Abstract
This study improves an approach for Markov chain-based photovoltaic-coupled energy storage model in order to serve a more reliable and sustainable power supply system. In this paper, two Markov chain models are proposed: Embedded Markov and Absorbing Markov chain. The equilibrium probabilities of the Embedded Markov chain completely characterize the system behavior at a certain point in time. Thus, the model can be used to calculate important measurements to evaluate the system such as the average availability or the probability when the battery is fully discharged. Also, Absorbing Markov chain is employed to calculate the expected duration until the system fails to serve the load demand, as well as the failure probability once a new battery is installed in the system. The results show that the optimal condition for satisfying the availability of 3 nines (0.999), with an average load usage of 1209.94 kWh, is the energy storage system capacity of 25 MW, and the number of photovoltaic modules is 67,510, which is considered for installation and operation cost. Also, when the initial state of charge is set to 80% or higher, the available time is stable for more than 20,000 h.
Highlights
In [17,18,19], the state of Markov chains is defined to represent the level of the energy storage system and the transition probabilities are derived from the load demand and the power generation data
Since the model proposed in this paper is based on the power usage of the university campus, the optimal number of PV modules and energy storage systems (ESSs) capacity is proposed based on the given target power supply availability of 3 nines
While various Markov chain models have been developed and employed as stochastic models to analyze the steady-state behavior of the energy storage systems, the variation caused by diurnal and the seasonal cycle is not considered in few researches
Summary
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. The load demand and power supply of the systems behave randomly and vary over time Such variation is not reflected in the existing Markov chain models. In [17,18,19], the state of Markov chains is defined to represent the level of the energy storage system (battery) and the transition probabilities are derived from the load demand and the power generation data. Note that the existing models has limitation to incorporate the fluctuation of load and PV generation within a day To resolve this issue, two-dimensional Markov chain is proposed. With the proposed Absorbing Markov chain model, the time duration until the PV-coupled ESS system fails to supply power to loads can be estimated.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
