机构地区: 广东工业大学
出 处: 《自动化仪表》 2012年第8期51-54,共4页
摘 要: 针对PID神经元网络(PIDNN)将静态神经元扩充到动态神经元的特性,通过分析PIDNN控制算法存在的局限性,对变量系统的激励函数提出了改进方案,扩大了PID神经网络控制的适用范围。Matlab仿真测试表明,改进后的PIDNN系统具有良好的稳态性,试验验证了算法的有效性。 Due to PID neural network (PIDNN) possesses the characteristics of expanding static neuron to dynamic neuron, thus through analyzing the limitation of the PIDNN control algorithm, the improved strategy for the excitation function of variable parameters is proposed to extend the applicable range of PID neural network control. The Matlab simulation test shows that the improved PIDNN system possesses excellent stability ; the effectiveness of the algorithm is verified through experiments.
领 域: [自动化与计算机技术] [自动化与计算机技术]