摘要
为实现等效能耗最小策略中等效因子的实时调整,提出一种基于自适应等效能耗最小的能量管理策略。首先,设计一种基于多种群自适应协同粒子群优化算法的最优等效因子提取方法,该方法为双层优化的结构。在上层优化中,以船舶的运行成本、储能系统最终电量和初始电量误差最小为目标函数,求解燃料电池系统和储能系统的最优运行轨迹;在下层优化中,建立等效因子的优化模型,提取最优等效因子的分布。然后,建立以系统状态参数为输入、等效因子为输出的神经网络模型。利用最优的等效因子作为训练样本,对神经网络模型进行训练。最后,将神经网络模型与等效能耗最小策略相结合,可实现等效因子的实时调整。在Matlab/Simulink中搭建船舶混合能源系统的仿真模型,对基于自适应等效能耗最小的能量管理策略进行验证。仿真结果表明,与基于恒定等效因子的等效能耗最小策略相比,储能系统的最终电量更接近初始值,氢气的总消耗量降低1.98%。
In order to realize the real-time adjustment of the equivalence factor in the equivalent consumption minimization strategy,an energy management strategy based on adaptive equivalent energy minimization is proposed.Firstly,an optimal equivalent factor extraction method based on multiple swarm adaptive cooperative particle swarm optimization algorithms is designed,which is structured as a two-layer optimization.In the upper-layer optimization,the optimal operation trajectories of the fuel cell system and the energy storage system are solved with the objective functions of minimizing the operating cost of the ship and minimizing the error between the final charge and the initial charge of the energy storage system;In the lower level optimization,the optimization model of the equivalence factor is established to extract the distribution of the optimal equivalence factor.Then,a neural network model with system state parameters as inputs and equivalent factors as outputs is established.The neural network model is trained using the optimal equivalent factors as training samples.Finally,the neural network model is combined with the equivalent energy minimization strategy to realize the real-time adjustment of the equivalence factor.A simulation model of the ship hybrid energy system is built in Matlab/Simulink to verify the energy management strategy based on adaptive equivalent energy minimization.The simulation results show that the final charge of the energy storage system is closer to the initial charge and the total hydrogen consumption is reduced by 1.98%compared with the equivalent energy minimization strategy based on a constant equivalence factor.
作者
许晓彦
曹伟
韩冰
Xu Xiaoyan;Cao Wei;Han Bing(Logistics Engineering College,Shanghai Maritime University,Shanghai 201306,China;College of Physics and Electronic Information Engineering,Minjiang University,Fuzhou 350108,China;Shanghai Ship and Shipping Research Institute Co.,Ltd.,Shanghai 200135,China)
出处
《太阳能学报》
北大核心
2025年第3期108-115,共8页
Acta Energiae Solaris Sinica
基金
福建省自然科学基金(2022J011128)
上海市优秀学术和技术带头人计划(23XD1431000)。
关键词
燃料电池船
能量管理策略
神经网络
等效因子
多种群自适应协同的粒子群优化算法
fuel cell ship
energy management strategy
equivalent factor
neural network
multiple swarms self-adaptive and cooperative particle swarm optimization algorithm
