ASHRAE ST-16-014-2016 Data and Interfaces for Advanced Building Operations and Maintenance.pdf

1、134 2016 ASHRAEThis paper is based on findings resulting from ASHRAE Research Project RP-1633.ABSTRACTThis paper provides recommendations for data-driveninterfaces for advanced building operations and maintenancedeveloped through ASHRAE Research Project 1633. Inform-ing operations and maintenance wi

2、th data-driven informationis critical to achieve high-performance buildings. Substantialguidance illustrating how to measure and convey buildingperformance information has already been created, such as inASHRAE Guideline 13-2015 (ASHRAE 2015) and Perfor-mance Measurement Protocols for Commercial Bui

3、ldings(ASHRAE 2010). RP-1633 focused attention on operationsand maintenance stakeholders, including control technicians,HVAC technicians, service providers, commissioning agents,and facility managers by conducting literature reviews,commercial interface reviews, and stakeholder interviews tocreate g

4、uidance about data-driven metrics and visualizationsthat clearly quantify and communicate building operationalperformancetothesestakeholders.Theresultsofthisresearchare presented here, with recommendations to provide metricsandvisualizationsatmultiplescales,includingportfoliowide,whole building, and

5、 for specific building areas, systems, andequipment.Metricsspancategoriesrelatedtooperatingcosts,utility consumption, carbon emissions, system performance,controllability, faults, and energy savings. Metrics may bevisualized on maps, system graphics, and in floor plans; astime-series line charts; in

6、 calendar plots, bar charts, and piecharts; and relative to expected performance, past perfor-mance,orarelevantbenchmark.Wewillpresentfeedbackfromoperationsandmaintenancepersonnelandourresearchaboutthe types of metrics, at each scale, in which visualizationformat is most useful for advanced operatio

7、ns and mainte-nance.INTRODUCTIONAnalyzing and interpreting building performance data toinform operations and maintenance is critical to the success ofhigh-performance buildings. With the advance of technologyhardware and software for buildings, there is an increasingamount of available data to infor

8、m building operations, main-tenance, and management. However, facility managementpersonnel have limited time and resources and need concisemetrics, visualizations, and information to support their dailyoperations and decision making. Recent works, such asASHRAEs Performance Measurement Protocols for

9、Commercial Buildings (ASHRAE 2010, have focused atten-tion on the metrics relevant to tracking building performance.The research described in this report seeks to expand suchinvestigations to consider graphical visualization of opera-tionalmetrics,focusingonanaudienceincludingcontroltech-nicians,HVA

10、Ctechnicians,serviceproviders,commissioningengineers, and facility managers.This research is based on a literature and product reviewof metrics and interfaces for operations and maintenance,interviews with 79 people directly responsible for operationsand maintenance, and surveys of a mock interface

11、presentingvarious metrics and visualizations. Recommendations foradvanced building operations and maintenance interfaces areprovided based on these activities.Data and Interfaces forAdvanced Building Operations andMaintenanceNicholas Gayeski, PhD Sian Kleindienst, PhD Jaime Gagne, PhDMember ASHRAESt

12、ephen Samouhos, PhD Ryan Cruz Bradley WerntzNicholas Gayeski is cofounder and CEO, Sian Kleindienst is chief scientist, Jaime Gagne is principal building scientist, Stephen Samouhosisacofounder,RyanCruzisaseniorbuildinganalyst,andBradleyWerntzisaformerseniorbuildinganalystatKGSBuildings,LLC,Somer-vi

13、lle, MA.ST-16-014 (RP-1633)Published in ASHRAE Transactions, Volume 122, Part 2 ASHRAE Transactions 135STATE OF THE TECHNOLOGYMetrics About Building Performance. Building stake-holders have access to an increasing amount of data frombuilding automation systems and metering. Performancemetrics use da

14、ta and other building information to produceinformation through which to assess performance. Examplesof metrics include energy use intensity (EUI, or energy perbuilding area), chiller kW/ton (kWe/kWth), and occupantcomplaints per day. Hitchcock (2003) defines performancemetrics as representing “the

15、performance objectives for abuilding project, using quantitative criteria, in a dynamic,structuredformat.”AsapartofASHRAESpecialProject115:Performance Monitoring Protocols, McNeill et al. (2007)completed a comprehensive review of literature relevant tobuilding performance measurements. Based on this

16、 work,ASHRAEpublishedPerformanceMeasurementProtocolsforCommercial Buildings (ASHRAE 2010) in an effort to stan-dardize building performance claims and measurement prac-tices. The book identifies the metrics and appropriatemeasurement practices for building performance for six typesof building inform

