Benchmark of Intelligent Schedulers in an Adaptive PI by Tabulated Gains for DC Motor Speed Control

In the field of electric motor control, especially for direct current (DC) motors, speed control effectiveness is a critical aspect. Over the past years, proportional-integral (PI) controllers have been primarily responsible for controlling these machines. However, these controllers face limitations in nonlinear operating ranges, particularly at low speeds. This work addresses this issue by proposing a significant improvement through the use of an adaptive PI controller that dynamically adjusts the gains Kp and Ki through intelligent schedulers, which recalibrate these gains in response to changing motor operating conditions. Thus, this research focuses on the development and comparison of: a planner based on fuzzy logic, one employing neural networks, and a hybrid system that combines both approaches, for speed control in a direct current (DC) motor using programmed gain PI control. The algorithms for the planners were implemented in Python, using various specialized libraries. The control execution is carried out using an Arduino Nano board, chosen for its versatility and accessibility. Communication between the control system in Python and the Arduino hardware is performed through a serial connection, enabling effective integration between both software. Performance analysis is conducted using the Integral of Absolute Error (IAE) of the system response. This research reveals interesting results for control applications in industrial machinery.