Market Performance and Planning Forum Welcome to the Inaugural - PowerPoint PPT Presentation

Market Performance and Planning Forum Welcome to the Inaugural Meeting Karen Edson Vice President, Policy and Client Services February 4, 2010

Objective: Enable dialogue on implementation planning and market performance issues  Review key market performance topics  Share updates to 2009-2011 release plans, resulting from stakeholders inputs  Provide information on specific initiatives  to support Market Participants in budget and resource planning  Focus on implementation planning; not on policy  Clarify implementation timelines  Discuss external impacts of implementation plans  Launch joint implementation planning process Slide 2

Agenda TIME TOPIC PRESENTER 9:00-9:15 Overview, Objectives Karen Edson 9:15 – 9:45 Power System Technology Architecture & Development Khaled Abdul- Rahman, Hani Alarian 9:45-11:00 Market Performance Mark Rothleder -Minimum Online Commitment Constraint -Improve Imbalance Requirements in RT Market -RTUC to RDT Commitment Consistency and Financial Implications -Load Distribution Factor -HASP/RTD Price Convergence -Market Model Improvements 11:00 -12:00 Policy Update Margaret Miller -Current Stakeholder Initiatives -Market Evolution -Market Initiatives Roadmap 12:00 -1:00 Lunch – Provided by ISO 1:00 – 2:15 Release Planning Khaled Abdul- -Spring 2010 Release Rahman, Hani -Summer 2010 Release Alarian, Janet Morris -Fall 2010 Release -Early 2011 Release -Upcoming Implementation Efforts 2:15 – 3:00 Other Forum Updates – SIUG, SaMC, Team Slide 3

Power System Technology Architecture & Development Khaled Abdul-Rahman, PhD. Principal, Market and Performance Initiative Owner Hani Alarian Director, Advanced Power Network Technology Slide 4

Power System Technology Architecture & Development  New Department  Involved from inception through after production  Works with all departments to balance the needs of policy, business units, design complexity, implementation, integration, testing, transition to production, maintenance and support after production  Evaluates the feasibility and balance of the design to reach the goal of engineering, economic, grid operation, optimization, transmission, performance, and IT needs. Slide 5

Our release plan derives from the strategic plan, FERC orders, and stakeholder input. Continue the Integration of Renewable Resources Comparable Treatment of Generation and Non-Generation Transparent Develop Market Design Enhancements and well functioning Implement Planned Market Improvements markets Long term Improved resource Market Infrastructure planning Efficiency MSG guided by PLR Reliability and LDF MOC Reliability Economics Reduce Greater FOR CB PDR credit risk Demand Continue to SCP participation ASHASP SP MWP advance State, Improved Regional and price Federal Economic Efficiency signals priorities 1 year 2-3 years 5 years 2010 Slide 6 Slide 6

Market Performance Mark Rothleder Director, Market Analysis & Development Slide 7

Current Market Improvements Short-Term Market Improvements – Current Market • Minimum Online Commitment Constraint • Improve Imbalance Requirements in RT Market • RTUC to RDT Commitment Consistency and Financial Implications • Load Distribution Factor • HASP/RTD Price Convergence • Market Model Improvements 1 year 2-3 years 5 years During 2010 Slide 8 Slide 8

Minimum Online Commitment Constraint Objective:  Obtain correct and consistent unit commitment in IFM/RUC to satisfy procedural and outage-based reliability requirements  Reduce the need for exceptional dispatch  Trade-Date February 5, 2010 start using constraint for G-217/G- 219 Slide 9

Load Distribution Factor (LDF) Objective:  Analyze the quality and develop some metrics of LDF accuracy that can be tracked  Analyze the LDF correlation between weather patterns and changes to weather patterns  Develop and implement improvements to the LDF accuracy and forecasting  Track improvements and impacts of LDF accuracy improvements Slide 10

Day-Ahead System LDFs Improvement Overview  Perform Day-Ahead hourly load forecast for each of the 23 sub-LAPs within the CAISO system  Adjustment factors, one for each sub-LAPs for each hour are determined on the basis of the sub-LAPs DA load forecast.  Adjustment factors are used to adjust DA system LDFs under the current approach so that the total LDFs within each sub-LAP will reflect sub-LAP forecast loads relative to each other.  The adjusted DA system LDFs are fed to the market system to run DA market. Slide 11