17、ation (energy, water, thermal comfort,indoor air quality, lighting, and acoustics) from basic toadvanced levels. At all levels, the energy metrics recom-mended include energy consumption and cost by source, EUI,and energy normalized by weather and/or occupancy.Thereisanongoingefforttodevelopframewor

18、ksofstan-dardizedmetrics,particularlyforenergy-relatedperformance.The Performance Metrics Project through the U.S. Depart-ment of Energys (DOE) Commercial Building Initiative, theNational Renewable Energy Laboratory (NREL), and PacificNorthwest National Lab (PNNL) has defined a set of perfor-mance m

19、etrics with the goal of standardizing the “measure-ment and characterization of building energy performance”(Deru and Torcellini 2005; Barley et al. 2005; Fowler et al.2010).Several other studies have considered the use of metricsfor building performance assessment. Hitchcocks (2003)research involve

20、d the development of a model for buildingperformance metrics thatis consistent withthe Industry Foun-dation Classes (IFC) for use across a buildings life cycle(McNeill et al. 2007). OSullivan et al. (2004) used an IFC-basedmodelofabuildingatUniversityCollegeCorkasacasestudy for a building energy mon

21、itoring, analyzing, andcontrolling (BEMAC) framework for life-cycle buildingperformanceassessment,andMorrisseyetal.(2004)proposeda building information model to support this BEMAC frame-work. Neumann and Jacob (2008) defined the performancemetrics that would be required for different steps or levels

22、 ofcontinuous commissioning, including benchmarking (opera-tional rating), certification (asset rating), optimization, stan-dard analysis, and regular inspection.Building performance rating systems provide an addi-tional way of assessing building performance. There existseveral different approaches

23、to producing a rating or score fora building. Glazer (2006) evaluated a wide variety of energyrating systems and identified three broad categories of proto-cols: statistical (the building is rated based on where it falls ina statistical distribution of actual buildings), points (the build-ing is rat

24、ed based on how many points it gets in a long list ofcriteria), and prototypical (the building is rated based oncomparison with good conceptual buildings, using simula-tions). A more recent examination of rating systems focusedon benchmarking, rating, and labeling as the three differenttypesofrating

25、sclassifications,wherelabelingisdefinedastheequivalent to assigning percentile intervals to energy classes(Perz-Lombard et al. 2009). Many rating systems exist, suchas the ENERGY STARlabel for buildings (EPA 2014),LEED(USGBC 2016), BREEAM (BRE 2016), BOMA 360(BOMA 2016), and many others. ASHRAEs Bui

26、ldingEnergy Quotient or bEQ (ASHRAE 2016) is a letter-basedgrading system based on the actual and/or designed buildingEUI versus the median EUI for similar buildings.VisualizingBuildingPerformanceInformation.Whilebuilding data, metrics, and ratings all provide valuable infor-mation about a buildings

27、 operations and performance, theway in which this information is provided to a building stake-holdermaybeequallyimportant.Whilebuildingperformancemetrics and rating systems offer ways in which raw data canbe processed into more condensed nongraphical forms,display of both raw data and metrics in gra

28、phical formats suchasscatterplotsanddailyorweeklyprofilescanhelpabuildingstakeholder view and analyze large amounts of building datavery efficiently (Abbas and Haberl 1994). Graphical displayof data in plots and graphs can also be helpful for diagnosingbuilding equipment faults (Meyers et al. 1996).

29、One important consideration for the visualization ofbuilding information is the target audience of the tool. Mariniet al. (2011) conducted a study in which a dashboard wasinstalled in a federal building. Five different user categorieswere considered, with different granularity of informationavailabl

30、e to the different user groups. Some of the lessonslearned included: information should match the user, dash-boards should transform data to information, and dashboardscan help knowledge lead to action. While most control systeminterfacesaregearedtowardbuildingoperatorsandengineers,othertypesofdashb

31、oardshaveemergedthatareaimedtowarddifferentstakeholderssuchasregionalmanagersandfinancialstakeholders.Interfaces for Building Operations. Interfaces provideinteractivesettingsinwhichdata,metrics,andgraphicalinfor-mation about a building may all be displayed. Building auto-mation systems represent on

32、e of the more common types ofsystems that building operators may interact with regularly inbuildings today. However, a variety of other systems, such asenergy monitoring dashboards, enterprise energy manage-ment systems, energy information systems (EIS), advancedanalytics or fault detection and diag

