Preprint of paper for World Renewable Energy Congress, Firenze 1998







TRADING WIND POWER AT THE NORDIC POWER POOL





PETER MEIBOM, TORBEN SVENDSEN and BENT SØRENSEN

Roskilde University, Institute 2, Energy and Environment Group

P.O. Box 260, DK-4000 Roskilde, Denmark

fax: +45 4674 3020, E-mail: meibom@ruc.dk, Web: http://mmf.ruc.dk/energy/









ABSTRACT



The official Danish energy plan "Energy 21" calls for a very high penetration of wind energy in the electricity sector. This will put issues of fitting a variable energy source into a stable supply system in focus. The present study investigates the role that collaboration with the hydro-based Scandinavian countries can offer, and particularly looks at the conditions of the Nordic power pool, asking whether it will be possible profitably to trade wind power in such a pool system.





KEYWORDS



Wind power, hydro power, renewable energy systems, electricity markets.





INTRODUCTION



The background for the current work was the need to update previous work on combination of large-scale wind power with seasonal reservoir-based hydro power, as it exist in Scandinavia (Sørensen, 1981). The conditions have changed in one important respect, as the electric power production in several Scandinavian countries is no longer undertaken by publicly

controlled companies, but has been privatized in the expectation of creating a market-like situation with several competing producers serving the power demanding community.



The Nordic Power Pool started operating in Norway 1993, Sweden joined 1996, and Denmark and Finland has been accorded certain rights of trading. The Power Pool consists of a diurnal market, where contracts that imply physical deliverance of power are entered, and a seasonal option' market allowing actors to buy security for supply or sale over longer periods. The diurnal market is a spot market, where actors auction desired amounts for sale (stating minimum price) or desired purchases (stating maximum price) during each hour of a 24h period, starting at midnight following the closing time at 12h noon for bidding. The auctioneer Nordpool ASA then matches supply and demand in a sequence of priorities giving the cheapest solution. Nordpool further operates an "adjustment market" for the system responsible company Statnett, and a similar "balancing service" is in Sweden operated by Swedish Kraftnät. This allows adjustments to be made in case a bidder cannot deliver or in case of unforeseen load change. (Nordpool, 1996; Andersen, 1996).



Our primary interest is whether or not it is feasible to sell wind power in the diurnal market, with consideration of possible penalties arising due to the price difference between the adjustment market and the diurnal market, in case of incorrect forecasts of the amounts of wind power that will be available for delivery over the 24h auction period.





MODEL STRUCTURE AND ASSUMPTIONS



We investigate this by performing an hour-by-hour simulation of a simplified but realistic model of the future Scandinavian electricity supply system (Fig. 1.), using actual time series of wind speeds (Jensen 1996) and water influx into reservoirs (Andersen 1997), together with established load variation patterns (Petersen 1996; Mørk 1997), and simulating the sale of wind power to the Nordic Power Pool by performing the daily noon bidding required by the present setup, while restricting the information available to the simulation program at each instant to the backwards part of the time series. We use a prediction model (Landberg et al.1996) to determine the amount of wind-based power to offer to the pool, and keep track of any deviation of the actual production

(known to the program during its subsequent time steps), allowing us to calculate the additional cost of trading



Fig. 1. Structure of Scandinavian energy system model. Connections are labeled with the transmission capacity in MW, taken from (Nordel 1996) or in case of transmission between internal regions estimated. Dashed connections are decided but not yet implemented.



at the pool (i.e. the cost of delivering less power, or having to sell more power, than originally promised). For information about the costs assumptions see (Meibom et al. 1997).



We investigate a scenario regarding the system structure, modeling the actual plan for the future (Danish) energy system contained in the government plan for year 2030 (Danish Department of Environment & Energy, 1996).

The capacity situation in year 2030 for Norway and Sweden is taken from their current published energy plans, although they are essentially business-as-usual plans (Hansen et al., 1996; Nutek). We add the assumption that CPH will reach a level in Southern Sweden similar to that in Denmark.



The electricity production and demand in the scenario are shown for all the Scandinavian countries in Fig. 2.



