An autonomous, adaptive optimization control system with cooperative fuzzy reasoning and a neural network for large scale process operations

A control system which can adapt to varying process dynamics due to varying operating conditions is proposed. This system finds optimum control solutions by cooperative fuzzy reasoning and a neural network, based on actual operational data obtained through an iterative process. The fuzzy rules represent qualitative knowledge as expertise on causal relations between process dynamics and desirable modification of control set points. The varying operational conditions are taken into consideration by the neural network to modify the control set points. The system is applied to a simulation of a gas and steam combined cycle power plant and the following points are concluded: 1) the system can optimize the start-up schedule under varying conditions of ambient temperature and humidity; 2) start-up time and energy loss are reduced by 35.6% and 26.3%, respectively, compared with the conventional off-line start-up scheduling method; and 3) the cooperative optimization can make good use of the operational margins, i.e. machine thermal stress and NOx emission