The energy transition is not without its challenges. It requires significant investments in new infrastructure and a shift in the way we consume energy. The move to renewables has increased the peak loads in the grid and has made energy production less predictable.
One of our customers, an energy retailer, wanted to know if large grid scale batteries could be a profitable solution for this challenge to balance the grid. Could we more accurately (and profitably) match supply and demand of energy and reduce congestion on the grid?
Our solution to the request was to build a trading algorithm that can trade with various energy markets such as the “day-ahead market,” “intra-day market,” and “imbalance market.”
For this, we started on our Azure data lake and with Python we built a model that can take in battery data (in our case: a Lithium-ion batterij of 1 MWh) and market data to simulate battery behavior across time. This simulator runs trading scenarios based on historical data of 2022, to assess the performance of various trading approaches.
By taking into account the demand for energy and the price volatility of energy, we could compute the payback periods for each market.
The results were satisfactory. The simulator indicates that trading on the imbalance market alone could cover battery purchase costs within a few years. The energy retailer has therefor decided to go ahead with large grid scale batteries and implement them with the simulator on 2 to 4 locations.
Our own next step is actually proving within Itility that this works - by managing energy for our own building. For this we will place 1 MW+ solar arrays and a battery on the roof and integrate that data into the simulator.