Data Acquisition Methods: Lessons Learned from Deployment of Cloud - - PowerPoint PPT Presentation

▶
data acquisition methods lessons learned from deployment
SMART_READER_LITE
LIVE PREVIEW

Data Acquisition Methods: Lessons Learned from Deployment of Cloud - - PowerPoint PPT Presentation

Feb 15, 2024 282 likes •452 views

Data Acquisition Methods: Lessons Learned from Deployment of Cloud HVAC Analytics at UCSB Tuesday, June 27 th 2017 Michael Georgescu (Ecorithm) Jordan Sager (UCSB) California Higher Education & Sustainability Conference (CHESC 2017)

slide-1
SLIDE 1

Data Acquisition Methods: Lessons Learned from Deployment of Cloud HVAC Analytics at UCSB

Michael Georgescu (Ecorithm) Jordan Sager (UCSB)

California Higher Education & Sustainability Conference (CHESC 2017)

Tuesday, June 27th 2017

slide-2
SLIDE 2

Tuesday, June 27, 2017 California Higher Education & Sustainability Conference (CHESC 2017) 2

Motivation & Project Background

  • Today’s talk:

§ Geared to FM teams interested in or at early stages of FDD

implementation

  • This effort:

§ Mid 2015 - late 2016: UCSB FM deploys analytics software

  • n two campus buildings as part of MBCx project

§ Early 2017 - present: project extends to two laboratory

buildings with bandwidth for full campus rollout

§ Important aspects of the project are:

  • Integration approach & system architecture
  • Scalability
  • Persistence of savings vs. traditional retro-Cx projects
  • Cost-effectiveness & time-effectiveness
slide-3
SLIDE 3

Tuesday, June 27, 2017 California Higher Education & Sustainability Conference (CHESC 2017) 3

Building Analytics at UCSB

  • Deployment of monitoring-based commissioning

through advanced analytics of trended BMS data

  • At the outset:

§ BAS integration § Analytics § Reporting

  • Moving forward:

§ Mechanical drawings/schedules § Spatial data § Work order system interface § Rethinking data acquisition

slide-4
SLIDE 4

4

Education and Social Science Building

  • ESSB buildings selected for initial deployment due to a history
  • f temperature complaints
  • 5,686 points trending on 5-minute intervals using native

functionality in BMS and stored in cloud database

  • Advanced analytics software interacting with database to

monitor building plant, AHU (6), and terminal units (175) for MBCx project

Tuesday, June 27, 2017 California Higher Education & Sustainability Conference (CHESC 2017)

slide-5
SLIDE 5

5

Education and Social Science Building

Tuesday, June 27, 2017 California Higher Education & Sustainability Conference (CHESC 2017)

Monthly TCI Nov Dec Jan Feb Mar Apr May 42.7% 64.6% 77.6% 83.9% 80.2% 88.2% 87.8%

6.4 36.6 30.6 39.1 59.6 29.6 53.6 78.3 29.1 12.1 31.4 41.1 48.3 30.6 44.6 37.1 34.1 34 27.4 42 32.7 11.1 17.3 19.3 9.9 18 3.7 7.1 12.6 6.4 1 14 10.1 O C TO B E R WEEK 3 WEEK 5 WEEK 7 WEEK 11 WEEK 11 J A N U A R Y WEEK 15 FEBRUARY WEEK 19 M A R C H WEEK 23 APRIL WEEK 27 WEEK 29 WEEK 31 WEEK 33

TCI: Avg. % of Spaces within comfort band (70F - 76F)

slide-6
SLIDE 6

MBCx Results

  • 1st Year Outcomes:

§ Frequency of new issues lowered by 50% § Occupant comfort improved by 45% § HVAC Maintenance requests lowered by 60%

  • Approved incentive amount of $40,289 with energy

savings of 137,368 kWh & 7,321 therms

  • Data integration and consolidation facilitated easy

information access

  • Feedback to Ecorithm helped improve analytics, FDD

definitions and workflow integration…

6 Tuesday, June 27, 2017 California Higher Education & Sustainability Conference (CHESC 2017)

slide-7
SLIDE 7

7

Energy Efficient Operations

National Renewable Energy Laboratory

Innovation for Our Energy Future

A national laboratory of the U.S. Department of Energy Office of Energy Efficiency & Renewable Energy NREL is operated by Midwest Research Institute ! Battelle Contract No. DE-AC36-99-GO10337

