Energy consumption prediction based on FB prophet -GRU method
Abstract
In modern commercial medium industries, electric energy consumption prediction plays a
significant role in including the energy expenditure in the process of making the annual
financial budget of the company. In this project, we predicted electric energy consumption
using a model which combines FbProphet and GRU that we developed. We collected raw
data from Igara Growers Tea Factory (IGTF) and refined using the MinMax scaler for
effective training. We then developed a hybrid model with the integration of Fbprophet and a
five-layer gated recurrent unit (GRU) and trained it by exploiting the transfer learning
technique. The evaluation techniques used in the trained model were; the root mean square
error (RMSE), mean square error (MSE), and mean absolute error (MAE) metrics. From the
experimental results, we found that, in comparison with other existing models, the hybrid
model we developed provided more accurate results with relatively lower errors which were:
(RMSE = 0.0103), (MSE = 0.0001) and (MAE = 0.0076). The use of the model developed in
this project will guide the industry in the proper allocation of resources by helping to estimate
the correct amount to be spent on power in the stipulated duration of time in the future up to
one year period. Further work on the prediction of Uganda’s future energy tariffs basing on
the available past quarterly tariff data as determined by the energy regulatory bodies is
necessary in order to evade the task of predicting future expenditure on electrical energy by
the industry basing on the current tariff.