Related Concept Videos
Light Acquisition
Precipitation Processes
Precipitation and Co-precipitation
Prediction Intervals
However, the point estimate is most likely not the exact value of the population parameter, but close to it. After calculating point estimates, we construct interval estimates, called confidence intervals or prediction intervals. This prediction interval comprises a range of values unlike the point estimate and is a better predictor of the observed sample value, y.
Multiple Regression
Farmers can use multiple regression to determine the crop yield based on more than one factor, such as water availability, fertilizer, soil properties, etc. Here, the crop yield is the response or dependent variable as it depends on the other independent variables. The analysis requires the construction of a scatter plot...
Survival Tree
Building a Survival Tree
Constructing a...
You might also read
Related Articles
Articles linked to this work by shared authors, journal, and citation graph.
Ferroptosis and Wnt/β-Catenin Signaling Triggered by Environmentally Relevant Nanoscale Polypropylene Plastics in Human Intestinal Models.
Forecasting migraine with time-series machine learning from mobile health data.
Related Experiment Video
Updated: Jun 30, 2025

A Method of Trigonometric Modelling of Seasonal Variation Demonstrated with Multiple Sclerosis Relapse Data
Published on: December 9, 2015
Attention-Based Models for Multivariate Time Series Forecasting: Multi-step Solar Irradiation Prediction.
Sadman Sakib1, Mahin K Mahadi1, Samiur R Abir1
1Department of Electrical and Electronic Engineering, Islamic University of Technology, Gazipur, 1704, Bangladesh.
Accurate solar irradiance forecasting in Bangladesh is crucial for managing solar power. Attention-based models, particularly the Temporal Fusion Transformer (TFT), significantly improve prediction accuracy for photovoltaic systems.
More Related Videos
00:10X-ray Beam Induced Current Measurements for Multi-Modal X-ray Microscopy of Solar Cells
Published on: August 20, 2019
09:00Indoor Experimental Assessment of the Efficiency and Irradiance Spot of the Achromatic Doublet on Glass ADG Fresnel Lens for Concentrating Photovoltaics
Published on: October 27, 2017
Area of Science:
- Renewable Energy Systems
- Artificial Intelligence in Energy
- Time Series Forecasting
Background:
- Bangladesh's subtropical climate offers high solar panel efficiency.
- Accurate solar irradiance forecasting is vital for grid-connected photovoltaic systems to manage power variability.
- Existing forecasting models struggle with the long-term sequential dependencies inherent in solar irradiation data.
Purpose of the Study:
- To develop and evaluate an attention-based model framework for multivariate solar irradiance time series forecasting.
- To assess the performance of Attention-based encoder-decoder, Transformer, and Temporal Fusion Transformer (TFT) models.
- To compare these advanced models against traditional forecasting methods.
Main Methods:
- Utilized 30-minute resolution solar irradiance data from two locations in Bangladesh.
- Trained and tested Attention-based encoder-decoder, Transformer, and Temporal Fusion Transformer (TFT) models.
- Evaluated model performance for predicting solar irradiance 24 steps ahead.
Main Results:
- Attention mechanisms significantly enhanced prediction accuracy.
- The Temporal Fusion Transformer (TFT) demonstrated superior precision and robustness compared to other models.
- TFT achieved a Mean Squared Error (MSE) of 0.151, Mean Absolute Error (MAE) of 0.212, and R-squared (R²) of 0.815.
Conclusions:
- Attention-based models, especially TFT, are highly effective for solar irradiance forecasting.
- TFT offers substantial improvements over benchmark and sequential models, reducing MSE by up to 47.9% and MAE by up to 22.3%.
- The capability of attention models to capture long-distance dependencies boosts predictive power for solar energy applications.