Fault Diagnosis of Bearings Based on the Fusion of 1D-CNN and Random Forest Algorithm

As industrial automation continues to advance, bearing faults have become one of the significant causes leading to mechanical equipment failures, potentially resulting in substantial economic losses. Traditional fault diagnosis methods often struggle to effectively address increasingly complex fault patterns. This study proposes a method for bearing fault diagnosis based on the fusion of one-dimensional convolutional neural network (1D-CNN) and random forest algorithm. Our approach combines the ability of 1D-CNN to capture local patterns in time-series data and the high performance of random forest in classification tasks. We conducted experiments on the bearing dataset from the Case Western Reserve University. The experimental results demonstrate that our method outperforms other popular diagnostic methods across multiple evaluation metrics, particularly in terms of recall, where it significantly surpasses other methods. This provides an efficient and reliable tool for bearing fault diagnosis.