摘要
纯模糊控制具有一定的主观性,缺乏自适应能力,使其在燃料电池混合汽车(FCHV)能量管理中的应用受限,由此提出一种基于自适应神经模糊推理系统(ANFIS)的能量管理策略(EMS)。首先利用MATLAB/simulink建立FCHV动力系统简化模型,然后设计ANFIS控制器,采用BP算法和最小二乘估计法对ANFIS的前后件参数进行自适应调节,最后对所提出的EMS进行仿真验证和分析。结果表明,在基于ANFIS的EMS下,蓄电池soc全程波动范围仅为功率跟随策略的70%,且soc初值与最终值的差值更小;此外,相比于功率跟随策略,燃料电池全程工作于低功率点之上,工作于高功率点的时间仅为功率跟随策略的40%,工作于高功率变化率的时间减少了26%。因此,所提出的EMS不仅能有效管理FCHV动力系统中能量的流动,满足负载功率需求,而且有助于延长蓄电池和燃料电池寿命,从而降低整车成本。
Pure fuzzy control has certain subjectivity and lack of self-adaptive ability,which limits its application in the energy management of fuel cell hybrid vehicles(FCHV).Therefore,this work presents an energy management strategy(EMS)based on adaptive neuro-fuzzy inference system(ANFIS).Firstly,the simplified model of FCHV power system is established by using MATLAB/simulink.Then to design ANFIS controller BP algorithm and least squares estimation method are used to adaptively adjust the parameters of ANFIS.Finally,the proposed EMS is simulated and analyzed.The results show that the battery soc’s full-range fluctuation range is only 70%of the power following strategy and what is smaller that the difference between the initial and final values of the soc under ANFIS-based EMS.In addition,compared to the power following strategy the fuel cell works throughout above the low power point,the time working at the high power point is only 40%of the power following strategy,and the time working at the high power change rate is reduced by 26%.Therefore,the presented EMS can not only effectively manage the energy flow in the FCHV power system and meet the load power demand,but also help to extend the lifetime of the battery and fuel cell,thereby reducing the overall vehicle cost.
作者
刘琦
詹跃东
Liu Qi;Zhan Yuedong(Department of Information Engineering and Automation,Kunming University of Science and Technology,Kunming 650504,China)
出处
《电子测量技术》
2020年第7期6-11,共6页
Electronic Measurement Technology
基金
国家自然科学基金项目(51667012)资助
作者简介
刘琦,硕士研究生,主要研究方向为燃料电池混合汽车的能量管理策略研究。E-mail:qliu1272258637@163.com;詹跃东,教授,硕士生导师,主要研究方向为燃料电池混合汽车的能量管理策略研究、燃料电池的建模与优化研究、新能源(氢能、风能、太阳能)发电系统的建模与控制研究等。E-mail:ydzhan@163.com
