| CPC H02J 3/381 (2013.01) [H02J 3/003 (2020.01); H02J 3/004 (2020.01); H02J 3/28 (2013.01); H02J 3/388 (2020.01); H02J 2203/20 (2020.01); H02J 2300/24 (2020.01)] | 20 Claims |
|
1. A system for the resilient artificial intelligence (AI)-based data-driven dispatch of a battery energy storage system (BESS), the system comprising:
a processor; and
a machine-readable medium in operable communication with the processor and having instructions stored thereon that, when executed by the processor, perform the following steps:
a) collecting a weather forecast based on geographical coordinates of a microgrid via an Application Programming Interface (API);
b) collecting a market forecast from available energy market data accessed via the API;
c) retrieving internal system parameters comprising State of Charge (SOC) directly from the BESS;
d) generating a photovoltaic (PV) generation prediction of an on-site solar canopy using collected data and a first machine learning (ML) model;
e) generating a load consumption prediction of on-site loads using collected data and a second ML model;
f) generating a regional outage prediction of a surrounding distribution system using collected data and a third ML model;
g) calculating charge requirements to address detected outage event durations using a rule-based (RB) method;
h) determining a dispatch profile for the BESS to achieve optimal economic return using a Mixed-Integer Linear Programming (MILP)-based method;
i) transmitting recommended control setpoints based on the dispatch profile generated at a tertiary control level of a microgrid control system to a secondary control level; and
j) upon detecting an energy deficit or an energy surplus within the microgrid, effectuating energy purchase from or sale to, respectively, a main grid through a Point of Common Coupling (PCC).
|