Energy Price Prediction on the Romanian Market using Long Short-Term Memory Networks

Transition to a market-based economy has reached, eventually, the production of electrical energy in Romania. Historically considered a never-ending resource, the producers did not have to interact with the consumer and more specific with the demands of the consumers. This paper proposes a neural algorithm based on Long Short-Term Memory (LSTM) architecture able to assess the price of energy as a time sequence application and predict trends based on the interaction between resources availability and demand.