Reliability assessment of distribution system with the integration of photovoltaic and energy storage systems

With the current focus on energy and the environment, efficient integration of renewable energies, especially solar energy into power systems, is becoming indispensable. Moreover, to fully capture solar potentials and to recognize the unique characteristics associated with solar energy in power systems reliability assessment, a profound investigation is needed. Accordingly, this paper attempts to establish a comprehensive analytical approach for modeling the reliability of a hybrid system (Photovoltaic (PV) system with Energy Storage System (ESS)). To this end, the output of the PV system is modeled by a multi-state model, and the ESS system is modeled as a two-state Markov model. Moreover, the failure of the hybrid system's components is considered by using the Failure Modes and Effects Analysis (FMEA) method. It then continues with the integration of the PV and ESS system model and the model associated with the failure of their components, thereby allowing the hybrid system to be represented as a multi-state model in analytical studies. Also, the conditional probability approach is used and related indices have been extracted. Finally, a discussion of a methodology for optimal hybrid system allocation and sizing in distribution systems, to minimize the electrical network losses and to guarantee acceptable reliability level and voltage profile is presented. The results obtained from the implementation of the case study, validate the major features of the proposed analytical model along with optimal placement of the hybrid system more efficiently. (C) 2021 Published by Elsevier Ltd.

Automated summary

This paper aims to establish a comprehensive analytical approach for modeling the reliability of a hybrid system, including Photovoltaic (PV) system and Energy Storage System (ESS). The research uses multi-state model and Markov model for modeling, considers component failure using FMEA method, and extracts relevant indices using conditional probability approach.