基于PID的中温型固体氧化物燃料电池稳态控制

Steady State Control of Medium Temperature Solid Oxide Fuel Cell Based on PID

为了解决传统稳态性控制能力较差的问题,提出一种基于模糊PID动态反馈调节的中温型固体氧化物燃料电池稳态控制模型。利用模糊变结构PID神经网络构建被控对象模型,在电池输出增益稳定条件下,采用模糊PID动态反馈调节方程进行控制目标函数分析,根据电池的电压输出权重进行模糊自适应学习,构建电池隐含层权重学习的时滞双曲比例微分调节反馈单元,在边值控制节点中进行电池稳态性输出的鲁棒性训练,求取电池电压输出的最优解,完成稳态控制模型的构建。仿真结果表明,该方法的稳态控制误差平均约为0.06。得出使用该方法进行中温型固体氧化物燃料电池输出控制的稳态性较好,降低了中温型固体氧化物燃料电池的输出振荡。

In order to solve the problem of poor steady-state control ability of traditional solid oxide fuel cell, a steady-state control model of medium temperature solid oxide fuel cell based on fuzzy PID dynamic feedback regulation was proposed. Fuzzy variable structure PID neural network is used to construct the controlled object model. Under the condition of stable output gain of batteries, the fuzzy PID dynamic feedback regulation equation is used to analyze the control objective function. Fuzzy adaptive learning is carried out according to the voltage output weight of batteries, and the time-delay hyperbolic proportional differential control feedback unit for the weight learning of hidden layer of batteries is constructed, which is carried out in the boundary value control node. Robustness training of battery steady-state output can obtain the optimal solution of battery voltage output and complete the construction of steady-state control model. The simulation results show that the steady-state control error of this method is about 0.06 on average. It is concluded that the output control of moderate temperature solid oxide fuel cell using this method has good stability and reduces the output oscillation of moderate temperature solid oxide fuel cell.