽  
Vol 543-547 ◽  
pp. 1246-1249
Author(s):  
Liang Zhang ◽  
Bin Jiao ◽  
Lei Li
Keyword(s):  
Electric Vehicles ◽  
Control Strategy ◽  
Power Distribution ◽  
Hybrid Electric Vehicles ◽  
System Structure ◽  
Modeling And Control ◽  
Powertrain Control ◽  
Hybrid Electric ◽  
And Control ◽  
Made In

The modeling method and control strategy for series hybrid electric vehicles were presented in this paper. Firstly, the system structure and operation principles are discussed systematically; and then a control strategy is proposed based on the modeling of powertrain. Control strategy focus on the multi-modes switch logic and power distribution. In the last part of this paper, the simulation made in MATLAB/Simulink was introduced, which results indicate that the model and control strategy are correct.

A Dynamic Programming Control Strategy for HEV

2012 ◽  
Vol 263-266 ◽  
pp. 541-544 ◽  
Author(s):  
Babici Leandru Corneliu Cezar ◽  
Onea Alexandru
Keyword(s):  
Dynamic Programming ◽  
Electric Vehicles ◽  
Control Strategy ◽  
Hybrid Electric Vehicles ◽  
The State ◽  
State Of Charge ◽  
Battery Pack ◽  
Matlab Simulink ◽  
Hybrid Electric

Dynamic programming is a very powerful algorithmic paradigm which solves a problem by identifying subproblems and tackling them one by one. First the smallest are solved, and then using their answers, it can be figured out larger ones, until the whole lot of them is solved. This paper presents a control strategy for hybrid electric vehicles, based on the dynamic programming, applied in MATLAB, Simulink environment, using ADVISOR. It was tried this method due to the calculation speed of the suitable torque and speed required from the engine, considering the driver power request (torque and speed), and the state of charge (SOC) of the batteries. Using the fuel converter (FC) fuel map, and the remaining SOC of the battery pack, it was designed an algorithm that will chose at each time the required torque and speed from the first and second source of power.

scholarly journals Optimal Control Strategy for Parallel Plug-In Hybrid Electric Vehicles Based on Dynamic Programming

2021 ◽  
Vol 12 (2) ◽  
pp. 85
Author(s):  
Ying Tian ◽  
Jiaqi Liu ◽  
Qiangqiang Yao ◽  
Kai Liu
Keyword(s):  
Optimal Control ◽  
Dynamic Programming ◽  
Electric Vehicles ◽  
Control Strategy ◽  
Hybrid Electric Vehicles ◽  
Dynamic Programming Algorithm ◽  
Programming Algorithm ◽  
Optimal Control Strategy ◽  
Parallel Hybrid ◽  
Hybrid Electric

In this paper, the dynamic programming algorithm is applied to the control strategy design of parallel hybrid electric vehicles. Based on MATLAB/Simulink software, the key component model and controller model of the parallel hybrid system are established, and an offline simulation platform is built. Based on the platform, the global optimal control strategy based on the dynamic programming algorithm is studied. The torque distribution rules and shifting rules are analyzed, and the optimal control strategy is adopted to design the control strategy, which effectively improves the fuel economy of plug-in hybrid electric vehicles. The fuel consumption rate of this parallel hybrid electric vehicle is based on china city bus cycle (CCBC) condition.

scholarly journals A Study on the Anomaly Detection of Engine Clutch Engagement/Disengagement Using Machine Learning for Transmission Mounted Electric Drive Type Hybrid Electric Vehicles

2021 ◽  
Vol 11 (21) ◽  
pp. 10187
Author(s):  
Yonghyeok Ji ◽  
Seongyong Jeong ◽  
Yeongjin Cho ◽  
Howon Seo ◽  
Jaesung Bang ◽  
...  
Keyword(s):  
Machine Learning ◽  
Anomaly Detection ◽  
Electric Vehicles ◽  
Electric Drive ◽  
Hybrid Electric Vehicles ◽  
Short Term Memory ◽  
Support Vector ◽  
Clutch Engagement ◽  
Hybrid Electric ◽  
Engine Clutch

Transmission mounted electric drive type hybrid electric vehicles (HEVs) engage/disengage an engine clutch when EV↔HEV mode transitions occur. If this engine clutch is not adequately engaged or disengaged, driving power is not transmitted correctly. Therefore, it is required to verify whether engine clutch engagement/disengagement operates normally in the vehicle development process. This paper studied machine learning-based methods for detecting anomalies in the engine clutch engagement/disengagement process. We trained the various models based on multi-layer perceptron (MLP), long short-term memory (LSTM), convolutional neural network (CNN), and one-class support vector machine (one-class SVM) with the actual vehicle test data and compared their results. The test results showed the one-class SVM-based models have the highest anomaly detection performance. Additionally, we found that configuring the training architecture to determine normal/anomaly by data instance and conducting one-class classification is proper for detecting anomalies in the target data.

Sign in / Sign up

Export Citation Format

Share Document