System identification of a nonlinear continuously stirred tank reactor using fractional neural network

• Enhanced CSTR Modeling with Novel Fractional Activation Functions: This study introduces and applies newly developed fractional-order activation functions within a neural network framework for improved modeling of continuous stirred tank reactors (CSTRs) using real-time process data. • Superior Performance Compared to Conventional and Existing Fractional Approaches: The proposed fractional activation functions demonstrate superior performance in terms of accuracy, generalization, and robustness to data reduction and noise compared to conventional activation functions and existing fractional approaches in CSTR modeling tasks. • Efficient and Robust Black-box Modeling: The developed fractional neural network model offers an efficient and robust black-box approach for CSTR modeling, effectively capturing complex nonlinear dynamics even with reduced datasets and noisy measurements. • Faster Convergence and Adaptability: The proposed fractional activation functions exhibit faster convergence during training and improved adaptability to varying process conditions, making them ideal for real-time CSTR modeling applications. Chemical processes are vital in various industries but are often complex and nonlinear, making accurate modeling essential. Traditional linear approaches struggle with dynamic behaviour and changing conditions. This paper explores the advantages of the new theory of fractional neural networks (FNNs), focusing on applying fractional activation functions for continuous stirred tank reactor (CSTR) modeling. The proposed approach offers promising solutions for real-time modeling of a CSTR. Various numerical analyses demonstrate the robustness of FNNs in handling data reduction, achieving better generalization, and sensitivity to noise, which is crucial for real-world applications. The identification process is more generalized and can enhance adaptability and improve industrial plant management efficiency. This research contributes to the growing field of real-time modeling, highlighting its potential to address the complexities in chemical processes.