Research on LPF-GSMC Control of Molten Metal Crane Based on Multi-mass-spring-damping Model

• Published in: International journal of control, automation, and systems 2024, Automation, and Systems, 22(1), , pp.136-150
• Main Authors: Wang, Tianlei, Tan, Nanlin, Zhou, Jing, Qiu, Jiongzhi, Zhang, Xianwen, Liang, Yanyang
• Format: Article
• Language: English
• Subjects: Control; Control methods; Controllers; Cranes; Damping; Dynamic characteristics; Engineering; Hardware; Liquid metals; Liquid sloshing; Load; Low pass filters; Mathematical models; Mechatronics; Regular Papers; Robotics; Robustness (mathematics); Sliding mode control; 제어계측공학
• ISSN: 1598-6446; 2005-4092
• DOI: 10.1007/s12555-021-0377-5

The molten metal crane faced with high difficulty to control can be mainly attributed to the features as a typical underactuated system and liquid sloshing phenomenon is often ignored in the existing control methods. Consequently, this paper firstly establishes the multi-mass-spring-damping mechanics model of liquid sloshing, followed by deducing the dynamics equation of the molten metal crane system based on the model in detail, which verifies the correctness of the system model from the physical sense. Then, a low pass filter global sliding mode control (LPF-GSMC) is introduced to effectively eliminate the high frequency signals of the system and keep the system on the sliding mode surface. Lastly, a hardware experiment and three kinds of numerical simulations were designed for the molten metal crane system. The hardware experiment shows that compared with the traditional solid-load model, the proposed model is more consistent with the dynamic characteristics of liquid load during transportation. And the numerical simulations show that compared with the conventional sliding mode controller and low-pass filter sliding mode controller, the proposeed LPF-GSMC controller possesses higher response speed and better control performance. Also, under the interference of wind and the non-zero initial angular, the proposed controller can ensure that the maximum load swing angle is less than 2° degrees and converges within 7 seconds, which fully demonstrats favorable global robustness.