Riassunto analitico
Independent cart conveyor system is a new technology that could replace rotary driven mechanical chains and belts in the field of automatic machines. This technology uses advance linear motors for moving several carts on a close-loop path. Each cart contains permanent magnets and can be moved independently one from the other through a variable magnetic field produced by drivers. The carts are connected to a frame by means of several bearings placed on and under a mechanical guide. The bearings may be subject to wear and the condition monitoring of this system is challenging due to the non-stationary working conditions, in fact each cart can have a different motion and load profile that can also change during the production. A multibody model has been developed for healthy and faulty Independent cart conveyor system by using the commercial software Simcenter 3D Motion, which takes into account the motion profile of the cart, the mechanical design of the cart, the expected loads and the type of faults on the roller bearings. In this paper, the multibody model is used to train a machine learning Independent cart conveyor system. The multibody model provides several simulations both of healthy and faulted bearings, which are used to create the training dataset. The input features of the machine learning algorithm are statistical parameters that proved to be effective in the analysis of real vibration data. The final tests were carried out on real vibration data recorded on a test rig and the fault detection algorithm was validated both in faulted and healthy cases in an industrial environment. The aim of this activity is to virtualize the training of a machine learning system for fault detection and to test its accuracy in a real environment. The virtualization could lead to evident advantages for industries, reducing the time to get sufficient historical datasets for machine learning purposes.
|