la transformation digitale du monde de l nergie data
play

LA TRANSFORMATION DIGITALE DU MONDE DE L'NERGIE Data science in - PowerPoint PPT Presentation

LA TRANSFORMATION DIGITALE DU MONDE DE L'NERGIE Data science in energy industry Data science in energy industry 40% of the mondial energy consumption Data science in energy industry 40% of the mondial energy consumption Data science in


  1. LA TRANSFORMATION DIGITALE DU MONDE DE L'ÉNERGIE

  2. Data science in energy industry

  3. Data science in energy industry 40% of the mondial energy consumption

  4. Data science in energy industry 40% of the mondial energy consumption

  5. Data science in energy industry 40% of the mondial energy consumption

  6. Data science in energy industry

  7. Data science in energy industry

  8. Data science in energy industry

  9. Data science in energy industry Machine learning

  10. Machine learning challenges

  11. Machine learning challenges 80% of the job

  12. Machine learning challenges 20% of the job

  13. Machine learning challenges 20% of the job Suitable for competitions

  14. Why ? • Critical role in energy efficiency • Optimize operations of chillers, boilers and energy storage systems • Baseline for flagging potentially wasteful discrepancies ⇒ Forecasting the use of the electrical energy is the backbone of effective operations

  15. Forecasting building energy consumption Com Competitio ion Da Data • Energy consumption historic for ~200 buildings • Temperature

  16. Forecasting building energy consumption Com Competitio ion Da Data ??? • Energy consumption historic for ~200 buildings • Temperature Com Competitio ion Ob Objective • Forecast Energy consumtption through different horizons

  17. Winner solution

  18. Winner solution

  19. Winner solution

  20. Feature engineering

  21. Feature engineering ~ X ~ y

  22. Feature engineering Engineered features ~ X ~ y

  23. Feature engineering Cyclical time encoding D= 20h

  24. Feature engineering Cyclical time encoding

  25. Feature engineering Cyclical time encoding

  26. Feature engineering Cyclical time encoding

  27. Feature engineering Cyclical time encoding

  28. Feature engineering Cyclical time encoding ???

  29. Feature engineering Cyclical time encoding 𝒕𝒋𝒐 𝟑𝝆𝒊 & 𝐝𝐩𝐭 𝟑𝝆𝒊 𝟑𝟓 𝟑𝟓

  30. Winner solution

  31. Winner solution Boosted trees

  32. Boosted trees Boosted trees BT BT You

  33. Boosted trees Decision trees

  34. Boosted trees Decision trees Depth

  35. Boosted trees Decision trees Prediction

  36. Boosted trees Boosting Prediction

  37. Boosted trees Boosting Prediction Reality

  38. Boosted trees Boosting Prediction - Reality Error

  39. Boosted trees Boosting Prediction - Reality Error

  40. Boosted trees Boosting Original data

  41. Boosted trees Boosting Original data Decsion tree

  42. Boosted trees Boosting Original data Decsion tree Error

  43. Boosted trees Boosting Original data Decsion tree Error

  44. Boosted trees Boosting … Original data Decsion tree Error

  45. Boosted trees Boosting … Original data Decsion tree Error Nb of trees

  46. • Improve the state of the Art • Create a community • Provide a solution to a typical Energy problematic ➔ This solution can now be used in other context

  47. Why ? • Flexibility in energy management is essential for secure supply and increasing the penetration of renewable sources. • Energy storage and local production can increase smart building flexibility. • Time of use tariffs can incite use of energy when it is the most available. ⇒ Algorithms can help battery charging systems to be as efficient as possible

  48. Competition Description Com Competitio ion Da Data • Actual Consumption and Production (for 11 buildings) • Forecast for next 24h • Grid energy price (sell and buy)

  49. Competition Description Com Competitio ion Da Data • Actual Consumption and Production (for 11 buildings) • Forecast for next 24h • Grid energy price (sell and buy) Com Competitio ion Ob Objective • Plannify a battery usage to save money How to use the battery for the next 15 minutes ?

  50. Competition Results Perf erformance Metric ic

  51. Competition Results Perf erformance Metric ic Be Best Co Competiti tion score: drives 19% savings with a battery.

  52. Linear Programming

  53. Linear Programming

  54. Linear Programming

  55. Linear Programming Iss Issue: Future consumption and prediction are unknown. We only have forecastings.

  56. Forecasting Error

  57. Scenario based stochastic programming

  58. Scenario based stochastic programming

  59. Scenario based stochastic programming

  60. Results Scores es Method Percentage of saving with a battery Our method 19,6 % 1 st competition method 19,4 % 2 nd competition method 19,2 % 3 rd competition method 19,1

  61. Results Scores es Method Percentage of saving with a battery Our method 19,6 % 1 st competition method 19,4 % 2 nd competition method 19,2 % 3 rd competition method 19,1 Want to go further ? https://github.com/kaizen-solutions/power-laws-optimization

  62. • Algorithms driving 19% of savings with a battery • Algorithms and comparison code are on github

  63. Conclusion Business Continuous Open Sources needs Improvement • Business • Understand • Compare context Solutions with existing • True dataset • Formation • Community

  64. Conclusion Business Continuous Open Sources needs Improvement • Business • Understand • Compare context Solutions with existing • True dataset • Formation • Community

  65. Conclusion Business Continuous Open Sources needs Improvement • Business • Understand • Compare context Solutions with existing • True dataset • Formation • Community

  66. Any questions ?

  67. Winner solution

  68. Data Collection Feature engineering Final model Model building Data viz - QC - Transfo Problem formulation

Recommend


More recommend