Event-triggered Proportional-Integral Control for Flexible Linkage Servo System Using Genetic Algorithm
Year: 2024
Type: article
Abstract: A traditional servo system regards the coupling of the motor and load as a rigid connection, it ignores the influence of system flexibility, and thus it hardly meets the requirements of high-precision industrial production.In order to improve the resonance problem caused by flexible coupling in flexible linkage servo systems, this study proposed an eventtriggered proportional-integral control for flexible linkage servo system using genetic algorithm.First, a dynamic model of flexible linkage servo system was established, and the system parameters were identified using a multi-innovation stochastic gradient (MISG) algorithm.Second, a genetic algorithm optimized proportional-integral controller with an event-triggered (ET-GA-PI) mechanism was designed, and the genetic algorithm was introduced to optimize the control parameters.Moreover, an event-triggered mechanism with a fixed threshold was adopted to control the update times of the controller.Finally, the simulation experiment was compared with traditional control methods, such as pole placement proportional-integral control, second-order system engineering design proportional-integral control, and multiplecapacity process proportional-integral control.Furthermore, the effectiveness of the control method was verified.Results demonstrate that, the MISG identification algorithm combined with the current and historical speed input and output information of the flexible linkage servo system effectively improves the accuracy of parameter identification (5.8 times higher than the traditional stochastic gradient identification, and the quantization error is reduced by 21.6 times) and the speed of convergence (21.6 times higher than the traditional stochastic gradient identification).The update times of the proposed ET-GA-PI controller are 1/20 of those of the traditional genetic algorithm proportional-integral control, and the overshoot is only 0.1731%.Compared with traditional control, the proposed ET-GA-PI controller needs less communication resources to meet the needs of complex network control systems.The proposed method has practical implications on meeting the requirements of flexible servo systems in high-precision robots and computerized numerical control machine tools and other engineering applications.
Authors: Zechong Lu, Chuansheng Tang, Hui Wang, Tao Li
Language: English
Subfield: Mechanical Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W4396842772