Non-linear Dynamic System Identification on Application of Industrial Control

System identification is a vital aspect of modern control and automation, offering a pathway to develop mathematical models from observed input-output data of dynamic systems. This paper explores methodologies, applications, and advancements in system identification. Applications such as digital twin integration, industrial automation, and transportation systems highlight the practical relevance of these methodologies. A focus is given to integrating advanced machine learning techniques with traditional modelling approaches to enhance accuracy and scalability for nonlinear dynamic systems. Deep Random Vector Functional Neural Network is used for experimentation. Practical applications, including digital twin integration for real-time monitoring and control, are discussed to underline the relevance of system identification in Industry 4.0.