Multisensor-Based Heavy Machine Faulty Identification Using Sparse Autoencoder-Based Feature Fusion and Deep Belief Network-Based Ensemble Learning
نویسندگان
چکیده
Faulty identification plays a vital role in the area of prognostic and health management (PHM) industrial equipment which offers great support to maintenance strategy decision. Owing complexity machine internal component-system structure, precise prediction heavy is hard be obtained, thus full uncertainty. Moreover, even for single component, feature representation acquired conditional monitoring signal can different due deployment sensor location environmental inference, causing difficulty selection uncertainty faulty identification. In order improve model reliability, novel hybrid approach based on sparse autoencoder- (SAE-) deep belief network- (DBN-) ensemble learning proposed this paper. First, six kinds statistical features are extracted normalized from multiple sensors same target component. Second, fused by two-stage SAE paper dimension dimension, respectively. The composite regarded as comprehensive corresponding Finally, containing components utilized predict condition classifiers. effectiveness method validated two case studies wind turbine gearbox port crane. experimental result shows that outperforms other traditional approaches terms accuracy stability when dealing with multisensor fusion machine.
منابع مشابه
Deep neural network-based bottleneck feature and denoising autoencoder-based dereverberation for distant-talking speaker identification
Deep neural network (DNN)-based approaches have been shown to be effective in many automatic speech recognition systems. However, few works have focused on DNNs for distant-talking speaker recognition. In this study, a bottleneck feature derived from a DNN and a cepstral domain denoising autoencoder (DAE)-based dereverberation are presented for distant-talking speaker identification, and a comb...
متن کاملEnsemble Learning Based Sparse High-Order Boltzmann Machine for Unsupervised Feature Interaction Identification
Identifying interpretable high-order feature interactions is important specially for biomedical applications. In an unsupervised setting, it is a challenging structure learning task. In this paper, we propose an Ensemble Learning based Sparse High-order Boltzmann Machine (ELSHBM) to identify interpretable high-order feature interactions. By converting the problem of data log-likelihood maximiza...
متن کاملMultisensor Images Fusion Based on Feature-Level
Until now, of highest relevance for remote sensing data processing and analysis have been techniques for pixel level image fusion. So, This paper attempts to undertake the study of Feature-Level based image fusion. For this purpose, feature based fusion techniques, which are usually based on empirical or heuristic rules, are employed. Hence, in this paper we consider feature extraction (FE) for...
متن کامل[Article] EEG Based Eye State Classification using Deep Belief Network and Stacked AutoEncoder
متن کامل
Sparse Feature Learning for Deep Belief Networks
Unsupervised learning algorithms aim to discover the structure hidden in the data, and to learn representations that are more suitable as input to a supervised machine than the raw input. Many unsupervised methods are based on reconstructing the input from the representation, while constraining the representation to have certain desirable properties (e.g. low dimension, sparsity, etc). Others a...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Journal of Sensors
سال: 2022
ISSN: ['1687-725X', '1687-7268']
DOI: https://doi.org/10.1155/2022/5796505