Long short-term memory based semi-supervised encoder—decoder for early prediction of failures in self-lubricating bearings

Abstract

The existing knowledge regarding the interfacial forces, lubrication, and wear of bearings in real-world operation has significantly improved their designs over time, allowing for prolonged service life. As a result, self-lubricating bearings have become a viable alternative to traditional bearing designs in industrial machines. However, wear mechanisms are still inevitable and occur progressively in self-lubricating bearings, as characterized by the loss of the lubrication film and seizure. Therefore, monitoring the stages of the wear states in these components will help to impart the necessary countermeasures to reduce the machine maintenance downtime. This article proposes a methodology for using a long short-term memory (LSTM)-based encoder—decoder architecture on interfacial force signatures to detect abnormal regimes, aiming to provide early predictions of failure in self-lubricating sliding contacts even before they occur. Reciprocating sliding experiments were performed using a self-lubricating bronze bushing and steel shaft journal in a custom-built transversally oscillating tribometer setup. The force signatures corresponding to each cycle of the reciprocating sliding motion in the normal regime were used as inputs to train the encoder—decoder architecture, so as to reconstruct any new signal of the normal regime with the minimum error. With this semi-supervised training exercise, the force signatures corresponding to the abnormal regime could be differentiated from the normal regime, as their reconstruction errors would be very high. During the validation procedure for the proposed LSTM-based encoder—decoder model, the model predicted the force signals corresponding to the normal and abnormal regimes with an accuracy of 97%. In addition, a visualization of the reconstruction error across the entire force signature showed noticeable patterns in the reconstruction error when temporally decoded before the actual critical failure point, making it possible to be used for early predictions of failure.

Metric Learning-Guided Semi-Supervised Path-Interaction Fault Diagnosis Method for Extremely Limited Labeled Samples under Variable Working Conditions

“…. , z (L) , h (L) , (16) where Decoder(•) denotes the 1D transposed convolution operation TransConv(•) followed by the batch normalization operation BN(•) and an element-wise activation operation g(•, •) as follows:…”

“…To fully use the more abundant unlabeled data, Wu et al [ 14 ] designed a hybrid classification autoencoder as a one-input two-output configuration consisting of the reconstruction of the input and the prediction of the health condition. Analogously, encoder–decoder network architectures based on CNNs [ 15 ] and LSTM [ 16 ] are established to distinguish the abnormal regime from the normal operating regimes by the magnitude of the reconstruction loss. As is common practice, a skipped connection was introduced in the encoder–decoder architectures, which was known as a vanilla ladder network (LAN) [ 17 ].…”

Boosting UHMWPE Self-Lubricating Performance via Embedding Porous Cobalt-Based Zeolitic Imidazolate Framework Oil-Nanocontainers

“…Self-lubricating spherical plain bearings are becoming more and more crucial in the field of heavy-duty machinery, especially for the development of marine ships and aerospace devices, due to their unique advantages of maintenance-free, needlessness of external lubrication, and low friction. − And embedding oil-containing microcapsules into polymer matrices is gradually becoming an appealing strategy to produce pivotal friction materials with high self-lubricating performance, because lubricating oil can be automatically released to provide lubrication in response to friction force changes without adding any extra oil into the contacting interface. − In general, fabricating various core–shell structures is an effective approach for storing more lubricating oil. Among them, polymer shells (like polysulfone), silica microcapsules, and carbon hollow nanospheres have been explored as oil containers used for the construction of self-lubricating materials and respectable examples have been proven to possess good performance.…”

LGMA-DRSN: a lightweight convex global multi-attention deep residual shrinkage network for fault diagnosis

“…Specifically, deep learning possesses the capability to automatically extract pertinent fault features through its multi-layer nonlinear transformation capabilities. Deep learning-based fault diagnosis methods mainly include convolutional neural networks (CNNs) [12][13][14], auto-encoders (AEs) [15][16][17], and deep belief networks (DBNs) [18][19][20]. For example, Zhou et al [14] proposed a hybrid approach that combines nonlinear auto-regressive networks with CNNs to achieve effective fault diagnosis for imbalanced data sets.…”