Pattern recognition of rigid hoist guides based on support vector machine

Abstract

In order to evaluate the serviceable condition of rigid hoist guides and recognize their status patterns, vibration and angle signals were collected from an test-rig which can simulate different sorts of defects. Characteristic parameters by which status patterns can be judged were selected and designed. Multi-classifier based on support vector machine was applied during recognition. The collected samples were divided into two parts which were used as training data and testing data, respectively. The numerical result verified the validity of support vector machine multi-classifier on pattern recognition of rigid hoist guides.