汽包锅炉燃烧的PID型神经网络解耦控制系统仿真研究

The Simulation Study on PID Neural Network Decoupling Control System for Steam Drum Boiler Combustion

介绍了汽包锅炉燃烧控制系统的控制任务、物理量间关系、原则性控制方案、传统控制方法,以及解耦控制方法的发展,引入了PID型神经网络控制系统的结构、学习算法、初始权值参数选取.MATLAB仿真结果表明,该控制系统对多变量强耦合时变的锅炉燃烧控制对象具有良好的解耦性能和自学习控制特性.

The control tasks, the relations between physical variables, the fundamental control schemes, the conventional control methods of the control system for steam drum boiler combustion, and the development of decoupling control method are discussed, and the structure, the learning algorithm, the initial weight selection for PID neural network system are introduced. The MATLAB simulation results prove that the control system has perfect decoupling and self-learning control performance for multi-variable, strong-coupled, time-varying boiler combustion control system.