Rethinking the Energy Infrastructure from a Cyber-Physical Perspective Xiaofan Jiang, Randy Katz, David Culler, and Seth Sanders University of California, Berkeley Emerging CPS Application.
Download ReportTranscript Rethinking the Energy Infrastructure from a Cyber-Physical Perspective Xiaofan Jiang, Randy Katz, David Culler, and Seth Sanders University of California, Berkeley Emerging CPS Application.
Rethinking the Energy Infrastructure from a Cyber-Physical Perspective Xiaofan Jiang, Randy Katz, David Culler, and Seth Sanders University of California, Berkeley Emerging CPS Application Workshop, ICCPS April 15, 2010, Stockholm “Energy permits things to exist; information, to behave purposefully.” W. Ware, 1997 The Grid: Marvel of Industrial Age Design • • • • • Deliver high quality low-cost power To millions of customers over thousands of miles Synchronized to <<16 ms cycle (60 Hz) With no orders, no forecasts, no plans No inventory anywhere in the supply chain • To enable rapid economic & industrial growth through oblivious consumption 2 A New Reality … • Energy becoming increasingly dear – increased cost of acquisition – inclusion of environmental costs • Improvements in energy efficiency cause high dynamic variability in the load – high peak-to-ave ratio, bursty • Limitations of existing grid present transmission and distribution bottlenecks • Incorporation of renewable resources reduces control over supply – most are non-dispatchable (solar, wind) 3 Towards an “Aware” Energy Infrastructure Nearly Oblivious Loads Baseline + Dispatchable Tiers Generation Transmission Distribution Non-Dispatchable Sources Demand Interactive Dispatchable Loads Communication Communication 4 Where to Focus? • Buildings … • 72% of electrical consumption, 40% of total consumption, 42% of GHG footprint • 370 B$ in US annual utility bill • 9.5% of GDP in bldg construction/renovation • Primarily Coal generation • Primary opportunity for renewable supplies Our Buildings Do Nothing very poorly! 11/7/2015 6 Internet 101 • Intelligence at the edge, not the core – Smart Grid => Dumb Grid with Smart End Points – Reliability and performance by buffering and continuous measurement and adaptation – Lower cost, incremental Deployment, Greater Resilience • Horizontal Layering not Vertical Integration – Technology agnostic protocols – Application agnostic protocols • Create the new as an overlay on the old 7 Energy Network Architecture • Information exchanged whenever energy is transferred – “packetized” • Loads are “Aware” and sculptable – Forecast demand, adjust according to availability / price, self-moderate • Supplies negotiate with loads • Storage, local generation, demand response are intrinsic 8 Start from Scratch? • No! 9 Grid Exists Generation Transmission Distribution Load Conventional Electric Grid 10 Internet Exists Generation Transmission Distribution Load Conventional Electric Grid Conventional Internet 11 Intelligent Energy Network as Overlay on Both Intelligent Energy Network Source IPS energy subnet Load IPS Intelligent Power Switch Generation Transmission Distribution Load Conventional Electric Grid Conventional Internet 12 Aware Co-operative Grid • Availability • Pricing • Planning • Forecasting • Tracking • Market • Monitor, Model, Mitigate • Deep instrumentation • Waste elimination • Efficient Operation • Shifting, Scheduling, Adaptation 13 A MultiScale Study Price profile w Load profile w $ now IPS CT now comm Internet power IPS IPS IPS Grid Bldg Energy Network AHU IPS Chill w Data center IPS IPS Actual load IPS Power proportional kernel IPS M/R Energy Net IPS now Power proportional service manager QualityAdaptive Service 14 Intelligent Power Switch Host Load Intelligent Power Switch (IPS) Intelligent Power Switch (IPS) Power Generation Host Load Energy Storage Energy Storage energy flows PowerComm Interface Intelligent Power Switch (IPS) Energy Storage Intelligent Power Switch (IPS) Energy Network information flows Intelligent Power Switch (IPS) Energy Storage Energy Storage • PowerComm Interface: Network + Power connector • Scale Down, Scale Out 15 Questions… • Where does the energy go? – how much is wasted? – how can the rest be optimized? • How much slack is there? – Can it be exercised? – Energy storage? Electrical Storage? • What limits renewable penetration? – vs storage, scheduling, cooperation • What are the protocols involved? • What is the System and network