ISS4E or how can the Internet help smarten the grid? S K S. S. Keshav Keshav and C. Rosenberg K h h and C. Rosenberg d C R d C R b b May 2011 May 2011 http://blizzard.cs.uwaterloo.ca/iss4e http://blizzard.cs.uwaterloo.ca/iss4e
Smart Grid Bi ‐ directional energy flows Integration of renewables renewables Smart Grid Consumer’s Communication Participation Storage Elastic Loads: EVs + 2 smart appliances
Smart Grid: large scale, heterogeneous, distributed system distributed system • Millions of sources Millions of sources • Communication C i ti • Stochastic sources • Maintaining • New loads: elasticity, l d l i i reliability variable demand • Incentivization • Two ‐ way flows • Security • Dealing with storage g g • Non ‐ traditional utility • Non ‐ traditional utility players 3
The smart grid will require massive change A highly dynamic system A relatively static, with elastic loads and predictable, stable millions of points of control millions of points of control system with inelastic t ith i l ti loads and a few points of control A A paradigm shift di hift 4
Our Research Our Research Use Internet concepts to smarten the grid h id 5
Why Is This Possible: Similarities y • Large ‐ scale Large scale • Heterogeneous • Critical infrastructure C iti l i f t t • Both match geographically distributed demands with distributed generation • Distributed sources that are highly variables • Hierarchical • Balance centralization and decentralization • Balance centralization and decentralization 6
Differences Differences • Primarily one ‐ way vs primarily two ‐ way flows Primarily one way vs. primarily two way flows • Grid has practically no storage • Consumers are used to see their electrical bill C d h i l i l bill reflect what they really use 7
Our Vision To apply our expertise in Information Systems and Sciences to find innovative solutions to problems in energy f p gy systems . We work within Waterloo Institute for Sustainable Energy (WISE) in collaboration with • researchers in related disciplines • researchers in related disciplines • partners in industry Initial focus is smart grids , where energy systems converge with information systems 8
ISS4E Faculty S. Keshav, Canada Research Chair, Computer Sciences C Rosenberg Canada Research Chair El C. Rosenberg, Canada Research Chair, Electrical & Computer Eng. t i l & C t E Post ‐ Doctoral Fellow Weihong Wang PhD students PhD students Pirathayini Srikantha Tommy Carpenter Masters students Omid Ardakanian Ryan Case Bo Hu Theodosios Tzoutzas h d Hadi Zarkoob Laboratory facilities include sensors for building monitoring, smart power strips for home monitoring and control, ENVI systems for data collection, wireless sensors for solar panel monitoring, etc. 9
Our expertise p Modeling, mathematical analysis, and system building using techniques from: using techniques from: Internet and information technology (planning, design, implementation, deployment, and management) p , p y , g ) Telecommunications (wireline and wireless communication systems) Distributed systems Stochastic analysis Large scale sim lation Large ‐ scale simulation Data mining and machine learning Economics and game theory Economics and game theory 10
Ongoing projects 1. Modeling and control of grid energy storage and DG 11
Ongoing projects 1. Modeling and control of grid energy storage and DG 2. Demand Response: a revisit based on • Internet views (allows fine grained DR) Internet views (allows fine grained DR) • Elasticity 12
Ongoing projects 1. Modeling and control of grid energy storage and DG 2. Demand Response: a revisit based on • Internet views (allows fine grained DR) Internet views (allows fine grained DR) • Elasticity 3. Smart Home • GW to appliances (control measurement) • GW to appliances (control, measurement) • Applications 13
Ongoing projects 1. Modeling and control of grid energy storage and DG 2. Demand Response: a revisit based on • Internet views (allows fine grained DR) Internet views (allows fine grained DR) • Elasticity 3. Smart Home • GW to appliances (control measurement) • GW to appliances (control, measurement) • Applications 4. EV Integration • Charging control • Billing and roaming • Fleet integration Fleet integration 14
Ongoing projects 1. Modeling and control of grid energy storage and DG 2. Demand Response: a revisit based on • Internet views (allows fine grained DR) Internet views (allows fine grained DR) • Elasticity 3. Smart Home • GW to appliances (control measurement) • GW to appliances (control, measurement) • Applications 4. EV Integration • Charging control • Billing and roaming • Fleet integration Fleet integration 5. Impact of context • Developed countries vs. developing countries (e.g., smart EPS) 15
Ongoing projects 1. Modeling and control of grid energy storage and DG 2. Demand Response: a revisit based on • Internet views (allows fine grained DR) Internet views (allows fine grained DR) • Elasticity 3. Smart Home • GW to appliances (control measurement) • GW to appliances (control, measurement) • Applications 4. EV Integration • Charging control • Billing and roaming • Fleet integration Fleet integration 5. Impact of context • Developed countries vs. developing countries (e.g., smart EPS) 6 Prototype systems & measurements 6. Prototype systems & measurements • ENVIs, Sensors, I ‐ smart, HomeOS 16
Fine grained Measure
Fine grained Measure
Fine grained Measure
Measure Model
Gain from storage g Measure Analyze Analyze Model Trends
Gain from storage g Measure Analyze Analyze Model Trends
Gain from storage g Measure Analyze Analyze Model Trends
Measure Controller Model Broadcast Analyze Analyze House Control EV charging Lateral PHEV PHEV Pole-Top Transformer
Measure Model Analyze Analyze Control EV charging DR: • Elasticity
Measure Model Analyze Analyze Control EV charging DR: • Elasticity
Measure Model Analyze Analyze Control EV charging DR: • Elasticity
Measure Model Analyze Analyze Control EV charging DR: • Elasticity
Measure Model Analyze Analyze Control EV charging DR: • Elasticity A 15% decrease in peak without noticeable decrease in decrease in comfort
Gridlab-D Measure Simulate Analyze Analyze Control Model
Measure Model Analyze Analyze Control Simulate Build Prototype • Solar panel anomaly p y detection
Measure Model Analyze Analyze Control Simulate Build Prototype Weather • Solar panel anomaly reports p p y detection Applications: • Alert and weather report IESO
Conclusions Conclusions • 2010 ‐ 2020 will decide the grid of 2120 2010 2020 will decide the grid of 2120 • Internet ≈ Grid • 40 years of Internet research {could, should, 0 f h { ld h ld may} help • Rich area for impactful research 33
Recommend
More recommend
