基于FNN的直驱式波浪发电系统PID控制

PID Control of Direct Drive Wave Power Generation System Based on FNN

本文在分析浮子水动力模型和永磁同步线性发电机模型的基础上,得到了最大功率捕获条件。然后,将改进的模糊神经网络PID (RFNNP)控制算法应用于最大功率点跟踪控制。通过对模糊神经网络结构的分析,在模糊神经网络中加入Relu激活函数,构建了RFNNP控制。将模糊规则与神经网络的自学习能力相结合,根据海况变化调整PID参数,使RFNNP算法更准确地逼近非线性目标模型。仿真结果表明,采用该策略的波能转换系统是有效可行的。

Based on the analysis of float hydrodynamic model and permanent magnet synchronous linear generator model, the maximum power capture conditions are obtained. Then, the improved fuzzy neural network PID (RFNNP) control algorithm is applied to the maximum power point tracking control. By analyzing the structure of fuzzy neural network, adding Relu activation function into the fuzzy neural network, RFNNP control is constructed. The fuzzy rules are combined with the self-learning ability of neural network, and PID parameters are adjusted according to the change of sea state, so that RFNNP algorithm approximates the nonlinear target model more accurately. The simulation results show that the wave energy conversion system using this strategy is effective and feasible.