Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.

2012 Tutorial: Markets for Differentiated Electric Power Products

5,807 views

Published on

ACC 2012 Tutorial
http://accworkshop12.mit.edu

The talk will review the many services needed in today's grid, and those that will be more important in the future. It will also review recent competitive equilibrium theory for the highly dynamic markets that may emerge in tomorrow's grid. In particular, to combat volatility from increasing penetration of renewable energy resources, there will be greater need for regulation services at various time-scales. There is enormous potential to secure these ancillary services via demand response. However, there is an obsession today with the promotion of real time prices to incentivize demand response. All evidence strongly suggests that this is a bad idea: 1) In 2011, massive price swings in the real-time market generated anger in Texas and New Zealand 2) Our own research shows that this is to be expected: in a completive equilibrium real-time prices will reach the choke up price (which was recently estimated at 1/4 million dollars). With transmission constraints, our research concludes that prices can go much higher. 3) A recent EIA study shows that consumers are scared of smart meters - they do not trust utility companies to experiment with their meters, or their power bills. We must then ask, is there any motivation to focus on markets in a real-time setting? The speaker believes there is none. Explanations will be given, and alternative visions will be proposed.

Published in: Education

2012 Tutorial: Markets for Differentiated Electric Power Products

  1. 1. Markets for Differentiated Electric Power Products Smart Grid Markets Integration of Renewables, Pricing, Modeling, and Optimization Emerging Topics in Interconnected Systems Sean P. Meyn Joint work with: In-Koo Cho, Anupama Kowli, Matias Negrete-Pincetic, Ehsan Shafieeporfaard, Uday Shanbhag, and Gui Wang Laboratory for Cognition & Control in Complex Systems Department of Electrical and Computer Engineering University of Florida Thanks to NSF, AFOSR, and DOE / TCIPG June 26, 2012
  2. 2. Markets for Differentiated Electric Power ProductsConclusions in advanceTraditional fossil fuels will be history to our great grandchildren 2 / 34
  3. 3. Markets for Differentiated Electric Power ProductsConclusions in advanceWe need renewable energy, but how do we create a new energyinfrastructure to support this? 2 / 34
  4. 4. Markets for Differentiated Electric Power ProductsConclusions in advanceWe need renewable energy, but how do we create a new energyinfrastructure to support this?Some required elements: 2 / 34
  5. 5. Markets for Differentiated Electric Power ProductsConclusions in advanceWe need renewable energy, but how do we create a new energyinfrastructure to support this?Some required elements: Electricity must treated as a service and not a commodity: Gas turbine generation provides regulatory service. So could HVAC Smart Grid programs have helped to create a framework for greater service differentiation Real time control will be an essential element to combat volatility and ensure reliability 2 / 34
  6. 6. Markets for Differentiated Electric Power ProductsConclusions in advanceTraditional fossil fuels will be history to our great grandchildrenWe need renewable energy, but how do we create a new energyinfrastructure to support this?Some required elements: Electricity must treated as a service and not a commodity: Gas turbine generation provides regulatory service. So could HVAC Smart Grid programs have helped to create a framework for greater service differentiation Real time control will be an essential element to combat volatility and ensure reliabilityReal time prices have little or no value here: This is supported by theory and history. 2 / 34
  7. 7. Outline1 Smart Grid in 20122 Some Science3 Conclusions & Suggestions4 References 3 / 34
  8. 8. Smart Grid in 2012 Nodal Power Prices$20,000 per MWh Otahuhu Stratford$10,000 $0 0 4am 9am 2pm 7pm Smart Grid 2012 4 / 34
