Optimal cost versus efficiency configuration of a grid-connected photovoltaic system exploiting the weighted-sum method with focus on Kuwaiti National Grid

Abstract

An overview of the production and consumption of Kuwait electrical energy, installed capacity and peak loads is presented in this research. The results show that Kuwait has a serious problem because of insufficient electrical energy installed and load peaking, which is considered unacceptable. The research also identifies and analyses the geographical and temporal variability of solar energy inside Kuwait. The fundamental solar models are modified to estimate and identify daily and hourly global (direct-beam) and total solar radiation (SR) on horizontal surfaces on the basis of the more readily available meteorological data such as latitude angle, longitude angle, clearness index, solar time and corresponding hour angle. The presented results demonstrate that Kuwait has an abundance of solar energy capability in terms of almost cloudless atmosphere for nine months and twelve hours solar time a day throughout the year. The daily global and monthly averaged solar intensity have been computed. This research shows that the knowledge of SR data is essential for design and sizing of the photovoltaic (PV) systems. A specific type of PV module has been modelled and its characteristics such as I-V and P-V curves for each month of the year have been calculated and analysed using MATLAB/Simulink to determine the amount of DC current, voltage and power. These results form the basis of the grid-connected PV system (GCPV) design from array construction to the reliability of electrical supply. A technical sizing procedure based on sizing algorithm using iterative manual approach (SAIMA) for meeting specific amount of GWh output required by a potential PV system sponsor in Kuwait is presented. SAIMA has been implemented to determine the configuration of the PV array, inverter-to-PV array sizing factor and efficiency of the system according to previous PV module and inverter database.

A novel methodology for approximating Pareto front multi-criteria cost-efficiency optimization problem for a proposed GCPV system has been constructed using system planning constraints. The proposed algorithm is based on bi-objective weighted-sum (BoWS) method to maximize the system efficiency and minimize the system cost. A main objective function of both GCPV system cost and efficiency has been stated as function of PV output power and inverter rated power. The proposed function is performed with the Sequential Quadratic Programing (SQP).

The results presented in this research have been acquired through simulation of the proposed GCPV to a specific section of Alsabyia generation station part of Kuwait national grid with efficient maximum power point tracking (MPPT) algorithm incorporated into a DC-DC boost converter. The simulations were performed using Power Simulation Software (PSIM). The analytical model of the PV module has been combined with a ‘perturb and observe’ (P&O) method so that MPP is achieved with the external temperature and SR also considered. An inverter is used to track the output voltage of the converter and interface the PV array with the grid. The results show that the model not only achieves the MPP function but also improves the output of the inverter by reducing the ripples in the sine waveforms. Moreover, this research involved using the software package ERACS to analyze the impact of penetrating approximately 100 MW of the proposed PV generation to a part of the generation unit at Alsybia electrical station in Kuwait. The one-line diagram of the network was modeled in ERACS and it’s used to conduct power flow and fault studies. Four network locations were chosen as potential sites to connect the PV system. Power flow studies were conducted on the network for every hour that the PV array contributed power to the network and for 35 different network configurations for each daylight hour. Computer programs were created to conduct all of these power flow studies and to help analyze the data. Fault studies were then carried out on the network, with the PV array connected at all of the potential locations. There were a few faults that caused a fault level greater than 40 kA to flow through the 13.8 kV busbars.

More information

Identifiers

ORCID for J.K. Sykulski: orcid.org/0000-0001-6392-126X

Catalogue record

Export record