Abstract:Deep neural networks has been widely used in industrial equipment fault diagnosis. The accuracy of deep neural network is usually proportional to the complexity, but the high inference delay and energy consumption caused by the complex model make it difficult to be applied in the industrial environment of real-time demand. At the same time, in the diagnosis of industrial equipment, different categories of samples have unbalanced characteristics in terms of number, difficulty of identification, and demand of identification. In order to solve this problem, this paper designs Multi-Branch Neural Network (MBNN), which is a new type neural network architecture that can use the unbalance of sample categories in industrial equipment fault diagnosis for fast inference. MBNN has multiple sub-networks with different complexity, and each branch is responsible for processing different categories of samples. Categories with large numbers, easy to process, and high demand of identification are processed through simple branches, such as normal samples. Categories with small numbers, difficult to identification, and low demand of identification are processed through complex branches, such as potential failure samples. The feasibility of MBNN has been verified on motor bearing fault diagnosis and gearbox fault diagnosis, and its performance has been evaluated on multiple computing platforms. The results show that MBNN can greatly improve the inference speed while ensuring the recognition accuracy, especially on resource-constrained platforms.

What problem does this paper attempt to address?

The paper primarily addresses the challenges faced in the application of deep neural networks (DNN) for industrial equipment fault diagnosis, particularly issues related to real-time performance and resource constraints. The authors propose a new architecture called the Multi-Branch Neural Network (MBNN). ### Problems the Paper Aims to Solve: 1. **Improving Inference Speed**: Although deep neural networks have high accuracy in industrial equipment fault diagnosis, their complexity leads to high inference latency and energy consumption, limiting their application in industrial environments that require real-time response. 2. **Utilizing Sample Imbalance Characteristics**: In industrial equipment fault diagnosis, there is an imbalance in the number, recognition difficulty, and recognition needs of different categories of samples. The authors leverage this imbalance to design a new neural network structure capable of fast inference. ### Solutions: - **MBNN Architecture Design**: MBNN includes multiple sub-networks (branches), each responsible for handling different types of data. Simple branches handle easily recognizable and more numerous samples (such as normal samples), while complex branches are used for difficult-to-recognize or less numerous samples (such as potential fault samples). - **Fast Convergence Training Algorithm**: To ensure effective training of MBNN, the paper also proposes a training algorithm that utilizes the regularization terms naturally provided by the simple branches and closer backpropagation gradients, helping to address issues like overfitting and gradient vanishing. - **Multi-Level Resource Adaptability**: The different complexity branches of MBNN enable it to perform inference and training on various levels of hardware resources, such as in cloud-edge collaborative computing environments. ### Experimental Validation: - **Motor Bearing and Gearbox Fault Diagnosis**: The effectiveness and performance of MBNN are validated through fault diagnosis cases of motor bearings and gearboxes. - **Multi-Platform Evaluation**: The paper conducts experimental evaluations on various computing platforms, including laptops and embedded devices (such as Jetson Nano and Raspberry Pi 3B+), to demonstrate the reliability and practicality of the MBNN method. ### Main Contributions: 1. By leveraging sample imbalance through the design of multiple branches with different complexities, the model's inference speed is improved, and computational load is reduced. 2. The effectiveness of MBNN and its feasibility on multiple platforms are validated through real-world cases. 3. A training algorithm that aids in the fast convergence of MBNN is proposed, addressing issues like overfitting and gradient vanishing. 4. MBNN can perform inference and training on devices with different resource levels, supporting applications in cloud-edge collaborative computing environments.

MBNN: A Multi-Branch Neural Network Capable of Utilizing Industrial Sample Unbalance for Fast Inference

Multibranch and Multiscale Dynamic Convolutional Network for Small Sample Fault Diagnosis of Rotating Machinery

Fault Diagnosis Method for Rotating Machinery Based on Multi-scale Features

A multi-scale collaborative fusion residual neural network-based approach for bearing fault diagnosis

A Multidepth Step-Training Convolutional Neural Network for Power Machinery Fault Diagnosis Under Variable Loads

Deep Transfer Network With Multi-Kernel Dynamic Distribution Adaptation for Cross-Machine Fault Diagnosis

Intelligent Health Monitoring of Machine Tools Using a Bayesian Multibranch Neural Network

Fault Diagnosis for Modular Multilevel Converter (MMC) Based on Deep Learning: An Edge Implementation Using Binary Neural Network

A motor bearing fault diagnosis method based on multi-source data and one-dimensional lightweight convolution neural network

A Multi-Layer Spiking Neural Network-Based Approach to Bearing Fault Diagnosis

One-Dimensional Multi-Scale Domain Adaptive Network for Bearing-Fault Diagnosis under Varying Working Conditions

Double Branch Rebalanced Network for Class Imbalance Learning in Process Fault Diagnosis

Multi-Source Information-Based Bearing Fault Diagnosis Using Multi-Branch Selective Fusion Deep Residual Network

An Efficient Lightweight Neural Network using BiLSTM-SCN-CBAM with PCA-ICEEMDAN for Diagnosing Rolling Bearing Faults

A novel fault diagnosis method for imbalanced datasets based on MCNN‐Transformer model in industrial processes

MBSNN: A Multi-Branch Scalable Neural Network for Resource-Constrained IoT Devices

An Anti-Noise Convolutional Neural Network for Bearing Fault Diagnosis Based on Multi-Channel Data

Toward End-to-End Bearing Fault Diagnosis for Industrial Scenarios with Spiking Neural Networks

Fault Diagnosis Method for Bearing Based on Attention Mechanism and Multi-Scale Convolutional Neural Network

A Novel Meta-Learning Network With Adaptive Input Length and Attention Mechanism for Bearing Fault Diagnosis Under Different Working Conditions