BTO Program Peer Review Autotune Building Energy Models

BTO Program Peer Review Autotune Building Energy Models

BTO Program Peer Review Autotune Building Energy Models 1 | Program Name or Ancillary Text Joshua New Oak Ridge National Laboratory [email protected], 865-241-8783 April 2, 2013 eere.energy.gov Purpose & Objectives Problem Statement: All (building energy) models are wrong, but some are useful 22%-97% different from utility data for 3,349 buildings More accurate models are more useful Error from inputs and algorithms for practical reasons Useful for cost-effective energy efficiency (EE) at speed and scale Calibration is required to be (legally) useful ASHRAE G14 (NMBE<5/10% and CV(RMSE)<15/30% monthly/hourly)

Manual calibration is risk/cost-prohibitive Development costs 10-45% of federal ESPC projects <$1M Need robust and scalable automated calibration for market Adjusts parameters in a physically realistic manner Scales to any available data and model (audit) 2 | Building Technologies Office eere.energy.gov Purpose & Objectives Impact of Project: Reduces transaction cost of developing and selling EE improvement projects in existing buildings Enables the ESCO business model to reach smaller buildings and projects Enables speed and scale deployment approaches based on every building in served area having a continuously maintained calibrated model (audit) Enables tracked actual performance of implemented EE measures to improve model (audit) over time Project endpoint is an automated calibration package that

users of simulation tools can deploy as they choose. 3 | Building Technologies Office eere.energy.gov Purpose & Objectives Project Focus: Objective: Develop a generalized, automated model (audit) tuning methodology that enables the model (audit) to reproduce measured data as best it can, by selecting best-match input parameters in a systematic, automated, and repeatable fashion. BTO Goals: supports the BTO overarching goal of reducing building energy use 50% by 2030 BTO strategic programs: Autotune is listed as a key service within the BTO Strategic BEM Portfolio 4 | Building Technologies Office eere.energy.gov Approach Approach: Multi-objective optimization algorithms to minimize error

between simulation output and measured data by intelligently adjusting building model inputs Sensitivity analysis and uncertainty quantification to determine importance of individual parameters Suite of machine learning algorithms to generate calibration functions based on building dynamics Quantify trade-off between tuning accuracy and amount of data available Creation of intuitive Autotune application on users PC or website with database, software tools, and accelerated tuning agents in the background 5 | Building Technologies Office eere.energy.gov Approach Approach: Demonstrations of end-to-end Autotune prototype on: ORNLs fleet of research houses and light commercial test buildings (flexible research platforms) Weatherization and audit buildings in the wild Key Issues: How well does it reproduce measured data?

How long does it take? How well does this represent the actual building? 6 | Building Technologies Office eere.energy.gov Approach Distinctive Characteristics: Method scalable to available data Methods employed are model (audit) agnostic Can be used to speed up model (audit) runtime Capabilities in place for big data mining Interactive dashboard for Autotune progress Repeatable tuning results 7 | Building Technologies Office eere.energy.gov Accomplishments and Progress Accomplishments: End-to-end Windows desktop prototype created

Overnight tuning of envelope-only parameters, 61% as accurate as 4 man-months of effort Autotune 156 EnergyPlus inputs in 3 hours on desktop was within 30/day (actual use $4.97/day) Autotune for National Energy Audit Tool (NEAT) 15 inputs (30%), reduced error 48%, 20mins on netbook Experts would have tuned same way; 9,154 buildings Trinity test shows G14 compliance and realistic tuning Outputs: CV(RMSE)<2.5%, NMBE<1% both hourly and monthly Inputs: For 60% range, Autotune is close to real value (within 8% when tuning to hourly data, 15% when tuning to monthly data) 8 | Building Technologies Office eere.energy.gov Accomplishments and Progress Accomplishments: Titan scalability 65k cores, 262,144 EnergyPlus (9TB), 44mins MLSuite allows easy use of http://autotune.roofcalc.com http://autotune.roofcalc.com software on supercomputers Tableau and Google Vis API

