Power grid operators rely on solar irradiance forecasts to manage uncertainty and variability associated with solar power. Meteorological factors such as cloud cover, wind direction, and wind speed affect irradiance and are associated with a high degree of variability and uncertainty. Statistical models fail to accurately capture the dependence between these factors and irradiance. In this paper, we introduce the idea of applying multivariate Gated Recurrent Units (GRU) to forecast Direct Normal Irradiance (DNI) hourly. The proposed GRU-based forecasting method is evaluated against traditional Long Short-Term Memory (LSTM) using historical irradiance data (i.e., weather variables that include cloud cover, wind direction, and wind speed) to forecast irradiance forecasting over intra-hour and inter-hour intervals. Our evaluation on one of the sites from Measurement and Instrumentation Data Center indicate that both GRU and LSTM improved DNI forecasting performance when evaluated under different conditions. Moreover, including wind direction and wind speed can have substantial improvement in the accuracy of DNI forecasts. Besides, the forecasting model can accurately forecast irradiance values over multiple forecasting horizons.
more »
« less
Short-term solar irradiance forecasting using convolutional neural networks and cloud imagery
Access to accurate, generalizable and scalable solar irradiance prediction is critical for smooth solar-grid integration, especially in the light of the accelerated global adoption of solar energy production. Both physical and statistical prediction models of solar irradiance have been proposed in the literature. Physical models require meteorological forecasts—generated by computationally expensive models—to predict solar irradiance, with limited accuracy in sub-daily predictions. Statistical models leverage
- NSF-PAR ID:
- 10362276
- Publisher / Repository:
- IOP Publishing
- Date Published:
- Journal Name:
- Environmental Research Letters
- Volume:
- 16
- Issue:
- 4
- ISSN:
- 1748-9326
- Page Range / eLocation ID:
- Article No. 044045
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
null (Ed.)With the recent discoveries and engineering solutions emerging in nanomaterials and nanostructures, independent band modulation of solar radiation on building envelopes, including glazing systems, has become increasingly viable as a potential means of improving building energy savings and indoor visual comfort. However, when it comes to the prediction of these new materials’ potential energy performance in buildings, most studies utilize a simple solar irradiance (e.g., global horizontal solar irradiance, direct beam solar irradiance) or a rough estimation of solar infrared (e.g., 50% solar irradiance) as input, which may cause significant errors. Consequently, there is a pressing need for reliable performance estimations of the solar infrared control and response at the building’s scale. To assess this, we need a solar spectral irradiance model, or at least a wideband (visible or infrared) solar irradiance model, as input. To develop this new type of model, one needs to understand the modeling-related key elements, including available solar spectral irradiance datasets, data collection methods, and modeling techniques. As such, this paper reviews the current major measurement methods and tools used in collecting solar spectral irradiance data with a focus on the solar infrared region, identifies the available related resources and datasets that particularly encompass the solar spectral irradiance data with a sufficient wavelength range, and studies existing solar irradiation modeling techniques for building simulations. These investigations will then form the background and backbone for a study scheme of solar infrared radiation modeling and indicate future research paths and opportunities.more » « less
-
Abstract Shortwave solar irradiance through building windows may have significant impacts on indoor thermal comfort, especially in near-window zones. Such effects change with intensity and spectral variations of the solar irradiance incident on building windows, which is related to the day of the year, time of day, orientation and dimension of the window, and atmospheric conditions. To assess the effects on thermal comfort, we derived a variable - mean radiant temperature delta based on a proposed spectrally-resolved method to represent the quantity of shortwave solar irradiance incident on occupants and be incorporated into PMV (predicted mean votes)-based thermal comfort models. By characterizing the variations of the calculated PMV values under different solar conditions, the influencing factors to indoor thermal comfort by shortwave solar irradiance were obtained and analyzed. Last, upon a series of parametric settings and numerical analysis, simplified statistical regression models were also established to directly predict spectrally-resolved mean radiant temperature delta and PMV values. This could be convenient and extensively to estimate the solar effects on indoor thermal comfort within the near-window zones.more » « less
-
more » « less
Abstract The Asian summer monsoon (ASM) is teleconnected to the El Niño Southern Oscillation (ENSO), but this relationship is nonstationary and has shifted significantly in recent decades. Characterizing the drivers of such shifts is crucial for improving ASM prediction and extreme event preparedness. Paleoclimate records indicate a link between ASM strength and solar activity on multidecadal‐to‐centennial timescales, but 20th‐century data are too short to test mechanisms. Here we evaluate how solar irradiance influences the ASM‐ENSO relationship using last‐millennium paleoclimate data assimilation reconstructions and model simulations. We find that high solar irradiance weakens the ENSO‐East Asian summer monsoon (EASM) correlation, but strengthens the ENSO‐South Asian summer monsoon (SASM) correlation. Solar irradiance likely influences the strength of the ENSO‐EASM and ENSO‐SASM teleconnections via changes in the Western Pacific Subtropical High and the amplitude of ENSO events, respectively. We suggest a need for considering solar activity in decadal ASM rainfall predictions under global warming scenarios.
-
Variation in solar irradiance causes power generation fluctuations in solar power plants. Power grid operators need accurate irradiance forecasts to manage this variability. Many factors affect irradiance, including the time of year, weather and time of day. Cloud cover is one of the most important variables that affects solar power generation, but is also characterized by a high degree of variability and uncertainty. Deep learning methods have the ability to learn long-term dependencies within sequential data. We investigate the application of Gated Recurrent Units (GRU) to forecast solar irradiance and present the results of applying multivariate GRU to forecast hourly solar irradiance in Phoenix, Arizona. We compare and evaluate the performance of GRU against Long Short-Term Memory (LSTM) using strictly historical solar irradiance data as well as the addition of exogenous weather variables and cloud cover data. Based on our results, we found that the addition of exogenous weather variables and cloud cover data in both GRU and LSTM significantly improved forecasting accuracy, performing better than univariate and statistical models.more » « less