Sub-LAP Day-Ahead Hourly Load Forecast  Currently, CAISO production system is set up to perform DA load forecast for 10 load forecast zones  Load forecast model is based on neural network and auto regression and moving average methodologies  Weather forecast and actual load and weather data of pre-specified lag time are fed to the load forecast model to perform load forecasting  Sub-LAP DA load forecasting is on “indirect” basis through the load forecast of the 10 load forecast zones  Using the 2009 hourly load data of sub-LAPs and load forecast zones thr o ugh statistical approach, the CAISO identifies for each sub-LAPs, one or a group of LF zones of no more than 2 to be best correlated with Slide 12

Sub-LAP Day-Ahead Hourly Load Forecast cont  For each sub-LAP, a functional mapping from the load(s) of the identified forecast zone (group of forecast zones) to the sub-LAP load is determined on the basis best fit between the historical hourly load data of the sub-LAP and the forecast zone.  In future operation, using the forecast loads of forecast zones, the forecast load of each sub-LAP is determined by applying the pre-determined functional mapping of the sub-LAP to the forecast load(s) of its corresponding forecast zone (group of forecast zones). Slide 13

Proposed Adjustments of DA LDF from Current Production System by Example  A system consisting two sub-LAPs A and B with sub-LAP A comprised of Nodes 1 and 2 and sub-LAP B comprised of Nodes 3 and 4.  Normalized LDFs from current production system adaptation method are given in the table below  LDF values implies that Node Name LDF ratio of the two sub-LAP N1 (Sub-LAP A) 0.4 loads be 7 : 3 N2 (Sub-LAP A) 0.3 N3 (Sub-LAP B) 0.2 N4 (Sub-LAP B) 0.1 Slide 14

Proposed Adjustments of DA LDF from Current Production System by Example  Forecasted loads of sub-LAPs are 80MW for sub-LAP A and 20MW for sub-LAP B  Adjustment factors are 8/7 for sub-LAP A and 2/3 of sub- LAP B. Each sub-LAP specific adjustment factor is applied to all nodes of the sub-LAP. Node Adj LDF  Final LDFs after adjustments are listed in the table on the N1 0.4*8/7 = 0.457 left. The adjusted LDFs are N2 0.3*8/7 = 0.343 normalized. N3 0.2*2/3 = 0.133 N4 0.1*2/3 = 0.067 Slide 15

Improve Imbalance Requirements of RT Market Objective:  Direct 15 minute and 5 minute load forecast  Analyze the need for regulation/AGC feedback into the imbalance energy needs with objective of returning regulation resources  Propose solution and potential approach to dispatching in recognition of regulation energy dispatched Slide 16

The ISO improved the consistency between RTPD (HASP) and RTD load forecast. Slide 17

RTPD load forecast lags behind RTPD load forecast before the improvement. Hourly Average of 12/29/09 – 01/01/10 Slide 18

There is no more such pattern after the improvement. Hourly Average of 01/02 – 01/05/10 Slide 19

The ISO will monitor RTM load forecast and its impacts on other market issues.  Causes for remaining differences are under review  The improvement may help other known issues  Accuracy of RTPD unit commitment decision  Divergence between HASP and RTD energy prices Slide 20

HASP/RTD Price Convergence Objective:  Continue to analyze the sources of the HASP/RTD price divergence  Evaluate the impact of HASP load forecast for any systemic time shifts or load forecasting differences  Implement and evaluate the Hourly Intertie Ramp enhancements which are currently in test  Monitor improvement to RT Energy Offset Slide 21

RTUC to RTD Commitment Consistency Objective:  Understand the drivers to the RTUC and RTD price differences that leads to different expected results  Analyze the source of if issue is more a problem with STUC (5 hour run) and RTD versus RTUC and RTD  Review STUC SIBR bid replication rules  Analyze effect load forecast shifting has on RTUC/RTD price/commitment  Analyze effectiveness of BCR to make resources whole Slide 22

Market Model Improvements Objective:  Model Trans-Bay DC cable  Review opportunity for enhancing external network to  Improve accuracy of responsiveness to Lugo – Victorville Congestion. Development of a new nomogram is underway  Compensating Injections:  Return automated compensation  Analyze impact on flow accuracy improvement  Analyze impact on imbalance Slide 23

Model: Trans-Bay Cable General:  53 Mile Direct – Current (DC) Cable  Connects Pittsburg (East Bay) to Potrero (San Francisco)  400 MW Unidirectional Control from Pittsburg to Potrero under CAISO control  +/-300 MVAR reactive capability under PG&E control  PTO: Cost recovery via Transmission Access Charge Slide 24