33、nostic systems, and othertools have emerged in recent years.In 2015, ASHRAE released an updated version of Guide-line 13, Specifying Building Automation Systems (ASHRAEPublished in ASHRAE Transactions, Volume 122, Part 2 136 ASHRAE Transactions2015). This guideline is meant to help someone construct

34、 aneffective specification for a building automation system, andit specifies capabilities such as open protocols, systeminteroperability, custom reporting, data trending and trendvisualization (both time-series and scatter plot), remote orportable terminals, and applications such as demand limiting,

35、energy calculations, and antishort cycling, as well as moretraditional building automation system features. Annex D ofGuideline 13 identifies three levels of performance monitor-ing, from simple data trending to sophisticated diagnostics ofequipment faults, operational issues, and power quality, and

36、calls fault detection “a natural enhancement to monitoring theperformance of an HVAC system” (ASHRAE 2015).Granderson et al. (2009) created a framework to charac-terize and classify EIS tools. From an overview of existingtools, they found that visualization and analytical features aredistinguishedby

37、theirflexibilityandthatrigorousenergyanal-yses (baselining, forecasting, anomaly detection) are notuniversal. They also conducted a small number of case studiesin which the use of EIS tools in real buildings was evaluated.Some of the conclusions from the case studies were that dataquality has signif

38、icant impact on EIS usability and that, whileEIS may offer a wide range of features, actual use of thosefeatures may be limited. In a recent cost-benefit analysis of 26EIS case studies, Granderson et al. (2013) found that 21 of 23in-depth cases attributed significant savings to the installationof EI

39、S. Among the factors associated with greater energysavings were pre-EIS site EUI (how wasteful the building wasbefore the EIS), length of time since EIS installation, higher-granularity instrumentation, consumption benchmarking,regular load profiling, and consumption anomaly detection.Also on the li

40、st of operational efficiency best practices werethe use of time-series visualizations to study load profiles andthe use of xy scatter plots to assess load versus outdoortemperature.In addition to EISs, energy monitoring dashboards are agrowing trend. Lehrer and Vasudev (2010) interviewed build-ing m

41、anagers and design professionals and found that suchtoolsarecurrentlybeingusedinsimilarwaystobuildingauto-mation system. The authors found that some of the users keyneeds were high-level overviews with drill-down capabilities,integrationofenergyvisualizationfeatureswithdataanalysis,and compatibility

42、 with existing building automation systems.ExistingOperationalInterfaces.Asignificantaspectofthisresearchwastoidentifyandcompilealistofexistingtoolsfor building operations, maintenance, and decision-making.These tools included general building automation and controlsystems, energy or resource monito

43、ring systems, enterpriseenergy management systems, and systems with moreadvanced analytics, such as optimization, fault detection, ordemand response (DR) functionalities. A database of toolswascompiledcontaininginformationabout70differenttools.Thesetoolswereidentifiedusingpreviousstudies,recommen-da

44、tions by industry stakeholders, Internet searches, and stake-holderinterviews.Foreachexistingtool,thedatabaseincludesa short summary, categorization by intended audience, cate-gorization by content or functionality, a link to a folder ofexample interface graphics (if available), and a website link.T

45、hefunctionalcategoriesconsideredforthetoolswereasfollow: educational content or public display (such as energymonitoring kiosks), enterprise or campus-level views (data orinformationovermultiplebuildingsavailableatonce),energyor utilities monitoring, ENERGY STAR or LEED informa-tion,real-timeequipme

46、ntdata(suchasthattypicallyavailablein a building controls system), optimization features, equip-ment fault detection and diagnosis (FDD), DR, and retrofitrecommendations or calculated return on investment (ROI).The most common feature in the considered tools and dash-boards was energy or utilities m

47、onitoring (90%). While suchsystems are typically found only in high-performance build-ings today, it remains to be seen if such tools will eventuallybecomecommonplaceforbuildingoperations.Othercommonfeatures offered by existing tools were real-time equipmentdata(57%)andenterpriseorcampus-levelinform

48、ation(56%).The least common features were educational/public contentand retrofits or ROI (both 14%), followed by FDD and DR(both 17%).Metrics and Graphics Database. In addition to identi-fying existing tools, we developed databases of metrics andgraphics used to evaluate building performance and aid

49、 inoperational and financial decision-making. The metrics data-base attempts to provide a comprehensive overview of thetypesofdata,metrics,andotherinformationthatisorcouldbemade available in building automation systems, energy dash-boards, and other analytics systems. The graphics databasesummarizes the types of graphical representations that can beused to present these metrics and information to the user fromwithin an interf