Control Design and Experiment of a Microwave–Hot Air Vibrating Fluidized Bed Dryer Based on an Improved Tuna Swarm Optimization Algorithm

Conventional PID controllers exhibit unsatisfactory parameter-tuning performance when applied to microwave–hot air vibrating fluidized bed drying systems, which are characterized by nonlinearity, large time lag, and multi-source coupling. To address this issue, this paper proposes a self-tuning PID control system based on a multi-strategy improved Tuna Swarm Optimization (TSO) algorithm. First, a hardware control platform for the dryer is constructed on a 32-bit STM32 microcontroller embedded with the RT-Thread real-time operating system. Second, three enhancement strategies—Fuch infinite-folding chaotic initialization, Lévy flight search, and nonlinear convergence factor optimization—are integrated to overcome the inherent defects of the original TSO, namely low solution precision and premature convergence to local optima. Comparative numerical simulations of the original and improved TSO are conducted on MATLAB R2021b using six benchmark functions. Third, transfer function models of three core actuators (magnetron, PTC electric heater, and vibration motor) are established, and the improved TSO is employed to realize automatic PID parameter tuning with supplementary simulation verification. Finally, bench tests on a physical drying prototype validate the optimized control parameters. Comparative experiments with the relay feedback algorithm, particle swarm optimization (PSO), and the original TSO demonstrate that the proposed improved TSO achieves superior optimization performance for PID tuning. Practical drying tests on bitter melon slices further confirm that the improved TSO-PID control system significantly enhances drying quality, reducing average moisture content by 21.2% and total color difference by 38.7% relative to relay feedback PID control, thereby satisfying industrial production requirements.