Abstract
Nonlinear time-varying systems are changing frequently. The precise control of the nonlinear time-varying system is a research hotspot in the machinery area. Digital twins (DTs) can accurately describe or predict the physical world, which is expected to improve the adaptability of the controller for nonlinear time-varying systems. As the starting controller cannot recognize the change of clutch friction coefficient, the starting quality of the power shift tractor (PST) gradually deteriorates. Taking PST as the representative of nonlinear time-varying systems, a controller design method based on DTs was proposed. Firstly, a five-dimensional model of PSTDTs was developed. Then, according to the mechanism model of the starting process, the surrogate model and the updating strategy of the clutch friction coefficient, a virtual PST was established. Secondly, a DTs reference adaptive linear quadratic regulator (DTsRALQR) was designed based on the virtual PST. Finally, a PST digital twin system was built including a physical PST, a virtual PST, a DTsRALQR, and a DTs service platform. Data synchronization, motion synchronization, and adaptation tests were carried out. The test results showed that the virtual PST could synchronize with the physical PST, and its accuracy was increasing. DTsRALQR can adapt to the physical PST state and environmental changes, and it enables the tractor to have a continously good starting quality. In short, this research improves the adaptability of the nonlinear time-varying system controller, and provides a reference for the application of DTs in the field of machinery.
Get full access to this article
View all access options for this article.