الفهرس | Only 14 pages are availabe for public view |
Abstract The speed control of induction motor (IM) drives is a very important research topic for many industrial applications. In practice, IM drives are applied and coupled to external mechanical systems to deliver the mechanical energy required for the operation. The mechanical system, in this context, is the uncertain polymer extruder system. It is maybe considered a non-positive pump. In general, the main function of a polymer extruder is the continuous feeding of a high-quality melt output maintaining desired and consistent temperature, pressure, and flow rate. To achieve this function, it is required to effectively manipulate the barrel zone temperatures and the rotating screw speed. Currently, many polymer extruders are equipped with V/f controlled IM drives due to their valuable advantages of easy implementation and low cost. Hence, having accurate speed control of the V/f controlled IM drive applied to the rotating screw is highly required to accomplish stable operation and highquality products. The challenge is to consider variability, nonlinearity, and uncertainty of material, machine, and process variables, which strongly influence the performance of the IM drive. Given the lack of research regarding speed control approaches for V/f controlled IM drives applied to uncertain and perturbed systems, this study aims to develop and validate effective robust, adaptive and intelligent-based speed control approaches for V/f controlled three-phase squirrel cage IM drive applied to polymer extruder system. The proposed control approaches combine features of both fuzzy logic control (FLC) and the robust sliding mode control (SMC) structures to overcome the aforementioned challenges. The main objectives are ensuring stable and accurate tracking performance of the V/f controlled IM drive in the presence of high uncertainty due to modeling errors, parametric variations, and external disturbances. Second, improving the performance in terms of tracking accuracy and dynamic response. Finally, extending the applicability to applications where accurate speed control and moderate dynamic response are required.II In this study two intelligent-based control approaches on basis of a robust SMC approach are proposed and validated. The first proposed controller is the fuzzy SMC (FSMC), which is developed to overcome the chattering problem of the conventional SMC. A fuzzy inference system (FIS) is developed to replace the discontinuous sign function and it acts like a weighing function to reduce the constant and large reaching control gain. However, the control design of FSMC causes undesired overshooting and requires the bound of uncertainties and the parameters of the nominal dynamic model to compute the equivalent control. The second proposed controller is the model-free adaptive gain FSMC, which can eliminate the chattering and it can avoid the requirements of the mathematical modeling of the real system and the knowledge of the upper bounds of the uncertainties in advance. Besides the high performance of the proposed control approach, it has simple structure, easy to implement, and low cost. For the proposed control approach, an improved FIS is developed to minimize the number of fuzzy rules. Second, a new fuzzy-based adaptation law is developed using the improved FIS to estimate the adaptive reaching control gain such that the sliding mode exists. Third, and finally, the equivalent control part is approximated to avoid the requirement of the mathematical model. The asymptotic stability and convergence of the proposed control approaches are proved using the Lyapunov stability method. Second, the proposed controllers are implemented practically using (Arduino Due) embedded system, and then it is realized and tested in real-time using the experimental setup. Third, the experimental results are analyzed using statistical methods, specifically, MAE, MAPE, and RMSE. Finally, the statistical quantitative comparison between the proposed controllers, secondorder SMC, and the conventional PID controller is presented. The experimental and statistical results show and verify the good performance of the proposed controllers in presence of uncertainties. |