L. Alfredo Fernandez-Jimenez

Short-Term Forecasting Models for Photovoltaic Plants: Analytical versus Soft-Computing Techniques

We present and compare two short-term statistical forecasting models for hourly average electric power production forecasts of photovoltaic (PV) plants: the analytical PV power forecasting model (APVF) and the multiplayer perceptron PV forecasting model (MPVF). Both models use forecasts from numerical weather prediction (NWP) tools at the location of the PV plant as well as the past recorded values of PV hourly electric power production. The APVF model consists of an original modeling for adjusting irradiation data of clear sky by an irradiation attenuation index, combined with a PV power production attenuation index. The MPVF model consists of an artificial neural network based model (selected among a large set of ANN optimized with genetic algorithms, GAs). The two models use forecasts from the same NWP tool as inputs. The APVF and MPVF models have been applied to a real-life case study of a grid-connected PV plant using the same data. Despite the fact that both models are quite different, they achieve very similar results, with forecast horizons covering all the daylight hours of the following day, which give a good perspective of their applicability for PV electric production sale bids to electricity markets.

A comparative study of electric load curve changes in an urban low-voltage substation in Spain during the economic crisis (2008-2013).

This paper presents a comparative study of the electricity consumption (EC) in an urban low-voltage substation before and during the economic crisis (2008–2013). This low-voltage substation supplies electric power to near 400 users. The EC was measured for an 11-year period (2002–2012) with a sampling time of 1 minute. The study described in the paper consists of detecting the changes produced in the load curves of this substation along the time due to changes in the behaviour of consumers. The EC was compared using representative curves per time period (precrisis and crisis). These representative curves were obtained after a computational process, which was based on a search for days with similar curves to the curve of a determined (base) date. This similitude was assessed by the proximity on the calendar, day of the week, daylight time, and outdoor temperature. The last selection parameter was the error between the nearest neighbour curves and the base date curve. The obtained representative curves were linearized to determine changes in their structure (maximum and minimum consumption values, duration of the daily time slot, etc.). The results primarily indicate an increase in the EC in the night slot during the summer months in the crisis period.