No abstract

No abstract

AbstrakSistem double inverted pendulum digunakan untuk mengkaji metode kontrol dari sistem nonlinear yang kompleks. Pada sistem tersebut ditemukan tiga permasalahan, yaitu swing-up, stabilisasi, dan tracking. Pada penelitian ini digunakan metode LQR-Bat Algorithm untuk mengoptimasi kestabilan pada sebuah double inverted pendulum. Hasil penelitian menunjukkan penggunaan Bat Algorithm pada metode LQR dapat menyelesaikan permasalahan stabilisasi. Pada iterasi ke 300-400 terjadi pengurangan nilai fitness yang menghasilkan fungsi SSE bernilai semakin kecil. Kata-kata kunci: double inverted pendulum, Bat-Algorithm, LQR AbstractThe double inverted pendulum system is used to examine the control methods of complex nonlinear systems. In the system three problems were found, namely swing-up, stabilization, and tracking. In this study, the LQR-Bat Algorithm method is used to optimize stability in a double inverted pendulum. In this study, the LQR-Bat Algorithm method is used to optimize stability in a double inverted pendulum. The results show that the use of Bat Algorithm in the LQR method can solve stabilization problems. n the 300-400 iteration there is a decrease in fitness value which results in a smaller value SSE function.

No abstract

Particle swarm optimization (PSO) is an optimization algorithm that is simple and reliable to complete optimization. The balance between exploration and exploitation of PSO searching characteristics is maintained by inertia weight. Since this parameter has been introduced, there have been several different strategies to determine the inertia weight during a train of the run. This paper describes the method of adjusting the inertia weights using fuzzy signatures called signature PSO. Some parameters were used as a fuzzy signature variable to represent the particle situation in a run. The implementation to solve the tuning problem of linear quadratic regulator (LQR) control parameters is also presented in this paper. Another weight adjustment strategy is also used as a comparison in performance evaluation using an integral time absolute error (ITAE). Experimental results show that signature PSO was able to give a good approximation to the optimum control parameters of LQR in this case.