Research On Energy Management Strategy For Hybrid Electric Vehicle Based On Driving Intention Recognition

With the aggravation of energy crisis and environmental problem,energy-saving and environmentally friendly vehicles have become a critical step for automotive industry’s facing to this challenge.The development of battery technology restricts the substitution of pure electric vehicles for traditional fuel vehicles.Then,hybrid vehicles that are more energy-efficient than traditional fuel vehicles have become an important development direction for the automotive industry to deal with energy problems.As the core of hybrid vehicles,the development of vehicle control strategies is the focus of hybrid vehicle research and development.Demand torque as the input of control strategy affects the division of driving model and the load of powertrain,thus affects the power and economy of the vehicle.Therefore,it is of great significance to introduce driving intention recognition to obtain more accurate driving demand torque.This paper takes a power split hybrid vehicle as the research object to carry out research on driving intention recognition and drive control strategy.The main research contents are as follows:First,according to the layout of the hybrid vehicle’s powertrain,a vehicle model including the engine,power battery,motor/generator,drive train and vehicle dynamics model was built under Matlab/Simulink.A simple rule control strategy was used to verify the accuracy of the model.Under NEDC,the speed error range-0.4km/h～+0.4km/h,Maximum error speed-0.53km/h,indicate that the model accuracy is good and can be used as a basis for energy management strategy research.Secondly,collect vehicle dynamics test data as sample set and test set data,extract accelerator pedal opening and pedal opening rate as characteristic parameters,build driving intention recognition model based on support vector machine theory。Then the model parameter was optimized by grid optimization algorithm.Using 75 sets of test data to verify the accuracy of model classification,the results show that only 3 sets of sample data are misclassified,and the accuracy of driving intention recognition reaches 96%.Then,a driving force fuzzy control strategy is proposed.Based on the driving intent recognition result and vehicle speed constraints,the pedal position was calculated by the virtual pedal displacement coefficient.Thus calculating the driving force required to meet the driver’s expectations.Comparing the start-up accelerated real-vehicle test and simulation results shows that the introduction of driving intention recognition and driving force fuzzy control,the speed change can be more in line with driver expectations.Finally,a mode switching strategy based on driving intent recognition is designed according to the driving intent recognition result,vehicle speed,and battery SOC threshold.For the simultaneous driving conditions of the engine and the motor,the equivalent fuel consumption minimum strategy theory is introduced to coordinate the torque distribution of the engine and the motor.Under Matlab/Simulink environment,the simulation test of NEDC,CLTC and WLTC standard cycle conditions was carried out.The comparison of simulation and actual vehicle test results showed that the vehicle has good speed following characteristics and the fuel consumption of vehicles under NEDC conditions decreases by 8.85%per 100 kilometers,the fuel consumption of vehicles under CLTC conditions decreases by 8.68%,and vehicles under WLTC conditions achieves 100 kilometers fuel consumption decreased by 6.41%,indicating the effectiveness of the energy management strategy and effective fuel economy improvement in this paper.