Table of Content
Below is a list of my research keywords:
- Ensemble Weather Forecasting
- Uncertainty Quantification of Renewable Energy
- Artificial Intelligence
- High Performance Computing
My dissertation title is Uncertainty Quantification for Photovoltaic Energy Production Using Analog Ensemble. I’m investigating the predictability of photovoltaic solar energy production over the Continental US. The uncertainty of solar energy production not only originates from the variation of solar irradiance, but also other weather conditions, including wind speed and ambient temperature, together with the configuration of the particular energy performance system. I’m using Analog Ensemble to study the spatial and temporal predictability of photovoltaic solar energy generation.
There are several key features of the project:
- The evaluation of uncertainty via ensembles: Analog Ensemble is used to generate ensembles from a deterministic model and then ensembles are used for uncertainty evaluation.
- Coupled power simulation system: The weather forecast system has been coupled with a solar power performance system to study the predictability of the power generation.
- High-performance computing: The research domain is the continental US with a 12 km spatial resolution and an hourly temporal resolution. This requires computing capabilities from supercomputers.
Here is a curated list of posts that showcase my research and document my research experiences. The summary of the research project is followed by an excerpt. Comments are well welcome.
 Scholarly Poster Guidelines
Excerpt: My curated list of guidelines for making scholarly posters
 Predictability Index for Renewable Energy and Uncertainty Quantification with Analog Ensemble
Excerpt: Uncertainty quantification of solar photovoltaic energy production with ensemble forecasts and performance system simulations. AGU poster. Publicly available after January 2021.
Excerpt: An introduction to the open-source implementation of the Analog Ensemble technique
 Using Long Short-Term Memory (LSTM) and Internet of Things (IoT) for Localized Surface Temperature Forecasting in an Urban Environment
Excerpt: Surface temperature time sequence forecasts with high spatio-temporal resolution via a deep network. AGU poster. Publicly available after January 2021.
 Machine Learning and Analog Ensemble Techniques for Temporal Extraction of Wind Resource Uncertainty
Excerpt: A comparative study on the performance of Analog Ensemble and a gradient boosting model for modeling wind uncertainty. AGU poster. Publicly available after January 2021.
Excerpt: Using a Genetic Algorithm for Optimal Location Finding in Weather Predictions
Excerpt: Running an efficient ensemble forecast technique with Ensemble Toolkit on HPC
Excerpt: Using Analog Ensemble to forecast photovoltaic energy production on a household level
Excerpt: A performance comparison of different implementations of Kosaraju's algorithm
Excerpt: Using an unstructured grid for better weather predictions
I’m passionate in programming and I’m an active contributor to open-source projects and the community.
I have been actively developing and maintaining the following packages and code repositories. They are developed during my current and previous courses of research. Some of them were converted into packages for a broader research community.
- Parallel Analog Ensemble (C++/R)
- RAnEnExtra (R)
- Solar Photovoltaic Energy Simulator (Python)
- EITrans: Empirical Inverse Transform Function (R)
- More Parallel R (R)
For more about my technical skills, please see my posts organized by tags.