International Journal of Mechatronics and Manufacturing Systems (5 papers in press)

Regular Issues

• Intelligent Condition Monitoring of PMSM-Based Drive Systems using a Swin-CNN Fusion Network for Manufacturing Applications  
  by Hao Yu, Hao Zhang, Xiaojing Liu 
  Abstract: Permanent Magnet Synchronous Motors (PMSMs) are widely employed in advanced manufacturing and mechatronic systems, where early detection of mechanical faults is vital for system reliability and productivity. This paper presents a novel current-based radial misalignment diagnosis framework for PMSM-driven systems using a dual-branch Swin-CNN fusion network. The proposed method integrates time-frequency and spectral domain analysis through Continuous Wavelet Transform (CWT) and Fast Fourier Transform (FFT), respectively. A two-stream deep learning architecture is introduced, combining a 2D Swin Transformer for global time-frequency feature modeling and a 1D CNN enhanced with Convolutional Block Attention Module (CBAM) for frequency-domain emphasis. Experimental results on a lab-scale PMSM setup demonstrate an average classification accuracy of 98.125% under varying fault levels. The proposed approach offers a cost-efficient, non-invasive solution for condition monitoring in intelligent manufacturing systems, supporting predictive maintenance and enhanced equipment availability in Industry 4.0 environments.
  Keywords: Permanent Magnet Synchronous Motor (PMSM); Sliding Window Attention Network; Convolutional Neural Network (CNN); Attention Mechanism; Fault Diagnosis.
  DOI: 10.1504/IJMMS.2025.10073259
   
  
• Identification of Tangential and Normal Forces in Micro End Milling Through Machine Learning Analysis of Force Signals  
  by Yiğit Karpat 
  Abstract: Developing digital twins of manufacturing processes, like computer numerical control (CNC) machining, is vital due to their importance for creating high value-added parts. Tool condition monitoring has been an important research topic within this context where a major focus is on analysing machining force signals. Micro-milling is a complex process due to contributing factors like tool runout, deflection, edge radius, elastic recovery of materials, microstructure effects, and machining dynamics. This paper focuses on machine learning analysis of force signals to identify normal and tangential forces acting on the micro end mill. A machine learning algorithm based on Gaussian Process Regression (GPR) has been used to identify normal and tangential forces as a function of uncut chip thickness. The novelty of this approach is that identified normal force variation as a function of uncut chip thickness reveals information on minimum uncut chip thickness and edge radius. Monitoring the variation of these characteristic points on the force curves can be used to identify tool wear and predict remaining useful tool life.
  Keywords: Micro-milling; Machine Learning; Gaussian Process Regression.
  DOI: 10.1504/IJMMS.2025.10074068
   
  
• Correlating Weight Assignment and Ranking Approaches: The Versatility of the Best Holistic Adaptable Ranking of Attributes Technique  
  by Kartik Bajaj, Ravipudi Venkata Rao 
  Abstract: In multi-attribute decision-making (MADM), selecting an appropriate method for both weight determination and ranking of alternatives plays a crucial role in ensuring accurate and consistent results. This study employs the best holistic adaptable ranking of attributes technique (BHARAT) across two industrial case studies 3D printer selection and cutting fluid selection. The BHARAT framework simultaneously determines attribute weights and ranks alternatives, thereby obviating the need for hybrid or multistage approaches. Comparative analysis is conducted against conventional weighting techniques: equal weight, rank reversal, rank sum, rank order centroid, analytic hierarchy process, decision analysis table, and pairwise comparison, alongside with ranking techniques including TOPSIS, WPM, and PROMETHEE. Correlation analysis indicates that TOPSIS exhibits instability in the cutting fluid case, while PROMETHEE shows moderate consistency in the 3D-printer case. In contrast, BHARAT delivers stable, interpretable, and consistent rankings across both cases, demonstrating its transparency, computational efficiency, and capability as a robust standalone MADM approach.
  Keywords: Decision making; 3D Printing; 3D Printer Ranking;.
  DOI: 10.1504/IJMMS.2025.10074195
   
  
• Building Fault-Specific Decision Trees for Quality Control in Steel Plate Manufacturing through Machine Learning Approach  
  by Rajendra Bhange, A.S. Chatpalliwar 
  Abstract: The quality control of steel plates is vital in manufacturing, where identifying faults can enhance productivity and reduce costs. This study uses the publicly available Steel Plates Faults dataset from the UCI Machine Learning Repository to propose a fault-specific decision tree framework for fault classification and analysis. The dataset includes multiple fault types, such as Pastry, Z_Scratch, K_Scatch, Stains, Dirtiness, and Bumps, along with features like plate thickness, luminosity indices, and perimeter measurements. Interpretable decision tree models were trained for each fault type, enabling the extraction of actionable rules that highlight feature thresholds contributing to classification. Analysis of feature importance revealed critical factors influencing predictions, offering insights for proactive measures in steel production. The framework bridges the gap between automated fault detection and decision-making, demonstrating high accuracy with interpretability. These results underscore the value of explainable AI for industrial applications and highlight the importance of open datasets in advancing quality control research.
  Keywords: Fault detection; steel plate defects classification; decision tree framework; surface defects; machine learning; industrial applications.
  DOI: 10.1504/IJMMS.2025.10074541
   
  
• Improving Trajectory Tracking For Quadrotors Under Wind Disturbances by a Neural Network-Based Control Strategy  
  by Jinxing Zhao, Yuhao Fan, Haohao Liu, Zinuo Zeng, Haolan Zheng 
  Abstract: Accurate trajectory tracking for a quadrotor is challenging in windy environments. This study proposes a novel control framework that combines the model predictive control (MPC) and a neural network state space model (NNSSM) to improve trajectory tracking drift under wind disturbances. The aerodynamic effects are explicitly modelled with a small Multi Layer Perceptron (MLP) neural network by introducing the influences of the aerodynamic disturbances on the control input and state. Then the NNSSM could be achieved by combing the quadrotor's simple dynamic model and the MLP model, and incorporated into the MPC framework as the predictive model. In this way, a neural network MPC (NNMPC) capable of compensating the wind disturbances is achieved. A hardware-in-the-loop simulation has been performed to evaluate the trajectory tracking performance, and shows that the NNMPC greatly reduces trajectory tracking errors compared to the MPC neglecting the aerodynamic disturbances in a windy environment.
  Keywords: Quadrotor Drones; Trajectory Tracking; Aerodynamic Disturbances; Neural Network State Space Model; Model Predictive Control.
  DOI: 10.1504/IJMMS.2025.10074546