Lessons Learned from Case Studies of Six High-Performance Buildings

  • P. Torcellini, S. Pless, M. Deru, B. Griffith,
  • N. Long, and R. Judkoff

Technical Report

NREL/TP-550-37542 June 2006

  • “Properly applied off the shelf or state-of-the-

art technologies are available to achieve low-energy buildings. However, these strategies must be applied together” NEED FOR INTEGRATION OF BEST-IN- CLASS COMPONENTS

  • “There was often a lack of software to allow

the technologies to work well together” NEED INTEGRATED CONTROL SOFTWARE AND UNCERTAINTY ANALYSIS

Tuesday, June 27, 2017 California Higher Education & Sustainability Conference (CHESC 2017)

slide-8
SLIDE 8

Tuesday, June 27, 2017 California Higher Education & Sustainability Conference (CHESC 2017) 8

Information Architecture

Native Name from BMS DeviceType DeviceName PointName Point1 City Center/TU_VAV/L12/_1211/DMPR COMD FPB FPB_L12_1211 DamperCommand Point2 City Center/HP/L15/_1501/DAT HP HP_L15_1501 DischargeAirTemp

Architectural Information BMS Trends / Data Model Web Reporting Work Orders Fault Detection Central Data Store

slide-9
SLIDE 9

9

Gathering BMS Trend Data

Tuesday, June 27, 2017 California Higher Education & Sustainability Conference (CHESC 2017)

UCSB Network FM Network Bacnet Traffic BMS Local Volttron Host Central Data Store

Data Push

VPN Firewall

  • Original trending at ESSB using BMS functionality was cumbersome
  • BMS data collection now performed with DOE’s Volttron* software
  • Compared to native BMS functionality, Volttron* allows trending of data

with no measurable congestion on BMS network

  • 10,000 data points trended on a 5-minute interval

*https://github.com/VOLTTRON/volttron

slide-10
SLIDE 10

Data Onboarding: Contextualizing Building Data

10

Equipment and point hierarchies setup correlating systems/devices that are physically interconnected

FPB_L12_1211 FPB_L12_1212 AHU_01 DamperCommand Occupied AirFlow Chiller_01.ChilledWater SupplyTemp HP_L15_1501 Etc.

Relationship Building

  • Data onboarding allows buildings with dissimilar systems to be

uniformly represented within the analytics system

Tuesday, June 27, 2017 California Higher Education & Sustainability Conference (CHESC 2017)

  • Device-relational mapping allows data to be converted into a

logical representation enabling analysis, visualizations, and

  • ther complex behaviors to be designed
slide-11
SLIDE 11

Floor Plan Generation

11 Tuesday, June 27, 2017 California Higher Education & Sustainability Conference (CHESC 2017)

Space / Equipment Adjacency Information Floorplans Equipment Schedules Riser Diagrams

slide-12
SLIDE 12

12

Building Output (ex. Temperature)

Advanced Analytics

Tuesday, June 27, 2017 California Higher Education & Sustainability Conference (CHESC 2017)

Automated Fault Detection and Diagnostics (AFDD) Learning algorithms used to identify and diagnose faults in buildings or in any dynamical system with the goal of bringing the system back to “Intended Design”

slide-13
SLIDE 13

Web Reporting

13 Tuesday, June 27, 2017 California Higher Education & Sustainability Conference (CHESC 2017)

slide-14
SLIDE 14

Work Orders

14 Tuesday, June 27, 2017 California Higher Education & Sustainability Conference (CHESC 2017)

slide-15
SLIDE 15

Moving Forward

15 Tuesday, June 27, 2017 California Higher Education & Sustainability Conference (CHESC 2017)

  • As building managers and operators, we know a

building’s energy systems can become inefficient quickly!

  • Lessons learned at UCSB in sustaining performance:

§ Defining a plan for collecting and integrating BMS, energy,

and other data sources enables analytics to be applied to everyday operations

§ Involving campus stake holders (FM, IT, Occ.) ensured that

all approvals are met cost effectively and without delays

  • Making analytics mainstream: we hope to uncover

additional opportunities and improve building performance as adoption grows

slide-16
SLIDE 16

16

Thank You

Jordan Sager UCSB Campus Energy Manager Jordan.sager@pf.ucsb.edu (805) 705-5630 Michael Georgescu, Ph.D. Ecorithm Director of Engineering & Research mgeorgescu@ecorithm.com (818) 231-4900

Tuesday, June 27, 2017 California Higher Education & Sustainability Conference (CHESC 2017)