The control of main-steam temperature is one of the problems at the process control in power plant, the main reasons are: Firstly, the object of main-steam temperature always has the characteristic of time-delayed, and the larger the unit capacity is, the longer the delayed time is , which always makes feedback control seemed too late to act. Secondly, the dynamic characteristic of main-steam temperature will change when the condition changes, which will cause the real controlling effect of the controller with fixed parameters worse. On the other hand, the dynamic characteristic of main-steam temperature object is nonlinear, which also increases the complexity of control.According to these characteristics of main-steam temperature object, such as long time-delayed, nonlinear and model changeable, we have done the research works on comprehensive applying of multiple intelligent control technologies at the control system of main-steam temperature.We brings forward a new kind of state-observer and realize the reconstruction of the ahead temperature signal. The new state-observer is different with the usual gradually-approaching state-observer or reduced-order state-observer,it is attached with the pure time-delayed block, and its structure is more simple and direct. The simulating result shows that: the state on this state-observer is more closer to the state on the real system.By detailedly analyzing the control problems caused by the time-delayed characteristic of the main-steam temperature system, it is clearly stated hereby that: when the delayed-time exceeds a certain level, any form of feedback control is unable to gain a satisfied control result. Meanwhile the fact that a particularly designed intelligent feed-forward effect may solve the long time-delayed problem is brought forward. According to the experience of the skilled operators, we designed a feed-forward controller which is based on the neural network technology for enhancing the main-steam temperature control system.Aiming at the characteristic of uncertainty of the main-steam temperature object, we apply the Fuzzy control technology on the main-steam temperature control system. Above this, we design a dynamic...
|