基于径向基函数神经网络的燃料电池温度非线性建模与预测

针对现有的熔融碳酸盐燃料电池（MCFC）模型过于复杂的弊端，本文应用RBF神经网络辨识方法建立了MCFC的温度非线性模型。简要分析了MCFC电堆的温度特性，讨论了应用RBF神经网络进行多输入／多输出非线性系统建模的主要问题，并详细给出了其辨识结构，算法和模型训练方案，应用仿真对建模的有效性和建模精度进行了检验，并与BP神经网络辨识的效果进行了对比，仿真结果证明RBF神经网络远比BP神经网络收敛得快，应用RBF神经网络辨识方法对MCFC电堆建模是可行的，它避免了用复杂的微分方程组来描述MCFC，通过神经网络可快速地得到其输入同特性，它为MCFC温度的在线预测和在线控制奠定了基础。     

Development and Experimental Research of kW—scale Molten Carbonate Fuel Cells

A kW-scale moten carbonate fuel cells stack was developed and 800-hours‘ operating test and performance experimental research had been done.Utilizing domestic materials completely,we developed NiO cathode and Ni-Al aonde with the active area of 336cm^2 and γ-LiAlO2 electrolyte tile and bipolar plate with the area of 900cm^2,The stack was composed of thirth cells,with 62% Li2CO3 38% K2CO3 as its electrolyte,During the 800hours continuous operating,the performance of the stack was stable.With 99.7%(mole fraction)H2 as fuel and O2 from air as oxidant,the average operating voltage of a cell was about 0.72V.The maximal current density attained to 165mA/cm^2.and the maximal output power attained to 1080Watt.The whole perfomance of the stack approached to the international level in the early 90‘s ,This paper gives the main works and experimants results.     

Neural—networks—based Modelling and a Fuzzy Neural Networks Controller of MCFC

Molten Carbonate Fuel Cells(MCFC) are produced with a highly efficient and clean power generation technology which will soon be widely utilized.The temperature characters of MCFC stack are briefly analyzed.A radial basis function (RBF) neural networks identification technology is applied to set up the temperature nonlinear model of MCFC stack,and the identification structure,algorithm and modeling training process are given in detail.A fuzzy controllery of MCFC stack is designed.In order to improve its online control ability,a neural network trained by the I/O data of a fuzzy controller is designed.The neural networks can memorize and expand the inference rules of the fuzzy controller and substitute for the fuzzy controller to control MCFC stack online,A detailed design of the controller is given,The validity of MCFC stack modilling based on neural networks and the superior performance of the fuzzy neural networks controller are proved by Simulations.     

基于神经网络辨识的熔融碳酸盐燃料电池建模

针对熔融碳酸盐燃料电池（MCFC）电堆系统难以建模以及已建立的模型过于复杂，难以满足工程上对MCFC系统控制设计特别是实时控制需要的情况，绕开MCFC的内部复杂性，提出利用神经网络可能逼近任意复杂非线性函数的性能，将神经网络辨识方法应用到MCFC这种高度非线性系统的建模中。以燃料气和氧化剂气体的流速的输入量，MCFC电堆温度响应为输出量，根据输入输出数据用神经网络辨识建立MCFC电堆系统的温度模型，给出了辨识系统的结构及改进BP算法。仿真结果证明了这种方法的可行性，所建立的模型精度较高，使得设计MCFC的实时控制器成为可能。     

对边简支加筋三明治板隔声性能研究

针对加筋三明治板声学特性的研究主要集中于无限大结构、很少考虑边界条件影响的问题,结合脉冲函数的使用,提出了对边简支、层芯为平行筋板的三明治板的动态响应力学模型,利用模态函数法推导了结构的传声损失计算公式.将理论和模拟结果进行了对比,二者吻合较好,验证了模型的有效性.所提出的计算方法相对于有限元数值方法具有更高的计算效率,能在较宽的频段范围内给出精确的结果,且便于分析结构传声的物理机制和结构参数对传声性能的影响.研究结果表明:简支边界在隔声曲线上引入了结构的共振隔声波谷;加筋三明治板结构中同样存在明显的吻合共振效应;随着入射声波频率的提高,筋板惯性效应的影响越发明显.