基于RBF神经网络的智能负载控制策略研究

Intelligent load control strategy based on RBF neural network

传统用于电力弹簧(ES)控制的PI控制器调节性能较差,且控制方法中未考虑非关键负载突然变化的问题,为解决该问题,根据ES的数学模型和控制电路提出了一种基于径向基函数(RBF)神经网络的智能负载控制方法。利用RBF神经网络算法弥补传统PI控制器参数固定即无法更改的缺点,通过对控制器参数的实时在线调整来减少智能负载失稳情况,确保系统母线电压稳定。在Matlab/Simulink仿真环境中进行仿真验证,结果表明,与传统PI控制相比,文中所提控制策略下的智能负载对关键负载两端电压的调节性能更优。因此,在基于RBF神经网络的PI新型控制策略下的智能负载具有更好的鲁棒性和系统控制能力。

Aiming at the problem that the traditional PI controller used for the control of electric springs has poor adjustment performance and the control method does not take into account the sudden changes of non-critical loads,an smart load control method is proposed based on RBF neural network of the network on the basis of the mathematical model and control circuit of electric spring. The RBF neural network algorithm is used to make up for the shortcomings of the traditional PI controller that the parameters are fixed and cannot be changed. The real-time online adjustment of the controller parameters reduces the intelligent load instability and ensures the stability of the system bus voltage. Simulation verification in the simulation environment of Matlab/Simulink shows that,compared with traditional PI control,the intelligent load under the proposed control strategy has better performance in regulating the system. Therefore,the smart load under the new PI control strategy based on RBF neural network has better robustness and system control capability.