International Journal of Simulation and Process Modelling (4 papers in press)

Regular Issues

• Open Access

  Multimodal simulation for diagnosing microstrabismus integrating electrooculogram and electroencephalographic signals ( Free Full-text Access )

  by Tong Yang, Shaoyun Bai 
  Abstract: Microstrabismus is difficult to diagnose using conventional methods due to its subtle presentation and high risk of misdiagnosis. To overcome this limitation, we propose a multimodal diagnostic framework that integrates electrooculogram (EOG) and electroencephalogram (EEG) signals. The framework first preprocesses both signal types and applies principal component analysis to remove redundancy while preserving essential information. A dual-branch architecture then extracts spatio-temporal features from each modality, which are subsequently fused through a multi-layer interaction mechanism designed to capture cross-modal complementarity. To handle inter-sample variability, a gating-based module dynamically adjusts the fusion ratio between modalities according to individual sample characteristics. Experimental results indicate that the proposed model improves diagnostic accuracy by at least 6.35% over baseline methods, demonstrating strong potential for aiding the precise diagnosis of micro-degree strabismus.
  Keywords: strabismus diagnosis; multimodal feature fusion; EOG signal; EEG signal; attention mechanism.
  DOI: 10.1504/IJSPM.2026.10079108
   
  
• Simulation-driven deep learning framework for early poultry disease detection using faecal image classification with optimised CNN architectures  
  by Nonita Sharma, Monika Mangla, Manik Rakhra, Baljinder Kaur, Raj Kumar Mohanta, Bijay Kumar Paikaray 
  Abstract: The present investigation aims to devise an optimized Convolutional Neural Network (CNN) framework to identify prominent poultry diseases based on faecal images. The proposed research model uses a tri-convolutional layer architecture to enhance the accuracy of poultry disease classification and improve feature extraction. Three major diseases, namely coccidiosis, salmonella, and Newcastle, have been considered. The prime objective of current research is to achieve early detection by employing advanced deep-learning techniques. The proposed model uses fecal images to identify pathological conditions using the TriConvLayer architecture accurately. In this work, custom CNN models, viz. SoloConvLayer, TriConvLayer, and FiveConvLayer models are used to achieve an accuracy of 97%, 98%, and 98%, respectively. The achieved result advocates the efficacy of the proposed approach. It thus has the potential to revolutionize early disease detection in poultry farming, a major step towards improving animal health and farm productivity.
  Keywords: deep learning; poultry disease detection; convolutional neural networks; CNN; faecal image analysis; poultry health monitoring; disease classification; simulation framework.
  DOI: 10.1504/IJSPM.2025.10073525
   
  
• Modelling and simulation of a non-contact electromagnetic damping process for enhanced band saw performance  
  by Xuexuan Tao, Zixuan Li, Qingbo Yu, Yundong Chen, Yi Wu, Zhiyi Wei 
  Abstract: This paper presents a simulation-driven methodology to address the issues of reduced machining precision, poor surface quality, and shortened service life in horizontal band saws caused by high-speed blade vibration. A coupled electromagnetic-structural process model was developed using finite element analysis to simulate the interaction between a novel non-contact electromagnetic vibration damper and the saw blade. Systematic parametric simulations identified an optimal relative air gap ratio (a=0.4) that maximizes the electromagnetic damping effect. Analysis indicated that a single-damper configuration alone could achieve a magnetic flux density of 1.48 T in the blade's vibration zone and reduce vibration displacement by 66.7%. The optimized symmetrical dual-damper configuration, derived from simulation insights, generated a higher magnetic flux density and induced an eddy current distribution that decreased from the surface inward, resulting in faster vibration decay and significantly lower residual amplitude. Experimental validation confirmed the model's accuracy. The study demonstrates how simulation-based process modelling can replace trial-and-error in damper design and optimization, providing a transferable framework for vibration suppression in high-speed cutting tools.
  Keywords: eddy current; manufacturing processes; non-contact damping; parametric optimisation; process modelling; vibration suppression.
  DOI: 10.1504/IJSPM.2026.10078820
   
  
• A Bayesian multimodal deep learning model for process simulation and anomaly diagnosis in electrofusion magnesium furnaces  
  by Qiuxia Qu, Hong-Yang Wan, Jia-Qi Zhai 
  Abstract: This paper aims to develop a high-reliability process simulation framework for partial melting anomaly diagnosis in electrofusion magnesium furnace operation process. The proposed Multimodal Deep Neural Network-Bayesian Model integrates multimodal sensor data comprising Infra-red thermal imaging and three-phase current signals to overcome the limitations of traditional monitoring methods. Spatiotemporal features are extracted from image sequences using 3D Convolutional Neural Network in conjunction with Long Short-Term Memory networks, while non-stationary current characteristics are captured using time-frequency analysis. Furthermore, attention mechanism is designed to realize adaptive multimodal feature fusion, and robust classification with uncertainty quantification is achieved based on Bayesian neural network. Simulation-based experiments with industrial data verify that the model accuracy achieves to 95.8%. Higher confidence is further proved through uncertainty analysis in normal conditions, providing reliable early warning for anomaly diagnosis. Thus, this approach offers a viable digital twin solution for complex industrial processes.
  Keywords: process modelling; simulation framework; partial melting anomaly diagnosis; multimodal deep neural network; Bayesian inference; convolutional neural network; electrofusion magnesium furnace.
  DOI: 10.1504/IJSPM.2026.10079195