STANDALONE PHOTOVOLTAIC SYSTEMS SIZING OPTIMIZATION USING DESIGN SPACE APPROACH: CASE STUDY FOR RESIDENTIAL LIGHTING LOAD

Journal of Engineering Science and Technology Vol. 10, No. 7 (2015) 943 - 957 © School of Engineering, Taylor’s University STANDALONE PHOTOVOLTAIC SYSTEMS SIZING OPTIMIZATION USING DESIGN SPACE APPROACH: CASE STUDY FOR RESIDENTIAL LIGHTING LOAD 1, 2 2 D. F. AL RIZA *, S. I. U. H. GILANI , M. S. ARIS 1 Department of Agricultural Engineering, Faculty of Agricultural Technology, University of Brawijaya, Malang, Indonesia 2 Department of Mechanical Engineering, Universiti Teknologi Petronas, Bandar Seri Iskandar, Malaysia *Corresponding Author: Abstract This paper presents a sizing optimization methodology of panel and battery capacity in a standalone photovoltaic system with lighting load. Performance of the system is identified by performing Loss of Power Supply Probability (LPSP) calculation. Input data used for the calculation is the daily weather data and system components parameters. Capital Cost and Life Cycle Cost (LCC) is calculated as optimization parameters. Design space for optimum system configuration is identified based on a given LPSP value, Capital Cost and Life Cycle Cost. Excess energy value is used as an over-design indicator in the design space. An economic analysis, including cost of the energy and payback period, for selected configurations are also studied. Keywords: Sizing, Standalone PV system, LPSP, Design space, Economic analysis. 1. Introduction In the near future, Malaysia is expected to be a net importer of oil, and the nation will have to live up to issues related to the security of supply and the economic consequences of a commodity which is highly volatile [1]. Realizing the situation, it is important that further emphasis is given into diversifying of the energy resources with a special consideration into renewable energy systems such as photovoltaic systems. Photovoltaic (PV) systems, especially the standalone configuration, have been widely used for renewable energy applications in Malaysia. More than 90% of 943 944 D. F. Al Riza et al. Nomenclature Cbat DoD E B( n) Battery capacity, Ah Depth of discharge, % Energy stored in the battery on the day n, Wh E Bhalf ( n ) Battery energy condition during sunset on the day n, Wh Minimum energy stored in the battery, Wh Maximum energy stored in the battery, Wh Nominal load energy of the day, Wh PV panels power output, Watt Global total solar irradiation, Watt/m2 Daily average of irradiation, Watt/m2 Loss of Power Supply Probability Loss of Power Supply, Wh Number of days Nominal peak power of solar panel, Watt peak Peak Sun Hour, hour Peak sun hour on the tilted surface, hour State of charge, % PV panel temperature, oC E B min E B max Eload (n ) EPV(n) GTt H LPSP LPS n PPV PSHt PSHt(n) SOC T panel Tamb Vrated Daily average of ambient temperature, oC Rated voltage/voltage of the system, Volt Greek Symbols Negative temperature coefficient of power ρ Total system efficiency, % ηsystem Daily self-discharge factor, % ηself Inverter efficiency, % ηinv installed PV systems in Malaysia are standalone applications and installed mainly for rural electrification schemes and non-building structures [2]. However, advances in PV and storage systems technology have improved the commercial viability and point towards a larger market share in residential applications. To convince potential investors into financing PV system installation, an appropriate supply and demand matching exercise is required. Here, a sizing and analysis tools are important features, which can strengthen the reliability of a PV system. Sizing is an essential part of PV system design to ensure reliability of the system by ensuring the number of solar panels used to capture solar energy and the capacity of the batteries for energy storage is sufficient [3]. Among the current approaches used for modelling, and sizing PV system includes the use of a deterministic, stochastic, empirical and statistical model. Deterministic methods are sometimes chosen because of its simplicity, but it will be only suitable for an initial design or quick estimate. Due to the stochastic nature of solar radiation as the resource of the system, more sophisticated results will be obtained by using a statistical method. Through the deterministic method, the storage system capacity is determined by the “autonomy-day” to ensure Journal of Engineering Science and Technology July 2015, Vol. 10(7) Standalone Photovoltaic Systems Sizing Optimization using Design Space . . . . 945 reliability of the system. This concept, however, does not ensure a direct relationship between reliability and PV systems component capacity [4]. Stochastic method considers the random nature of solar radiation as the most important data input for PV system sizing. Usually this method is carried out by calculating Loss of Power Supply Probability (LPSP) as the design parameter representing the reliability of the system [5]. Some author uses a different term such as Loss of Power Probability (LPP) [6], Loss of Load Probability (LOLP/LLP) [7] and Loss of Energy Probability (LOLE) [8]. Loss of Power Supply Probability (LPSP) is the total time when the power supply is not able to supply the load [4]. Another definition of LPSP is given by T. Markvart [9] as the ratio between the estimated energy deficit and the energy demand over the total operation time of the installation. The stochastic method is frequently used and studied by many researchers. Shen [10] carried out solar array and battery sizing for stand-alone PV system in Malaysia using daily PSH data and an energy balance concept to calculate LPSP. Posadillo and Luque [11, 12] developed sizing method for stand-alone PV system using the LLP concept and proposed the annual number of system failures and standard deviation of the annual number of failures as new sizing parameters. T. Markvart et al. [13] proposed a sizing method based on LLP concept using a graphical method of solution and the sizing curve was obtained in combination with a climatic cycle line. Fragaki and Markvart [14] presented the results of a new sizing approach and recommend a combination of a large array and smaller storage size to reduce the loss of load. Some authors also implemented simulation methods using hourly data to calculate LLP [15, 16]. Balouktsis et al. [17] used stochastic time series model as the data input for PV system sizing for locations where no actual data exist. Based on the above mentioned researches, stochastic method can use hourly or daily data, actual data or even artificial data. In this study, we are interested to develop LPSP calculation procedure using daily data and incorporate parameters that were not considered in the previous research [10] such as tilted angle and LCC. The optimum size of a PV array and battery system can be obtained based on a minimum cost of the system [8]. A.N. Celik et al. [18] used a Life Cycle Cost (LCC) concept to find the optimal size of a PV system configuration. J.K. Kaldellis et al. [1 (...truncated)