부산대학교 도서관

The Model Order Reduction has indeed gained significant attention due to the advancements in automation and control system engineering. The need of more accurate reduced-order models arises for achieving effective control of the process or the plant. The objective of MOR is to reduce the complexity of higher order model into lower order model while preserving the system's critical characteristics. The two commonly used methods i.e. Factor Division and Dominant Pole Retention are employed for reducing higher-order models. Dominant pole retention selects the dominant poles of the higher-order complex system and transfers them to the desired lower-order system to obtain the characteristic equation of the desired system. Unknown parameters of numerator are computed using factor division algorithm. The effectiveness of the proposed method is verified by comparing it with other recently published works through examples. Further, PID (Proportional-Integral-Derivative) controller design is a well-established technique for controlling dynamic systems. Once a reduced-order model is obtained, a PID controller is designed based on the model's characteristics. Particle Swarm Optimization (PSO) approach has been used to find the unknown controller parameters, resulting in an optimized controller that can provide superior performance to the system. The time domain characteristics and the response of the original and the reduced systems have been examined to test the efficacy of the method proposed in this article. It was observed that the use of PID controller improves overall time-domain performance of the system with negligible steady-state error.