interactive visualization and comparison of all Autotune experiments Progress on Goals: Tuning accuracy satisfies ASHRAE Guideline 14 Less than 3 hours on standard computer Physically realistic results 9 | Building Technologies Office eere.energy.gov Accomplishments and Progress Awards/Recognition: 2+ million core-hours, 4 competitive awards (free cost share) Extreme Science and Engineering Discovery Environment (XSEDE) Nautilus 30k core-hours (CY11), 200k (CY12), 500k (CY13) Oak Ridge Leadership Computing Facility (OLCF) Jaguar 500k core-hours (CY12), Titan 500k (CY13), Frost 200k (CY13), Lens/EVEREST (CY12&13) 10 | Building Technologies Office

eere.energy.gov Project Plan & Schedule Project initiation date: Oct. 2011 (FY12) Project planned completion date: Sept. 2014 (FY14) Schedule and Milestones: FY12 10 DOE deliverables on time and budget FY13 7 of 14 DOE deliverables so far, all on time and budget FY2014 Q4 (Jul-Sep) Q3 (Apr-Jun) Q2 (Jan-Mar) Q1 (Octt-Dec) FY2013 Q4 (Jul-Sep)

Work completed Active Task Milestones & Deliverables (Original Plan) Milestones & Deliverables (Actual) Q3 (Apr-Jun) Q2 (Jan-Mar) Q1 (Octt-Dec) FY2012 Q4 (Jul-Sep) Q3 (Apr-Jun) Task / Event 312856DC, 312856DF FY12-08, FY13-02 BT0201, BT0305 Q2 (Jan-Mar) WBS Number or Agreement Number

Project Number Agreement Number Legend Q1 (Octt-Dec) Summary Project Name: Autotune E+ Building Models D1) Sensitivity analysis of ZEBRAlliance E+ building model D2) Mapping mechanism for E+ to sensor data D3) Heterogeneous learning systems for directed variable tuning D4) Initial Autotune status and performance report 1) Autotune Reporting 2) Monthly util ity bill calibration 3) Building templates 4) Autotune capabilitie s 5) Reporting and Travel 11 | Building Technologies Office eere.energy.gov

Project Budget Budget History Cost to Date: FY12: fully costed FY13: $119k (45%) FY2012 DOE $650k FY2013 Cost-share $980k DOE $264k Cost-share $1,225k Funding Sources: FY12 Autotune Resources

Interns; 3.07% Hardware; 5.52% DOE BTP; 39.88% FY13 Autotune Resources DOE BTP; 17.73% LDRD (Weinberg); 10.74% OLCF (Jaguar); 34.05% 12 | Building Technologies Office UT (Nautilus); 6.75% eere.energy.gov Project Integration, Collaboration & Market Impact Partners, Subcontractors, and Collaborators:

Karpay Associates Jacksonville State University The University of Tennessee Technology Transfer, Deployment, Market Impact: Autotune Invention Disclosure filed 5 software systems to copyright and release Plan to deploy Autotune in FY13 13 | Building Technologies Office eere.energy.gov Project Integration, Collaboration & Market Impact Communications (selected): 1 PhD Dissertation, 2 journals, 5 conference, 6 submitted soon, 5 internal reports (250+ pages) Published: New, Joshua R., Sanyal, Jibonananda, Bhandari, Mahabir S., Shrestha, Som S. (2012). "Autotune EnergyPlus Building Energy Models." In Proceedings of the 5th National SimBuild of IBPSA-USA, International Building Performance Simulation Association (IBPSA), Aug. 1-3, 2012. [PDF pre-print]

