Satellite positioning watch solar power generation

How do satellite images affect PV power forecasting?

Furthermore, the predicted satellite images can track the movement of clouds, which significantly affects the PV power forecasting. For example, a moving cloud can cover the PV panel directly, and PV power can be temporarily lowered.

Can a solar power forecasting network be trained with satellite images?

To address this problem, we propose a PV power forecasting model, including a cloud amount forecasting network trained with satellite images. In addition, our proposed model adopts convolutional self-attention to effectively capture historical features, and thus acquire helpful information from weather forecasts.

How can remote-sensing technology-based satellite images improve PV power plant forecasting?

For the latter one, remote-sensing technique-based satellite images usually provide forecasts ranging from half an hour to 6 h, and offers a wider view of observation. Therefore, it is conducive to achieve forecasts in the environment of growing number of PV power plant clusters .

Can a satellite image predict photovoltaic power?

Conclusions In this paper, a novel satellite image-based approach for photovoltaic power forecast is proposed. The hourly satellite image variation is predicted using Conv-LSTM, which takes the nonlinear cloud motion into account, providing more accurate cloud movement trajectories.

How a low update frequency satellite image is used in PV power forecast?

Therefore, estimating the intra-hour satellite images through the low update frequency satellite images plays an indispensable role in PV power forecast. To match the general temporal resolution requirements of the ultra-short-term PV power forecast, the SCRS algorithm is developed based on the ACRS rule.

What is a satellite image – PV power mapping model?

A “satellite image – PV power” mapping model is established by using the Back-Propagation neural network (BPNN) with model output solar data records and other factors identified above as model inputs.