Electricity market models Design of the National Electricity Market CEEM 2006 Gross pool (eg NEM): Temporal & location risk managed collectively: Ancillary services, spot market, PASA, SOO Net pool (eg UK NETA): Long term & location risk managed bilaterally Network not modelled in trading arrangements Capacity markets commonly used Short-term operational risk managed collectively: System operator given only one day s notice of bilateral trades 2 (ERIG discussion papers, 11/06) National Electricity Law: Overall objective for the National Electricity Market (NEM) NEL Section 7: The national electricity market objective is to promote efficient investment in, and efficient use of, electricity services for the long term interests of consumers of electricity with respect to price, quality, reliability and security of supply of electricity and the reliability, safety and security of the national electricity system A difficulty with this objective: Ambiguity of the wording with respect to interpretation of terms & trade-offs between sub-objectives 3 4
Scope of the NEM Features of National Electricity Rules (NER) Queensland New South Wales & ACT Victoria South Australia Tasmania (Basslink in 2006) NEM regions are indicated, and their boundaries need not be on state borders (e.g. two regions in NSW) NEM covers all participating states: A multi-region pool with intra-regional loss factors Ancillary services, spot market & projections Auctions of inter-regional settlement residues Operated by NEMMCO (owned by states) Compulsory participants in NEM: All dispatchable generators & links > 30 MW Network service providers & retailers Contestable consumers may buy from NEM 5 6 NER categories of generators Either market, non-market or exempt Market implies sell to NEM Can then also sell ancillary services Non-market or exempt: sell to local retailer Either scheduled or non scheduled: Scheduled implies centrally dispatched: Must then participate in the NEM processes of bidding, predispatch & PASA Default category for generation projects > 30 MW Not appropriate for intermittent generation, eg wind 7 Electricity industry structure in SE Australia Generation Sector:- large generators Gen 1 Gen 2 Gen 3 Gen X Electricity Financial instrument & REC (emission) trading Intentions offers & payments Tx network pricing Multi-region National Electricity Market (NEM) Intentions bids & payments Transmission Tx network Sector pricing NSW Victoria South Aust. Electricity Queensland & Tasmania Retail sector Retailer 1 Retailer 2 Retailer Z Distribution sector Distributor 1 Distributor 2 Distributor Y Retail Markets Network access Electricity Embedded generators Contestable end-users Franchise End-users End-use sector End-use Equipment & Distributed resources 8
Spot market offers & bids Offer (or Bid) structure in the NEM Generators, retailers & consumers: Price-quantity curve (sell/buy) for each half hour: 10 daily prices, quantities changeable until dispatch Demand forecasts bid in at $10000/MWH (VoLL) Dispatchable links between regions: Flow offer curve based on price difference 10,000 $/MWh up to 10 price levels that apply all day Demand forecast for un-bid demand Bids & offers ranked to give dispatch stack: Considering loss factors & inter-tie constraints 5 minute prices set by economic dispatch: Half-hourly averages are calculated in real time -1000 A set of offer quantities for each 5 min dispatch interval MW 9 10 Single node spot market (all offers & bids) $/MWh p Aggregate demand curve D consumers surplus producers surplus VOLL & forecast if no demand bidding Aggregate generation curve G local monopoly price band MW one marginal price p Spot price formation - central dispatch Maximise value of spot market trading, subject to: Dispatch offers & bids; ancillary service offers Constraints due to availability & commitment Non-scheduled load requirements in each region Power system security requirements Intra-regional network constraints & losses Inter-regional network constraints & losses Constraints consistent with registered bid & offer data Current levels of dispatch Ancillary service requirements Pro-rata loading of tied bid and/or offer data Minimising the impact of a direction or reserve contract dispatch 11 12
Region boundaries & inter-connectors Regions boundaries selected so that: Transmission constraints are rare within a region Frequently-occurring constraints are placed on region boundaries Region boundaries to be reset as required: Whenever a constraint occurs > 50 hours/year An unregulated inter-connector is allowed if: dispatchable so that it can bid like a generator: Directlink the first (operating since July 2000): 180 MW DC link between NSW & Queensland regions 13 NEM regional spot market model (Based on NEMMCO, 1997) South Aust SANI (proposed regulated AC) Murraylink 220 MW DC 500 MW 250 MW 1,100 MW Victoria 3,000 MW Snowy 1,500 MW 750 MW 300 MW NSW 850 MW Tasmania 600 MW pk Basslink DC link MNSP (2006) Queensland 1000 MW Directlink 180 MW DC thermal or stability flow limits 14 Node NQ CQ SWQ SEQ NNS NCEN CAN SNY MEL LV POR SESA RIV ADE NSA TAS Pk Ld (MW) 1250 1900 200 4350 10000 5750 900 650 100 500 2100 200 1500 Gen Cap (MW) 4150 2150 1450 150 11650 300 3900 7000 150 50 2250 1100 2500 0 Net Gen (MW) - 450 2250 1950-2900 - 650 1650-500 3100-4950 6100-650 50-450 150 900 1000 16 region NEM model (NEMMCO SOO, 2004) load >gen 15 Tasmanian electricity & gas industries & Basslink (Govt. of Tasmania, 2003) Transmission: Transend ~1200 MW ~350 MW Distribution Melbourne Latrobe Valley Basslink Northern Tasmania Southern Tasmania 16
