Summary
Solargis aimed to transform its solar energy data analysis capabilities. By leveraging both statistical methods and machine learning, we provided comprehensive solutions that automated data quality control and improved processing efficiency. Our work included building statistical and machine learning models, enhancing the data infrastructure, and reducing manual data operation time from hours to minutes. The project utilized tools such as Docker, Python, and Scikit-learn, resulting in significant improvements in data handling and operational efficiency.
Challenge
Motivation to change
Solargic wanted to create an effective transformation of their know-how in analysis of solar energy data.
Solution
Change delivery
By tackling the problem from both statistical and machine learning perspective we proposed comprehensive solutions covering a large spectrum of aspects.
We have automatized the quality control of various data related to the solar energy sector. We have built statistical models and simple machine learning models and helped to extend the data infrastructure and processes of this project.
Tools & means
- Docker
- Python
- Pandas, NumPy, SciPy
- Matplotlib
- Scikit-learn
- Keras, TensorFlow
Outcomes
Change outcomes
The project resulted in manual work reduction of data operators from hours to minutes.
Used services
-
7 areasfound to be candidates for automation and optimization