Optimization of hybrid renewable energy system in radial distribution networks considering uncertainty using meta-heuristic crow search algorithm

Abstract

In this paper, probabilistic design and placement of hybrid wind–photovoltaic​ system based on battery storage (PV/WT/Batt) in distribution networks with the aim of minimizing active losses and voltage deviations considering renewable units generation uncertainty and network load demand. The uncertainties are modeled with Monte Carlo simulation (MCS) using probability distribution function (PDF). It is assumed that the total load of hybrid system is supplied via the PV/WT/Batt system without the use of the power grid and the hybrid system extra power is transferred to the distribution network. The decision variables include the optimal installation location and hybrid system components size, i.e. the number of PVs, WTs and batteries, which is determined using an optimization method named improved crow search algorithm (ICSA) using behavior of crows searching for food hidden by them and the pursuit of other crows. In the ICSA, performance of traditional CSA is improved based on decreasing inertia weight method to increase the exploration capability. The proposed method is simulated on IEEE 33 and 69 bus distribution networks. The net present cost of the hybrid system is also assessed for its load supply as well as the improvement of the distribution network characteristics. The deterministic results indicated that design and placement of PV/WT/Batt system in the networks, optimally causes reduction of active losses and voltage deviations significantly. The superiority of the ICSA is compared with traditional CSA, well-known method of particle swarm optimization (PSO) and manta ray foraging optimization (MRFO) that the proposed method is resulted less power losses and less voltage deviations and more net saving the other methods. The results cleared that the designing cost of hybrid system is increased, reduced losses and voltage deviations considering uncertainty. So the probabilistic results indicated that considering uncertainty determines the worst possible events of the network for the operator and creates the possibility of logical decision making to improve the network characteristics.