ANALYSIS OF DATA-DRIVEN REDUCED ORDER MODEL WITH TEMPORAL CONVOLUTIONAL APPROACH USING AI

An analysis of a data-driven reduced order model (ROM) with a temporal convolutional approach using artificial intelligence (AI) techniques. Reduced order modelling is a popular technique for capturing the essential behaviour of complex systems with reduced computational costs. However, traditional reduced order models often struggle to capture the temporal dependencies present in dynamic systems. To address this challenge, we propose a data-driven approach that leverages temporal convolutional neural networks (CNNs) within the reduced order modelling framework. The temporal convolutional approach allows for the efficient and accurate modelling of temporal dependencies by learning directly from the available data. The AI algorithms enable the ROM to effectively capture the system's dynamic behaviour, leading to improved accuracy and computational efficiency. We demonstrate the effectiveness of the proposed method through a case study where we model a complex dynamic system. The results show that the data-driven reduced order model with temporal convolutional approach outperforms traditional reduced order models in terms of accuracy and computational efficiency. The findings of this study highlight the potential of AI techniques in enhancing reduced order modelling for dynamic systems. The proposed approach has promising applications in various fields, including structural analysis, fluid dynamics, and system control. Further research can explore the integration of other AI techniques and algorithms to advance data-driven reduced order modelling for complex dynamic systems.