A Hybrid Intelligent-Based Speed Tracking Controller of Cement Rotary Kiln Motor Drive System (Case Study: Mugher Cement Factory)

Abstract

The cement rotary kiln is the core equipment in the manufacturing of the cement industry, which is a large cylindrical furnace used for the production of clinker, it is a key ingredient in the cement production process. It has been driven by a high performance of an asynchronous motor and it plays a crucial role for efficient operation of the cement kiln. In this thesis, a case study has been taken from Mugher cement factory, the existing control technique of the kiln motor drive system has a conventional vector control or field orientated control technique using proportional, integral and proportional, integral and derivative controllers. In MCF the existing major problems in the factory such as challenging to handle the uncertain, nonlinear and complex dynamic of the system, leading to suboptimal control performance and in addition it has a speed sensor directly measures the motor speed using a physical sensor. Those all the above mentioned characteristics are degrade the performance of controllers. As a result many controllers fail to smoothly track the speed trajectory. Therefore, appropriate and real time speed controllers are required to attain the required level of control performance. In this thesis, the Model reference adaptive system control and artificial neural network control technique is used for the speed control of the kiln motor drive system. To overcome these challenges, the proposed control technique is designed and its performance evaluated against the model reference adaptive and artificial neural network controller individually. In this thesis, the simulation works have been carried out by using MATLAB/Simulink software setup (Ra2018). Hence, the results showed that the Model reference adaptive system control a tracking control of speed and torque is good, adaptability is good, response time is good, but disturbance handling capability is moderate and rise time is 0.170sec, settling time is 0.134sec and overshoot is 2.24%. On the other hand, artificial neural network controller outperforms speed and torque tracking control accuracy is excellent, response time is excellent, disturbance rejection capability is good, adaptability is excellent and rise time is 0.123sec, settling time is 0.016sec and overshoot is 1.64%. Finally, a hybrid Model reference adaptive system control and Artificial neural network control has a superior performance of an accurate speed tracking control, adaptability is excellent and response time is excellent and rise time is 0.106sec, settling time is 0.014sec and overshoot is 1.25%. In this study, the finding results have been proven that a combined control techniques are effectively regulated and improved that both the required speed control performance for kiln motor drive system.