For fault diagnosis of automotive assembly, the experience knowledge of experts is still the primary means which is inconvenient and time-consuming. This paper proposes a novel fault diagnosis method for automotive assembly based on optical coordinate data and machine learning. First, after obtaining large amounts of measured data from sensors, a feature selection method which is based on the subset-level score is performed. Because high-dimensional data will result in high computational cost and irrelevant and redundant features may also degrade the performance of fault diagnosis. The feature selection method which can be classified into filter methods needs to choose a certain evaluation criterion first. By using a fast iterative algorithm, we can finally find the optimal subset of features. Second, for the nonlinear relationship between measured data, we introduced well-known kernel methods to efficiently improve feature extraction effect. In machine learning, kernel methods are a class of algorithms for pattern analysis, whose best known is the support vector machines. Last, Bayesian inference classifier is built to recognize the fault pattern. The experimental on real production process data of automotive assembly show that the proposed method can greatly enhance the diagnosis accuracy.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.