Seminar - Development of a method to optimally size micro-grids based on meta-heuristic optimization algorithms

ECS PhD Proposal

Speaker: Soheil Mohseni
Time: Friday 22nd February 2019 at 10:30 AM - 11:30 AM
Location: Cotton Club, Cotton 350

Add to Calendar Add to your calendar

Abstract

Distributed renewable energy systems have been proposed as an intervention to overcome many of the challenges associated with large centralized power generation, such as: high power losses, high emissions (if based on fossil fuel) and expensive capital infrastructure. However, the inherent intermittency associated with renewable energy technologies, particularly from solar and wind resources, lowers the reliability of the power supply, especially in off-grid systems. A potential solution for dealing with the intermittent nature of renewable energy sources (RES) is the concept of a micro-grid that facilitates the use of demand response programs. Micro-grids are discrete, small-scale power grids that provide a platform for the integration of distributed energy resources and loads that can be operated in both grid-connected and islanded modes. The main goal of this PhD is to develop a method to optimally size the components of micro-grids in the presence of new demand challenges, such as electric vehicles. The method incorporates a demand-side management (DSM) strategy that shifts an appropriate percentage of the loads from peak to off-peak consumption hours, and considers several uncertainties associated with the input parameters of the problem. A direct load control-based demand response program is considered to implement the DSM strategy and the uncertainties are modelled using Monte Carlo simulations. The method also incorporates a meta-heuristic optimization algorithm to calculate the optimal sizes of the components of micro-grids through minimizing their life-cycle costs subject to reliability and operational constraints. Furthermore, in order to reduce the computational burden and make the optimal sizing procedure computationally feasible, a scenario reduction technique, based on an appropriate algorithm in the uncertainty analysis phase, as well as a data compression-based model reduction technique, is employed in the method. The method will be applied to both grid-connected and islanded micro-grid case studies in at least three locations across New Zealand.

Go backGo back to the seminar list