摘要 |
PROBLEM TO BE SOLVED: To provide an operation control method and an operation controller for a treatment plant, introducing prediction model control of the plant, optimally and automatically updating a change of resposiveness or controllability, maintaining stable operation for a long term without executing parameter setting or readjustment by an experienced operator, and capable of exerting a control purpose and performance. SOLUTION: In this operation control method for treatment plant facilities having a multi-input/multi-output system, various kinds of process data 34 by operation of the waste treatment plant facilities are introduced to a neural network (a boiler generation steam amount prediction neural network 31), a model of operation control of the waste treatment plant facilities is learned by the neural network to produce a learning model, an operation value of the prediction operation control (for making a difference between a generation steam amount prediction vale S1 and a control target generation steam amount S2 zero) after a prescribed time is calculated by use of the learning model by a primary combustion air flow rate calculation part 32, and the prediction operation control is performed. COPYRIGHT: (C)2005,JPO&NCIPI
|