نوع مقاله : مقاله پژوهشی
عنوان مقاله English
نویسندگان English
The study of chaotic dynamical systems is a fundamental problem in many scientific fields, as their complex behavior cannot always be readily characterized using classical methods. In this research, in addition to examining existing methods, such as Lyapunov exponents, sensitivity analysis, and contraction and expansion, a geometric approach based on path curvature is studied in order to detect chaotic behavior in dynamic systems. In this approach, after numerically solving the differential equations of each system, the curvature of the trajectory is calculated using the velocity and acceleration vectors. Then, in order to quantitatively analyze the behavior of the systems, we calculate statistical quantities including the number and frequency of events, and compare them in two chaotic and non-chaotic states. In addition, the occurrence times of curvature peaks and significant changes in their values are investigated as indicators of dynamical changes in the system. These statistical quantities can effectively distinguish between chaotic and non-chaotic behaviors. The proposed approach is applied to three dynamical systems: the Lorenz system, the T-system, and the generalized Sprott-C system. The results obtained for all three systems demonstrate that trajectory curvature, together with the proposed statistical measures, can serve as an effective geometric approach for detecting chaotic behavior in nonlinear dynamical systems.
کلیدواژهها English