Abstract
This paper presents a first-phase project toward designing a digital twin for a smart crane on a ship. Digital twin technologies create a virtual model of a physical system to simulate its behavior and aid in its operation. The long-term goal of this project is to integrate the finite element method, the multi-body dynamics method, artificial intelligence, and 3D visualization for the creation of digital twins. The short-term objective of this first phase, and the focus of this paper, is to use the moving frame method (MFM) in dynamics as a foundation for such an integration. The MFM relies on Lie group theory (distilled to the simplicity of matrix algebra) and Cartan’s notion of Moving Frames. A compact notation simplifies the kinematics. In addition to the underlying method, this paper presents a systematic and coherent way to create the frame connection matrices for multi-bodies. Then, with an appropriate restriction on the variation of the angular velocity, the MFM extracts the equations of motion using the Principle of Virtual Work. These are solved using an adaptive Runge-Kutta-4. Following this, the paper presents a systematic way to isolate each body in the linked system and solve for all contact forces (the latter in preparation for the deployment of the finite element method). The solution of the system drives a software simulation of the crane’s motion using WebGL and ThreeJS. Finally, the method and analysis is validated. A crane scale model was constructed using 3D-printed parts and electric motors for motion. Load cells were integrated into the base to measure the reaction forces. This paper reports on the first phase.
