Workshop 2: Fuzzy-Neural-Network


Ahmed Rubaai, Fellow IEEE

Professor and Chairperson

Electrical Engineering and Computer Science

Howard University, Washington, DC, USA


AHMED RUBAAI received the M.S.E.E degree from Case Western Reserve University, Cleveland, OH, and the Dr. Eng. degree from Cleveland State University, Cleveland, OH, in 1983 and 1988, respectively. In 1988, he joined Howard University, Washington, DC, as a faculty member, where he is presently a Professor and Chairperson of the Electrical Engineering and Computer Science Department. Dr. Rubaai has been named an IEEE Fellow in 2015.

 As an Educator, Dr. Rubaai has been an acknowledged educator and leader of curriculum development at Howard University for more than two decades. He is the Founder and Lead Developer of Motion Control and Drives Laboratory that provides engineering students with valuable hands-on and “real-world” experiences.” In recognition of his scholarly work and dedication to the improvement of engineering education, his work is recognized by the larger community of engineering educators, as verified by his receipt of the 2011 ASEE Robert G. Quinn Award and the Distinguished Educator Award of the Middle-Atlantic Section of the American Society for Engineering Education. This recognition is a clear demonstration and confirmation of his peers’ high regard for his contributions to engineering education.

 As a researcher, Dr. Rubaai has made significant contributions to the development and control of electric motor drives for industrial system applications in a variety of roles including scientist, research engineer, university professor, and as IEEE volunteer and leader. Most of these contributions are heavily oriented towards industrial applications that IEEE serves. Of importance is his development of control technologies by way of intelligence; laying the technological foundations for the production versions of high-performance drives used in an expansive array of industrial, commercial, and transportation applications today. His work covers a broad range of manufacturing and product applications and exemplifies his ability to bridge between academic research and the application to industrial applications. The bridges that Dr. Rubaai has built between industry and academia represent a uniquely valuable contribution that can be matched by very few others in the academic world today.

Development and Implementation of Fuzzy-Neural-Network Structure-Based Self-learning Controls of Industrial Drives

A fuzzy-neural-network Promotional-Integral (PI)-and Promotional-Derivative (PD)-type control design is offered to replace the industrial Promotional-Integral-Derivative (PID) controller, adds a self-learning capability to the initial fuzzy design for operational adaptively, and implements the solution on real hardware using an industrial test-bed system. First, the proposed solution is to use fuzzy logic-based decision structures to mimic the PI and PD elements of a PID controller in parallel. The fuzzy decision engines seek to improve response by executing custom actions per the combinations of fuzzy sets of the input parameters to the logic. Operational knowledge of the physical meaning of the fuzzy input set combinations and the necessary control response forms the basis for the fuzzy rules design. Second, a fuzzy-neural-network (FNN) structure that replaces the fuzzy logic in the control design and allows for the capability for self-tuning of the weights and memberships of the input parameters is introduced. This leads to selection of a learning algorithm for training the networks. The design implements the novel use of the extended Kalman filter (EKF) to train FNN structures as part of the PI-/PD-like fuzzy design. The benefits of the proposed control providing access to the fuzzy rules online and the proper execution of the updates are improved control law maintenance operations.

A test bench enables design implementation in the laboratory on hardware using a dSPACE DS1104 DSP and MATLAB/Simulink environment. Experimental testing results show that the proposed controller robustly responds to a wide range of operating conditions in real time. The motor drive over time is subject to experience degradation of mechanical parts or even electrical characteristics, and the ability to automatically adapt the control laws to these changes is a feature that a non-adaptive fuzzy controller does not have.

Important Dates:
  • Papers submission deadline: April 01, 2023
  • Acceptance notification: April 15, 2023
  • Final version due: April 30, 2023
Submission Link:

Download Call For Papers

Latest Posts
May 4, 2022
November 23, 2018