A Power Forecasting Method for Ultra-Short-Term Photovoltaic Power Generation Using Transformer Model

Table 1

Advantages and disadvantages of several PV forecasting hybrid methods.

	Advantages	Disadvantages
ACO-SVM	It has good accuracy and good global search ability	The ant colony algorithm is easy to fall into the local optimum
ABC-SVM	It has fewer control parameters, stronger optimization ability, and higher prediction accuracy	There are difficulties with the initialization parameter settings
LSTM-FC	It further studies the time correlation for improving the prediction accuracy	It only considers the time correlation and did not catch the correlation between the weather data and the PV power generation.
ELM-LSTM	This model has the characteristics of fast running speed and low complexity	It has the defects of easy overfitting and easy to be affected by outliers.
PSO-ANFISA	It improves the performance, stability, and reliability of model extraction data features	The algorithm needs to develop new models to improve the prediction accuracy