Study On Adaptive Equivalent Consumption Minimum Strategy Based On Driving Intention Recognition

With the energy crisis and environmental pollution caused by the burning of fossil fuels in traditional cars becoming more and more serious,people have to turn their eyes to clean and efficient electric vehicles.However,due to the current battery energy density and charging speed and other technical constraints,the transition from traditional vehicles to pure electric vehicles is not achieved overnight.As the transition from traditional vehicles to electric vehicles,hybrid vehicles have become the best choice for energy saving and emission reduction at nowadays.Thanks to the cooperation of internal combustion engine and motor,hybrid electric vehicle has an excellent capacity of energy saving and emission reduction.However,the realization of this cooperation mainly depends on the energy management strategy to distribute the torque of each power source reasonably.Therefore,the research of energy management strategy and the developing effective controller are very important to improve the energy saving and emission reduction capacity of hybrid vehicles.In addition,as the vehicle is driven mainly by artificial operation,the subjective consciousness of the driver will affect the driver’s operation of the vehicle during the driving process of the vehicle,thus affecting the energy management strategy.Based on the above analysis,it is of great practical significance to design and develop the energy management strategy according to the driver’s intention for the improvement of hybrid vehicles.In this paper,the energy management strategy based on driving intention identification is studied with parallel hybrid vehicles as the research object and fuel economy as the optimization goal.(1)Parallel hybrid electric vehicle was choosed as the research object.The structural characteristics and energy flow of hybrid electric vehicle has been analyzed and forward simulation model including engine,motor,power battery,clutch,transmission and other main components has also been established.Because this paper mainly takes fuel economy as the optimization goal,it ignores the dynamic characteristics of the transverse and vertical directions of the whole vehicle,and only models the longitudinal dynamics of the whole vehicle.Morever,a driver model based on the PID controller also been built in the forward simulation model.The driver model takes the deviation between the actual speed and the target speed as the input,follows the speed after PID adjustment,and outputs the pedal command to simulate the real driving situation.The model of vehicle forward simulation provides a model basis for the subsequent development of control strategy based on driving intention identification.(2)In order to identify the driving intention accurately,cloud model algorithm was choosed to identify the driving intention.Driving intention is a qualitative concept expressed in natural language.The fuzzy theory only considers the fuzziness of driving intention and neglects the randomness generated by drivers according to random road conditions.After analyzing cloud model algorithm,it proves that cloud model algorithm is more suitable for driving intention recognition than fuzzy theory.Therefore,in order to obtain some driving simulation data,the typical working condition is obtained by clustering the working condition library and run in the forward simulation model.Then,by synthesizing the existing research,selecting the appropriate driving intention identification parameters and clustering them to get the real-time intention under typical working conditions.Finally,taking the real-time intention and the corresponding identification parameters as the sample parameters of the cloud model to establish the driving intention recognition system based on cloud model.(3)In order to establish an optimal control strategy that conforms to driving intentions,the Adaptive-Equivalent Consumption Minimum Strategy(A-ECMS)based on driving intention recognition is proposed.As a pattern recognition method,driving intention recognition is used to adjust the equivalent factor by matching real-time look-up table.Considering the essence of the equivalent factor is the average fuel-electric conversion efficiency,the weighted average method is used to obtaining the corresponding equivalent factor of each intention,thus establishing a look-up table of equivalent factor.At the same time,use the penalty function to modify the matching equivalent factor to ensure that the SOC does not exceed the allowable limit.(4)In order to verify the proposed A-ECMS based on driving intention recognition,it was simulated and analyzed under different operating conditions,then compared with the ECMS and A-ECMS based on SOC feedback.The results show that under NEDC operating conditions,the fuel economy of A-ECMS based on driving intention recognition is improved by 1.3% and the frequency of SOC fluctuations also be reduced compared to the A-ECMS strategy based on SOC feedback.Under combined operating conditions,A-ECMS based on driving intent recognition has achieved a fuel economy control effect similar to the offline optimized ECMS and the SOC deviation is significantly reduced,proving the effectiveness of the proposed strategy.