Abstract:Cutting state monitoring in the milling process is crucial for improving manufacturing efficiency and tool life. Cutting sound detection using machine learning (ML) models, inspired by experienced machinists, can be employed as a cost-effective and non-intrusive monitoring method in a complex manufacturing environment. However, labeling industry data for training is costly and time-consuming. Moreover, industry data is often scarce. In this study, we propose a novel adversarial domain adaptation (DA) approach to leverage abundant lab data to learn from scarce industry data, both labeled, for training a cutting-sound detection model. Rather than adapting the features from separate domains directly, we project them first into two separate latent spaces that jointly work as the feature space for learning domain-independent representations. We also analyze two different mechanisms for adversarial learning where the discriminator works as an adversary and a critic in separate settings, enabling our model to learn expressive domain-invariant and domain-ingrained features, respectively. We collected cutting sound data from multiple sensors in different locations, prepared datasets from lab and industry domain, and evaluated our learning models on them. Experiments showed that our models outperformed the multi-layer perceptron based vanilla domain adaptation models in labeling tasks on the curated datasets, achieving near 92%, 82% and 85% accuracy respectively for three different sensors installed in industry settings.

What problem does this paper attempt to address?

### What problems does this paper attempt to solve? This paper aims to address the challenges of cutting - sound detection in the metal - cutting process, especially in the case of scarce data and high - cost labeling in industrial environments. Specifically, the researchers propose a new Adversarial Domain Adaptation (ADA) method to make use of abundant laboratory data to compensate for the lack of industrial data, thereby training an effective cutting - sound - detection model. #### Main problem description 1. **Scarcity of data and labeling cost in industrial environments**: - In actual industrial environments, collecting and labeling cutting - sound data is both expensive and time - consuming. Therefore, it becomes impractical to train machine - learning models directly on industrial data. - In laboratory environments, a large amount of cutting data can be collected and labeled automatically, but these data are different from those in actual industrial environments, resulting in poor performance of the model on industrial data. 2. **Domain - adaptation problem**: - The differences between the source domain (laboratory data) and the target domain (industrial data) make it difficult for traditional machine - learning models to be applied directly. To solve this problem, the researchers propose an adversarial - domain - adaptation method. By learning the common feature representations between the source domain and the target domain, the model can make effective predictions on the target domain. #### Specific expressions of research problems - **RQ1: What type of sound - feature transformation can improve learning performance?** - Evaluate the influence of different types of feature transformations on single - domain classification performance through experiments, and select the best feature - transformation method to optimize the transfer - learning process. - **RQ2: How does the domain - adaptation method perform when dealing with sufficient source - domain data and limited target - domain data?** - Compare the performance of two different adversarial - domain - adaptation models (FARADAy and DIRAC) with the baseline model (Vanilla domain - adaptation model) on different sensors, and evaluate their classification performance on the target domain. ### Summary The core problem of this paper is to use the adversarial - domain - adaptation method to train a model that can effectively detect cutting sounds in actual industrial environments by using abundant laboratory data, thereby improving manufacturing efficiency and tool life. The focus of the research lies in exploring how to overcome the differences between the source domain and the target domain and improve the generalization ability of the model through feature transformation and adversarial - learning mechanisms.

Adversarial Domain Adaptation for Metal Cutting Sound Detection: Leveraging Abundant Lab Data for Scarce Industry Data

Metric Learning-Based Few-Shot Adversarial Domain Adaptation: A Cross-Machine Diagnosis Method for Ball Screws of Industrial Robots

Building Manufacturing Deep Learning Models with Minimal and Imbalanced Training Data Using Domain Adaptation and Data Augmentation

Learning Distinctive Margin Toward Active Domain Adaptation

Adversarial Domain Adaptation for Cross-user Activity Recognition Using Diffusion-based Noise-centred Learning

Contrastive Adversarial Domain Adaptation for Machine Remaining Useful Life Prediction

A novel adversarial domain adaptation transfer learning method for tool wear state prediction

Monitoring of Laser Powder Bed Fusion process by bridging dissimilar process maps using deep learning-based domain adaptation on acoustic emissions

Adversarial-Learned Loss for Domain Adaptation

Joint Discriminative Adversarial Domain Adaptation for Cross-Domain Fault Diagnosis

A Hybrid Adversarial Domain Adaptation Network for Bearing Fault Diagnosis Under Varying Working Conditions

Category-level selective dual-adversarial network using significance-augmented unsupervised domain adaptation for surface defect detection

Open-Set Domain Adaptation in Machinery Fault Diagnostics Using Instance-Level Weighted Adversarial Learning

Multi-level joint distributed alignment-based domain adaptation for cross-scenario strip defect recognition

Adversarial Experts Model for Black-box Domain Adaptation

Addressing appearance change in outdoor robotics with adversarial domain adaptation

Deep-Learning-Based Metal Artefact Reduction with Unsupervised Domain Adaptation Regularization for Practical CT Images.

Bearing Fault Diagnosis Based on Deep Discriminative Adversarial Domain Adaptation Neural Networks

A Mixed Adversarial Adaptation Network for Intelligent Fault Diagnosis

Cross-Machine Fault Diagnosis with Semi-Supervised Discriminative Adversarial Domain Adaptation

Double-level Adversarial Domain Adaptation Network for Intelligent Fault Diagnosis.