  9. 9. Smart Grid in 2012EIA 2011 StudySmart grid legislative and regulatory policies and case studiesMany success stories: Millions of smarter meters installed all over the globe PNNL study: Automation of water heaters and other appliances provided ancillary service in the Olympic peninsula Large buildings such as hotels, and energy-intensive companies such as IBM, Google, and ALCOA have contracts in place to help stabilize the grid, encouraged by FERC Ruling 745∗ ∗ Market-Based Demand Response Compensation Rule: Electric utilities and retail market operators are now required to pay demand response resources the market price (LMP) for energy 5 / 34
  10. 10. Smart Grid in 2012Increasing Leverage of FlexibilityConstellation Energy & NJP&L: Awards gift cards and rate reductions to residents for control of air conditioners; company sells flexibility as ancillary service. www.eia.gov/analysis/studies/electricity/pdf/smartggrid.pdf, December 12, 2011Energy department to launch new energy innovation hubfocused on advanced batteries and energy storage.www.energy.gov, February 7, 2012 Honeywell And Hawaiian Electric To Use Demand Response To Integrate Renewables And Reduce Fossil Fuel Dependence. www.honeywell.com, February 2, 2012Axion Power’s PowerCube Battery Energy StorageSystem Integrated Into PJM Utility Grid.www.axionpower.com, November 22, 2011 First ’small-scale’ demand-side projects in PJM providing frequency regulation. www.sacbee.com, November 21, 2011 6 / 34
  11. 11. Smart Grid in 2012EIA 2011 StudySmart grid legislative and regulatory policies and case studiesMany success stories ... and failuresResidential consumers have high expectations, Predictable cost savings They may distrust those tampering with their appliances. They distrust meters they believe interfere with appliances. 7 / 34
  12. 12. Smart Grid in 2012EIA 2011 StudySmart grid legislative and regulatory policies and case studiesMany success stories ... and failuresResidential consumers have high expectations, Predictable cost savings They may distrust those tampering with their appliances. They distrust meters they believe interfere with appliances.Moreover, the value of ancillary service obtained via demand response maybe reduced because of uncertainty of the level of consumer response. 7 / 34
  13. 13. Smart Grid in 2012EIA 2011 StudySmart grid legislative and regulatory policies and case studiesMany success stories ... and failuresResidential consumers have high expectations, Predictable cost savings They may distrust those tampering with their appliances. They distrust meters they believe interfere with appliances.Moreover, the value of ancillary service obtained via demand response maybe reduced because of uncertainty of the level of consumer response. ... yet, prices to devices are coming our way!∗ ∗ Terry Boston, CEO PJM, ISGT 2012 7 / 34
  14. 14. Smart Grid in 2012EIA 2011 StudySmart grid legislative and regulatory policies and case studiesMoreover, the value of ancillary service obtained via demand response maybe reduced because of uncertainty of the level of consumer response. ... yet, prices to devices are coming our way!∗ ∗ Terry Boston, CEO PJM, ISGT 2012 My concern: real-time pricing not TOU or contracts 7 / 34
  15. 15. Smart Grid in 2012EIA 2011 StudyCase studies ... very little to say on real-time prices "The active participation of final demand in the wholesale market is essential to managing the greater intermittency of renewable resources and in limiting the ability of wholesale electricity suppliers to exercise unilateral market power. A demand that is able to reduce its consumption in real-time in response to higher prices limits the ability of suppliers to exercise unilateral market power in a formal wholesale market such as the California ISO" (http://www.stanford.edu/group/fwolak/cgi- bin/sites/default/files/files/little_hoover_testimony_wolak_sept_2011.pdf) -F. Wolak "Virtually all economists agree that the outcome [of the California crisis] was exacerbated by the inability of the demand side of the market to respond to real or artificial supply shortages. This realization prompted my research stream on real-time electricity pricing." - S. Borenstein My concern: real-time pricing not TOU or contracts 8 / 34
  16. 16. Smart Grid in 2012Winds Cause Price SpikesMidwest ISO today: Friday afternoon, March 4, 2011 3:30 p.m. -2000.00 9 / 34
