کنفرانس مهندسی برق ایران

صفحه اصلی / سی و یکمین کنفرانس بین المللی مهندسی برق

An Improved Real-Time Implementation of Adaptive Neuro-fuzzy Controller

نویسندگان :

Iman Gholizadeh¹ Haniye Raziyan² Reza Javidan³

1- دانشگاه صنعتی اصفهان 2- دانشگاه شیخ بهائی اصفهان 3- دانشگاه صنعتی شیراز

کلمات کلیدی :

Artificial Neural Networks،fuzzy logic،Adaptive Neuro-fuzzy Controller،intelligent control،LabVIEW

چکیده :

Adaptive neural network-based control method is a highly recommended solution for the control of nonlinear systems that operate under various operational conditions. The Neural Controller has the capability to train and adapt the controlled system to attain the desired outcome. The backpropagation algorithm is utilized to optimize the neural controller parameters and reach the desired system reference. Compared to other neural control methods, the proposed neuro-fuzzy controller boasts a quick response time of its learning algorithm and is feasible for real-time implementation. To enhance the adaptive ability of the neural controller, fuzzy rules are established to select optimal values for the momentum and learning rate parameters. The method has been implemented in LabVIEW software and tested in a real-time laboratory position control system, demonstrating that optimizing the neuro-fuzzy controller with optimal parameters of the neural network model leads to improved control and faster response times in nonlinear controller systems.