A new approach with adaptive network-based fuzzy inference systems (ANFIS) based on experimental designs was used to model and characterize the tribological behaviors of diamond-like carbon (DLC) films deposited by a magnetron sputtering system. An orthogonal array experiment was introduced and the effects of deposited parameters on the films were systematically explored. The films were analyzed by X-ray photoelectron spectroscopy (XPS) and scanning electron microscopy (SEM). In this study, a group of highly developed hillock-like textures appeared and a lower wear volume loss became visible in the DLC films. Furthermore, the predicted values and experimental results, in which the ANFIS effectively predicts the tribological behaviors of the DLC films, are similar. It was experimentally confirmed the ANFIS predictions agreed with the experiments. Therefore, the experimental results demonstrate the tribological properties on DLC multilayer films are accurately predicted by ANFIS, thereby justifying the reliability and feasibility of the approach.
Probabilistic Risk Assessment of Rotor Angle Instability Using Fuzzy Inference Systems