  17. 17. Smart Grid in 2012Winds Cause Price SpikesMidwest ISO today: Friday afternoon, March 4, 2011 3:50 p.m. -762.55 10 / 34
  18. 18. Smart Grid in 2012Winds Cause Price SpikesMidwest ISO today: Friday afternoon, March 4, 2011 4:15 p.m. -1881.07 11 / 34
  19. 19. Smart Grid in 2012Winds Cause Price SpikesMidwest ISO today: Friday afternoon, March 4, 2011 4:30 p.m. 12 / 34
  20. 20. Smart Grid in 2012Cold Causes Price SpikesTexas today: Winter of 2011 Power Prices in Texas $/MWh 3000 80 $/MWh January 31, 2011 February 2, 2011 2000 60 40 1000 20 10 0 0−10 5am 10am 3pm 8pm −500 5am 10am 3pm 8pm 13 / 34
  21. 21. Smart Grid in 2012Cold Causes Price SpikesTexas today: Winter of 2011 Power Prices in Texas $/MWh 3000 80 $/MWh January 31, 2011 February 2, 2011 2000 60 40 1000 20 10 0 0−10 5am 10am 3pm 8pm −500 5am 10am 3pm 8pmThere will be multiple autopsies of the causes for the latest power breakdowns ... Who profitedoff this near-meltdown and what can be done to incentivize power producers to maintainadequate reserve capacity for emergencies rather than waiting for emergency windfalls? –HOUSTON CHRONICLE, Feb 12, 2011 New report hits ERCOT, electricity deregulation: A report released Monday concludes that electric deregulation has cost Texas residential consumers more than $11 billion in higher rates... – Dallas Morning News, Feb 14, 2011 13 / 34
  22. 22. Smart Grid in 2012Day-Ahead Market OutcomesTexas today: Summer of 2011 ERCOT North Zone - August 1-30, 2011 Hourly day-ahead, daily on-peak, and monthly weighted average prices 3,000 hourly, day-ahead price 2,500 wholesale price ($/MWh) daily, on-peak price 2,000 weighted average monthly price ($188/MWh) 1,500 1,000 500 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 Source: U.S. Energy Information Administration, based on the Electric Reliability Council of Texas (ERCOT). Source: U.S. Energy Information Administration, based on the Electric Reliability Council of Texas (ERCOT). Note: ERCOT North Zone includes Dallas/Fort Worth metrometro region and surrounding areas of Northeast Texas. On-Peak Note: ERCOT North Zone includes Dallas/Fort Worth region and surrounding areas of Northeast Texas. On-Peak refers to the 16-hour16-hour time block from hours ending to 10:00 p.m. 10:00 p.m. CDT on weekdays, excluding NERC holidays refers to the time block from hours ending 7:00 a.m. 7:00 a.m. to CDT on weekdays, excluding NERC holidays. 14 / 34
  23. 23. Smart Grid in 2012Madness in New ZealandNew Zealand today: March 25, 2011A typical day in the New Zealand power market on the N. Island Nodal Power Prices in NZ: $/MWh Otahuhu Stratford 100 50 0 4am 9am 2pm 7pm http://www.electricityinfo.co.nz/ 15 / 34
  24. 24. Smart Grid in 2012Madness in New ZealandNew Zealand today: March 26, 2011$25 million dollars extracted by the generators in just six hours Nodal Power Prices in NZ: $/MWh Otahuhu 20,000 Stratford 10,000 0 4am 9am 2pm 7pm http://www.electricityinfo.co.nz/ 16 / 34
  25. 25. Smart Grid in 2012Madness in New ZealandNew Zealand today: March 26, 2011>$20 million dollars demanded back from Genesis Nodal Power Prices in NZ: $/MWh Otahuhu 20,000 Stratford 10,000 0 4am 9am 2pm 7pm http://www.electricityinfo.co.nz/Preliminary view of NZ Electrical Authority: Genesis was not guilty of“manipulative” ... or “deceptive” conduct. However, high prices threatened to 16 / 34
  26. 26. Smart Grid in 2012Madness in New ZealandNew Zealand today: March 26, 2011>$20 million dollars demanded back from Genesis Nodal Power Prices in NZ: $/MWh Otahuhu 20,000 Stratford 10,000 0 4am 9am 2pm 7pm http://www.electricityinfo.co.nz/Preliminary view of NZ Electrical Authority: Genesis was not guilty of“manipulative” ... or “deceptive” conduct. However, high prices threatened toundermine confidence in, and ... damage the integrity and reputation of thewholesale electricity market 3:59 PM Friday May 6, 2011 www.nzherald.co.nz 16 / 34
  27. 27. Smart Grid in 2012PNNL Prices to Devices ProjectsAutomation in the market Transactive Controls: Market-Based GridWiseTM Controls for Building Systems $$$ Bid Price P + kσ P Mean Price Bid Curve Clearing Price Current Zone P – kσ Desired or Temperature Idea Set Point Minimum Maximum Set Point Set Point Tset = 72oF Tmax = 77oF Adjusted Zone Tmin = 67oF Tset,a = 70oF Tcurrent = 75oF Comfort Set Point Temperature 17 / 34
