• Skip to primary navigation
  • Skip to main content
  • Skip to primary sidebar
  • RCR Wireless News
  • Enterprise IoT
  • Editorial Calendar
  • Advertise
  • Webinars
  • Reports
  • White Papers
  • Subscribe

In-Building Tech

Connecting CRE building technology buyers with CRE tech sellers

720×90

  • Industry
    • Office & Commercial
    • Data Center, Network Hotels
    • Government
    • Healthcare
    • Higher Ed
    • Hospitality
    • K-12
    • Laboratory & Scientific
    • Manufacturing
    • Multi-Family
    • Transportation: Airports, Rail, Ports
    • Smart City
    • Stadiums, Arenas, Venues
  • Tech
    • Drones
    • AI-Machine-Learning
    • Wi-Fi
    • Augmented Reality
    • IoT (platform, gateway)
    • Networks
    • 5G Resources
    • Microcontrollers
    • Microprocessors
    • Data Analytics
    • Wired Networks, Fiber
    • Wireless (Cell, DAS, BDA, Repeaters, Boosters)
    • Positioning, GPS, Navigation
    • Security
    • Sensors
  • Systems
    • Energy
    • Lighting
    • HVAC
    • Security
  • Functions
    • Automation
    • Building Management
    • Construction
    • Asset Management (EAM)
    • Materials
    • Maintenace (MRO)
  • Smart Buildings
  • News & Event Coverage
  • In-Building Wireless
  • About In-Building Tech
  • Qualcomm 5G Insights

PNNL highlights use of deep learning to reduce energy use in buildings

     
smart glass

Deep learning can be used for reducing energy use in buildings across the U.S. according to Pacific Northwest National Laboratory?(PNNL).

PNNL is one of the United States Department of Energy national laboratories, managed by the Department of Energy’s (DOE) Office of Science.

PNNL noted that American buildings consume roughly 40% of U.S. energy, much of which is expended on heating, cooling, and ventilation. Enhanced control methods can help reduce energy consumption. Model Predictive Control (MPC) has shown potential for substantially reducing energy use in buildings. However, it has not been widely adopted due to a number of implementation challenges.

Recently, PNNL has demonstrated that deep learning can be used to overcome some of these challenges, paving the way for wider adoption of MPC in buildings.

?For successful application in buildings, the method must be less expensive and easier to implement, and that has been the focus of our work,? said?Jan Drgona, a PNNL postdoctoral research associate.?

According to the PNNL, MPC optimizes control over a receding time horizon, and in a building the method may optimize control for the next 24 hours in 15-minute intervals. MPC would use a model of the building to evaluate its performance over the next 24 hours under different control strategies for fixed occupancy and weather assumptions. The control settings for the first 15 minutes are implemented, the building?s response is measured, and the process is repeated with updated initial conditions and weather predictions.

PNNL also highlighted that a critical part of MPC is the model itself. MPC was initially used to optimize industrial chemical?processes through?physics-based models. Physics-based MPC has also proven effective in buildings. In fact, field testing conducted by KU Leuven University in an office building in Belgium showed energy savings of up to 50%. MPC also improved thermal comfort in the building by keeping the temperatures closer to prescribed bounds,?which can improve occupant productivity and well-being.

Deployment of MPC in a large portion of the building stock has not been possible because of high installation costs. Every building is unique and requires its own custom physics-based model. Physics-based models are computationally expensive, limiting the number of control strategy alternatives that can be explored and often requiring dedicated hardware.

A research team of Jan Drgona, Draguna Vrabie of PNNL, and Lieve Helsen of KU Leuven has developed an approach that overcomes the computational challenges of MPC. The team used physics-based MPC to train deep learning neural network models.

The neural networks models yield control actions that closely approximate those produced by physics-based MPC but do so much more quickly while using significantly less computing power. In lay terms, the researchers are teaching the cheap apprentice (neural network) to imitate the behavior of the much more expensive expert (physics-based MPC).

?We end up with a high-performing intelligent controller with just a fraction of the execution cost of classical Model Predictive Control,? Drgona said.

He added, ?There is much work that remains to be done to achieve robust, scalable methods applicable to large-scale building systems. By applying these methods, we are on track to reduce engineering costs and achieve a generic solution that is broadly available to the building control community.?

Related

About Juan Pedro Tom?s

Juan Pedro covers Global Carriers and Global Enterprise IoT. Prior to Arden Media Company, Juan Pedro worked for Business News Americas, covering telecoms and IT news in the Latin American markets. He also worked for Telecompaper as their Regional Editor for Latin America and Asia/Pacific. Juan Pedro has also contributed to Latin Trade magazine as the publication's correspondent in Argentina and with political risk consultancy firm Exclusive Analysis, writing reports and providing political and economic information from certain Latin American markets. He has a degree in International Relations and a master in Journalism and is married with two kids.

Contact Juan Pedro at [email protected]

Primary Sidebar

Sponsors

Search

300×350

300×100

CommScope forsees CBRS taking shape

Categories

Top Posts & Pages

  • In-Building Tech: Technology Insights for Commercial Real Estate Professionals
    In-Building Tech: Technology Insights for Commercial Real Estate Professionals
  • What is a distributed energy system?
    What is a distributed energy system?
  • What are the benefits of power-over-Ethernet (PoE)?
    What are the benefits of power-over-Ethernet (PoE)?

RSS Enterprise IoT Insights

  • Vodafone installs 5G private network at Skoda Auto manufacturing plant
  • Vodafone rigs-up 1.5 million Wimbledon strawberries with IoT monitoring, tracking
  • Libelium buys Spanish smart cities firm HOPU, sets sights on future IPO

Recent Posts

  • Honeywell invests in RapidSOS emergency response data platform
  • ‘Buildings have to be programmable,? says Cisco?s smart building lead
  • View to install smart windows at Skanska office project in Seattle

Archives

Tweets by InBuildingTech
  • RCR Wireless News
  • Enterprise IoT
  • Editorial Calendar
  • Advertise
  • Webinars
  • Reports
  • White Papers
  • Subscribe

Copyright © 2022 —In-Building Tech • All rights reserved.

Genesis Framework • WordPress • Log in

 

Loading Comments...
 

    This site uses cookies to improve and personalize your experience and to display advertisements. This site may also include cookies from third parties. By using this site you consent to the use of cookies.AcceptPrivacy Policy