A Probabilistic Tool for Modeling Smart Microgrids with Renewable Energy and Demand Side Management

Smart meters allow even remote villages to act as smart cities by enabling high-granularity sensing and control of power demand in individual buildings. Recent smart meters equip buildings or specific loads to perform demand side management (DSM) based on signals for electricity consumers to shed load or build load. For emerging contexts serviced by isolated microgrids and limited generation capacity, DSM is an important solution to advancing sustainable energy access. Quantitatively comparing the impact of different microgrid operating strategies and DSM schemes before implementation is needed to ensure dynamic, reliable microgrid service while keeping costs and emissions low. This paper addresses that need by presenting a simulation tool, a software platform that employs probabilistic models to simulate the full range of microgrid scenarios (e.g., occasional power cuts and service reliability). To accurately characterize the range and variability of potential results in a real world microgrid, the modeling software is stochastic, using Monte Carlo methods with a Markov model. The methods are demonstrated in MATLAB to simulate renewable energy installations and loads common in isolated microgrids of rural India and much of Africa. The tool is called the Probabilistic Load-Attenuating Smart Microgrid Simulator (PLASMiS). Users can define probability distributions for loads and supplies while defining control strategies for deploying energy storage and dispatchable generation. This functionality is demonstrated in case studies that use PLASMiS to model a typical isolated microgrid in Rwanda.