Parallel Machine Learning Algorithm Using Fine-Grained-Mode Spark on a Mesos Big Data Cloud Computing Software Framework for Mobile Robotic Intelligent Fault Recognition

Abstract

An accurate and efficient intelligent fault diagnosis of mobile robotic roller bearings can significantly enhance the reliability and safety of mechanical systems. To improve the efficiency of intelligent fault classification of mobile robotic roller bearings, this paper proposes a parallel machine learning algorithm using fine-grained-mode Spark on a Mesos big data cloud computing software framework. Through the segmentation of datasets and the support of a parallel framework, the parallel processing technology Spark is combined with a support vector machine (SVM), and a parallel single-SVM algorithm is realized using Scala language. In this approach, empirical mode decomposition (EMD) is applied to extract the energy of the acceleration vibration signal in different frequency bands as features. The parallel EMD-SVM approach is applied to detect faults in mobile robotic roller bearings from fault vibration signals. The experimental results show that it can accurately and effectively identify the faults, and it outperforms existing methods based on parallel deep belief network (DBN) and parallel radial basis function neural network under different training set sizes. Fault classification tests are conducted on outerrace and inner-race faults: in both cases, the proposed parallel EMD-SVM outperforms the serial EMD-SVM in terms of the classification accuracy and classification time under different test sizes. For a small number of nodes, the processing time of the proposed Spark model is less than that of Hadoop but close to that of Storm. For 17 slave nodes, the average precision, average recall, and average F1 score of Spark on Mesos in the fine-grained mode reach 97.3, 97.8, and 97.9%, respectively. The parallel EMD-SVM algorithm in the big data Spark cloud computing framework can improve the accuracy of intelligent fault classification, albeit by a small margin, with higher processing speed and learning convergence rate.

A mini-review of machine learning in big data analytics: Applications, challenges, and prospects

“…Another thrilling revelation is that most studies [70,77,79,84,94,97,106,108,[116][117][118][119][120][121] used Deep Neural Networks (DNN), such as LSTM and Convolutional Neural Network (CNN), as shown in Fig. 6.…”

“…We observed that several studies [70,83,85,87,88,92,95,97,104,108,117,119,122,124,125] combined two or more metrics to evaluate their models and framework.…”

A Method for Constructing Automatic Rolling Bearing Fault Identification Model Based on Refined Composite Multi-Scale Dispersion Entropy

“…Bearing fault diagnosis method is based on inner product continuation local mean decomposition and support vector machine (SVM) [12]. Online fault analysis method is based on improved multi-scale fuzzy entropy [13], while intelligent fault identification is based on relatively big data software and parallel machine learning algorithms [14]. Another example of diagnosis method is the fault diagnosis method founded on noise reduction technology and improved convolutional neural network (CNN) [15].…”

Rolling Bearing Fault Diagnosis Method Based on Parallel QPSO-BPNN Under Spark-GPU Platform

“…With the rapid development of big data technology, many scholars have carried out extensive research on fault diagnosis in industrial big data scenarios [19], [20], [21], [22], [23], [24], [25]. For example, some studies combine big data technology and data-driven fault diagnosis methods to diagnose faults of mobile robot [21], sulfur hexafluoride electrical equipment [22], power grid equipment [23], wind turbine gearbox [24], and reciprocating air compressor [25]. Most of the above-mentioned studies use MapReduce [26] or Spark [27] to parallelize the fault diagnosis models to improve the performance of industrial equipment fault diagnosis in the big data environment.…”