Hybrid Electric Vehicle Control Strategy Based On The Markov Prediction

In the last 100 years, the automotive industry has brought about great advances for human civilization, but at the same time it makes the situation of energy and environment is getting from bad to worse. Hybrid Electric Vehicle (HEV) is taken as one of the most promising technologies for clean vehicles through worldwide. Facing the problem of energy and environment, some of the nations introduce policy which shows that Hybrid Electric Vehicle are paid great attention to than ever before, such as China introduced the plan of"Ten Cities 1000 Cart", and the United States advocated"Green Economic Recovery Plan".HEV's control strategy is the key technology of HEV exploitation, which directly affects energy flow inner the vehicle and the whole vehicle's capability. So when researching the HEV, the investigation of control strategy is of extreme importance, and also is the key technology of HEV exploitation and design.The most familiar control strategy in parallel HEV is the logistic threshold value control. This control strategy is concise and has high code conversion. This control strategy can get better control effect under the condition that engine's and motor's characteristic and the drive cycle are defined. But this control strategy has bad adaptive ability, the thresholds have already been set and they are fixed values, so the complexity and randomness of the drive cycle aren't fully taken into account to HEV's automobile fuel economy and the Balance of SOC. to address these limitations of existing control strategy.Considering the effect of randomness on driving cycle, this paper proposes a new control strategy which based on Markov prediction theory. In the control strategy, researches are emphasized as follows:(1)The first inconsistency: the representation and determinacy of the driving cycle in developing process and the diversity and randomness of the real driving cycle respectively.(2) Predicting the future changing trend of SOC by Markov chain. (3) Optimizing the capacity of the SOC balance and fuel economy.Firstly, by studying the driving cycle, we have found the randomness of the real driving cycle. We also can find that the control strategy which is established according to the determinacy l has many disadvantages under this real drive cycle, and HEV can't show its superiority.Secondly, considering the randomness of the driving cycle, we set the working mode of the hybrid electric bus as the state to predict the future changing trend of SOC by Markov prediction. when it need more electric in the future,it use less electric and add more electric now, when it need less electric in the future ,it use more electric and add less electric now. By this way, it can improve the ability of the balance of SOC. As for ways to improve fuel economy, it predicts the future model and adjusts the corresponding control parameters, So that the power sources work more efficient in the future.Finally,establish prediction model under Matlab/Simulink,then send it to HEV model and adjust the relationship between motor and engine relate to information of prediction. The results show hybrid electric vehicle control strategy based on the Markova prediction can improve fuel economy and maintain the balance of SOC by improving the efficiency of motor and engine. Analysising the result of this control strategy improves fuel economy and promote the capacity of SOC balance.Hybrid electric vehicle control strategy based on the Markov prediction is about guiding energy Distribution by learning the law of using electric under different drive cycles on line. It not only learns drive cycles but also optimizes parameters on line. So, this method can adaptive the stochastic of drive cycles. This thesis is about the hybrid electric vehicle control strategy based on the Markov prediction and settles a foundation of the research of this control strategy,this thesis can advance the control strategy and control method of the HEV in a way,which forms the meanings of this thesis.