Modelling regulated interconnectors & intra-region location Regulated interconnector between 2 regions Modelled by a linearised marginal loss function: A dynamic network loss factor that depends on flow Flow limits (security or thermal criteria) Locational effects within regions Modelled by static network loss factors (LFs) Annual average of estimated half-hour marginal losses for each generator node & group of consumer nodes Intra-regional constraints not modelled but a constrained-on generator cannot set price 17 Effect of intra-regional network loss factors on spot market outcomes Dispatch offer price = DOP g Referred offer price = DOP/LF g Reference Node (price p r ) Generator Qg, LFg Consumer Qc, LFc Generator produces Q g & is paid p r x LF g x Q g Consumer consumes Q c & pays p r x LF c x Q c Net income is given to network service provider(s) National Metering Identifier determines location & thus the network loss factor that will be applied 18 Combining dynamic inter- regional & static intra-regional loss factors (unconstrained link) Region A ~ 1 If B RRP = 35.4 Then A RRP = 34.0 Region B LF 1.05 ~ 3 Bid 35 $/MWh»33.3 $/MWh (B) LF 1.05 Bid 35 $/MWh»33.3 $/MWh (A)»34.7 $/MWh (B) A RRN power B RRN ~ 2 Assume 4 ~ LF 0.99 inter-regional Bid 35 $/MWh dynamic loss factor»35.4 $/MWh (B) = 0.96, Region A wrt B LF 0.98 Bid 35 $/MWh»35.7 $/MWh (A)»37.1 $/MWh (B) If 4 marginal generator Then B RRP = 35.4 Supply-demand balance in the electricity industry Generator input power Thermal power stations Hydro generators Other, eg wind farms + Frequency is a measure of supply-demand balance: always varying due to fluctuations in power flows _ Load electrical power plus network losses Industrial Commercial Residential 19 20
Managing supply-demand balance in NEM Spot market forecasts & derivative markets Energy markets Ancillary services & security time Security projections & FCAS derivative markets Spot market for period t increasing uncertainty looking forward spot period t Frequency control ancillary service markets, period t Spot market for period t+1 spot period t+1 FCAS markets for period t+1 21 NEMMCO processes for managing supply-demand balance NEMMCO Forecasts of supply & demand 10 year (annual) 2 year (weekly) 1 week (hourly) day-ahead spot price & dispatch (5 min) Power system reliability & security standards Spot & FCAS Markets Derivative Markets NEMMCO operation: Participant bid/offers Network data Demand forecast Reserve threshold Security constraints Reliability safety net 22 NEM Pre-dispatch, Dispatch & AGC Day-ahead offers from participants Continuously updated offers from participants Security-based operating constraints (NEMMCO) SCADA Pre-dispatch (24 hr projection) Offer & constraint database Constrained economic dispatch (5 min load forecast) AGC (2 second cycle) Forecast sets of 5-min spot prices Set of spot prices for next 5 min Instructions to participants 23 Bidding & dispatch (source: NEMMCO) Commitment notices Capacity Energy/RoC Day -2 Day -1 Day 0 Day 1 ST PASA 1400hrs Initial Offers/Bids 1230hrs then re-bids until dispatch time 1st Pre-dispatch 1600hrs Updated 3hrly Dispatch Day 5-minute & 30-minute prices as set; previous days data at 0hrs 24
Dispatch, Pre-dispatch, PASA, SOO & ANTS (source: NEMMCO) 0 day 1 day 2 week 1 month 1 year 1 year 2 Medium Term PASA to 2 yr, updated weekly SOO & ANTS (10 yr) Short Term PASA to 7 days @ 30 min res, 2 hr update Pre-dispatch 1 to 1.5 days, re-bid & final dispatch schedule ST & MT Projected Assessment of System Adequacy support reserve assessment & participant operating decisions. ST PASA projects region demand & reserve for 7 days @ 5 min resolution, updated every 2 hours. MT PASA projects region daily peak demand & reserve for 2 yrs, updated weekly. Statement of Opportunities (SOO) & Annual National Transmission Statement (ANTS) are intended to inform generation, demand & network investment decisions (10 year horizon, issued annually) 25 PASA & reserve contracts Invitation to provide more capacity Purchase of more capacity Energy constraints Demand forecasts Network capacity Registration of available capacities PASA Available capacity (long term expected USE < 0.002%) NEMMCO Expected inadequacy Day-ahead Offers & bids Payment for purchased capacity Settlements Purchased capacity offered to market at price that would have otherwise applied (usually VOLL) Pre-dispatch then dispatch 26 Derivative trading in support of NEM Derivative trading in the NEM (Bach Consulting & SIRCA, Report for NEMMCO, 2002) Trading in swap & cap contracts: Bilateral trading Over-the-counter instruments Exchange-traded CFDs (swaps) Inter-regional hedges: Specialised form of financial instrument: to manage regional price difference risks funded by interconnector settlement residues NEMMCO inter-regional settlement residue auctions: Commenced in 1999 27 Key issues: Price discovery Aggregate volume Liquidity & depth of trading Little end-user participation 28
Key derivative markets Forward contracts (futures) (~$6 billion pa) Expected spot price for a defined load shape & period (eg flat annual demand) Either OTC or exchange traded Call options Renewable energy certificates Available to qualifying generators Increasing to 9,500 GWH pa at 2010 then constant to 2020 (~$300 million pa) Relationship between wholesale gas & electricity markets Market design simpler for gas than electricity: Gas macro-molecular; electricity sub-molecular Flows on pipelines individually controllable but not on transmission lines Intermediate storage in linepack, not in electrical networks Gas network linepack is a shared resource: Gas market design should be a pool not physical bilateral trading if linepack constraining: It does when gas is used for electricity generation 29 30 Gas wholesale trading in Australia Victoria (~8 hours linepac): Pool day-spot market with bilateral physical trading Pool market does not set sub-day prices are derivative markets adequate? Other states (several days linepack) Bilateral physical trading Status of gas wholesale trading: inadequate to support use of gas for peak electricity generation 31 Many of our publications are available at: www.ceem.unsw.edu.au 32