Quantification of HVAC Energy Savings for Occupancy Sensing in Buildings through an Innovative Testing

Methodology

Zheng O’Neill, PhD, PEThe University of Alabama, Tuscaloosa, AL

July 11th, 2019

ARPA-E SENSOR Program PI meeting, Denver, CO

2

Problem Statement and Our Solution

Solution:• A novel protocol with an eight-level diversity and a simulation

suite to investigate various types of occupancy sensors fordetection failure rates and energy saving potentials in real-world retrofit scenarios.

• Demonstrated at least 30% savings of HVAC energy usage

Occupancy Sensors: Human Presence Sensing, People Counting, CO2 Sensor

b) Simulation Suite

c) System-level Controlled Lab Testing

d) System-level Field Trial Testing and Simulation

Validation

a) Testing and Validation Protocols (Sensors and Sensor-driven HVAC Control) Eight-level Diversity, Incorporate IEEE and ASHRAE Standards/Guidelines

Occupant profile

Sensor Error

Single building and Nationwide savings potential

• Dynamic testing of sensors and an entire HVAC system within a controlled environment

• Side-by-side comparison with baseline

• 4 homes with different floor plans

• 4 commercial buildings (hotel, office, and academic classroom)

• Building automation system from three different vendors

Failure model

Sensor Specs.

UncertaintyWell Living LabTM

at DelosLab Homes at

PNNL

Metrics State of ArtTesting Accuracy

No 100% accurate ground truth evaluation method

Ensuring AdoptionDiversity

No standardized occupant and sensor diversity requirements exist currently

EnsuringAdoption Flexibility

Flexibility not currently established

Currently there is no widely-accepted method to test and validate performance of occupancy presence, people counting, and CO2 sensors, and the associated HVAC energy savings due to occupancy-based controls

3

Novelty of The Proposed Work • Testing protocol with eight-level diversity

• 1) occupant profile diversity, 2) building type and floor plan diversity, 3) sensor modality, 4) deployment location diversity, 5) software diversity, 6) diagnostic diversity, 7) HVAC controls and mode diversity, and 8) OBC testing diversity.

4

Novelty of the Proposed Work

• Simulation suite with uncertainty quantifications of sensor errors

• Integration with Well Building Standard concept • Testing data analytics with advanced Machine Learning algorithms• Market transition through proactive engagements with HVAC standards

committees, control consulting and building automation companies

Medium Office

Large Office Large Hotel

Primary School Single family home

System level testing protocol and simulation

System level controlled lab

testing

System level field testing and simulation

validation

5

System Level Controlled Lab Testing

• Two identical 1,500 ft2 all-electric 3 BR/2BA homes

• Both homes contain full end-use metering, a suite of environmental sensors, and remote data collection

• Well Living Lab (WLL) - the first scientific research center exclusively committed for human-centered research to understand human-building interactions.

Testing at PNNL’s Lab HomesScheduled: Fall 2020

Testing at Delos’ Well Living Lab Scheduled: Q3 2020

6

System Level Field Trial Testing

Field Trial Testing Starting Q4 2020 (Tentative)

Buildings Building Characteristics Scope of Test PlanTuscaloosa City Hall Complex, Tuscaloosa, AL

Office building in climate zone 3A; Floor plan: open office with conference room, close offices with conference room; Control: WebCTRL

2nd floor with 27 VAV boxes served by one AHU

South Engineering Research Center Building, Tuscaloosa, AL

Academic building in climate zone 3A; Floor plan: classroom with office rooms; Control: Schneider Electric

Classrooms and offices served by two AHUs on the 2nd floor

Marriott Residence Inn on Riverwalk, San Antonio, TX

A hotel building in climate zone 2A; Floor plan: hotel rooms with public space; Control: WebCTRL

Public space with six rooftop units

Center of Excellence Building, Syracuse, NY

Academic building in climate zone 5B; Floor plan: offices, meeting room;Control: Siemens + Schneider Electric

5th floor with reconfigurable floor plan

2 residential houses in Tuscaloosa, AL

Residential home built in 1970 and 2010 (modern open floor plan)

4 bedroom single family house

2 residential houses in Syracuse, NY

Residential home built in 1940 and 2000 (modern open floor plan)

3 and 2 bedroom single family house

7

Current Status and Plan• Testing Protocol V2 is being finalized

• Diversity profile/scenario definition • Occupancy ground truth• Energy end use determination (M&V)

• Preliminary testing plan • PNNL’s Lab Homes• Well Living Lab

• Simulation suite• Communication/integration with

building HVAC control systems• Organize an annual workshop• Move lab tests to 2020 to better

coordinate with Category A an Category B teams

Raspberry Pi

8

End Goals and Stakeholder Entanglement • New Standard/Guideline development for

occupancy sensing system • Application Guideline: Aid the HVAC industry understand

occupant sensing technologies

• Existing Standards/Guideline integration • ANSI/ASHRAE/IES 90.1, IECC, California Title 24 - Expand

requirements for occupancy sensing/control• Guideline 36 - High Performance Sequences of Operation

for HVAC Systems

• Stakeholders• Manufacturers• Building Owners• General Contractors• HVAC Designers/Installers/Contractors• Facility Managers• Architects• Control System Designers/Installers/Contractors

New Sensor System Information

Promote and enable occupancy-sensor driven controls through demonstrated at least 30% savings of HVAC energy usage.

9

Thank You! Questions and Discussions

Zheng O’Neill, PhD, PEDepartment of Mechanical EngineeringThe University of AlabamaBox 870276Tuscaloosa, AL 35487-0276 USAPh: (205) 348-6982 | Fax: (205) 348-6419Email: ZONeill@eng.ua.eduhttp://hvac.ua.edu/