In real industrial applications, intelligent methods are recently emerging for remaining useful life (RUL) prediction. However, their development is hindered by two obstacles. First, it is hard for an ordinary edge client to achieve RUL prediction owing to its weak computing capacity and limited data volume. Second, as for all edge clients, it is unrealistic to share data with each other due to the potential conflict of interests. Therefore, a federated learning-based RUL prediction method namely FedRUL is proposed to solve these problems. In this method, multiple edge clients and a cloud server are applied to train a global encoder and an RUL predictor without data sharing. Each client includes a convolutional autoencoder (CAE) comprised of an encoder and a decoder while the server includes a same encoder and an RUL predictor. During each epoch, CAEs are trained in all clients through their corresponding local training datasets first. Then, all local encoders are uploaded to the server and aggregated to a global encoder through assigning weights for all clients according to their performance on a validation dataset in the server. At last, this global encoder is sent back to all clients for extracting low-level features from their datasets and uploading these features to train the RUL predictor for each client one by one. Two experiments suggest FedRUL offers a promising solution on confidential decentralized learning for RUL prediction.

With the coming of the industrial Big Data era, data-driven fault diagnosis models emerge recently and show potential results in many studies. However, it is impractical to collect massive high-quality data in actual industrial applications, which strongly reduces the performance of fault diagnosis models. Therefore, multiple data owners are motivated to share their own data for training a brilliant model with the aggregated dataset. Notwithstanding, it is unrealistic to share data due to the potential conflict of interest. For addressing the issue of data island, a new federated framework namely FedCAE is proposed for fault diagnosis. In this framework, each client is equipped with a convolutional autoencoder (CAE) trained by its corresponding local data. Then, all CAEs are uploaded to a server and aggregated to a global CAE according to an adaptive weighted selection mechanism. Afterwards, the global CAE is downloaded to all clients for extracting low-level features from local datasets by its encoder and uploading these features to train a global fault diagnosis classifier in the server. At last, the classifier is downloaded to all clients for completing their own diagnosis tasks. Experiment results on two bearing datasets suggest FedCAE offers a promising solution on confidential decentralized learning.

Artificial intelligence-based machinery fault diagnosis techniques have been increasingly considered in many industrial fields. The convolutional neural network (CNN) is able to learn features from raw signals because of its filter structure. Thus, several studies have applied CNN based methods for machinery fault recognition and classi-fication. However, most of these studies are based on a balanced dataset, while ignoring that normal data and fault data tend to be highly imbalanced in real-world applica-tions. Conventional CNNs do not work well for highly im-balanced fault diagnostics tasks and often lead to the degradation of performance. Therefore, in this paper, a learning framework called deep focus parallel convolu-tional neural network (DFPCN) is proposed to overcome the weakness. It has powerful feature learning capabilities due to its parallel convolutional architecture. A new loss function named adaptive cross entropy (ACE) loss is de-signed for the DFPCN to focus training on minority health condition samples which are hard to classify. The effec-tiveness and superiority of the proposed DFPCN are vali-dated by a highly imbalanced dataset constructed from bearing vibration signals. The diagnostics results demon-strate that DFPCN outperforms the state-of-the-art CNN based methods in terms of accuracy and stability, and avoids adding computational burden with the redundant samples compared with oversampling methods.