  28. 28. Smart Grid in 2012PNNL Prices to Devices ProjectsAutomation in the market Transactive Controls: Market-Based GridWiseTM Controls for Building Systems $$$ Bid Price P + kσ P Mean Price Bid Curve Clearing Price Current Zone P – kσ Desired or Temperature Idea Set Point Minimum Maximum Set Point Set Point Tset = 72oF Tmax = 77oF Adjusted Zone Tmin = 67oF Tset,a = 70oF Tcurrent = 75oF Comfort Set Point Temperature Proportional control: Comfort = k × Price 17 / 34
  29. 29. Smart Grid in 2012PNNL Prices to Devices ProjectsAutomation in the market Transactive Controls: Market-Based GridWiseTM Controls for Building Systems $/MWh Mean Price Zone Bid Price Market Clearing Price 300 200 100 Hour 0 0 5 10 15 20 25 Proportional control: Comfort = k × Price 18 / 34
  30. 30. Smart Grid in 2012PNNL Prices to Devices ProjectsAutomation in the market Transactive Controls: Market-Based GridWiseTM Controls for Building Systems ConsumerAnger Mean Price Zone Bid Price Market Clearing Price 300 200 100 Hour 0 0 5 10 15 20 25 Proportional control: Comfort = k × Price 19 / 34
  31. 31. Smart Grid in 2012MIT Prices to Devices ProjectsAutomation in the market Market-Based Control @ MIT Fig. 4. Stochastic evolution of prices and demand Roozbehani et. al. 2010 $$$ Demand Consumer Anger? 100 600 Ask Munzer et. al. Mean Price Zone Bid Price Demand 400 50 200 0 Hour 0 50 100 150 Real Time Prices Can Be Ugly 20 / 34
  32. 32. Smart Grid in 2012MIT Prices to Devices ProjectsAutomation in the market Market-Based Control @ MIT Fig. 4. Stochastic evolution of prices and demand Roozbehani et. al. 2010 $$$ Demand Consumer Anger? 100 600 Ask Munzer et. al. Mean Price Zone Bid Price Demand 400 50 200 0 Hour 0 50 100 150 Real Time Prices Can Be Ugly Seen in Theory & Practice 20 / 34
  33. 33. Smart Grid in 2012The Future as seen by FERC TodayFERC Order 755: RTMs on the microsecond scale Regulation Required @ MISO 600 400 200 0 -200 -400 TimeNew rules for fair treatment of resources participating in regulation marketsCurrent method of regulation compensation does not fairly account for theregulation service provided. 21 / 34
  34. 34. Smart Grid in 2012The Future as seen by FERC TodayFERC Order 755: RTMs on the microsecond scale Regulation Required @ MISO 600 400 200 0 -200 -400 TimeNew rules for fair treatment of resources participating in regulation marketsCurrent method of regulation compensation does not fairly account for theregulation service provided.Requires ISOs to pay resources based on actual service provided 21 / 34
  35. 35. Smart Grid in 2012The Future as seen by FERC TodayFERC Order 755: RTMs on the microsecond scale Regulation Required @ MISO 600 400 200 0 -200 -400 TimeNew rules for fair treatment of resources participating in regulation marketsCurrent method of regulation compensation does not fairly account for theregulation service provided.Requires ISOs to pay resources based on actual service provided Sounds fair enough! 21 / 34
  36. 36. Smart Grid in 2012The Future as seen by FERC TodayFERC Order 755: RTMs on the microsecond scale Regulation Required @ MISO 600 400 200 0 -200 -400 TimeNew rules for fair treatment of resources participating in regulation marketsPossible payment plan, consider 1 of storage regulation, T d Payment ∝ S(t)| dt 0 dt 22 / 34
  37. 37. Smart Grid in 2012The Future as seen by FERC TodayFERC Order 755: RTMs on the microsecond scale Regulation Required @ MISO 600 400 200 0 -200 -400 TimeNew rules for fair treatment of resources participating in regulation marketsPossible payment plan, consider 1 of storage regulation, T d Payment ∝ S(t)| dt 0 dt Sounds game-able enough! 22 / 34
  38. 38. Some Science Torque Cost Low SpeedSome Science 23 / 34
  39. 39. Some Science Torque Cost Low Speed Some ScienceConcerning real-time pricing not TOU or contracts 23 / 34
  40. 40. Some ScienceEquilibrium with Dynamics & Network ConstraintsEntropic pricesTheorem 1: When dynamics (temporal constraints) are taken intoaccount, price is never equal to marginal cost [5, 4, 3, 1] 24 / 34