Sanyal, Jibonananda, Al-Wadei, Yusof H., Bhandari, Mahabir S., Shrestha, Som S., Karpay, Buzz, Garret, Aaron L., Edwards, Richard E., Parker, Lynne E., and New, Joshua R. (2012). "Poster: Building Energy Model Calibration using EnergyPlus, Machine Learning, and Supercomputing." In Proceedings of the 5th National SimBuild of IBPSA-USA, International Building Performance Simulation Association, Aug. 1-3, 2012. [PDF] Accepted: Garrett, Aaron, New, Joshua R., and Chandler, Theodore. Evolutionary Tuning of Building Models to Monthly Electrical Consumption. ASHRAE Conference in Denver, CO, June 22-26, 2013. Sanyal, Jibonananda and New, Joshua R. Simulation and Big Data Challenges in Tuning Building Energy Models. IEEE Workshop on Modeling and Simulation of Cyber-Physical Energy Systems, May 2013. Planned/submitted: Edwards, Richard E., New, Joshua R., and Parker, Lynne E. Constructing Large Scale EnergyPlus Surrogates from Big Data. To be submitted to Energy & Buildings Journal, 2013. Garrett, Aaron and New, Joshua R. Scalable Evolutionary Tuning of Building Models to Multiple Channels of Sub-Hourly Data. To be submitted to ASHRAE, New York City, NY, Jan. 18-22, 2014. Internal: Edwards, Richard E., and Parker, Lynne E. (2013). MLSuite - FY2012 Final Report. 68 pages Garrett, Aaron and New, Joshua R. (2012). An Evolutionary Approach to Parameter Tuning of Building Models. 68 pages Edwards, Richard E., New, Joshua R., and Parker, Lynne E. (2012) Approximate l-fold cross-validation with Least Squares SVM and Kernel Ridge Regression. 9 pages 14 | Building Technologies Office eere.energy.gov

Next Steps and Future Plans Next Steps and Future Plans: BTO finish Autotune project as detailed BTO deploy in residential and commercial building integration program elements Weatherization Autotune NEAT comparison of human vs. computer calibration Federal Energy Management Program ESPC ENABLE program 15 | Building Technologies Office eere.energy.gov

Recently Viewed Presentations

  • Overlapping Direct &amp; Indirect Treaty Claims

    Overlapping Direct & Indirect Treaty Claims

    "an investor who controls several entities in a vertical chain of companies may commit an abuse if it seeks to impugn the same host state measures and claims for the same harm at various levels of the chain in reliance...
  • Chapter Seven: Abbasid Decline and the Spread of Islamic ...

    Chapter Seven: Abbasid Decline and the Spread of Islamic ...

    Indian achievements were passed to Arabs Colonies of Arabs settled along India's coasts, adopted local customs Provided staging points for Islamic expansion The Coming of Islam to South Asia 10th Century- a Turkish dynasty gained power in Afghanistan 12th Century-...
  • Writing Standardisation 2017 Year 6 Teachers  Spring term

    Writing Standardisation 2017 Year 6 Teachers Spring term

    using a range of cohesive devices*, including adverbials, within and across sentences and paragraphs. using passive and modal verbs mostly appropriately. using inverted commas, commas for clarity, and punctuation for parenthesis mostly correctly, and making somecorrect use of semi-colons, dashes,...
  • Dove Campaign - Weebly

    Dove Campaign - Weebly

    Dove Campaign for Real Beauty. The Dove Campaign for Real Beauty is a marketing campaign. The campaign's mission is "to create a world where beauty is a source of confidence and not anxiety" The advertisements use "regular" women instead of...
  • Hybridization - KFUPM

    Hybridization - KFUPM

    We assume that orbital order is the same as that for N2. The bond order is 2.5. ... An orbital energy-level diagram for sp2 hybridization. Note that one p orbital remains unchanged. When an s and two p orbitals are...
  • EOF Budget Workshop - New Jersey

    EOF Budget Workshop - New Jersey

    EOF Budget Workshop. February 12, 2016. ... be returned by e-mail that should then be attached to your FY 2017 contract for expenditure report reference andaudit purposes. ... The provision and maintenance of state of the art technological equipment (such...
  • CHAPTER 3: LEGAL ETHICS - Lancaster High School

    CHAPTER 3: LEGAL ETHICS - Lancaster High School

    Torts Torts are part of civil law. wrong committed against a person or the person's property. LEGAL ASPECTS Unintentional torts Negligence The negligent person failed to act in a reasonable and careful manner. harm was caused to the person or...
  • Problems in the Republic - Dearborn Public Schools

    Problems in the Republic - Dearborn Public Schools

    2. Working with your group, read the reason why Rome fell and write it in the classification chart. 3. You will be given five minutes per station. Then the paper will rotate. 4. The person with the most RED is...