Search

A large proportion of photovoltaic (PV) power generation is connected to the power grid, and its volatility and stochasticity have significant impacts on the power system. Accurate PV power forecasting is of great significance in optimizing the safe operation of the power grid and power market transactions. In this paper, a novel dual-channel PV power forecasting method based on a temporal convolutional network (TCN) is proposed. The method deeply integrates the PV station feature data with the model computing mechanism through the dual-channel model architecture; utilizes the combination of multihead attention (MHA) and TCN to extract the multidimensional spatio-temporal features between other meteorological variables and the PV power; and utilizes a single TCN to fully extract the temporal constraints of the power sequence elements. The weighted fusion of the dual-channel feature data ultimately yields the ideal forecasting results. The experimental data in this study are from a 26.52 kW PV power plant in central Australia. The experiments were carried out over seven different input window widths, and the two models that currently show superior performance within the field of PV power forecasting: the convolutional neural network (CNN), and the convolutional neural network combined with a long and short-term memory network (CNN_LSTM), are used as the baseline models. The experimental results show that the proposed model and the baseline models both obtained the best forecasting performance over a 1-day input window width, while the proposed model exhibited superior forecasting performance compared to the baseline model. It also shows that designing model architectures that deeply integrate the data input method with the model mechanism has research potential in the field of PV power forecasting. [ABSTRACT FROM AUTHOR]

• A quad-kernel convolutional neural network model for PV power forecasting. • Sequence-to-sequence learning method to capture timing-dependent features. • Multiple global-max-pooling algorithms to improve the learning efficiency. • The effect of dataset resolution on the model performance is studied. • Operation data prove the higher forecasting accuracy of the proposed model. Photovoltaic (PV) power is highly stochastic and volatile, and PV power forecasting is a key technology to guarantee the safe and economic operation of high-penetration renewable power systems. To improve the accuracy of PV power forecasting, a quad-kernel deep convolutional neural network (QK_CNN) model is proposed to perform intra-hour PV power forecasting for the next four timesteps: four CNNs with different kernel sizes are used to extract different local cross features between sequence elements of four timesteps; a single-kernel CNN is used to further feature extraction of these features, and then the target sequence forecasting results are obtained; global maximum pooling method is used to simplify the feature extraction process and improve model learning efficiency. Operation data from a 26.52 kW PV plant in Central Australia is selected as the experimental data. Compared with single-kernel CNN and hybrid models (CNN_LSTM) on 5, 10, and 15 min of resolution data, respectively, the proposed model shows better forecasting performance and is able to explain 96 % to 98 % of the total variation of the forecasted PV power. All these demonstrates that the CNNs with specific design have great potential to handle the task of PV power forecasting as well. [ABSTRACT FROM AUTHOR]