Regression analysis helps identify malfunction of controls in an industrial baking oven

Please download to get full document.

View again

of 16
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Categories
Published
1. Industrial Case Study: Weston Bakery Sudbury Anatoli Naoumov Managing Director, GreenQ Partners RETScreen PLUS Webinar – December 8,2014 2. *Grace Murray Hopper was…
  • 1. Industrial Case Study: Weston Bakery Sudbury Anatoli Naoumov Managing Director, GreenQ Partners RETScreen PLUS Webinar – December 8,2014
  • 2. *Grace Murray Hopper was an American computer scientist and United States Navy rear admiral. Let’s see what an accurate analysis is worth for a business
  • 3. Weston Bakery Sudbury  WBS Industrial bakery within Weston Foods, same owners as Loblaw's  Annual production:  14,600 ton of bread and  3,000 ton of rolls  Burn over 800,000 m3 of gas per year
  • 4. Energy Management situation  Corporate efficiency goal: 2% consumption decrease per year  Natural gas consumption metric: m3/ton of baked product per 28-day period  Metering Submeters at bread oven and at rolls oven Manual data collection Processed in Excel by …  A single Energy Engineer for 30+ facilities
  • 5. Issue: gas consumption grows  Rate of natural gas consumption (m3/ton of bread) has been increasing since 2010  Weston estimated waste of 65,000 m3 in 2011-13  No known reason
  • 6. What did bakery see? – A problem 15.0 20.0 25.0 30.0 35.0 20,000 25,000 30,000 35,000 1-Jan-10 1-Mar-10 1-May-10 1-Jul-10 1-Sep-10 1-Nov-10 1-Jan-11 1-Mar-11 1-May-11 1-Jul-11 1-Sep-11 1-Nov-11 1-Jan-12 1-Mar-12 1-May-12 1-Jul-12 1-Sep-12 1-Nov-12 1-Jan-13 1-Mar-13 1-May-13 1-Jul-13 1-Sep-13 1-Nov-13 1-Jan-14 1-Mar-14 1-May-14 1-Jul-14 1-Sep-14 Nat. gas, m3 m3/ton Consumption rate is unstable and grows Ovenconsumption,m3 Consumptionrate,m3/ton
  • 7. What is CUSUM? CUSUM is a cumulative sum of differences between Expected and Actual consumption over a period of time. Why is CUSUM important? CUSUM shows long-term changes in consumption – savings or waste – and cancels out short-term “noise”.  CUSUM goes down = savings, Actual < Predicted  CUSUM goes up = losses, Actual > Predicted How to calculate CUSUM? 1. Determine a formula that describes consumption during analysis period, e.g., year: Consumption = A*kg + B*temp + C ±error, per day or week Formula (model) is determined via statistical analysis 2. Calculate expected consumption during analysis period using this formula 3. Actual minus Expected equals difference for each measurement period – day, week, or month 4. Sum up the differences over the analysis period
  • 8. Example: AC consumption When more than 1 factor drives consumption, noticing changes in consumption pace is way more difficult
  • 9. What did analysis show? Saving opportunities and business benefits! *More specifically: 87,300±3,700m3 with 90% confidence We found Value to client • Oven efficiency dropped every April for 3 years, • Company wasted 87,000m3 by 2013* • Waste in actual operation conditions • Ability to monitor oven efficiency • Fixing maintenance procedure (no cost) eliminated waste worth about $10,000 Outside temperature materially affects oven efficiency Preheating combustion air offers major savings Errors in data collection process Clean data improves product costing, budgeting, quality of operational decisions
  • 10. CUSUM analysis April Baseline
  • 11. CUSUM analysis and result April Baseline Oven recalibration
  • 12. Replay: What did bakery see? – A problem 15.0 20.0 25.0 30.0 35.0 20,000 25,000 30,000 35,000 1-Jan-10 1-Mar-10 1-May-10 1-Jul-10 1-Sep-10 1-Nov-10 1-Jan-11 1-Mar-11 1-May-11 1-Jul-11 1-Sep-11 1-Nov-11 1-Jan-12 1-Mar-12 1-May-12 1-Jul-12 1-Sep-12 1-Nov-12 1-Jan-13 1-Mar-13 1-May-13 1-Jul-13 1-Sep-13 1-Nov-13 1-Jan-14 1-Mar-14 1-May-14 1-Jul-14 1-Sep-14 Nat. gas, m3 m3/ton Consumption rate is unstable and grows Ovenconsumption,m3 Consumptionrate,m3/ton
  • 13. What did bakery get? – A result 15.0 20.0 25.0 30.0 35.0 20,000 25,000 30,000 35,000 1-Jan-10 1-Mar-10 1-May-10 1-Jul-10 1-Sep-10 1-Nov-10 1-Jan-11 1-Mar-11 1-May-11 1-Jul-11 1-Sep-11 1-Nov-11 1-Jan-12 1-Mar-12 1-May-12 1-Jul-12 1-Sep-12 1-Nov-12 1-Jan-13 1-Mar-13 1-May-13 1-Jul-13 1-Sep-13 1-Nov-13 1-Jan-14 1-Mar-14 1-May-14 1-Jul-14 1-Sep-14 Nat. gas, m3 m3/ton After recalibration in Jan 2014 consumption rate dropped below 2010 level
  • 14. RETScreen Plus value Technology agnostic  Can be used for analysis of any utility Solution agnostic  Can be used for any energy conservation technology, including behavioural changes Scalable  Can be used for one machine or for a whole facility Smart  Sensitive to sustained changes and insensitive to short-term deviations (aka, noise) Convincing  After client has bought into the idea of statistical analysis
  • 15. Lessons learned  Validate data  Submetering is a must  Automated data collection is a bliss  Analysis results may upset plant manager  Collaborate with client to understand their process and validate findings  Beware of statistical analysis uncertainty: “CUSUM = 100,000” does not mean “company saved (lost) 100,000”
  • 16. Q&A and Final remarks Thank you for your attention. Questions?  Want more information? - Business cases If you want to receive a 1-page business cases on the discussed and other projects - please email me  Still more information? - http://greenq.ca/blog/ Irregular blog on energy conservation. ONLY valuable information. Just check previous posts.  Refer us If you know about a group that could benefit from the same kind of program we just did together - please contact me and thank you in advance for your generosity. Anatoli Naoumov MBA, MSc, CMVP Managing Director 416.728.7239 anaoumov@greenq.ca If a question comes to mind later – feel free to contact me www.greenq.ca
  • We Need Your Support
    Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

    Thanks to everyone for your continued support.

    No, Thanks