  41. 41. Some ScienceEquilibrium with Dynamics & Network ConstraintsEntropic pricesTheorem 1: When dynamics (temporal constraints) are taken intoaccount, price is never equal to marginal cost [5, 4, 3, 1]Equilibrium priceThe equilibrium price process is a function of equilibrium reserves: P ∗ (t) = p∗ (Re (t)) The marginal value of power to the consumer. 24 / 34
  42. 42. Some ScienceEquilibrium with Dynamics & Network ConstraintsEntropic pricesTheorem 1: When dynamics (temporal constraints) are taken intoaccount, price is never equal to marginal cost [5, 4, 3, 1]Equilibrium priceThe equilibrium price process is a function of equilibrium reserves: P ∗ (t) = p∗ (Re (t)) The marginal value of power to the consumer.Proof: Lagrangian decomposition, as in the static Second Welfare Theorem, 24 / 34
  43. 43. Some ScienceEquilibrium with Dynamics & Network ConstraintsEntropic pricesTheorem 1: When dynamics (temporal constraints) are taken intoaccount, price is never equal to marginal cost [5, 4, 3, 1]Equilibrium priceThe equilibrium price process is a function of equilibrium reserves: P ∗ (t) = p∗ (Re (t)) The marginal value of power to the consumer.Proof: Lagrangian decomposition, as in the static Second Welfare Theorem, or the proof of the Minimum Principle. 24 / 34
  44. 44. Some ScienceEquilibrium with Dynamics & Network ConstraintsEntropic pricesWhat is marginal value?It is not always obvious. With the introduction of network constraints, 25 / 34
  45. 45. Some ScienceEquilibrium with Dynamics & Network ConstraintsEntropic pricesWhat is marginal value?It is not always obvious. With the introduction of network constraints, Prices can go well beyond marginal value (as defined in static model) Prices can go well below zero [Dynamic competitive equilibria in electricity markets. Wang et. al. 2011] 25 / 34
  46. 46. Some ScienceEquilibrium with Dynamics & Network ConstraintsEntropic pricesWhat is marginal value?It is not always obvious. With the introduction of network constraints, Prices can go well beyond marginal value (as defined in static model) Prices can go well below zero [Dynamic competitive equilibria in electricity markets. Wang et. al. 2011]Without price-caps, Australia might look like an efficient equilibrium: 10,000 19,000 1,400 9,000 Victoria Demand 1,200 Tasmania 1,000 Price (Aus $/MWh) 8,000 Price (Aus $/MWh) Volume (MW) 1,000 Volume (MW) 7,000 Demand 6,000 800 10,000 5,000 0 Prices 600 4,000 3,000 400 2,000 Prices 200 - 1,000 1,000 1,000 0 - 1,000 0 - 1,500 00:00 01:00 02:00 03:00 04:00 05:00 06:00 07:00 08:00 09:00 10:00 11:00 12:00 13:00 14:00 15:00 16:00 17:00 18:00 19:00 20:00 21:00 22:00 23:00 24:00 00:00 01:00 02:00 03:00 04:00 05:00 06:00 07:00 08:00 09:00 10:00 11:00 12:00 13:00 14:00 15:00 16:00 17:00 18:00 19:00 20:00 21:00 22:00 23:00 24:00 25 / 34
  47. 47. Some ScienceSustainable business? Illinois: July 1998 California: July 2000 Spinning reserve prices PX prices $/MWh 5000 Purchase Price $/MWh 70 250 Previous week 60 4000 200 50 3000 150 40 2000 30 100 20 1000 50 10 0 0 Mon Tues Weds Thurs Fri Weds Thurs Fri Sat Sun Mon Tues Weds Ontario: November, 2005 Texas: February 2, 2011Marginal value of electricity, Forecast Demand Demand in MW Last Updated 11:00 AM Predispatch 1975.11 Dispatch 19683.5 $/MWh 21000 3000 Average price is usually $30 18000 2000 $250,000/MWh (?) 15000 Hourly Ontario Energy Price $/MWh Last Updated 11:00 AM Predispatch 72.79 Dispatch 90.82 2000 1000 Forecast Prices 1500 $/MWh 1000 500 0 0 3 6 9 12 15 18 21 3 6 9 12 15 18 21 3 6 9 12 15 18 21 Time −500 5am 10am 3pm 8pm Tues Weds Thurs 26 / 34
  48. 48. Some ScienceSustainable business? Illinois: July 1998 California: July 2000 Spinning reserve prices PX prices $/MWh 5000 Purchase Price $/MWh 70 250 Previous week 60 4000 200 50 3000 150 40 2000 30 100 20 1000 50 10 0 0 Mon Tues Weds Thurs Fri Weds Thurs Fri Sat Sun Mon Tues Weds Ontario: November, 2005 Texas: February 2, 2011Marginal value of electricity, Forecast Demand Demand in MW Last Updated 11:00 AM Predispatch 1975.11 Dispatch 19683.5 $/MWh 21000 3000 Average price is usually $30 18000 2000 $250,000/MWh (?) 15000 Hourly Ontario Energy Price $/MWh Last Updated 11:00 AM Predispatch 72.79 Dispatch 90.82 2000 1000 Forecast Prices 1500 $/MWh 1000 500 0 0 3 6 9 12 15 18 21 3 6 9 12 15 18 21 3 6 9 12 15 18 21 Time −500 5am 10am 3pm 8pm Tues Weds Thurs However, 26 / 34
