CNN-Based Fault Detection for Smart Manufacturing

Abstract

A smart factory is a highly digitized and networked production facility based on smart manufacturing. A smart manufacturing plant is the result of intelligent systems deployed in the factory. Smart factories have higher production volumes and are prone to machine failures when operating in almost all applications on a daily basis. With the growing concept of smart manufacturing required for Industry 4.0, intelligent methods for detecting and classifying bearing faults have become a subject of scientific research and interest. In this paper, a deep learning-based 1-D convolutional neural network is proposed using the time-sequence bearing data from the Case Western Reserve University (CWRU) bearing database. Four different sets of data are used. The proposed method achieves state-of-the-art accuracy even with a small amount of training data. For the sensitivity analysis of the proposed method, metrics such as precision, recall, and f-measure are determined. Next, we compare the proposed method with a 2-D CNN that uses two-dimensional image illustrations of raw data as input. This method shows the effectiveness of using 1-D CNNs over 2-D CNNs for time-sequence data. The proposed method is computationally inexpensive and outperforms the most complex and computationally intensive algorithms used for bearing fault detection and diagnosis.

Custom Lightweight Convolutional Neural Network Architecture for Automated Detection of Damaged Pallet Racking in Warehousing & Distribution Centers

“…State-of-the-art models such as those mentioned above are deployed across various domains including healthcare, autonomous vehicles and renewable energy [21]. However, due to the high computational demand of these architectures, deployment infrastructure is usually facilitated via a cloud architecture [22]. Although this is suitable for applications where computational requirements is not the top priority, domains such as manufacturing can have stringent data security and power consumption requirements, hence demanding edge device inference close to the data source.…”

“…Although this is suitable for applications where computational requirements is not the top priority, domains such as manufacturing can have stringent data security and power consumption requirements, hence demanding edge device inference close to the data source. The lack of device deployment within production is due to the fact that all algorithms mentioned above demand a significant number of computational resources [22].…”

Delamination Detection Framework for the Imbalanced Dataset in Laminated Composite Using Wasserstein Generative Adversarial Network-Based Data Augmentation

“…Over recent years, there has been significant research interest in the Two-Dimensional Convolutional Neural Network (2D CNN) model-based fault diagnosis because it has a remarkable ability to automatically extract features from the images [57,58]. However, 2D CNN requires various data preprocessing techniques, especially when used with vibration signals.…”

Moving toward Smart Manufacturing with an Autonomous Pallet Racking Inspection System Based on MobileNetV2

“…Researchers are continuously improving the architecture of these models with the fifth version of YOLO introduced in April 2021, YOLOv5 [21]. Dhiraj et al [22], present a 1-D CNN-based architecture for the detection and classification of bearing faults on time-series data. The research is based on four different datasets with the aim to provide a computationally light-weight solution to problem.…”