Control of a Fourth-Order Fluid Level System Using Reinforcement Learning and Radial Basis Neural Networks

LUCAS GUILHEM DE MATOS¹; Adolfo Bauchspiess¹

¹ Universidade de Brasília

Resumo

This work presents a proposal of an adaptive controller using reinforcement learning and neural networks in order to deal with nonlinearities and time-variance. To test the controller a fourth-order fluid level system was chosen because of tits high time constants and the possibility of varying the system parameters. Implementation was made on a computer connected to an Arduino as interface to the sensor and actuator. The controller is an adaptive PI with its gains defined by a Neural Network and its performance was better than the conventional PI controller used as reference and has shown adaptive features and improvement during execution. Also, the proposed controller needs no previous information of the system in order to be designed.