Modelagem e aplicação de compensação de tempo morto para malha de controle de taxa de minério de ferro em um alimentador de sapatas

Abstract

In the ore extraction companies, the mass control in each ore treatment unit is indispensable. From the data provided by instruments installed in the field, the control and management of the plant can be done through industrial controllers. Logical controllers are able to interpret the functional descriptive of the process, making automatic decisions to make commands on equipment and actuators. Among the programming algorithms of the industrial controllers, there are specific logics for process control. These logics include the use of PID controllers (proportional, integral and derivative), that in several applications only proportional and integral control (PI) can be used. Controller tuning requires prior knowledge of the dynamics of the process to be controlled, but several factors may hamper the stability of the control loop. For those cases where the pure process time lag is dominant, new control strategies must be developed. This work describes the development of a dead time compensation system based on Smith's predictor in a mass flow control mesh, where the manipulated variable is the speed of a Shoe Feeder. For the elaboration of the control structure with the dead time compensator, techniques of system identification were used to represent the process. The output signals of the control mesh and the process variable were collected to estimate the parameters of the model. The least squares method was used to estimate the parameters of the mathematical model. After the test results, the control structure considering Smith's predictor was implemented in the PLC (Programmable Logic Control). From the results obtained, it is concluded that Smith's predictor improves the performance of the control mesh and eliminates the overshoot of the transient response of the process. In this way, the tuning of the control mesh allows a strategy disregarding the existence of the pure time delay of the system.