With the increase of renewables in the energy mix, the supply-demand balance on the grid is becoming increasingly challenging to achieve. It results in higher volatility on electricity markets especially for imbalance prices . The digital revolution in particular the development of new Artificial Intelligence technologies allows industrial consumers to convert this challenging environment into a potential opportunity. Use of accurate forecasts based on Deep Learning techniques offers higher visibility on the imbalance price and allows industrial consumers to reduce their risks and capture additional market opportunities.
As N-SIDE’s mission is to bring our customers a more analytical, fact based-approach to take complex decisions, we developed new electricity price forecasts for the imbalance market.
The new electricity price forecasts module completes our Industrial Flexibility Optimization solution (ENERTOP) and counts three types of imbalance market price forecasts based on Deep Learning techniques:
- Real-time: predicts the imbalance price for the current fifteen minutes based on the system real-time data.
- Quarter-hour – 15 or 30 minutes : predicts the imbalance price and publishes it 15 or 30 minutes before the quarter-hour, every fifteen minutes.
- Risk Management 1 day-ahead: assess the level of imbalance costs for coming day.
Why use the N-SIDE imbalance market price forecasts ?
Based on cutting-edge technologies and accurate business know-how, our imbalance market price forecasts give a greater visibility on the electricity price to the industrial consumers. It allows them therefore to spot ahead periods where the energy market is particularly tense by giving extra care in their consumption predictions.
Our forecasts are based on Deep Learning algorithms and have more than 50K free parameters and use about 500 data sources among which the weather conditions, the load of the electrical grid and the prices observed in the recent past. These algorithms are well suited to deal with the non linearity of the electricity market by leveraging high number of hidden layers in the neural network.
As all N-SIDE solutions, these forecast modules can take into account the specificities of different industries and different levels of flexibility in the industrial processes.
Interested to know more about electricity price forecasts based on machine learning technologies ?
*Savings could be generated by optimally using the mix of N-SIDE Forecasts considering both upwards and downwards opportunities.