Optimal design of hybrid renewable energy systems in Lebanon -

Abstract

The objective of this thesis is to develop an optimization methodology based on Single Step Dynamic Programming (SSDP) and Ordinal Optimization (OO) to operate and size the components of a hybrid power system in Lebanon in order to fulfill the electricity demand in a reliable, affordable and sustainable manner with a cost effective solution. The system consists of three sources of energy: the unreliable utility supply of EDL, a diesel generator, PV solar panels and a storage consisting of a battery bank. Using SSDP, the model would simulate the operation of the system to deduce the levelized cost of electricity (LCOE) during one year of operation for a given set of source sizes. A simple model, based on operation simulated over two typical summer and winter weeks, would run much faster. Such simple model would be used to sample a large number of designs quickly and estimate the LCOE over the whole year considering that each one of these two weeks represents the behavior of the system over half a year. Ordinal optimization uses SSDP to simulate the operation of the system to sample N designs in a large search space ϴ, resulting from different source sizes. In order to select a finite set of “good enough” alternatives, the simple model is used to evaluate the LCOE of the N designs and sort them in an ascending order to select the top-S designs. As per OO theory, the rank is more robust than value so the use of the simple model in sampling the N designs and ranking them appropriately will speed up their initial evaluation. Then using an “accurate model” of yearly operation, the top-S designs will be evaluated and the best solutions will be identified. The model is tested to design a hybrid power system for Qaraoun, a village located in the West Beqaa district, where data on power consumption trends, available space and solar radiation were acquired. The model also investigates how the LCOE is affected when adding or removing a battery bank, varying energy prices and allowing a net mete