微生物燃料电池性能分析与控制方法研究

发布时间：2018-04-14 09:37

本文选题：微生物燃料电池 + 神经网络模型预测　；参考：《兰州理工大学》2017年硕士论文

【摘要】：生物质和污水中蕴藏着巨大的可再利用能源,这些可再利用能源在能源枯竭的今天隐藏着巨大的利益,微生物燃料电池(microbial fuel cell,MFC)为此能源的利用提供了一条可行的途径。为了提高系统的工作性能和工作效率,对微生物燃料电池进行控制方法的研究具有重要的意义。本文主要对微生物燃料电池的性能进行分析,通过性能的分析引出对微生物燃料电池优化控制的关键切入点:对阳极和阴极进料流量的控制;然后以负载电流作为一个重要影响因素,检验针对阴极、阳极进料流量所设计的控制器的有效性。本文主要研究内容如下:1)简述微生物燃料电池进行性能分析与控制方法研究的意义和背景。首先描述了微生物燃料电池机理与组成结构并简单描述了电池的关键指标,然后引出并搭建起微生物燃料电池系统模型,随后对系统模型的性能进行仿真分析,研究各个参数对系统输出性能的影响,得到对系统进行优化控制的关键切入点:阳极和阴极的进料流量。2)针对微生物燃料电池非线性、超调量大和响应速度慢的特性,提出了神经网络模型预测控制的方法,主要对阳极进料流量进行控制。利用随机阶跃信号激励系统模型得到训练数据,并用于训练神经网络,然后将得到的神经网络模型作为模型预测控制的参考模型,通过反馈校正和滚动优化实现对阳极进料量的神经网络模型预测控制,仿真研究表明相对于传统PID控制方法本文所提出的神经网络预测控制方法超调量小、鲁棒性较强。3)针对微生物燃料电池多运行于稳态操作点的特点,在稳态点处对系统进行线性化,利用分数阶PIλDμ控制器灵活、设计方便的优势,对微生物燃料电池进行分数阶PIλDμ控制的研究。通过分析分数阶PIλDμ控制控制器的性能,针对阳极和阴极线性化模型分别设计分数阶PIλDμ控制器,然后将设计好的分数阶PIλDμ控制器应用于微生物燃料电池系统,仿真研究表明微生物燃料电池系统在本文所设计的分数阶PIλDμ控制器的作用下具有超调量小、响应速度快等优点。4)针对微生物燃料电池稳态操作点发生变化时系统输出电压不稳定问题,提出了多模型切换的模型预测方法。对阳极和阴极在稳态点处进行线性化得到线性化状态方程,根据线性化模型设计模型预测控制器,并将其应用于非线性微生物燃料电池系统模型;然后,针对稳态操作点变动时系统电压输出不稳定的问题,提出非恒定负载电流下的多模型切换预测控制,通过跟踪系统运行状态,采用不同的切换策略切换至相应的模型预测控制器,仿真结果表明当负载电流变化时,所提出的多模型切换预测控制能够很好地实现不同稳态操作点间的切换,能较好的实现电压的平稳输出。
[Abstract]:Biomass and sewage contain huge reusable energy, which hides huge benefits in energy depletion today. Microbial fuel cell (microbial fuel) provides a feasible way to make use of this energy.In order to improve the performance and efficiency of the system, it is of great significance to study the control method of microbial fuel cell.In this paper, the performance of microbial fuel cell is analyzed, and the key point of optimization control of microbial fuel cell is introduced: the control of anode and cathode feed flow;Then the load current is taken as an important factor to verify the effectiveness of the controller designed for cathode and anode feed flow.The main contents of this paper are as follows: (1) the significance and background of performance analysis and control methods for microbial fuel cells are briefly described.Firstly, the mechanism and structure of microbial fuel cell are described, and the key parameters of the cell are briefly described. Then, the model of microbial fuel cell system is introduced and built, and then the performance of the system model is simulated and analyzed.The influence of each parameter on the output performance of the system is studied, and the key entry point for optimizing the control of the system is obtained: the feed flow rate of anode and cathode. 2) aiming at the characteristics of microbial fuel cell, such as nonlinear, large overshoot and slow response speed.A neural network model predictive control method is proposed to control the anode feed flow.The training data are obtained by using the stochastic step signal excitation system model and used to train the neural network. Then the obtained neural network model is used as the reference model of the model predictive control.The neural network model predictive control of anode feed is realized by feedback correction and rolling optimization. The simulation results show that compared with the traditional PID control method, the neural network predictive control method proposed in this paper has less overshoot than the traditional neural network predictive control method.In view of the fact that microbial fuel cells often run at the steady-state operating point, the system is linearized at the steady-state point, and the fractional Pi 位 D 渭 controller is used to make use of the advantages of flexible fractional Pi 位 D 渭 controller and convenient design.The fractional-order Pi 位 D 渭 control of microbial fuel cells was studied.By analyzing the performance of fractional Pi 位 D 渭 controller, a fractional order Pi 位 D 渭 controller is designed for anode and cathode linearization models, and then the designed fractional Pi 位 D 渭 controller is applied to microbial fuel cell system.The simulation results show that the microbial fuel cell system has a small overshoot under the action of the fractional Pi 位 D 渭 controller designed in this paper.(4) aiming at the instability of system output voltage when the steady operating point of microbial fuel cell changes, a multi-model switching model prediction method is proposed.Linearized the anode and cathode at the steady point to obtain the linearized equation of state. According to the linearization model, the predictive controller is designed and applied to the nonlinear microbial fuel cell system model.In order to solve the problem of unstable voltage output when the steady operating point is changing, a multi-model switching predictive control under unsteady load current is proposed to track the operating state of the system.Different switching strategies are used to switch to the corresponding model predictive controller. The simulation results show that the proposed multi-model switching predictive control can achieve the switching between different steady-state operating points well when the load current changes.It can realize the steady output of voltage.
【学位授予单位】：兰州理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP273;TM911.45

【参考文献】