  49. 49. Some ScienceSustainable business? Illinois: July 1998 California: July 2000 Spinning reserve prices PX prices $/MWh 5000 Purchase Price $/MWh 70 250 Previous week 60 4000 200 50 3000 150 40 2000 30 100 20 1000 50 10 0 0 Mon Tues Weds Thurs Fri Weds Thurs Fri Sat Sun Mon Tues Weds Ontario: November, 2005 Texas: February 2, 2011Marginal value of electricity, Forecast Demand Demand in MW Last Updated 11:00 AM Predispatch 1975.11 Dispatch 19683.5 $/MWh 21000 3000 Average price is usually $30 18000 2000 $250,000/MWh (?) 15000 Hourly Ontario Energy Price $/MWh Last Updated 11:00 AM Predispatch 72.79 Dispatch 90.82 2000 1000 Forecast Prices 1500 $/MWh 1000 500 0 0 3 6 9 12 15 18 21 3 6 9 12 15 18 21 3 6 9 12 15 18 21 Time −500 5am 10am 3pm 8pm Tues Weds Thurs However, Theorem 2: In this equilibrium, the average price is precisely the average marginal cost Proof: Lagrangian relaxation of initial condition. 26 / 34
  50. 50. Some ScienceSustainable business? Illinois: July 1998 California: July 2000 Spinning reserve prices PX prices $/MWh 5000 Purchase Price $/MWh 70 250 Previous week 60 4000 200 50 3000 150 40 2000 30 100 20 1000 50 10 0 0 Mon Tues Weds Thurs Fri Weds Thurs Fri Sat Sun Mon Tues Weds Ontario: November, 2005 Texas: February 2, 2011Marginal value of electricity, Forecast Demand Demand in MW Last Updated 11:00 AM Predispatch 1975.11 Dispatch 19683.5 $/MWh 21000 3000 Average price is usually $30 18000 2000 $250,000/MWh (?) 15000 Hourly Ontario Energy Price $/MWh Last Updated 11:00 AM Predispatch 72.79 Dispatch 90.82 2000 1000 Forecast Prices 1500 $/MWh 1000 500 0 0 3 6 9 12 15 18 21 3 6 9 12 15 18 21 3 6 9 12 15 18 21 Time −500 5am 10am 3pm 8pm Tues Weds Thurs However, Theorem 2: In this equilibrium, the average price is precisely the average marginal cost Proof: Lagrangian relaxation of initial condition. Is this a sustainable business? 26 / 34
  51. 51. Some ScienceMore Engineering: Where is the Missing Money?Addressing FERC Order 755 27 / 34
  52. 52. Some ScienceMore Engineering: Where is the Missing Money?Addressing FERC Order 755World-view, from eyes of a coal generator operator:Control must respect dynamics & costs 27 / 34
  53. 53. Some ScienceMore Engineering: Where is the Missing Money?Addressing FERC Order 755World-view, from eyes of a coal generator operator:Control must respect dynamics & costsCost of generation:Includes G and dt G, dshut-down, O&M, investment, ... 27 / 34
  54. 54. Some ScienceMore Engineering: Where is the Missing Money?Addressing FERC Order 755World-view, from eyes of a coal generator operator:Control must respect dynamics & costsCost of generation: TorqueIncludes G and dt G, d Cost Lowshut-down, O&M, investment, ...What is “marginal cost”? Speed 27 / 34
  55. 55. Some ScienceMore Engineering: Where is the Missing Money?Addressing FERC Order 755World-view, from eyes of a coal generator operator:Control must respect dynamics & costsCost of generation: TorqueIncludes G and dt G, d Cost Lowshut-down, O&M, investment, ...What is “marginal cost”? SpeedTheorem 3: If c(G, dt G) d = ce (G) + cw ( dt G) d then 27 / 34
  56. 56. Some ScienceMore Engineering: Where is the Missing Money?Addressing FERC Order 755World-view, from eyes of a coal generator operator:Control must respect dynamics & costsCost of generation: TorqueIncludes G and dt G, d Cost Lowshut-down, O&M, investment, ...What is “marginal cost”? SpeedTheorem 3: If c(G, dt G) d = ce (G) + cw ( dt G) d then E[P ∗ ] = E[ ce (G)] 27 / 34
  57. 57. Some ScienceMore Engineering: Where is the Missing Money?Addressing FERC Order 755World-view, from eyes of a coal generator operator:Control must respect dynamics & costsCost of generation: TorqueIncludes G and dt G, d Cost Lowshut-down, O&M, investment, ...What is “marginal cost”? SpeedTheorem 3: If c(G, dt G) d = ce (G) + cw ( dt G) d then E[P ∗ ] = E[ ce (G)] =⇒ lots of missing money 27 / 34
  58. 58. Some ScienceMore Engineering: Where is the Missing Money?Addressing FERC Order 755World-view, from eyes of a coal generator operator:Control must respect dynamics & costsCost of generation: TorqueIncludes G and dt G, d Cost Lowshut-down, O&M, investment, ...What is “marginal cost”? SpeedTheorem 3: If c(G, dt G) d = ce (G) + cw ( dt G) d then E[P ∗ ] = E[ ce (G)] =⇒ lots of missing moneyCompetitive equilibrium never compensates for “wear and tear”. 27 / 34
