Intelligent Hybrid Control for Conveyor Belt Speed Optimization: Integration of Fuzzy Logic and Neural Networks with PID Control

Abstract

This paper presents a novel intelligent hybrid controller for conveyor belt speed optimization that uniquely integrates three control paradigms: traditional proportional-integral-derivative (PID) control, fuzzy logic, and radial basis function (RBF) neural network (RBF-NN) backstepping (PID–fuzzy logic (FL)–RBF neural network, PID–FL–RBF-NN), through a performance-based adaptive weighting mechanism. Conventional fixed-gain PID controllers prove insufficient for modern variable-demand conveyor systems experiencing intermittent loading, belt elasticity effects, and dynamic operating conditions that alter system inertia and damping over time. The key innovation lies in the synergistic architecture where fuzzy logic supplies region-wise reasoning for nonlinear characteristics, the RBF neural network compensates adaptively for unmodeled dynamics, and the PID loop provides fundamental feedback regulation, all coordinated through dynamic weighting that adjusts each component's contribution based on real-time performance indicators. The research methodology encompasses comprehensive mathematical modelling of conveyor belt electrical-mechanical dynamics, systematic fuzzy rule base development with seven membership functions, RBF network training with backstepping integration for stability, and extensive MATLAB/Simulink simulations under no-load and varied-load conditions with disturbances. The proposed hybrid controller demonstrates measurable improvements of 15-34% across all performance metrics across performance metrics, showing 15-25% reduction in settling time, 20-30% decrease in overshoot, and 10-20% improvement in steady-state accuracy compared to baseline PID control under various simulation scenarios. While these improvements are promising for simulation environments, further experimental validation would be required to assess real-world performance and practical implementation challenges.