Distributed Load Algorithms LBNL Demand Response Automated Server Internet Open. ADR Client Weather data Siemens Smart Energy Box Energy Simulation Jay Taneja Nathan Murthy UC Berkeley DIADR Mid-Project Demonstration, April 27, 2011 BMS Adapter APOGEE BAS 3 rd Party Plug-in Watt. Stopper Distributed Load Control Gateway Air handlers/fans Chillers 1
Distributed DR Algorithms • Goal: Testing and evaluation of distributed DR strategies • Dense deployment of metering devices on appliances and office equipment, with actuation by the energy gateway – Thermostatically-controlled loads (e. g. refrigerators, space heaters, etc. ) – Battery-powered loads (e. g. laptop computers, desktop computers with UPS units, etc. ) – Lighting (e. g. overhead lights, lamps, etc. ) – Other office equipment (e. g. printers, routers, etc. ) 2
Smart Office (464 SDH) 3
Sensor Data Management: s. MAP • Interface for gathering and storing heterogeneous, unsynchronized physical data • Includes data from zone lights and two types of plug meters http: //green. millennium. berkeley. edu: 8080/media/graph/demo. html 4
Skipping Refrigerator Cycles 5
Devices with Onboard Batteries • Case Study: Laptops • Collected traces to build empirical model of charge and discharge behavior • Power delivered is a function of battery capacity • Developing metrics to design laptop charge schedule during DR period • Mix of known state (power consumption, maybe battery capacity) and unknown state (mobility, computation load) 6
power battery capacity 7
Curtailment of Battery Charging in a DR Event • Assume N laptops with uniform distributed capacity states • Assume laptops leave and enter zone both at a Poisson rate with λ=1 • Define duration of DR Event • Throughout DR event, set curtailment ratio c (% of baseline load) and select laptops to charge • Choose c to minimize projected peak power for remainder of DR event 8
9
Charging Curtailment Simulation Results • 30% curtailment possible • Choice of curtailment ratio is crucial to how load management throughout DR event • Aggressive initial curtailment may offset peak load reduction towards end of DR event • Aggregate distributed load in a zone can be shaped using device energy storage 10
Desktop Power Management • Desktop + UPS is similar to laptop • Collaboration with Dhaani Systems – Using network appliance to manage state (and power) of Windows machines – Machines put to sleep remotely when not in use – During DR event, aggressiveness can be increased 11
12 12
Lighting • Lighting zones on SDH 4 – Actuate using Wattstopper via BACnet – High (50 W) and low-power (25 W) ballasts in each zone 13
Other Loads • Printers – High peak-to-idle ratio (> 75: 1) – Idea: DR-aware print queue • Avoid concurrent printing (and resulting high peak load) • Modify existing print server 14
Next Steps • Application of techniques to similar loads • Integrated management of heterogeneous loads • Occupant light control 15
Questions? Acknowledgement This material is based upon work supported by the Department of Energy under Award Number DE-EE 0003847 Disclaimer This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government nor any agency thereof, nor any of their employees, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government or any agency thereof. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof. 16
Скачать презентацию Distributed Load Algorithms LBNL Demand Response Automated Server
17b9b464c43569dbd1dd7f7b56c3e916.ppt