Chaotic Motion Control Strategy for Gear System Based on RNN

To address the chaotic motion control problem in a single-degree-of-freedom gear system, a chaotic control method based on an external driving force controlled by a Recurrent Neural Network (RNN), optimized using the Harris Hawks Optimization (HHO) algorithm, is proposed. First, the relationship and characteristics between chaotic motion and variations in external driving forces are analyzed by bifurcation diagrams, Poincaré sections, and phase portraits, which reveal the evolution from bifurcation to chaos in the system. The temporal characteristics of the external driving force control method are explored. Acoording to this, the RNN-based chaotic controller was designed. Finally, a multi-objective optimization fitness function (MOFF) is constructed to guide the HHO algorithm optimizing the controller parameters. Simulation results demonstrate the effectiveness of the proposed control method.