Research On Start-stop Control Strategy Based On Velocity Prediction

As one of the key technologies for energy saving and emission reduction of vehicles,the intelligent idle start-stop system reduces the idle time of the engine without affecting the normal driving needs of the driver through automatic stop and quick start functions,thereby reducing the fuel consumption of the entire vehicle.However,due to the influence of road traffic,vehicles equipped with a start-stop system often start and stop the engine frequently in a short period of time,which not only accelerates the loss of parts,but also the noise and vibration problems during the start-up process will affect the driver’s driving experience.In order to make the intelligent idle start-stop system adapt to different driving condiction and reduce the phenomenon of invalid shutdowns and frequent start-stop,on the basis of identifying the current driving conditions of the vehicle,this paper propose a method of controlling the engine based on the predicted vehicle speed to optimize the logic controlled start-stop system.In order to reduce the number of invalid shutdowns and improve fuel economy,this paper develops an optimized control strategy for the idling start-stop system based on vehicle speed prediction.The main research contents of this paper are:(1)Based on the analysis of the technical solutions of different idling start-stop systems,the enhanced starting motor starting scheme was selected as the research object.Study the system components,control signals,and basic control strategies based on simple logic control.Aiming at the deficiency of invalid shutdown in this strategy,an optimized control strategy based on vehicle velocity prediction is proposed.(2)According to the average speed and idling times,three(a total of six)typical driveing cycle were selected as the research object,and a data set was prepared by extracting 10 parameters,such as the maximum and minimum speeds,that could reflect the characteristics of the driving conditions through Compound bisection.Part of the data is used to train the extreme learning machine and the bat algorithm is used to optimize the network parameters of the extreme learning machine.The remaining data is used as a test set to verify the effectiveness of the extreme learning machine in the recognition of operating conditions.The simulation results show that the extreme learning machine optimized by the bat algorithm has a recognition accuracy of 93% forvarious operating conditions.(3)Taking the vehicle speed time series as the input object,establish a NAR neural network model to predict the vehicle speed.Use the BP algorithm to adjust the network parameters in the open-loop mode,and finally determine the network delay order and the number of hidden layer neurons by trial and error.Different NAR neural networks are trained using different speed sequences,and based on the recognition of the operating conditions,the corresponding NAR neural network is used to predict the speed of the vehicles under different driving conditions.(4)In the end,this paper establishes a vehicle fuel consumption model,and simulates different strategies for the start-stop system under different driving cycle.The simulation results show that the start-stop control strategy based on the vehicle speed prediction can effectively avoid the occurrence of invalid shutdowns,and ultimately improve the overall fuel efficiency of the vehicle.