Data-driven systems have demonstrated significant value in real-time surface roughness evaluation and pre- diction for machining processes. This study presents a comprehensive methodology for evaluating three core elements critical to decision-making in machining monitoring: sensor-based dynamic signal selection and pro- cessing, sensory fusion scenarios, and Machine Learning (ML) models. Unlike previous research that often an- alyzes these elements in isolation, this study emphasizes their collective impact on unified datasets. The proposed methodology is validated using experimental data from machining trials on two distinct machines during the finishing process of hard turning. Hard turning is a widely employed finishing operation in manufacturing that directly affects dimensional accuracy, surface integrity, and surface finish, which are key characteristics of machined parts. Surface roughness, as an essential indicator of surface quality, plays a pivotal role in the functional performance of end products, necessitating accurate monitoring and assessment throughout the process. To optimize performance, Bayesian Optimization was employed for automatic hyperparameter tuning, facilitating efficient exploration of optimal parameters. The predictive capabilities of trained ML models were subsequently evaluated using performance metrics and quantitatively assessed for uncertainty through Predic- tion Intervals (PIs), calculated via non-parametric Kernel Density Estimation. Results indicate that the Wavelet Packet Transform method significantly enhances the predictive performance across all ML models. Among the evaluated models, Support Vector Regression and K-Nearest Neighbors demonstrated superior predictive accu- racy and minimal uncertainty across all signal processing methods. This work provides actionable guidelines for researchers and manufacturers in selecting optimal combinations of methodologies for developing accurate, reliable, and generalizable data-driven surface quality monitoring and prediction systems, particularly for hard part turning applications.

Surface roughness is an essential attribute of surface quality in the machining process to ensure the mechanical functionality of machined parts. Effective in-process monitoring of surface roughness is indispensable, which can be achieved through intelligent monitoring and prediction systems, utilizing data-driven models and dynamic process signatures. Notably, the performance of these systems, particularly in terms of prediction accuracy and robustness, is hindered by the limited application of data-driven models, especially deep learning models, which are intensive data-demanding. The inaccessibility of ample surface roughness data and corresponding process signatures leveraged for model building poses a significant challenge. In response to this challenge, as an emerging generative modelling approach, a classifier guidance diffusion model renowned for process stability and image generation with high fidelity and sufficient diversity is introduced to surmount data scarcity. The pre- trained diffusion is applied to synthesize audible sound signals collected from a machining process with the guidance of gradients from a classifier. After that, ground truth signal data mixed with synthesized ones in different proportions are leveraged to train and test the pre-trained model – VGG16, for surface roughness prediction in a transfer learning process. The results demonstrate that the proposed model can effectively generate synthesized 2D Mel-spectrogram images of audible sound signals with superior quality, further boosting the prediction performance of VGG16 in terms of its prediction accuracy and generalization under an enriched dataset. 

Surface roughness plays an indispensable and fundamental role as a leading indicator of the surface quality of machined parts in the manufacturing process. The precise and effective monitoring and prediction of surface roughness is crucial for surface quality control. In this regard, the development of an in-process surface quality monitoring system is necessary, which has the promising potential to achieve this goal. Such a system typically comprises data-driven models for decision-making and sensing techniques for detecting associated process information. However, some challenges still exist in building such systems. Firstly, the architecture design and deployment of data-driven models, specifically deep learning (DL)-based models, demand adequate domain knowledge. Secondly, most models trained on specific tasks with limited datasets are prone to suppressing their versatility and generalization across different machining conditions. Additionally, in most cases, reliance on handcrafted features to represent dynamic information on various signals during model training necessitates extensive expertise in selecting appropriate feature types. Furthermore, due to the nature of their low dimensionality, handcrafted features have difficulty in capturing of overall process-related underlying patterns from dynamics signatures, which is time-varying and often occurs in transient events. To address these challenges, this paper proposes the regression-based pre-trained convolutional neural network (pre-trained CNN) combined with Mel-spectrogram images based on the transfer learning method for surface roughness prediction. Within the context, the architecture of the transfer model is slightly adapted from already well-trained CNNs. Initial weights in each layer of the CNN model are directly inherited and then fine-tuned through the Bayesian optimization tuning method. Besides, the audible sound signals are captured and subsequently converted into 2D Mel-spectrogram images with variant time lengths, which are separately engaged to retrain and validate four existing pre-trained CNN models (VGG16, VGG19, ResNet50V2 and InceptionResNetV2). Eventually, the effectiveness of proposed models and comparison of their predictive capabilities are further validated through a case study in the turning process. The results demonstrate that each applied pre-trained CNN model is capable of effectively predicting surface quality with satisfactory prediction results. Therefore, the proposed method can facilitate the establishment of a machining monitoring system concerning its accuracy, reliability, and robustness.

In the turning process, the surface roughness of the machined part is considered a critical indicator of quality control. Provided the conventional offline quality measurement and control is time-consuming, with slow feedback and an intensive workforce, this paper presents an online monitoring and prediction system for the effective and precise prediction of surface roughness of the machined parts during the machining process. In this system, the audible sound signal captured through the microphone is employed to extract the features related to surface roughness prediction. However, owing to the nonlinear phenomena and complex mechanism causing surface quality in the whole process, the selection of statistical features of the sound signal in both the time and frequency domains varies from one case to another. This variation may lead to false prediction results as sufficient domain knowledge is required. Therefore, the versatile and knowledge-independent features extraction method is proposed, which exploits deep transfer learning to automatically extract sound signal features in the time-frequency domain through pre-trained convolution neural networks (pre-trained CNN). The performance of prediction models based on two feature extraction methods – statistical feature extraction and automatic feature extraction was further tested and validated in the case study. The results demonstrate that the performances of the prediction model built on the automatically extracted features outperformed that developed with the statistical feature method concerning the accuracy and generalization of the prediction model. In addition, this study also provides solid theoretical and experimental support for developing a more precise and robust online surface quality monitoring system.