| In recent years,major automobile manufacturers have successively launched strategic plans for the development of hybrid electric vehicles,and have continuously increased their investment in research and development of hybrid power systems.In the early stage of hybrid power systems research and development,the economic comparative analysis of configuration schemes based on energy management strategies is a pivotal basis,hence it has attracted much attention from enterprises.Considering the research and development demand of enterprises,this thesis focuses on the research of the multi-objective optimization method and its application in the energy management strategy of hybrid electric vehicles.The main research is stated as follows.Aiming at the fact that the existing multi-objective optimization algorithm lacks of convenience and effectiveness to solve the multi-objective and multi-stage decision-making problem,based on the dynamic programming theory and the basic multi-objective optimization theory,the optimality principle is extended from the traditional single-objective optimization to the research field of multi-objective optimization.A new optimization method,Non-dominated Dynamic Programming(NSDP),is proposed to solve multi-objective and multi-stage decisionmaking problems.However,the insufficient efficiency of NSDP makes it difficult to be applied to the solution of complex system.Therefore,an improved method of NSDP is proposed from three aspects: making full use of computer computing power,improving the efficiency of nondominated sequencing,and reducing the burden of non-dominated sequencing.On this basis,a solution process for applying NSDP to multi-objective and multi-stage decision-making problems is proposed.By applying NSDP to solve the FON test function in the test function set of commonly used multi-objective optimization problems,the feasibility of NSDP to solve multi-objective parameter optimization problems is proved.Applying NSDP and the current mainstream multiobjective optimization algorithms(weighted coefficient method and NSGA-Ⅱ)to solve the Multi-objective Unidirectional Travelling Salesman Problem,the solution results show that NSDP can effectively solve the multi-objective optimal control problem,the solution efficiency is higher,and the quality of the obtained non-dominated solution set is better.Based on the quasi-static backward simulation model of hybrid electric vehicles,taking fuel consumption and battery life as optimization goals,the multi-objective and multi-stage decision-making problems for the energy management of a series-parallel hybrid electric vehicle and a power-split hybrid electric vehicle are constructed respectively,with NSDP to facilitate numerical solution.Compared with the solution results of dynamic programming based on the weighting method,NSDP can easily and effectively obtain the non-dominated solution set of the multi-objective energy management optimization problem of hybrid electric vehicles.In addition,the NSDP is used to solve the multi-objective energy management optimization problem of the power-split hybrid electric vehicle and the series-parallel hybrid electric vehicle under different working conditions.Based on the theory of Multi-objective Optimization,a comparative analysis of the numerical solution results is conducted to analyze the comprehensive economy of the aforementioned hybrid power system.It suggests that NSDP can provide a decision-making basis for enterprises to select appropriate configuration schemes in the early stage of hybrid power system development. |
