This paper is based on iron and steel works sintering moisture control as the background, the application of neural network PID controller and a neural network identifier and Smith predictor of sintering mixture water automatic control model, simulated by MATLAB. The sintering mixture water automatic control lag problem is more serious in the process, through the Smith predictor to solve. The moisture measurement technology available, contact measurement of moisture measurement accuracy is more assured, but the contact sensor and material produced by the wear will shorten the service life of the sensor, so in order to meet the water content of sintering mixture measurement precision, prolong water detection interval, can effectively reduce the moisture of sinter the detection cost. In this paper, batching system working characteristics and moisture control requirements of detection, batching system in sinter mixture moisture control strategy is studied through simulation. From the result of the simulation shows that the neural network PID control algorithm can achieve better control effect than the conventional PID control, with no Smith predictive compensation control simulation results contrast can be obtained with Smith predictor is a good solution to the mixture of water control in the presence of lag problem, to automatically set the sampling interval is proposed in this paper, through the simulation to achieve the desired effect. |