Distributed Storage vs. Local Market Clive Tomlinson 05/06/2015 1
Two questions • Energy markets tend to be constrained • Historical/technical reasons • Local markets for many other goods are unconstrained • Would an unconstrained local energy market ‘work’? • Private storage benefits limited to deferring self- consumption • No benefit without own PV • No benefit for other users • How would private storage behave in an unconstrained local market? 05/06/2015 2
‘Unconstrained’ • Traders few and central • Home owner traders • Price set centrally • Price negotiated per trade • One trade at a time • Overlapping trades • Commit early • Commit late • Trade in 30 minutes++ • Trade in shrew’s heartbeat • Trade synchronously • Trade asynchronously • Bird’s eye view • Market stall view • Reliance on forecasting • Abandon forecasting • Distorted by FIT • No FIT 05/06/2015 3
Market model • Power network • Coincident with low voltage feeder • Varying DG and private storage • Energy volt gradient • Trading network • Traders = households & local businesses • Assumes conventional supplier of last resort • Trading down to 10s blocks • Data network • Domestic broadband / Internet • Home trading agent 05/06/2015 4
Economic model 05/06/2015 5
Discrete-event simulation Data Network Data Network Data Trading Network Data Network models x 200 Price setting heuristics Storage Simulated demand, management generation, heuristics storage 05/06/2015 6
Live network 05/06/2015 7
Does it work? • Trades happen • But not constantly • Households make/save money • At the expense of the incumbent supplier • Benefit to all local participants • Storage owners: price arbitrage • DG owners: market price > conventional supplier buy price • Non-owners: market price < conventional supplier sell price • Here are some pictures… 05/06/2015 8
Local price proxies local abundance demand response to price without price setting Prices shown are average – there’s no single market price 05/06/2015 9
Selfish control, network effects 2,500 2,000 1,500 1,000 500 kWh storage 0 0 300 600 900 1200 1500 1800 2100 2400 2700 3000 Consumption of energy from OFF-LF (kWh/d) 05/06/2015 10
Localism modest ICT • even for n * (n-1) second, average (whole LEM) 1.00E+05 Application bytes sent per Size of LEM 1.00E+04 1.00E+03 1.00E+02 200 1.00E+01 100 1.00E+00 20 10 60 600 10 3600 Trading interval (s) 86400 05/06/2015 11
Storage modifies price curve • No storage: price curve discontinuous • Some storage: price curve smooth, equilibrium in mid range, local trading increased • T oo much storage: • no shared energy availability pattern, • price unstable, • local trading decreased 05/06/2015 12
£££ manage flows • Intervene by selling cheap before evening demand peak 0 6 7 8 9 -2 Energy flow out of (into) LEM (kW) -4 -6 -8 -10 -12 -14 -16 -18 -20 Time of day (h) 05/06/2015 13
Conculsions Qualitatively tested • Numbers depend on so many things… energyBay? • T echnical • Regulation • Business ambition / consumer desire ? Good market for storage, but is it best? • So many market model nuances • That’s a puzzle 05/06/2015 14
Read the details • Proceedings of ICE – Energy, Volume 168 (2015) • Or from clive@swanbarton.com 05/06/2015 15
Recommend
More recommend
