电站锅炉汽包水位的CMAC神经网络与PID复合控制

提出了电站锅炉汽包水位系统的CMAC神经网络与PID的复合控制策略。将蒸汽流量信号引入到CMAC神经网络中，考虑负荷变化对汽包水位的影响，使其具有前馈补偿能力，以消除“虚假水位”现象。总的控制输出由CMAC神经网络的输出和一个常规PID控制器的输出通过变参数加权综合后作用于被控对象，这样保证在控制初期以及对象特性变化的情况下系统仍然具有良好的调节品质。仿真试验结果表明了该控制策略的有效性和实用性。图6参6

CMAC Neural Network and PID Combined Control of Water Level in Power Plant Boiler Drums

Aiming at water level control of boiler drums in power plant, a strategy of CMAC neural network and PID combined control is presented. Steam flow signals are transmitted to the CMAC neural network, to take charge of load change effects,enabling the system to take pre-feed compensation steps and therewith avoid "fictitious water level". Output signals from the CMAC network and from a conventional PID controler are weighted and integrated to constitute an overall output signal, which then acts on the controlled object. Thus, good regulating performance is guaranted in the initial control stage and also in case of characteristic deviations of the controlled object. Simulation results show that this control strategy is effective and practicable. Figs 6 and refs 6.