Adaptive PID Controller Design for Velocity Control of a Hydrogen Internal Combustion Engines using RBF Neural Networks

Achieving precise velocity control in ICEs is crucial for optimizing performance, fuel efficiency, and reducing emissions. However, the nonlinear dynamics and uncertainties inherent in ICE systems pose significant challenges for conventional control methods. In this paper, we propose an adaptive approach integrating a Proportional-Integral-Derivative (PID) controller with Radial Basis Function Neural Networks (RBFNN) to address these challenges effectively. The proposed controller architecture comprises two main components: a RBFNN designed to estimate modeling uncertainties, such as unknown friction and external disturbances impacting the ICE structure, and a primary PID controller responsible for regulating velocity. The RBFNN serves as a dynamic estimator, continuously learning and adapting to variations in system dynamics, thereby enhancing the controller's robustness and adaptability. By accurately capturing the nonlinearities and uncertainties inherent in ICEs, the RBFNN contributes to improved control performance and stability. To validate the efficacy of the proposed approach, extensive numerical simulations are conducted using MATLAB Simulink. The simulations involve various operating conditions and scenarios to comprehensively evaluate the controller's performance. Additionally, the proposed methodology is compared against conventional PID methods documented in the literature to assess its superiority in terms of robustness, tracking accuracy, and disturbance rejection. The results demonstrate that the adaptive PID controller utilizing RBFNNs outperforms traditional PID approaches, exhibiting superior velocity regulation and disturbance rejection capabilities. Moreover, the proposed methodology showcases promising potential for real-world implementation in ICE-based systems, offering enhanced control performance and efficiency. Overall, this study contributes to advancing the field of control engineering by introducing a novel adaptive control strategy tailored specifically for velocity control in internal combustion engines, leveraging the capabilities of RBFNNs to mitigate uncertainties and improve overall system performance.