Two-Layer Intelligent Learning Control Using Output Recurrent Fuzzy Neural Long Short-Term Memory Broad Learning System With RMSprop

Achieving precise and robust control in dynamic environments remains a challenge, particularly in nonlinear and uncertain systems. This paper proposes a novel two-layer intelligent learning control framework to address these issues, integrating an output recurrent fuzzy neural long short-term memory broad learning system (ORFNLSTM-BLS) with a root mean squared propagation (RMSprop) optimization algorithm. The first layer focuses on real-time system identification, leveraging ORFNLSTM-BLS to learn complex nonlinear system dynamics and effectively capture long-term dependencies in sequential data. The second layer implements an intelligent adaptive controller based on the ORFNLSTM-BLS, which dynamically adjusts control actions for time-varying setpoint tracking and disturbance rejection. The RMSprop optimization algorithm is applied across both layers to adaptively regulate learning rates, ensuring faster convergence and mitigating gradient issues. The convergence of the proposed method is provided through five theorems, ensuring gradient boundedness and learning rate conditions. Simulation and experimental results validate that, compared to conventional PID, the proposed method achieves superior tracking accuracy, with the best case showing an 82% reduction in RMSE, and enhanced control efficiency, with an 85% reduction in ITAE. Additionally, computational analysis shows a 10% increase in processing time for SISO tasks and 84% for MIMO cases. Despite the higher complexity, the method remains feasible for real-time applications. This work advances intelligent control systems, offering a robust solution for environments with sequential data and uncertainties.