研究目的
To estimate and forecast solar irradiation using satellite data, focusing on the impact of weather signatures like dust, aerosols, fog, and clouds on solar energy availability, especially in tropical countries like India.
研究成果
The hybrid model combining satellite data, image processing, and time series models effectively forecasts GHI for time horizons up to 5 to 7 hours, despite challenges like dust storms and cloud variability. The model's accuracy improves with data assimilation from ground measurements.
研究不足
The accuracy of physical models under cloudy conditions is limited. The resolution of Numerical Weather Prediction (NWP) models is too coarse to resolve microscale physics of cloud formation. The hybrid model's computational demand is high due to the large size of satellite images.
1：Experimental Design and Method Selection:
The study employs physical and statistical models to estimate Global Horizontal Irradiance (GHI) from satellite data, including the HELIOSAT method and All-Day Index (ADI) for capturing the effect of fog, dust, and clouds.
2：Sample Selection and Data Sources:
Utilizes satellite data from INSAT-3D and ground measurements from the Indian Meteorological Department (IMD) network stations.
3：List of Experimental Equipment and Materials:
Satellite data in various spectral bands, ground-based solar irradiance measurements, and image processing tools.
4：Experimental Procedures and Operational Workflow:
Includes image processing to derive indices like Brightness Temperature Difference (BTD), segmentation using k-means clustering, and tracking of cloud and dust movements.
5：Data Analysis Methods:
Employs Wavelet Transform for image compression, ARIMA and ANN models for time series forecasting, and Kalman Filter for data assimilation.
独家科研数据包，助您复现前沿成果，加速创新突破
获取完整内容
