Mill Load Optimal Control Based On DMC

As the main equipment of milling system in cement factory, Ball mill consumes lots of energy and has a low milling rate, so it is very urgent to achieve mill burthen automatic control and improve milling rate. Because grinding mill system has the characteristics of multi-variable, strong coupling, nonlinear serious, ball control mill load is very important and difficult in cement process. At present, the widely used mill load control methods is PID, which difficult to adapt to the conditions within the ball mill and cannot achieve the purpose.From 70'last century, predictive control developed rapidly and was bused in industry when people found new optimal control arithmetic with low quality model, high quality control affect and simple calculate online. Predictive control arithmetic has strong adaptability with complex system, can design system without complex identification, adopt incessant optimize substituted traditional optimal control, enhanced robustness in optimal process, all of those characters make the predictive control suitable to practical requirement in industrial process control.This dissertation is relied on the National High-Tech Research Development Plan (863) subject: soft measurement, optimization unit-level control system design, development and engineering applications (NPS integrated cement production process control applications) issues. Aim at the Pingyin product line of Sunnsy Group, the control and controlled variable were total feedstock, circle blower rev, weighing, hoisting current. Grinding process model is obtained with history data by least square method and DMC scheme is realized and regulated the parameters, besides analyzed the stability of. DMC control system, finally obtained the parameter condition for stabilization of system.Based on VS Studio 2005 the mill model identification software and control software were written with CPP programming language. In typical situation the SISO model reflect the relationship between input and output preferably. Roller loop and milling loop that strong coupling and effect with each other make up the mill control system, it is difficult too satisfy the requirement to control one loop lonely, so this dissertation adopt the strategy of harmonize the two loop. In abnormal situation chose the suit model to reduce the influence caused by model mismatch.With the OPC application in the DCS system, the control software read and wrote data successfully, and then delivery the control measure through DCS. The set point and bound of output were adjusted according the situation of field. The grinding circuit automatic control is realized.The operation result show that multi-variable, strong coupling and nonlinear problems were solved by using Dynamic Matrix Control, while ensuring product quality and stability ,which means the optimal control of milling load was established.