PID Control Tuning Technique for Temperature Control in Cotton Fiber Dyeing Using Neural Networks and PLC

This research work focuses on improving temperature control in the cotton fiber treatment process by developing an advanced PID control tuning technique. The proposed methodology uses a Programmable Logic Controller (PLC) and the integration of artificial neural networks to automatically optimize the PID control parameters based on the process conditions. The study explores classical PID tuning, software autotuning, and neural network autotuning. These methods are compared in terms of reference signal tracking, response time and system stability. The results indicate that the implementation of neural networks significantly improves the responsiveness to changes in the set point, achieving a faster settling time and better damping compared to classical and software autotuning methods. Although neural networks present challenges in terms of interpretability, the benefits in control performance suggest their viability in industrial environments for the control of nonlinear or complex systems.