Research On Calibration Method Of Vehicle Control Unit For Plug-in Hybrid Electric Vehicles

The Hybrid electric vehicle(HEV)is one of the most important driving forms and it has become a research hotspot in the current global automotive industry.The Vehicle Control Unit(VCU)is considered to be one of the core components of hybrid electric control system.In the actual development,it is necessary to develop and optimize the control strategy according to powertrains with different configurations.VCU calibration is a key step in the development process.Calibration quality affects vehicle performance,and calibration efficiency is related to the development cycle and cost control.At present,domestic and international research on the calibration of VCU is focused on the development of calibration tools and the re-development of supporting facilities;there are very few open studies of calibration method for VCU,which is not conducive to the development of the core technology of VCU.The development of calibration tools is different from that of calibration methods.Efficient calibration methods coupled with advanced calibration tools are the guarantee of calibration efficiency and calibration quality.Although VCU calibration is a practical project,its operation method can not only rely on the experience of calibration engineers by the Try-error way.How to scientifically improve the efficiency and accuracy of VCU calibration is the focus of this study.In order to explore the following calibration method,the VCU calibration system is firstly established,and the factors and responses of the system are determined.Taking a pre-research type PHEV as the research object,the vehicle simulation model is set up and the vehicle control system is developed.The one-dimensional control parameters are extracted based on analysis of vehicle energy management strategy,and the feasible region of each parameter is calculated by the way of theoretical analysis combined with experience.Energy consumption test methods of light hybrid electric vehicle are compared and analyzed.Based on which,an economic evaluation method suitable for this study is worked out.In this paper,the simulation model and vehicle control model are taken as the calibration system,the extracted control parameters are used as the input factor of calibration system,and the economic evaluation index is put forward as the output response of the calibration system.In order to identify the effective calibration factors and reduce the workload of calibration,a sensitivity analysis based on test scheme design and agent model is proposed.Firstly,the optimal Latin hypercube design method is used to sample the calibration factors in their feasible domain,and the experimental design is carried out.In this way,the overall design space characteristics are preliminarily explored.Secondly,in order to establish a mathematical model of calibration system and improve computational efficiency,an agent model of calibration system with higher accuracy is built by using the radial basis function neural network method;Finally,the sensitivity analysis of all calibration factors is carried out by using Monte Carlo method,and the main effect index and total effect index of each parameter are obtained.In order to provide theoretical guidance for the calibration engineers to adjust the direction of parameter calibration,and to reduce the optimal design space of parameters,the research on parameter calibration path and local sensitive confidence interval are proposed.The calibration path diagram of each parameter is obtained by using the master effect diagram method,and the difference of the response of each parameter is shown intuitively.Referring to the definition of confidence intervals in sample statistics,the concept of local sensitive confidence intervals for calibration parameters and its solution method are put forward to further reduce the global optimal feasible interval of each parameter.In order to meet the requirements of VCU calibration under different driving conditions,the influence of driving conditions on VCU calibration and adaptive calibration of driving conditions are put forward.The differences of calibration results under three different working conditions are studied based on the mechanism analysis of calibration process,and the suitability function of the calibration results is proposed to evaluate the applicability of the calibration results under the given conditions.At the same time,in order to explore a VCU calibration method suitable for multi-driving conditions,the calibration platform of driving-condition adaptive for VCU is set up,which facilitates the calibration operation for calibration engineers under different driving conditions.Finally,the differences between the adaptive calibration method and the traditional experimental calibration method are analyzed from the aspects of calibration efficiency,calibration cost and calibration effect.At the end of this paper,the calibration method of the VCU is verified by the drum test bed.Taking the most sensitive calibration parameter as an example,in order to verify the feasibility and practical value of the proposed VCU calibration method,the vehicle economy performance in the calibration model and the real vehicle calibration system is analyzed emphatically.