摘要
鉴于BUCK型降压斩波变换器的非线性和复杂性,采用传统的PID控制器难以获得理想的控制性能,利用神经网络的非线性函数逼近能力,构造神经网络PID自整定控制器。该控制器根据系统的运行状态,通过神经网络的自学习、加权系数调整,自动调节PID控制器参数,以期达到性能指标的最优化。将该控制器应用于BUCK型变换器,仿真结果显示该控制器在满足系统的性能指标方面取得了满意的效果,表明神经网络PID控制器是一种具有快速跟踪能力的自适应控制器。
In the light of nonlinear and complexity characteristics of BUCK chopper converters, the ideal control performance is difficult to he achieved using the conventional parameter tuning techniques for PID controller. Nonlinear function approximation capability of neural network is used to construct neural network PID self-tuning controller. According to the system operating state, the ability of self-learning and weights adjustment of neural network is used by the contruller to automatic adjust PID controller parameters in order to achieve the optimal performance. The controller is applied to BUCK converter. The simulation results show that satisfactory effect is achieved using the controller to meet system performance. This indicates that the neural network PID controller has the ability of fast tracking and self-tuning.
出处
《电气传动自动化》
2011年第4期20-23,共4页
Electric Drive Automation
