Simple moving average based capacity optimization for VRLA battery in PV power smoothing application using MCTLBO Chinmay Kumar Nayak a, *, Manas Ranjan Nayak b , Rabindra Behera a a Department of Electrical Engineering, Indira Gandhi Institute of Technology, Sarang, Dhenkanal, 759146, Odisha, India b Department of Electrical Engineering, Siksha ‘O' Anusandhan University, Bhubaneswar, 751030, Odisha, India A R T I C L E I N F O Article history: Received 21 September 2017 Received in revised form 2 December 2017 Accepted 12 February 2018 Available online xxx Keywords: Solar photovoltaic Battery energy storage system Power system reliability Techno-economic analysis Multi-course teaching learning based multi- objective optimization A B S T R A C T Rapid depletion of fossil fuel reserves and alarming increase of environmental pollution shift the researchers’ attention towards renewable energy sources technologies like solar photovoltaic (PV). But solar radiation being affected by natural factors, result in uncertain power generation which leads to lower power system reliability. This paper proposes a smoothing strategy of generated PV power using gelled electrolyte valve regulated lead acid (VRLA) type battery energy storage system (BESS). The BESS stores the excess energy and releases it to meet the load demand in case of power surplus and deficit, respectively. The IEEE-RBTS is considered as the basic system for the study. But using large BESS incur humungous cost. Hence Multi-course teaching learning based multi-objective optimization technique (MCTLBO) is utilized to find out the optimal size of the PV panel, the BESS and the smoothening duration. Here, the objectives are to obtain minimum financial loss due to power outage as well as minimum BESS life cycle cost. MCTLBO is proposed here to improve the performance of the traditional teaching learning based optimization technique and it shows promising results. Factors affecting the power output of the PV panels are also considered here. The simulation is performed considering real time solar irradiance and temperature data of a city located on the eastern coast of India and the results obtained are both technically and economically viable in Indian context. © 2018 Elsevier Ltd. All rights reserved. 1. Introduction Extending from 8  4 0 N. to 36  6 0 N., India receives abundant solar radiation of about 5000 trillion kWh per square meter per year. It has installed about 13.11 GW of solar PV plants of which Odisha, a state on the eastern coast of the country, contributes about 77.64 MW [1]. Odisha experience south-west summer monsoon for almost 4 months a year. Hence cloud cover, changes in ambient temperature along with variation of daylight hours affect the solar PV power generation to a greater extent. Additionally, constraints like thermodynamic limits, very low PV cell efficiency, effect of dust and shadow, power losses in various power electronic devices like converters and inverters used and in cables connecting the devices and very high and significant initial cost etc denigrates the technical as well as economical viability of the power generated. So, integrating solar PV panels into an existing conventional power distribution system lowers the power system reliability. In case of excess generation of solar power the system has to dump energy to keep the power system stable. And also high penetration of energy from PV units along with load variation arise many problems including voltage rise, high losses and low voltage stability [2]. These problems can be mitigated with the use of BESS which eventually leads to better power system reliability. A lot of research has been done to control the energy system in PV fed power systems. A cost effective optimal size of PV and BESS units are determined for a stand-alone PV system to electrify rural households in India [3]. Optimal sizing of a grid connected PV-BESS is done for a residential customer of India has been done [4]. Optimal size of battery energy storage system of a grid-connected PV system is found out considering peak load shaving in [5]. Multi- objective particle swarm optimization (MOPSO) is utilized to minimize the annual operating cost as well as to maximize correlation coefficient [6]. The performance of hybrid differential evolution algorithm and harmony search algorithm is presented for optimal operation of a micro-grid [7]. Hysteresis energy management of a grid-connected PV/BESS is proposed to avoid frequent change of state of the BESS in order to smooth the power transmitted [8]. A PV power smoothing strategy is presented to * Corresponding author. E-mail addresses: chinmaynayak009@gmail.com (C.K. Nayak), manasnayak@soa.ac.in, manasnk72@gmail.com (M.R. Nayak), b_rabindra@igitsarang.ac.in, b_rabindra@yahoo.co.in (R. Behera). https://doi.org/10.1016/j.est.2018.02.010 2352-152X/© 2018 Elsevier Ltd. All rights reserved. Journal of Energy Storage 17 (2018) 20–28 Contents lists available at ScienceDirect Journal of Energy Storage journal homepa ge: www.elsev ier.com/locate/est