Optimal Management of Renewable Generation and Uncertain Demand with Reverse Fuel Cells by Stochastic Model Predictive Control

TL;DR

This paper develops a stochastic model predictive control approach for managing renewable energy and demand uncertainties in a community using reverse fuel cells, optimizing economic operation amid forecast inaccuracies.

Contribution

It introduces a two-stage scenario-based MPC algorithm that accounts for forecast uncertainties and nonlinear fuel cell behavior in renewable energy management.

Findings

Effective handling of forecast uncertainties improves operational efficiency.

The control strategy enhances economic performance of renewable energy communities.

Validated on a community with PV and industrial buildings.

Abstract

This paper proposes a control strategy for a Reverse Fuel Cell used to manage a Renewable Energy Community. A two-stage scenario-based Model Predictive Control algorithm is designed to define the best economic strategy to be followed during operation. Renewable energy generation and users' demand are forecasted by a suitably defined Discrete Markov Chain based method. The control algorithm is able to take into account the uncertainties of forecasts and the nonlinear behaviour of the Reversible Fuel Cell. The performance of proposed approach is tested on a Renewable Energy Community composed by an aggregation of industrial buildings equipped with PV.