Empowering Plants with Python

Chaitra Vishweshwaraiah (~chaitra23)


In India, it’s a Government mandate, for the wind and solar farm owners to provide prediction of energy produced by the farm ahead in time. Hence, Forecasting plays an important role in reducing the uncertainty in demand and generation of electricity. Accurate forecast helps manage grid stability and also to reduce the penalties which the farm owners are liable to pay, in-case of Inaccurate prediction. The aim is to provide forecast as per government regulations, every 15 minutes using machine learning, statistics and physics-based models.

In our talk we would like to present, how python is used to

  • Handle data from solar and wind farms to forecast energy generation, day ahead and intraday

  • Forecast energy generation at different levels based on the granularity of data available

    Power Plant Level

    Turbine level or PV module level

  • Systematize handling multiple farms which

    provide different frequency of data

    have data stored in different data sources

    have different sensors' data associated with different assets

  • Devise mechanisms for data retrieval from different weather forecast providers in handling data which

    Have different units, tags

    Different forecast providers give better forecast in different locations

  • Implement various models and their performance comparisons

    Physics based model

    Statistical models

    Machine Learning models

Business Impact

  • For a farm in India, the revenue rose by about Rs 12 lakhs, in a month, after adoption of forecasting and scheduling solution. Annual cost savings and increased revenue generation brought about by the solution are significant.

  • The penalties payed to the government is drastically reduced.

  • Forecasts facilitate efficient planning and execution of maintenance activities at the farm.


  • Python basics
  • ML basics
  • Cron jobs basics
  • Curiosity to Learn

Speaker Info:

Ripunjoy Gohain

Ripunjoy is Analyst at BLP Clean Energy. He is a curious and driven data science professional with more than three year of experience in the IT industry. He has the ability to constantly mine for hidden gems located within large sets of structured and unstructured data. Master degree holder from Indian Statistical Institute with skills in Mathematics, Statistics and Quality Management. Background in computer programming language and databases learnt on the current job.

Malavika Peedinti

Malavika is the Assistant Manager, Data Analytics at BLP Clean Energy. She leads the data architecture validation team and the development activities in Analytics and Visualization. She has prior experience in handling enterprise data warehouses and crisp visualizations built using them. She has co-authored a White paper titled "Overcoming the challenges faced in a Data warehousing : A project perspective" which won a Platinum category award in an internal contest at Infosys. She has studied Data Analytics at IIIT Bangalore. She is a data enthusiast, with working knowledge of data storage, data analytics and data visualization.

Chaitra B V

Chaitra is Analyst at BLP Clean Energy. She works on Data Visualization and Analytics to create Intuitive dashboards upon analysis of solar, wind energy asset monitoring and management in the analytics practice. Previously, she worked in Risk Consulting for Financial Crime domain with PricewaterhouseCooper LLP. She Holds a Bachelor of Engineering in Information Technology from CMR Institute of Technology. She is been felicitated by the Ex-Governor of Karnataka H.R.Bharadwaj for Academic Performance.

Id: 1350
Section: Data Science, Machine Learning and AI
Type: Talks
Target Audience: Intermediate
Last Updated: