Solar Power Meteorological Bureau

How are solar power forecasts calculated?

The forecast is computed based on the selected parameters that are unique to your PV panels. To calculate solar power forecasts, our model combines several weather models and forecasting methods to generate the most accurate projections. The data presented on this website are for personal use and planning.

How can I find predicted solar output data?

Discover predicted solar output data based on your location, orientation, and other parameters of your solar panels. Fill out the form below and see the current solar production forecast or historical output up to 20 years in the past. Data are based on the machine learning combination of various different weather models and cover the whole world.

How does solar PV power generation forecasting work?

Solar PV power generation forecasting: Weather forecasting is selected based on data characteristics, and machine learning or optimization algorithms are added to the solar PV power generation prediction model, for example, optimization algorithms with RNN-LSTM, to optimize the superparameters and enhance the prediction accuracy.

How machine learning is used in solar PV power forecasting?

Neural networks (ANNs) are the most frequently used machine learning techniques in short-term solar PV power forecasting. Hybrid predictive models are designed by combining two or three deep learning techniques or combining optimization algorithms with AI methods.

Can a localised model predict solar power generation?

However, conditions impacting solar power generation, such as cloud cover or aerosols, can be much more localised. Localised modelling may be more effective for predicting solar power generation than traditional forecasting. As renewable generation capacity increases through expanding renewable infrastructure, the need for storage decreases.

Can GBRT-med-KDE predict global solar radiation?

Zhang et al. (2022) proposed the hybrid gradient boosting regression tree–median and kernel density estimation (GBRT-Med-KDE) models. This study proposes a short-term solar power interval prediction method for solar PV power generation, which effectively predicts global solar radiation.