SO Incentives Stakeholder Workshop Place your chosen image here. The four corners must just cover the arrow tips. For covers, the three pictures should be the same size and in a straight line. Monday 21 st January Jo Faulkner – Balancing Services Manager
In this session BSUoS and incentives SO Incentives journey How we have improved the current target setting methodologies Constraints Energy Advantages of a BSIS cost target incentive arrangement Our concerns on Ofgem’s cost disallowance/discretionary reward consultation 2
Contribution to BSUoS - 2011/12 2011/12 Wholesale/BM Prices 1000 SO Internal Costs 900 Other services 800 Contribution to BSUoS (£m) 700 Constraints 600 Generation Topology 500 Outages NETSQSS 400 Standards Reserve 300 Contracted Market Services / Length 200 Availability Response 100 0 -100 Residual Energy SO Loss Balancing 3
Journey to current incentive scheme 2013 - Pre Apr 2011: • Forecast target • Market volatility • Windfall gains & losses Learn the lessons from 2011-13: • Continue to move Drivers for change: forward • Focus on SO control • Multi year scheme 4
Methodologies and Relationships Forecast Actual Data Data Relationships Cost Accuracy Target Methodology 5
Steps already taken to enhance the constraint methodology Consultation carried out in July 2012 based on learning from scheme to date: Increased accuracy of a number of inputs Use actual data where we have limited control/forecasting capability e.g. generator availability and interconnector flows Allow for removal of erroneous data Correct identified issues with the model itself Accurate Appropriate Accurate and model Inputs focussed settings cost target 6
Impact of the methodology amendments Based solely upon forecast data, including Forecast forecasts for those ex post inputs Based on a combination of forecast inputs (ex ante) and real data inputs (ex post) as we Target move through the scheme – known at scheme end Actual cost of operating the system – Outturn compared against the target to determine National Grid’s performance 7
Constraint Cost Methodology Place your chosen image here. The four corners must just cover the arrow tips. For covers, the three pictures should be the same size and in a straight line. Iain McIntosh – Future Requirements Manager 8
Further constraint methodology enhancements identified Better representation of GB Transmission Network Ability to identify and directly map any constraint boundary into the model Improved wind modelling – including explicit modelling of embedded wind Improved hydro and pumped storage modelling Visualisation capability 9
Original constraint model 10
Enhanced constraint model 11
Wind representation Phase 1 model assumed generic load factor across GB Due to connection points of wind, increasingly localised constraints occur. Difficult to capture with SYS boundary definitions Therefore, even if wind is put into model ex-post, cost allocation is not necessarily accurate Forecast wind profiles reflect “typical year” 12
Enhanced wind representation Wind output can be modelled against actual transmission or GSP connection point Embedded wind modelled explicitly. Where no metered output exists, modelled with reference to most geographically proximate meteorological station Localised boundaries can now be modelled. High sensitivity 13
Energy Cost Methodology Place your chosen image here. The four corners must just cover the arrow tips. For covers, the three pictures should be the same size and in a straight line. Matt Magill – Trading and Assessment Manager 14
The 2011/13 Model Cummulative Scheme to date Target cost and outturn costs £1,200 25% £1,000 20% £800 15% ('000s) £600 10% £400 5% £200 0% £- -5% Apr- May- Jun- Jul- Aug- Sep- Oct- Nov- Dec- Jan- Feb- Mar- Apr- May- Jun- Jul- Aug- Sep- Oct- Nov- Dec- 11 11 11 11 11 11 11 11 11 12 12 12 12 12 12 12 12 12 12 12 12 Scheme to date Outturn Costs Scheme to date Target Costs Scheme to date deviation from target as percentage
The 2011/13 Model The theory behind the model is still the most applicable for development - statistical analysis of past behaviours to determine future outcomes The 2011/13 model was built in excel and required large amount of resource to apply and audit any changes Model development was challenging and resulted in: Model relationships that were not fully developed Some Ex-Ante inputs were inappropriate
Changes we have made for 2013 onwards We have developed a new modelling architecture A new inputs database that sources all the input data, creates auditable files and allows multiple selection and choices to be made Allows for rapid development of flexible component design including more volume, price and cost models Also allows for rapid development of input selections Improved, auditable and repeatable back testing The new modelling architecture gives us the opportunity to create more robust auditable models
Model Development Timeline Constraint Model Parallel Run Complete Continuing Parallel run Calibration of calibration Un- for Previous New Constraint constrained months Network Model Model Topology Jan Feb Mar Apr May New Further Further Parallel run Optimise and Architecture Model Model for previous refresh with complete Development Development months latest data with 11/13 phase starts finalised model baseline Constraints Energy Model Parallel Run Energy 18
BSIS vs Cost Disallowance Place your chosen image here. The four corners must just cover the arrow tips. For covers, the three pictures should be the same size and in a straight line. Katharine Clench – SO Incentives Development 19
Benefits of target based incentives ….. Align SO Align our and TO interests to those incentives of consumers Objective Encourage and framework within Incentives reward which to make innovation decisions = consumer Fair balance of risk and reward (upside & value downside in 20 equal measure)
Our concerns with cost disallowance Introduces risk and uncertainty to system operation Proposed limit on disallowance Insufficient detail and clarity as is disproportionate: to how disallowance may occur: • Circa £90m cap • Trigger for Ofgem investigation • Issue of hindsight Create inefficiencies rather than reduce them – foster a more cautious approach to balancing More actions managed through the Balancing Mechanism rather than through contracting ahead of time 21
Key Messages A cost target approach is required to maintain our current contracting activities and encourage further innovation Our view is that the quality of the inputs is vital to creating an accurate and robust cost target The 2011-13 scheme has enabled us to understand where improvements need to be focused on inputs and methodologies We should maintain momentum in this area rather than turn to developing a whole new approach to incentivisation Incentives should be designed such that they continue to deliver benefits to consumers 22
Recommend
More recommend
