Tyler McCandless - PhD Defense
(Penn State, Department of Meteorology)
"Artificial Intelligence Techniques for Short-Range Solar Irradiance Prediction"
|What||PhD Defense Homepage GR|
Aug 24, 2015 09:00 AM
Aug 24, 2015 10:00 AM
Aug 24, 2015
from 09:00 am to 10:00 am
|Where||529 Walker Building|
|Contact Name||Tyler McCandless|
|Add event to calendar||
"Advisors: Sue Ellen Haupt and George Young"
The world’s energy system will increasingly depend upon renewable energy sources, including solar power, due to the limitation of fossil fuel resources and their influence on global pollution and climate change. Solar power can provide substantial power supply to the grid; however, it is also a highly variable energy source. Changes in weather conditions, i.e. clouds, can cause rapid changes in solar power output, thus creating a challenge for utility companies to effectively use these renewable energy resources. Utilities will require solar irradiance forecasts to effectively balance the energy grid as the penetration of solar power increases. This study presents multiple nonlinear forecasting techniques to predict both the magnitude of the solar irradiance and its expected variability, which together are necessary for utility companies and systems operators to plan resources for efficient energy grid management.
In Part 1, the temporal irradiance variability is forecast for the temporal standard deviation of the Global Horizontal Irradiance (GHI) at eight sites in the Sacramento Valley of California and the spatial irradiance variability is forecast for the standard deviation across those same sites. A model tree with a nearest neighbor option was trained to predict the irradiance variability. Results indicate that the model tree technique can be applied in real time to produce solar variability forecasts to aid utility companies in energy grid management. In Part 2, a cloud regime-dependent short-range solar irradiance forecasting system is developed to provide 15-min average clearness index forecasts for 15-min, 60-min, 120-min and 180-min lead-times. A k-means algorithm identifies the cloud regime based on a combination of surface weather observations and irradiance observations. Then, Artificial Neural Networks (ANNs) are trained on each cloud regime to predict the clearness index. This regime-dependent system not only makes a more accurate deterministic forecast than a global ANN or clearness index persistence, but also the regime identification produces more accurate predictions of expected irradiance variability than assuming climatological average variability. In Part 3, regime-identification methods that also incorporate GOES-East satellite data both as inputs to the k-means regime algorithm and as predictors to the ANNs are explored. Several cloud-regime dependent short-range solar irradiance forecasting systems (RD-ANN) are tested to make 15-min average clearness index predictions for 15-min, 60-min, 120-min and 180-min forecast lead-times. The RD-ANN system that shows the lowest forecast error on independent test data classifies cloud regimes with a k-means algorithm based on a combination of surface weather observations, irradiance observations and GOES-East satellite data.
Using statistical techniques, such as those described in this work, allows for improved deterministic solar irradiance predictions as well as improved spatial and temporal solar irradiance variability predictions. The combination of deterministic solar irradiance predictions and predictions of the solar irradiance variability provides utility companies and systems operators with the necessary information to efficiently manage the grid.