基于非对称多变量PID神经网络的三相Z源光伏并网逆变器解耦控制研究

DECOUPLING CONTROL OF THREE-PHASE Z-SOURCE PHOTOVOLTAIC GRID-CONNECTED INVERTER BASED ON ASYMMETRICAL MULTI-VARIABLE PID NEURAL NETWORK

初步研究Z源逆变器用于光伏并网逆变系统中的强耦合和非线性问题,探索光伏阵列的最大功率跟踪(Maximum Power Point Tracking-MPPT)和直流侧稳压统一控制的方法,采用非对称输入输出多变量PID神经网络控制,抑制直流侧电压扰动,通过三相电流跟踪解耦控制实现MPPT和单位功率因数并网。仿真实验结果表明,该控制策略能有效抑制由光伏阵列输出变化、电网波动及其他干扰对直流侧电压的扰动。

Considering the nonlinear characteristics of the system, explore the solar array maximum power point tracking （Maximum Power Point Tracking-MPPT ） and DC voltage regulator unified control method, an asymmetrical input and output multi-variable PID neural network control strategy is used in this paper, with the ability to inhibit disturbance of DC voltage, and three-phase current decoupling control is applied to achieve MPPT tracking and high power factor gTid-connected. The simulation results show that this control strategy can effectively suppress the output from the PV array change, power fluctuations and other disturbances on the DC side voltage.