  59. 59. Conclusions & SuggestionsConclusions & Suggestions 28 / 34
  60. 60. Conclusions & SuggestionsEconomics and EngineeringArbitrage – between Nukes and Turbines? 29 / 34
  61. 61. Conclusions & SuggestionsEconomics and EngineeringArbitrage – between Nukes and Turbines?Any power engineer knows: Nuclear generation and gas turbines providedifferent services, and are not just delivering electrons. 29 / 34
  62. 62. Conclusions & SuggestionsEconomics and EngineeringArbitrage – between Nukes and Turbines?Any power engineer knows: Nuclear generation and gas turbines providedifferent services, and are not just delivering electrons. So FERC is on the right track ... 29 / 34
  63. 63. Conclusions & SuggestionsEconomics and EngineeringArbitrage – between Nukes and Turbines?Any power engineer knows: Nuclear generation and gas turbines providedifferent services, and are not just delivering electrons. So FERC is on the right track ... T dWhat about the other policy makers? Payment ∝ S(t)| dt 0 dt 29 / 34
  64. 64. Conclusions & SuggestionsEconomics and EngineeringArbitrage – between Nukes and Turbines?Any power engineer knows: Nuclear generation and gas turbines providedifferent services, and are not just delivering electrons. So FERC is on the right track ... T dWhat about the other policy makers? Payment ∝ S(t)| dt 0 dt “ ... One result of this divorce of the theory from its subject matter has been that the entitites whose decisions economists are engaged in analyzing have not been made he subject of study and in consequence lack any substance. ...consumers without humanity, firms without organization, and even exchange without markets” R. Coase, 1988 29 / 34
  65. 65. Conclusions & SuggestionsSummary of Research at Illinois & FloridaThe current RTM paradigm must be reconsideredConclusions of our research, under the most optimistic assumptions Volatile prices are to be expected in real time electricity markets in an efficient equilibrium. 30 / 34
  66. 66. Conclusions & SuggestionsSummary of Research at Illinois & FloridaThe current RTM paradigm must be reconsideredConclusions of our research, under the most optimistic assumptions Volatile prices are to be expected in real time electricity markets in an efficient equilibrium. Prices are never equal to marginal cost. 30 / 34
  67. 67. Conclusions & SuggestionsSummary of Research at Illinois & FloridaThe current RTM paradigm must be reconsideredConclusions of our research, under the most optimistic assumptions Volatile prices are to be expected in real time electricity markets in an efficient equilibrium. Prices are never equal to marginal cost. Risk to suppliers: Average prices coincide with average marginal cost. 30 / 34
  68. 68. Conclusions & SuggestionsSummary of Research at Illinois & FloridaThe current RTM paradigm must be reconsideredConclusions of our research, under the most optimistic assumptions Volatile prices are to be expected in real time electricity markets in an efficient equilibrium. Prices are never equal to marginal cost. Risk to suppliers: Average prices coincide with average marginal cost. Why stay in such a risky business? 30 / 34
  69. 69. Conclusions & SuggestionsSummary of Research at Illinois & FloridaThe current RTM paradigm must be reconsideredConclusions of our research, under the most optimistic assumptions Volatile prices are to be expected in real time electricity markets in an efficient equilibrium. Prices are never equal to marginal cost. Risk to suppliers: Average prices coincide with average marginal cost. Why stay in such a risky business?The real world 30 / 34
  70. 70. Conclusions & SuggestionsSummary of Research at Illinois & FloridaThe current RTM paradigm must be reconsideredConclusions of our research, under the most optimistic assumptions Volatile prices are to be expected in real time electricity markets in an efficient equilibrium. Prices are never equal to marginal cost. Risk to suppliers: Average prices coincide with average marginal cost. Why stay in such a risky business?The real world Volatile prices are observed all over the world Benefits to consumers are not clear, and innovation is slow 30 / 34
