The invention discloses a multi-power-source vehicle energy management strategy optimization method, system and device, and relates to the technical field of new energy automobiles, and the method comprises the steps: simulating a current vehicle state according to a standard working condition and a vehicle model; based on the genetic topology neural network, determining the proportion of the motor distribution demand power; calculating the motor torque based on the proportion, and calculating the consumed power of the motor according to the motor torque; calculating reward values of all individuals in the genetic topology neural network based on the consumed power; and determining an optimal individual based on the reward value to obtain an optimal energy management strategy. According to the invention, a neural network based on gradient update weight, which is commonly used in a multi-power-source vehicle energy management strategy (EMS), is replaced by a genetic topology neural network (NEAT) based on an evolutionary update topological structure and weight, so that the debugging period of the EMS can be shortened, energy flows among vehicle power sources are distributed through the optimal EMS, and the energy flow distribution efficiency is improved. And the best economical efficiency of the vehicle is achieved under the condition that the requirement for dynamic property is met.
本发明公开了一种多动力源车辆能量管理策略优化方法、系统及设备,涉及新能源汽车技术领域,该方法包括:根据标准工况和车辆模型模拟当前车辆状态;基于遗传拓扑神经网络,确定电机分配需求功率的比例;基于比例计算电机转矩,并根据电机转矩计算电机的消耗功率;基于消耗功率计算遗传拓扑神经网络中所有个体的奖励值;基于奖励值确定最优个体,得到最优能量管理策略。本发明将现在普遍用于多动力源车辆能量管理策略(Energy Management Strategy,EMS)中的基于梯度更新权重的神经网络替换为基于进化更新拓扑结构和权重的遗传拓扑神经网络(Neuro Evolution of Augmenting Topologies,NEAT),能减少EMS调试周期,通过最优EMS对车辆动力源之间的能量流进行分配,使车辆在满足动力性的要求下达到最佳的经济性。
Multi-power-source vehicle energy management strategy optimization method, system and equipment
一种多动力源车辆能量管理策略优化方法、系统及设备
2023-11-10
Patent
Electronic Resource
Chinese
IPC: | B60L PROPULSION OF ELECTRICALLY-PROPELLED VEHICLES , Antrieb von elektrisch angetriebenen Fahrzeugen / G06F ELECTRIC DIGITAL DATA PROCESSING , Elektrische digitale Datenverarbeitung / G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen |
European Patent Office | 2024
|Hybrid electric vehicle energy management strategy optimization method
European Patent Office | 2021
|POWER SOURCE MANAGEMENT SYSTEM, VEHICLE AND POWER SOURCE MANAGEMENT METHOD
European Patent Office | 2024
|European Patent Office | 2023
|European Patent Office | 2025
|