Implementación de un sistema viga y bola para análisis comparativo de algoritmo de control PID clásico y PD difuso

Abstract

This research addresses the design, implementation, and comparative analysis of a ball-and-beam system using two control strategies: the classical PID controller and the fuzzy logic-based PD controller. The ball-and-beam system constitutes a model widely used in control engineering due to its nonlinear behavior and inherent instability, making it an appropriate platform to evaluate the performance of different control algorithms under real operating conditions. The study begins with an analysis of the problem, highlighting the limitations of the PID controller when facing variations in the physical parameters of the system, such as changes in the mass of the ball, friction in the pivot axis, or external disturbances. PID control is recognized for its simplicity and low computational cost; however, its performance may degrade in nonlinear systems if proper tuning is not achieved. In contrast, the fuzzy PD controller offers greater flexibility in the presence of uncertainties, as it employs linguistic rules and membership functions that allow adaptation to dynamic changes without requiring an exact mathematical model. Methodologically, the project includes the mechanical design of the system using computer-aided design tools, the fabrication of the structure through manufacturing techniques such as 3D printing, and the assembly of components. In the electronic domain, a sensing subsystem based on an infrared distance sensor and an actuation subsystem using a servomotor are implemented, both controlled by a low-cost development platform. The microcontroller executes the control algorithms in real time, processing the sensor signal and generating the corresponding PWM signal for the actuator. Finally, a comparative performance analysis of both controllers is carried out using indices such as ISE and IAE, evaluating stability criteria, settling time, overshoot, and steady-state error. The results make it possible to determine under which conditions each strategy provides better performance, contributing both to the academic field and to the practical application of control techniques in nonlinear Systems.