| Process automation control system is the important part of computer control system for the tandem cold rolling, and it is the most effective control method to ensure the product quality and production efficiency of cold rolled strip. Based on a building project of 1450mm five-stand tandem cold mill, the study was carried out on the framework and function of process control system, online mathematical models, rolling model optimization and rolling schedule development and so on. The softwares for tandem cold rolling process control was developed and applied on line, which have exhibited good performance. The main work is as follows:(1) Process control system for tandem cold rolling was established. Taking the specific control and technological requirements of tandem cold rolling process into account, the structure frame, data flow and call logic for the process control system have been designed, and the process functions such as data communication, data management, coil tracking, model setup and model self-adaptation were developed.(2) Based on the numerical integration method, the micro-unit discrete model on the strip plastic deformation was established. Through mechanical analysis, numerical method of plastic stress distribution was deduced. By accumulating micro-unit in deformation zone, the formulas of rolling force, rolling moment and rolling power were established. According to the coupling of rolling force and roll flattening radius, rolling force parameters calculation process and iterative algorithm were given. Through an example, the pressure distribution of deformation area was analyzed and the accuracy of the model was verified. In order to improve the model calculation accuracy, a method of testing the motor power loss and a new mill stretch model were designed. Based on a five-stand tandem cold mill, mill motor power loss and bounce model coefficients have been achieved by using actual test data.(3) According to the fact that it’s difficult to adapt to the rolling condition changing with a fixed adaptive gain coefficient in the exponential smoothing algorithm of rolling force model, a dynamic adaptive coefficient adjustment method which is based on actual data was presented, and adaptive learning method of rolling force model, power model, forward slip model were focused. For the purpose of improving the rolling force model precision, a parameters adaptation method based on objective function was put forward. During adaptation process, the deformation resistance and friction coefficient were taken as the optimized parameters, by searching the optimal solution of the objective function, the adaptation coefficients of deformation resistance and friction coefficient can be received.(4) In terms of tandem cold mill productivity and product quality, a multi-objective optimization model of rolling schedule was proposed. For simplifying algorithm, a penalty item was brought into the multi-objective function, and with the penalty item, the multi-objective function was changed from constraint problem to unconstraint problem. Based on establishing rolling force, shape, electrical power, reduction ratio, tension and other single objective function, the total objective function consists of five single objective has been established. In order to obtain the stand reductions, inter-stand tensions, the objective function was minimized using the Nelder-Mead simplex method. The proposed method can make optimal rolling schedule and get rid of dependence on the value of experience.(5) Using the actual values, the control results of mathematical model, rolling schedule and thickness control accuracy were analyzed.The results of the study have practical significance for online control of tandem cold rolling process. The process control and model setup system have been applied successfully in five-stand tandem cold mill of a certain cold mill plant. It is beneficial to the development of cold rolling process control system with independent intellectual property rights in China. |
PDF Full Text Request