Research On Energy Management Strategy Of Fuel Cell Vehicle Based On Driving Condition Prediction

With the advantages of zero emission and high efficiency,fuel cell vehicle is considered as the ideal model of future automobile.Besides the fuel cell system,fuel cell vehicles generally have other energy sources,so in order to improve the economic performance of the vehicle,a reasonable energy management strategy is essential.In this paper,fuel cell vehicle with fuel cell and power battery are used as the research object,combined with the historical trip data of a driver,and the characteristics analysis and classification of historical trip data,the energy consumption analysis of fuel cell vehicle based on typical working conditions,and the energy management strategy based on driving condition prediction are studied.This paper relies on the provincial and university co-construction project “Research and Platform Development of Key Technologies of High Specific Power Fuel Cell Engine”,firstly,this paper analyzes the energy source and topological structure of fuel cell vehicles,determines the indirect structure of fuel cell and power battery as the vehicle configuration,analyzes the working principle of the fuel cell system,built a complete fuel cell system model in AMESim,and obtains the best output characteristics of the fuel cell.In order to establish the foundation for the follow-up research,the fuel cell vehicle model is built in the Matlab/Simulink platform,including driver model,fuel cell model and power battery model,motor model and vehicle dynamic model.Secondly,this paper collects a driver’s historical driving condition data,preprocesses the acquired data,mainly including bad data processing and filtering,divides the preprocessed condition data into segments,and calculates the characteristics of each segment parameters,uses the principal component analysis method to reduce the dimensionality of the characteristic parameters.According to the principal component analysis results,the improved K-means algorithm is used to cluster the working condition data,and the historical driving condition data is divided into four types.Thirdly,based on the typical driving cycles,the whole vehicle’s energy consumption of fuel cell vehicles is studied.For the convenience of subsequent research,combined with the classification results of typical driving conditions,the driving condition recognition method based on learning vector quantization is studied,which can identify the driving conditions.The correlation between the characteristic parameters of four types of typical working conditions and the energy consumption of the whole vehicle is studied,and the specific characteristic parameters related to various working conditions and energy consumption are obtained.Based on this,the method of multiple linear regression is used.The energy consumption fitting equations of various typical operating conditions are obtained,and the theoretical energy consumption curve of actual driving conditions can be generated in combination with the identification method of operating conditions.Finally,the energy management strategy research based on driving condition prediction is carried out.A trip condition prediction method that integrates traffic information and historical travel characteristics is studied,which can realize the prediction of future travel mileage and trip conditions,and the vehicle speed prediction of each road segment in the future is studied in a virtual traffic scenario.A global energy planning strategy based on the predicted conditions is proposed.The strategy plans the working time of the fuel cell system,which can further obtain the global predictive driving cycle’s reference SOC trajectory of the power battery.Taking the differences between actual driving condition and predicted driving condition into account,an online control strategy considering the reference SOC trajectory of the power battery is proposed.Through the simulation test of the strategy in this paper and the rule-based strategy,the results show that the strategy proposed in this paper can make the fuel cell work more in the high efficiency zone,the equivalent hydrogen consumption of the vehicle is reduced by 5.83% compared with the rule strategy,so as to achieve the goal of improving the economic performance of the vehicle.