基于isPSO的锅炉燃烧系统PIDNN解耦控制  

PIDNN Decoupling Control of Boiler Combustion System Based on IsPSO

作　　者：张勇[1] 朱晶[2] 

机构地区：[1]辽宁科技大学电子与信息工程学院,辽宁鞍山114051 [2]鞍山师范学院数学与信息科学学院,辽宁鞍山114051

出　　处：《控制工程》2014年第3期338-342,共5页Control Engineering of China

基　　金：国家自然科学基金重点项目(60874017)

摘　　要：针对传统的解耦方法对实际工业生产过程中的多变量、非线性、强耦合系统解耦效果不理想的问题,提出了改进的简化粒子群(isPSO)算法与PID神经网络(PIDNN)相结合的方法。PIDNN训练用于消除回路间的耦合,其连接权值由简化粒子群算法学习优化。该isPSO算法能克服PIDNN易陷入局部收敛的缺点,而且与基本PSO算法相比,搜索到最优值的概率更高。采用的小步长线性递减惯性权重和增加的极值扰动算子,则加速了对PIDNN权值的优化。通过对强耦合对象火电厂锅炉燃烧系统的仿真表明,该方法具有更好的控制品质:鲁棒性强、跟踪快、解耦效果好,为实际应用中强耦合系统的改进提供了理论依据。Aiming at the problem that the effect of the traditional decoupling method is not ideal for a muhivariable, nonlinear, strong - coupling system in the actual industrial production process, the application of an improved simple Particle Swarm （isPSO） algorithm to optimize PID neural network （PIDNN） is proposed. PIDNN training is used to eliminate the interactions between the loops, and the simplified particle swarm algorithm is used to optimize the weights. Different from the traditional back -propagation （BP） algorithm, the isPSO can overcome the shortcoming of reaching local optimum in early stage. Compared with the basic PSO, it can search the optimum value more accurately. Small step in a long - term decreasing weights and increased extreme disturbance operators are used to accelerate optimization of the PIDNN weights. It shows that the method has better control quality through the simulation of the strong coupling object of boiler combustion system： strong ro- bustness, fast track, good deeoupling effect , which provides a theoretical proof for strong coupling system in practical application.

关 键 词：锅炉燃烧系统 解耦控制 粒子群算法 PID神经网络 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]