Improving efficiency of Microgrids

Artificial Intelligence Solutions for Energy Microgrids Management

An electricity microgrid is an energy system consisting of local electricity generation, local loads (or energy consumption) and storage capacities.

Energy microgrids face a dual stochastic-deterministic structure: one of the main challenges to meet when operating microgrids is to find storage strategies capable of handling uncertainties related to future electricity production and consumption; besides this, microgrids also have the characteristics that their dynamics deterministically reacts to storage management actions.

This is a microgrid featuring photovoltaic (PV) panels with both short- and long-term storage capacities
The states were defined by the amount of energy in the storage devices, amount of energy in the battery, amount of energy in the hydrogen tank, storage sizing, efficiency, etc.
The actions were the amount of energy transferred into (if positive) or out of (if negative) of each of the storage devices.
The reward function of the system corresponds to the instantaneous operational revenues.
The parameters for the task are the following: size of the battery is xB = 15kW h, the instantaneous power of the hydrogen storage is xH2 = 1.1kW and the peak power generation of the PV installation is xPV = 12kWp.
Any useful information added as input to the agent helps improving the policy, such as accurate information about the production profile
Results

Microgrid Management with AI

Since the agent has no information about the future, it builds up (during the night) a sufficient reserve in the short-term storage device so as to be able to face the next day consumption without suffering too much loss load. It also avoids wasting energy (when the short-term storage is full) by storing in the long-term storage device whenever possible. Such algorithms can deliver on average 33% cost savings and 16% LEC improvements due to more optimized utilization of energy assets.

See Tableau Workbook