Application of Artificial Intelligence to Control a Nonlinear SIR Model

This article compares the control of artificially controlled nonlinear SIR system using artificial neural networks (ANNs) and traditionally controlled SIR system with Pontryagin’s Minimum Principle (PMP) and difference approximation method. The study focuses on the significance of these control techniques in the field of applied mathematics, emphasizing the importance of accurate control in nonlinear systems. While PMP is a commonly used method, it has limitations, such as the possibility of being trapped in local minima. To address these limitations, the exploration of alternative approaches, including artificial intelligence, has gained attention. The research aims to fill the existing gaps by comparing the performance of ANNs and PMP in controlling nonlinear systems, specifically investigating whether ANNs can overcome the limitations of PMP and offer superior control outcomes. The research approach involves simulating the nonlinear system and implementing both PMP-based control with difference approximation method and ANN-based control using machine learning algorithms. The key message of this study is that ANNs have the potential to mitigate the issue of local minima trapping associated with PMP, thereby advancing control methods and offering insights into the benefits and drawbacks of different approaches in controlling complex nonlinear systems within the realm of applied mathematics.