Multi-Objective Optimal Performance of a Hybrid CPSD-SE/HWT System for Microgrid Power Generation

Introduction

Nowadays, small-scale decentralised distributed generation systems are on track to become the foundation of the worldwide energy network and it is a promising alternative solution to the typical large-scale centralised power plants. Consequently, microgrid systems comprising of a hybrid renewable energy system (HRES) can play a significant role in satisfying the energy demands of remote regions and resolving the energy price inflation and environmental problems posed by the use of fossil fuels in energy production (Motevasel et al., 2013). Therefore, microgrids with HRES would help to eliminate over a hundred million tonnes of CO₂ emissions from the atmosphere each year by providing a reliable, sustainable, and cost-competitive renewable energy supply, thus meeting the global energy needs without compromising the planet’s well-being. For instance, integration of HRES with energy storage systems, such as batteries and traditional power systems, such as a boiler or diesel engine into microgrid is considered the popular way of increasing the reliability to meet the energy demand (Belfkira et al., 2011; Kumar et al., 2013), as depicted in Figure 1 (Kabalci, 2021). In consequence, the HRES microgrids are more reliable and economical when compared to the single renewable energy system (Dufo-López et al., 2011).

Figure 1.

Microgrid with its multiple demand and source connections (Kabalci, 2021).

Solar and wind energy resources are widely used in microgrids. Nevertheless, these resources need proper management in order to facilitate their power operations to mitigate the implications of the intermittent output. To fully utilise this process, the sizing optimisation methodologies should be applied by a suitable selection of the governing parameters to obtain an optimal hybrid system design. Therefore, the optimised system will be economical, efficient and reliable.

Generally, it has been found that two generic optimisation techniques have been adopted in the design of microgrid systems: heuristic and classical techniques (Singh et al., 2016). The former technique is employed once long-term weather data for a given location is known but this is not always available, whilst the latter is typically employed in the design of microgrids if sufficient information of long-term weather data for the design location is available (Badwawi et al., 2015). Traditional or classical optimisation methods that are deployed in the microgrids design are an iterative approach, probabilistic approach, graphical construction methods, linear programming, trade-off method, and the least square method (Khan et al., 2018), (Rojas-Zerpa & Yusta, 2015). Various heuristic approaches have been employed in the techno-economic design optimisation of microgrid systems, such as genetic algorithm (GA), particle swarm optimisation (PSO), simulated annealing (SA), ant colony algorithms (ACA), bacterial foraging algorithm (BFO), artificial bee colony algorithm (ABC), biogeography-based optimisation (BBO), and artificial neural networks (ANN) (Bernal-Agustín et al., 2006; Cristóbal-Monreal & Dufo-López, 2016; Diaf et al., 2008; Khan et al., 2018; Olatomiwa et al., 2016; Singh et al., 2016). Erdinc and Uzunoglu (Erdinc & Uzunoglu, 2012) elaborate various heuristic optimisation techniques reported in the literature that contribute significantly to size hybrid microgrid systems, such as GA, PSO, SA.