Seminar: Model-Based Optimization and Control for Real-time Applications

The importance of real-time optimization in today’s industrial applications of controls cannot be over-emphasized! Aside from the well known theoretical benefits, the case for the industry is more of the feasibility and ease of implementation. Some of the key enablers include; recent explosion in the understanding of convex optimization and availability of fast and robust algorithms for solving them, improvements in sensing technology which has led to a plethora of cost effective means of acquiring data in real-time, and an unprecedented exponential growth in computation speed accompanied with increase in memory at a significantly reduced relative cost.

In this talk, I will highlight some my recent industrial-academic collaborations to develop pragmatic control solutions using model-based control and optimization.

First, I will demonstrate the importance of physics-based modeling in building physical intuition into the ensuing algorithms by presenting the development of a Hysteresis tracking control for improving feel in electric power steering.

Next, I will talk in more details about the development of an adaptive control allocation scheme for a smooth and safe drive of an in-wheel electric vehicle. Again, real-time optimization is at the core of this technology.

Finally, I will talk about how GE is establishing its position as a leading digital industrial company by differentiation through advanced controls. I will discuss some of our recent model-based optimizing control solutions which are making a head way in the transportation and power generation industries.

About the Speaker

Dr. Olugbenga Moses Anubi is a Control Systems Engineer at the GE Global Research, where he is heavily involved in research activities on real-time optimization and control with applications to power generation, manufacturing and cyber-security. Before then he was a postdoctoral researcher at the Hyundai Center of Excellence in Vehicle Dynamic Systems & Control at UC Davis, where he carried out research in optimization, dynamical systems and control with applications to energy systems and vehicle system dynamics. He received his doctoral degree in Mechanical Engineering from the University of Florida with a minor in Mathematics. His work within GE has been recognized by a prestigious GE Technology Award, Physical+Digital, and also the Connected Controls Technical Achievement Award. He is also a key contributor to the recent DOE awarded GE proposal titled Cyber-Attack Detection and Accommodation for the Energy Delivery System. His research interests include convex optimization—as it applies to control methods, control theoretical development and applications, energy systems, modeling of dynamical systems, vehicle system dynamics and control, soft computing, and robotics.

Hosted by Professor Marcelo Dapino