  71. 71. Conclusions & SuggestionsSummary of Research at Illinois & FloridaThe current RTM paradigm must be reconsideredConclusions of our research, under the most optimistic assumptions Volatile prices are to be expected in real time electricity markets in an efficient equilibrium. Prices are never equal to marginal cost. Risk to suppliers: Average prices coincide with average marginal cost. Why stay in such a risky business?The real world Volatile prices are observed all over the world Benefits to consumers are not clear, and innovation is slow Market power is a reality, and symmetric information is absurd: Strategic behavior can lead to a new crisis each year! 30 / 34
  72. 72. Conclusions & SuggestionsThe current RTM paradigm must be reconsideredWhy have real time markets?No value has been demonstrated for real-time markets. 31 / 34
  73. 73. Conclusions & SuggestionsThe current RTM paradigm must be reconsideredWhy have real time markets?No value has been demonstrated for real-time markets.Empirical evidence: We cannot distinguish robbery from efficiency. 31 / 34
  74. 74. Conclusions & SuggestionsThe current RTM paradigm must be reconsideredWhy have real time markets?No value has been demonstrated for real-time markets.Empirical evidence: We cannot distinguish robbery from efficiency. Is This A Free Market For Fire Fighters? 31 / 34
  75. 75. Conclusions & SuggestionsThe current RTM paradigm must be reconsideredWhy have real time markets?No value has been demonstrated for real-time markets.Empirical evidence: We cannot distinguish robbery from efficiency. Is This A Free Market For Fire Fighters? Why then would you use real-time prices to control devices? 31 / 34
  76. 76. Conclusions & SuggestionsThe current RTM paradigm must be reconsideredWhy have real time markets?No value has been demonstrated for real-time markets.Empirical evidence: We cannot distinguish robbery from efficiency. Is This A Free Market For Fire Fighters? Why then would you use real-time prices to control devices? The EIA study shows that there are alternatives 31 / 34
  77. 77. Conclusions & SuggestionsThe current RTM paradigm must be reconsideredAlternatives TOU prices for peak shaving, and Contracts for real-time demand-response services 32 / 34
  78. 78. Conclusions & SuggestionsThe current RTM paradigm must be reconsideredAlternatives TOU prices for peak shaving, and Contracts for real-time demand-response services Successful contracts today: Constellation Energy, Alcoa, residential pool pumps, commercial buildings (see new work at Univ. of Florida), ... Efficiency loss, but utility and consumers each have reliable services 32 / 34
  79. 79. Conclusions & SuggestionsThe current RTM paradigm must be reconsideredAlternatives New Smart appliances that can facilitate these contracts Control theory to make this all work: We don’t know why the grid is so robust today. Introducing all of these dynamics will lead to new control challenges. 32 / 34
  80. 80. Conclusions & SuggestionsThe current RTM paradigm must be reconsideredAlternatives New Smart appliances that can facilitate these contracts Control theory to make this all work: We don’t know why the grid is so robust today. Introducing all of these dynamics will lead to new control challenges. Energy policy that is guided by an understanding of both physics and economics 32 / 34
  81. 81. Conclusions & SuggestionsThe current RTM paradigm must be reconsideredAlternatives New Smart appliances that can facilitate these contracts Control theory to make this all work: We don’t know why the grid is so robust today. Introducing all of these dynamics will lead to new control challenges. Energy policy that is guided by an understanding of both physics and economics Thank You! 32 / 34
  82. 82. Conclusions & SuggestionsPre-publication version for on-line viewing. Monograph available for purchase at your favorite retailer August 2008 Pre-publication version for on-line viewing. Monograph to appear Februrary 2009More information available at http://www.cambridge.org/us/catalogue/catalogue.asp?isbn=9780521884419 Control Techniques Markov Chains FOR and Complex Networks Stochastic Stability P n (x, · ) − π f →0 sup Ex [SτC (f )] < ∞ C π(f ) < ∞ ∆V (x) ≤ −f (x) + bIC (x) Sean Meyn S. P. Meyn and R. L. Tweedie References 33 / 34

×