Real time monitoring of manufacturing processes using a single sensor often poses significant challenge. Sensor fusion has thus been extensively investigated in recent years for process monitoring with significant improvement in performance. This paper presents the results for a monitoring system based on the concept of classifier fusion, and class-weighted voting is investigated to further enhance the system performance. Classifier weights are based on the overall performances of individual classifiers, and majority voting is used in decision making. Acoustic emission monitoring of tool wear during the coroning process is used to illustrate the concept. A classification rate of 87.7 % was obtained for classifier fusion with unity weighting. When weighting was based on overall performance of the respective classifiers, the classification rate improved to 95.6 %. Further using state performance weighting resulted in a 98.5 % classification. Finally, the classifier fusion performance further increased to 99.7% when a penalty vote was applied on the weighting factor.
Acoustic emission (AE) is introduced for tool condition monitoring during the coroning process. The frequency components of the AE signal were used as features for classification. Two different feature selection methods were investigated, namely visual observation and the class mean scatter criterion. The minimum error rate Bayesian rule was used to distinguish between two extreme tool conditions. Although the features from visual observation could result in 100% classification, features based on the class mean scatter criterion showed excellent monitoring capability of tool failure when fewer features were used.
It is often difficult for a single classifier to achieve perfect classification during process monitoring. Sensor fusion enables the final decision to be improved, but uses voting methods, which usually do not perform well when there is a tie vote. In this paper, classifier fusion with class-weighted voting is investigated to further enhance the performance of monitoring systems. The overall performances of individual classifiers are used as the weighting factors to classifier fusion based on majority voting. When applied to tool wear monitoring of the coroning process, the classifier that was based on overall performance weighting improved the classification rate to 95.6% and the one based on state performance weighting showed 98.5% classification, compared to 87.7% for classifier fusion with unity weighting. A classifier fusion further increased performance from 98.5% to 99.7% by applying a penalty vote on the weighting factor.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.