| Work Detail |
Spanish researchers have created a novel photovoltaic energy forecasting method that uses only direct radiation as a parameter. They found it to be “comparable, if not superior,” to four established forecasting techniques. The method could help homeowners with photovoltaic systems decide when to use electricity-intensive appliances and cleaning systems.
A research group led by the Polytechnic University of Valencia has developed a novel single-parameter power prediction method for residential photovoltaic installations.
The proposed approach defines forecast data in intervals rather than absolute figures, according to the scientists, who note that it transparently recognises and communicates natural variability in solar PV generation.
“Choosing a single-parameter model was a strategic decision aimed at simplifying the forecasting process,” the research group noted. “While multiparameter models can offer a more nuanced view, they often entail greater computational complexity and higher resource demands. Our simplified model promises ease of integration and user-friendliness, crucial for residential users and small-scale PV installations.”
The core of the novel method is the selection of days similar in the past in terms of direct radiation to forecast the power generation for a given day. For each prediction, a confidence level of 80% and a total of 10 similar days are selected. After identifying the similar days, the method uses a quantile-based approach to establish the prediction intervals, setting an upper and a lower limit. In statistics, quantiles are used to divide the range of a probability distribution into continuous intervals with equal probabilities.
The system was trained and tested using a case study of a residential installation in Spain, consisting of 12,450 W panels and a 5 kW inverter for self-consumption, all installed in 2018. Hourly PV generation was recorded for the years 2019, 2020, 2021 and 2022. Hourly meteorological data for the area was obtained from the Open Meteo database.
The forecasting technique was used to predict PV power generation in 2020, based on the algorithm of searching for similar days always within a two-year interval before the target day. In the same period, it was compared with four classical forecasting methods: linear regression model (Alt1); gradient boosting regressor (Alt2); gradient boosting with lags (Alt3); and long-term memory network (LSTM) (Alt4).
“The performance of the models was assessed using key performance indicators (KPIs) such as prediction accuracy, prediction interval width, true confidence level, and mean error. This comprehensive approach ensured a balanced evaluation, highlighting the strengths and limitations of each method,” the researchers say.
The proposed method obtained a mean absolute error (MAE) of 0.1490 kW, a root mean square error (MSE) of 0.0917 kW2, a root mean square error (RMSE) of 0.3029 kW, a mean bin width (AWI) of 0.3365 kW, a probability of coverage (CP) of 91.55% and an overall interval error (OIE) of 0.3789 kW. Alt1 showed a MAE of 0.3374 kW, a MSE of 0.2428 kW2, a RMSE of 0.4928 kW, an AWI of 0.9312 kW, a CP of 78.69% and an OIE of 0.4117 kW.
Alt2 recorded a MAE of 0.2558 kW, a MSE of 0.2044 kW2, a RMSE of 0.4521 kW, an AWI of 0.7464 kW, a CP of 80.12%, and an OIE of 0.4031 kW. Alt3 recorded a MAE of 0.1379 kW, a MSE of 0.0768 kW2, a RMSE of 0.2771 kW, an AWI of 0.4890 kW, a CP of 91.72%, and an OIE of 0.2355 kW. Alt4 showed a MAE of 0.1282 kW, a MSE of 0.0684 kW2, a RMSE of 0.2616 kW, an AWI of 0.3522 kW, a CP of 80.72%, and an OIE of 0.2642 kW.
After analyzing the numerical results, the researchers found that the proposed approach could help PV system owners achieve energy savings. According to their results, the average monthly energy bill was reduced from €44.3 ($47.96) to €37.48, as the energy imported from the grid decreased by 45.79 kWh, from 278 kWh to 232.21 kWh.
“By simply adjusting the operating times of the pool filtration system, washing machine and dishwasher to align with peak solar production hours, homeowners have been able to harness more solar energy, reducing grid dependency and lowering overall energy costs,” they conclude. “With advances in home automation technology, even greater results can be achieved.”
Their findings were presented in “ Interval-based solar photovoltaic energy predictions: A single-parameter approach with direct radiation focus ,” published in Renewable Energy . The group consisted of scientists from the Universidad Politécnica de Valencia, the University of Valencia, and the Universidad Politécnica Salesiana de Ecuador. |
