Abstract
With the rapid growth of photovoltaic (PV) power in recent years, the stability of system operation, the performance of system contingency analysis as well as the power quality of the power grid are threatened by the inherent uncertainty and fluctuation of PV output. It is necessary to have the knowledge of PV output characteristics for reliable power system dispatching. Day-ahead PV power forecasting is an effective support for achieving optimal dispatching. Probabilistic forecasting can describe the uncertainty that is difficult to depict by deterministic forecasting, and the forecasting results are more comprehensive. An ensemble nonparametric probabilistic forecasting model of PV output is proposed based on the traditional deterministic forecasting method. Quantile regression averaging (QRA) is used to ensemble a group of independent long short-term memory (LSTM) deterministic forecasting models for obtaining the probabilistic forecasting of PV output. Real measured data are used to verify the effectiveness of this nonparametric probabilistic forecasting model. Additionally, in comparison with the benchmark methods, LSTM-QRA has higher prediction performance due to the better forecasting accuracy of independent deterministic forecasts.
Highlights
In recent years, new energy power generation technologies such as photovoltaic (PV) and wind power have developed rapidly
In reference [20], quantile regression (QR) and numerical weather prediction (NWP) are used to obtain a set of probabilistic forecasting results for PV output, and ensemble probabilistic forecasting is achieved based on the optimal weights calculated by minimizing the continuous ranked probability score (CRPS)
To further analyse the relationship between the quantile regression averaging (QRA) ensemble probabilistic forecasting model and deterministic forecasting models, artificial neural networks (ANNs)-QRA, deep neural networks (DNNs)-QRA and Persistence-QRA are used as benchmark models
Summary
New energy power generation technologies such as photovoltaic (PV) and wind power have developed rapidly. F. Mei et al.: Day-Ahead Nonparametric Probabilistic Forecasting of PV Power Generation for meteorological inputs for time series model. Machine learning models mainly include neural network, support vector machine, random forest and long short-term memory (LSTM). A set of LSTM and deep neural network forecasting model combined with stationary wavelet transform is proposed in [14]. In reference [20], quantile regression (QR) and NWP are used to obtain a set of probabilistic forecasting results for PV output, and ensemble probabilistic forecasting is achieved based on the optimal weights calculated by minimizing the continuous ranked probability score (CRPS). To combine them and seek the connection between deterministic and probabilistic forecasting, an ensemble probabilistic forecasting model for PV output is proposed in this article. Quantile regression averaging (QRA) is used to combine independent deterministic prediction results to achieve probabilistic forecasting. Because the number of hidden layer units and output dimensions are usually different, a fully connected layer is usually added at the end of the output gate
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have