 |
Center for Nonlinear Studies |
In recent years, there has been increasing concerns about the susceptibility of modern electric power systems to extreme events that cause large-scale black outs, leaving millions of people without power for several days. As a result, the power engineering community has a vested interest in exploring solutions like microgrids that can mitigate the impacts of these events. One of the biggest obstacles for rapid deployment of microgrids is their high initial outlay. This makes optimizing the design and operation of microgrids a very important problem. However, the problem of planning and operating a microgrid over a finite time horizon,i.e., 5-10 years is computationally very difficult due to the underlying nonlinear, nonconvex physics and discrete decision variables. To address the problem of computational tractability, we propose a hierarchical predictive control algorithm that pushes the boundaries of scalability beyond what is possible for existing approaches. Initial computational experiments on a standard IEEE 13-node test feeder indicate that the solutions computed using this receding horizon heuristic are within 3 % of the optimal solution. This talk is a part of the student seminar series. Multiple ten minute talks starting at 2:00pm
Host: Angel Garcia