微型燃气轮机的新型神经网络控制的研究  

On New Neural Network Control of Micro Gas Turbine

作　　者：闫大朋[1] 闫士杰[1] 李爱平[1] 张化光[1] 

机构地区：[1]东北大学电气自动化研究所,辽宁沈阳110004

出　　处：《控制工程》2008年第5期541-543,548,共4页Control Engineering of China

基　　金：国家"863"重大专项基金资助项目(2002AA503020)

摘　　要：燃机控制系统是一种多变量、非线性、时变的系统,对微型燃气轮机的转速控制器进行了深入研究。PID控制应用广泛,但在实际应用中,其参数整定仍未得到较好的解决。因此,设计了一种新的神经网络PID控制器作为主控制器,通过神经网络所具有的任意非线性表达能力,可以通过对系统性能的学习来实现具有最佳组合的PID控制,确保系统的稳定性、快速性和准确性。大量的仿真证明,该算法具有良好的控制效果。To the turbine control system that is a kind of multivariable, nonlinear, time-varying system, the speed controller of micro gas turbine is studied. PID controllers are used in a large number of industries, particularly in process industries. However, up to now, there are no satisfactory solutions about the tuning of PID parameters. Therefore, a kind of new PID controller based on neural network is designed as main controller. Due to the fact that neural network has the ability of expressing nonlinearity, the optimal combination of PID controllers is gotten to assure system stability, celerity and veracity by studying the system performance itself. Simulation results show that the algorithm has good control performance.

关 键 词：微型燃气轮机 仿真 神经网络 调节器 

分 类 号：TP27[自动化与计算机技术—检测技术与自动化装置]