Ensuring reliability in 100% renewable microgrids: a scenario-based joint planning and operational design framework

TL;DR

This paper introduces a scenario-based optimization framework for designing 100%-renewable microgrids that meet strict reliability standards through joint capacity planning and operational dispatch.

Contribution

It develops a two-stage stochastic programming model that co-optimizes investment and operation of renewables and storage with reliability constraints, incorporating network and uncertainty modeling.

Findings

Achieves 99.998% supply reliability using only PV and batteries.

Provides resilient distributed resource allocation with power rerouting.

Demonstrates feasibility of remote, reliable, 100%-renewable microgrids.

Abstract

Off-grid microgrids powered entirely by renewable energy sources face substantial challenges in achieving utility-grade reliability standards. Existing microgrid planning frameworks often prioritize cost minimization while treating reliability as a secondary metric, thereby leading to suboptimal designs. This paper presents a comprehensive scenario-based optimization framework that simultaneously addresses long-term capacity planning and short-term operational dispatch in two stages for 100%-renewable microgrids. The developed two-stage stochastic programming model co-optimizes the investment and operation of photovoltaic generation and battery energy storage, while ensuring compliance with stringent reliability constraints following utility grid standards. Network modeling with operational constraints, such as line capacities and voltage limits, is incorporated to allow distributed resource placement leveraging power sharing between microgrid nodes. A novel scenario generation approach captures critical uncertainties, including seasonal demand fluctuations, solar output variations, and probabilistic equipment failures, through the statistical clustering of historical data. The optimization framework integrates utility-grade reliability constraints limiting the expected energy not served to below 0.002% of the annual demand while minimizing the total system costs. Numerical simulations demonstrate the effectiveness of the proposed framework, achieving 99.998% supply reliability using only photovoltaic power and battery energy storage. The optimized network-aware distributed resource allocation provides inherent resilience through power rerouting during component outages, maintaining load continuity even under simultaneous equipment failures. This study confirms the feasibility of 100%-renewable microgrids to support remote communities while meeting utility-grade reliability benchmarks.