Fig. 2. Summary of simulation results for the 2030 scenario based on published government energy plans. Production, use and transmissions are given in TWh/y. The transmission values state the energy flows out of the transmission lines. The energy flows into the lines are slightly bigger because of transmission losses.





MODEL RESULTS



The results of our simulation are summarized in Fig. 2.. There is never shortage of power in any of the regions,



and the Danish wind power surplus is mainly used to displace thermal power in Sweden. The hydro reservoir degree of filling is slightly lower at the end of the year as compared with the beginning, where it was above average, and continued simulation during a second year would shift some production from hydro to the Swedish thermal plants, according to the algorithm used. Increasing the transmission capacity already installed could avoid losing the 0.5 TWh of power not utilized in Denmark, but with the cost assumptions made, this expansion of the international exchange lines is not warranted.



The trade changes moderately from hour to hour, except for a few percent of the hours of the year, and large adjustment trade takes place in less than 5% of the hours of the year. The annual extra costs for Denmark due to adjustments are 316 M DKK, to which comes the value of power not sold or used, 77 M DKK. The average penalty for using the intermittent wind production is thus 23 DKK per MWh of wind power produced, or 12% of the average Pool trading price.





CONCLUSIONS AND ACKNOWLEDGMENTS



Trading Danish wind power at the Nordic Pool allows production surpluses and deficits to be fully taken care of, which at the high wind power penetrations assumed in the scenarios would have required additional system components (such as energy storage or backup allowing regeneration of electricity) and new regulation procedures to be added, had the solution been a strictly national one with no power exchange. The cost of using the Nordic Power Pool in this way is considered to be substantially lower than that of any storage or backup-based solution.



The work presented here has been financially supported by the Danish Energy Agency under its grant # 51191/96-0023. Support, data provisions and discussions with the following are gratefully acknowledged: Bjørn Godske, Torben Olsen and Charlotte Søndergreen (Elkraft Power Utility). Sigurd Lauge Petersen and Jan Bünger (Danish Energy Agency), Jan Andersen (Nordpool), Trygve Gotfred Borg and E Mørk (Statnett), Lars Landberg, H. Parbo and Lars Henrik Nielsen (Risø National Laboratory), and Søren Krohn (Danish Wind Industry Association).





REFERENCES



Andersen, J., 1996. De organiserte kraftmarkeder i Nord Pool. Nordpool internal report Jan. 8, 14 pp. Lysaker.

Andersen, J., 1997. Private communication of data from Nordpool ASA.

Danish Department of Environment & Energy, 1996. Energy 21: The Danish Government's Action Plan for Energy, 1996, 79 pp. Copenhagen.

Hansen, M., Johnson, T. and Oftedal, J., 1996. Det norske kraftmarket til år 2020: Nasjonale og regionale

fremskrivninger. Statistics Norway, Report 16, 39 pp. Oslo.

Jensen, G., 1996. Personal communication of wind speed data collected by Risø National Laboratory in collaboration with wind turbine operating utilities.

Landberg, L., Hansen, M., Vesterager, K. and Bergstrøm, W., 1996. Implementing wind forecasting at a utility (using HIRLAM and Wasp models). Report prepared by Risø, DMI, Elkraft and Nesa for the Danish Energy Agency Energy Research Program, EFP-1363/94-0005.

Meibom, P., Svendsen, T. and Sørensen, B., 1997. Import/export-politik som redskab til optimeret udnyttelse af el produceret på VE-anlæg. Final report for project performed for the Danish Energy Agency, Texts from IMFUFA, No. 343, 84 pp. Roskilde University.

Mørk, E., 1997. Communication of hourly values of the Statnett 1995 power usage.

Nordel, 1996. Annual Report 1995. 58 pp., SK Energy, Hellerup.

Nordpool, 1996. Døgnmarkedet/Ukemarkedet, Reports dated May 1, 21+13 pp., Lysaker.

Nutek, 1996. Swedish electricity market 1995. 25 pp. Swedish National Board for Industrial and Technical Development, Stockholm.

Petersen, S.L., 1996. Data from the energy model RAMSES of the Danish Energy Agency.

Sørensen, B., 1981. A combined wind and hydro power system. Energy Policy, March, pp. 51-55.