Power Demand Forecasting Method and Its Application in Parallel Hybrid Electric Vehicle Powertrain Control

Haidi Sun, Fengchun Sun, Junqiu Li, Qingyun Min, Chao Sun

Abstract


For hybrid electric vehicles (HEVs) , vehicle velocity and road gradient are essential. They can decide the driving power demand and have a great impact on its powertrain energy management performance. This paper presents a power demand forecasting method, aiming to predict the short-term future velocity and road gradient in real-time for the predictive energy management of HEVs. The artificial neural networks (ANNs) and the autoregressive integrated moving average(ARIMA) learning model are established with the data from expereiments to predict the vehicle velocity and the road gradient during each control horizon. Simulation results indicate the predictors can reflect the varying tendency of the velocity and the road gradient perfectly, validating the good performance of the method. Then we employ Model Predictive Control (MPC) to build the predictive energy management strategy, and solve the optimal powertrain control problem during each control horizon with Dynamic Programming (DP), getting the HEV fuel economy improved by about 5%.

Keywords


Hybrid Electric Vehicle; Power Demand; ANNs; ARIMA; Road Gradient Prediction


DOI
10.12783/dteees/iceee2018/27847

Refbacks

  • There are currently no refbacks.