design? • … 16 The 3 Views Operations and Environment CT: mains power monitoring A 1 5 9 13 17 21 25 29 33 37 41 Vibration B 3 7 11 15 19 23 27 31 35 39 2 6 10 14 18 22 26 30 34 38 42 Panel 1 4 8 A A Panel 2 1 A 16 5 9 12 B B 13 20 17 24 21 28 25 32 29 36 33 40 37 41 B 3 7 11 15 19 23 27 31 35 39 2 6 10 14 18 22 26 30 34 38 4 8 12 16 20 24 28 32 36 40 42 panel level power monitoring Humidity Temperature Pressure ACme: plug load energy monitor and controller 17 Applications A Narrow-Waste for Physical Information Personal Feedback Modeling Visualization Control Storage Location Continuous Commissioning Actuation Debugging Authentication Physical Information sMAP Water Structural Electrical Weather Geographical Environmental Occupancy Actuator IP Everywhere Applications sMAP Resources sMAP Gateway California ISO sMAP sMAP sMAP Google PowerMeter Weather AC plug meter Internet Cell phone sMAP sMAP Temperature/PAR/TSR Dent circuit meter Light switch Edge Router EBHTTP Translation EBHTTP / IPv6 / 6LowPAN Wireless Mesh Network Proxy Server Vibration / Humidity sMAP sMAP Every Building RS-485 Modbus sMAP Gateway sMAP Gateway Database Building Branch Level True Power (kW) Sunday True Power (kW) Monday Plug-Load Level Disaggregation: Understanding Diverse Load 22 Re-aggregation to Purpose 23 Energy Consumption Breakdown 24 Power Proportional Computing “The Case for Energy-Proportional Computing,” Luiz André Barroso, Urs Hölzle, IEEE Computer December 2007 Figure 1. Average CPU utilization of more than 5,000 servers during a six-month period. Servers are rarely completely idle and seldom operate near their maximum utilization, instead operating most of the time at between 10 and 50 percent of their maximum 25 Energy Proportional Computing “The Case for Energy-Proportional Computing,” Luiz André Barroso, Urs Hölzle, IEEE Computer December 2007 Doing nothing well … NOT! Energy Efficiency = Utilization/Power Figure 2. Server power usage and energy efficiency at varying utilization levels, from idle to peak performance. Even an energy-efficient server still consumes about half its full power when doing virtually no work. 26 Server Power Consumption Server Power Consumption Soda Machine Room Power Consumption 350 48 180 Active 87 300 Idle 160 15 13 200 13 14 19 140 31 120 10 287 150 190 100 17 190 200 161 9.5 530 Soda 420A Soda 80 50.9 50 HP Integrity rx2600 Compaq DL360 SunFire X2200 40 SunFire x2100 Cyber Switching 0 Dell PowerEdge 1950 60 PowerEdge 1850 50 18.9 19 26.5 30.6 31 est kW min est kW max kW meas 18.1 20 340 Soda 287 Soda 44.5 3-19-2004 290 Soda 288 Soda 100 KW 230 10.1 248 SunFire V60x Watts 250 0 • x 1/PDU efficiency + ACC • If Pidle = 0 we’d save ~125 kw x 24 hours x 365 … • … Do Nothing Well 27 Where the Power Goes Westmere Atom 333 Core i7 Power Transition Latency Power Proportional Design • New Trade-offs – power proportional @node vs @cluster • Workload & Benchmarking – idle is as important as active • Slack in Information Processing – interactive vs background workload • Power-constrained Service Degradation 30 Scaling Energy Cooperation Local Storage IPS IPS Energy Interconnect IPS Local Generation IPS IPS IPS Local Load Energy Interconnect Communications Interconnect • Hierarchical aggregates of loads and IPSs • Overlay on existing Energy Grid 31 Initial Steps 33 “Doing Nothing Well” • Existing systems sized for peak and designed for continuous activity – Reclaim the idle waste – Exploit huge gap in peak-to-average power consumption • Continuous demand response – Challenge “always on” assumption – Realize potential of energy-proportionality • From IT Equipment … – Better fine-grained idling, faster power shutdown/restoration – Pervasive support in operating systems and applications • … to the OS for the Building • … to the Grid 34 Cooperative Continuous Reduction User Demand High-fidelity visibility Facility Mgmt Automated Control Supervisory Control Community Feedback 3-19-2004 35 Aware Co-operative Grid • Availability • Pricing • Planning • Forecasting • Tracking • Market • Monitor, Model, Mitigate • Deep instrumentation • Waste elimination • Efficient Operation • Shifting, Scheduling, Adaptation 36 Energy Reduction and Support for Renewables thru Information Do Nothing Well Scheduling NonDispatchable Supply Storage NonDispatchable Supply Dispatchable Supply Dispatchable Supply Reduce Demand Increase Effectiveness of NonDispatchable Supply Thank You 38