基于多种群模糊遗传算法的PEMFC分数阶建模与控制

发布时间：2018-01-20 01:19

本文关键词： PEMFC 多种群模糊遗传算法分数阶建模分数阶PI~λD~μ控制器　出处：《南京理工大学》2017年硕士论文　论文类型：学位论文

【摘要】：随着社会的发展和人类生活方式的改变,能源短缺及环境污染已成为困扰人类社会可持续发展的重大问题。质子交换膜燃料电池(proton exchange membrane fuel cell,PEMFC)作为二十一世纪一种清洁高效的发电装置,具有发电效率高、启动速度快、噪音污染少、使用寿命长等优点,已被广泛使用。为了更好的提高PEMFC系统输出性能,在PEMFC建模、优化和控制等方面,还有待进一步的研究。本文以空冷型PEMFC为研究对象,针对燃料电池运行过程中所存在的分数阶特性,以氢气流量为输入、负载电流为扰动建立起了分数阶的PEMFC动态模型,并提出了多种群模糊遗传算法(multi-population fuzzy genetic algorithm,MPFGA)对模型中的未知参数进行寻优。基于辨识后得到的PEMFC分数阶模型,利用多种群模糊遗传算法对其进行分数阶PI~λD~μ控制器的设计,使电堆输出电压始终保持在最佳工作点附近。仿真结果表明该控制器能有效地控制氢气流量,使电堆在负载变化时的输出电压能一直保持在期望值附近。本论文主要创新点与内容如下:(1)针对分数阶系统的求解过程中存在的问题,提出了基于改进Oustaloup滤波器的框图法。该方法能通过在Simulink中搭建框图的形式,有效地求解分数阶微积分方程。(2)分析PEMFC运行过程中存在的分数阶特性,以氢气流量为输入、负载电流为扰动,建立起了分数阶PEMFC动态输出电压模型。(3)针对分数阶PEMFC动态输出电压模型中参数多,整定困难的问题,在遗传算法的基础上提出了一种多种群模糊遗传算法。基于多种群模糊遗传算法对分数阶PEMFC动态输出电压模型进行辨识,将辨识所得到的模型与实际电堆进行比较,结果表明基于该算法辨识后所得到的分数阶PEMFC动态输出电压模型能比较精确地描述实际电堆运行状态。(4)针对辨识后所得到的分数阶PEMFC动态输出电压模型,采用基于多种群模糊遗传算法设计的分数PI~λD~μ阶控制器对其进行控制,使PEMFC输出电压在实际运行过程中始终保持在理想电压点附近。仿真结果表明基于多种群模糊遗传算法所设计的分数阶PI~λD~μ控制器能有效地控制氢气流量,从而使PEMFC输出电压能迅速地达到并且保持在理想电压值附近。
[Abstract]:With the development of society and the changes of human life style, energy shortage and environmental pollution has become a major problem for the sustainable development of human society. The proton exchange membrane fuel cell (proton exchange membrane fuel cell, PEMFC) in twenty-first Century as a clean and efficient power generation device, has high power generation efficiency, high starting speed, less noise pollution, the advantages of long life, has been widely used. In order to improve the output performance of the system in PEMFC, PEMFC modeling, optimization and control, need to be further studied. In this paper, air cooling type PEMFC as the research object, according to the fractional characteristics of existing operation process of the fuel cell, hydrogen flow rate as input to load current disturbance, establish a dynamic model of PEMFC fractional order, and put forward the fuzzy multi population genetic algorithm (multi-population fuzzy genetic algorithm, MPFGA) Optimization of the unknown parameters in the model. PEMFC fractional order models are obtained after the identification based on the design, using a variety of groups of fractional order PI~ D~ controller of the lambda fuzzy genetic algorithm, the stack output voltage is always maintained at near optimal operating point. The simulation results show that the controller can effectively control the flow of hydrogen the stack, the output voltage when the load changes can be kept near the expected value. The main innovations of this dissertation are as follows: (1) in the solving process for the fractional order system of the problem, this paper proposes an improved Oustaloup filtering method based on the wave diagram. This method can build block diagram in Simulink in the form of effectively solving fractional differential equations. (2) analysis of fractional characteristics of PEMFC during the operation, the hydrogen flow rate as input, load current disturbance, establish the fractional order dynamic output power PEMFC Pressure model. (3) according to the multi parameter fractional PEMFC dynamic output voltage model, setting difficult problem based on genetic algorithm is proposed for multi population genetic algorithm. Fuzzy multiple fuzzy genetic algorithm for dynamic output voltage of the fractional PEMFC model were identified based on the identified model and the actual stack were compared. The results show that the fractional PEMFC output voltage dynamic model obtained based on the algorithm can accurately describe the actual stack operation. (4) the fractional PEMFC dynamic according to the identification of the output voltage model, a multi population genetic algorithm to design the fuzzy fractional PI~. D~ order controller to control the output voltage based on the PEMFC in the actual operation process is always maintained at near ideal point voltage. The simulation results show that the multi population genetic algorithm based on fuzzy design The number order PI~ lambda D~ controller can effectively control the flow of hydrogen, so that the output voltage of the PEMFC can be quickly reached and kept near the ideal voltage value.

【学位授予单位】：南京理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TM911.4

【相似文献】