Demand Prediction module deployment. The demand prediction module has been programmed with the same structure described in section 3.3 and validated in La Plana demo. Two prediction algorithms of the ones reported in D1.2 are required in the optimization problem defined for the cogeneration plant: • Osimo’s District Heating Network demand: hourly forecasting using clustering for prediction of the type of day and ARIMA models for the demand profile prediction. • Osimo’s Electricity demand: hourly forecasting using clustering for prediction of the type of day and ARIMA models for the demand profile prediction. The original prediction algorithms developed in R have been coded in Python and will be integrated in the algorithm of the demand prediction module.
Demand Prediction module deployment. As explained in the previous sections, an electric meter sensor is installed at each community member. This electric meter sensor computes the number of clockwise (consumption) or contraclockwise (production) rotations of the Ferrari meter or the pulse at the P1 port for the digital meter.
