Metaheuristic Approach for Bearing Fault Prediction

This paper presents a study on ball-bearing fault prediction using a combination of Genetic Algorithm and Ant Colony Optimization (ACO) algorithms. The selection of features used to develop machine learning plays an important role in building models. Ant Colony Optimization Technique is used for feature selection, Genetic Algorithm for hyper-parameter selection, and comparison of accuracies among different classifiers to choose the best model among others. Experimental results demonstrate the effectiveness of the proposed approach in improving fault prediction accuracy. Random Forest Classifier with Ant Colony Optimization gives an accuracy of nearly 95%.