Accurate tool wear prediction in machine tool operations plays a vital role in achieving high machining quality while contributing to green manufacturing by reducing waste, lowering energy consumption, and enabling sustainable tool management. However, existing deep learning-based approaches have difficulty predicting the remaining useful life (RUL) of tools when future sensor data are unavailable. In this paper, we propose a novel framework that combines a physics-based polynomial wear model with a transformer model to enable reliable RUL prediction even in the absence of additional input data. First, an adaptive labeling technique is used to correct non-monotonic measurements commonly found in industrial environments, thereby enhancing the reliability of the training data. Next, a transformer model— optimized via 25 Bayesian searches over the number of layers (1–6) and feedforward dimensions (128–8192)—learns from historical sensor data to precisely estimate tool wear. Building on this estimation, a physics-informed model is applied to predict the wear trend without requiring future sensor data. Finally, a kernel density estimation (KDE)-based ensemble of 10 independently trained models probabilistically aggregates multiple outputs, reducing sensitivity to outliers and further improving prediction accuracy. Validation using NASA’s open-source CNC milling dataset comprising 16 machining cases demonstrates that the proposed framework achieves RMSEs of 0.098 for tool wear estimation and 1.119 for RUL prediction, outperforming existing methods. By integrating physics-based modeling into a deep learning framework, this paper shows that tools’ RUL can be accurately predicted without additional sensor data, contributing to more efficient tool replacement scheduling in real-world processes.

Fault diagnosis of rotating machinery is essential to minimize damage and downtime in industrial fields. With the development of artificial intelligence, deep-learning-based fault diagnosis has gained significant attention. However, changes in the data distribution from machinery operating under different conditions have led to insufficient diagnostic accuracy. Additionally, the lack of labeled data in industrial settings hampers the performance of these deep-learning algorithms. To address these issues, unsupervised domain adaptation (UDA)-based fault diagnosis methods have been increasingly explored for robust diagnosis under varying conditions. Traditional UDA methods, however, struggle to adapt to hard-to-adapt classes as they focus only on reducing global distribution discrepancies, leading to misclassification and reduced performance for these classes. In this paper, we propose a latent space alignment based domain adaptation (LSADA) approach to overcome this limitation. LSADA reduces local distribution discrepancies by sequentially aligning minority regions and minimizing the distance between source and target data in high-dimensional latent space. Additionally, the feature extractor and predictor in LSADA are synchronized by generating reliable pseudo labels from unlabeled target data. The proposed method is validated using both open-source and experimental datasets, demonstrating that LSADA outperforms existing UDA-based fault-diagnosis algorithms. Moreover, a physical analysis of the method addresses the black-box issue, a common limitation of deep-learning approaches.

Fault diagnosis of rolling element bearings is essential to ensure the safety and reliability of industrial sites. However, changes in operating conditions can lead to variations in the distributions of the data that is collected for fault diagnosis. This, in turn, decreases the performance of deep-learning-based fault-diagnosis methods. In addition, most data in industrial settings are unlabeled, which leads to ineffectiveness of the supervised learning method. To address the issues of domain shift and unlabeled data, numerous studies have been conducted to reduce distribution discrepancies when using unlabeled data. Still, most of these studies assume that the number of labels in the training and test data are identical; this is not always true for data from industrial sites. Thus, the research outlined in this paper was pursued to address the partial domain adaptation problem, which occurs when there are fewer labels in the test data than in the training data. The proposed approach suggests two methods for applying partial domain adaptation in mechanical systems: i) a domain knowledge filter is proposed, which reflects fault characteristics in the original signal for effective feature extraction in the mechanical engineering domain, and ii) a gradient alignment module is defined to align the gradient of the statistical loss function. The method proposed herein was validated using two open-source datasets; the approach demonstrated high performance and low uncertainty, as compared to other prior methods. Additionally, physical analysis of the domain knowledge filter was conducted in this work. 

Data-driven fault diagnosis is an essential technology for the safety and maintenance of rock drills. However, since the signals acquired from a rock drill have different distributions, which arise due to their variable operating conditions, the classification performance of any data-driven method is diminished; this is called the domain-shift issue. This paper proposes a new domain-adaptation-based fault diagnosis scheme to solve the domain-shift problem. The proposed method introduces a data-cropping technique to mitigate the difference in the length of the data measured from a rock drill for each impact cycle. To extract invariant features for all operating conditions, the proposed method combines two methods: a domain adversarial neural network and minimization of the maximum mean discrepancy (MMD) between the features from different domains. In addition, a soft voting ensemble is used to reduce the model uncertainty. The proposed method shows superior performance when validated with a rock drill dataset; the proposed approach was ranked in 2nd place in the 2022 PHM Conference Data Challenge.