A memristor-based magnetized Hopfield neural network for hidden scroll chaotic attractors and control analysis with reduced input

This paper presents a novel memristor-based magnetized Hopfield neural network (HNN) designed for generating and controlling hidden scroll chaotic attractors with reduced input complexity. The integration of memristor introduces nonlinear dynamics, enhancing the network’s ability to produce complex chaotic behaviour. By magnetizing the Hopfield model, we further extend the system’s versatility and stability in representing intricate patterns inherent to chaotic systems. The proposed model is analysed using some of the numerical methods like bifurcation and phase plots. The proposed dynamics exhibits periodic, chaotic and quasiperiodic behaviour. Further, the attractor of the proposed memristive HNN is controlled using three approaches; (i) using its parameter control, (ii) normal integral sliding mode control SMC (NISMC) and (iii) adaptive barrier function-based integral SMC (ABFISMC). It is observed that the states are converging to origin just by changing the value of a parameter of the system in the absence of the disturbance. During externally added bounded unknown disturbances, ABFISMC performs better than the NISMC. In both the cases, only two control inputs are used out of four states.