Wind Turbines for the Power Supply for Offshore Fish Farms : A Case Study for the Norwegian West Coast

Abstract

In this thesis the power consumption of a fish farm is set in relation to the expected power production from a wind turbine in order to assess the feasibility of this combination. The fish farm, called Rataren, is located in Sør-Trøndelag and hourly data on power consumption is available in the period 1st February 2016 until 31st August 2016, as a result of change in energy source from diesel generators to the onshore power grid. This is one of Norway’s largest fish farms and has 14 cages during the studied period. There are some minor data losses that are removed from the data sets. Wind speed, air and water temperatures at 2.5 masl and 1 mbsl are measured by a Seawatch Midi 185 Buoy at the location, and the data are supplied by SINTEF Ocean. In view of the estimation of the power output of a wind turbine the data on wind speed have to be extrapolated from measuring height to turbine hub height. For this the atmospheric stability, estimated here by the temperature difference between the air and the sea, is taken into account. The classifications used are unstable, slightly unstable, near-neutral, slightly stable and stable, and are characterized by limits of the temperature differences set to -3°C, -1°C, 1°C and 3°C. Two methods of extrapolating wind speed from measuring height to hub height are considered; the logarithmic law with corrections according to stability and the power law parametrized according to stability. For both methods the complete calculation of key variables requires information that is unavailable in this case of study. The problems faced when using the log law were the determination of stability parameter z/L, but also variation in friction velocity or roughness height. When using the log law in combination with different values for the variables based on literature, there were inconsistencies in the results indicated by unreasonable wind profiles. The more simplified method, the power law, is therefore concluded as the best approximation when this many parameters are unknown. The power law exponents that are used to correspond to each of the atmospheric conditions mentioned are set to 0.07, 0.09, 0.11, 0.13 and 0.15. Hourly mean wind speeds were then calculated using the power law with the exponent for the stability class found to occur at that particular hour. The results show an average wind speed of 7.04 m/s at 65 meters, compared to a measured average of 5.01 m/s at 2.5 meters height. The hourly power production at site by a Hywind Demo 2.3 MW turbine and a GWP 750 kW turbine was estimated. To analyse the sensitivity of the results to the relation of consumption and production, the hourly consumption is also multiplied by three and by six in two other cases, while being compared to the power production by the Hywind turbine. The production by a wind farm of three GWP turbines are estimated and compared to the other cases, which is similar in capacity to the one Hywind turbine. The 750 kW turbines are considered to be slightly more favourable, as the ratio of energy import to export to the onshore power grid is lower. In all, the results show that wind turbine is not suitable without energy storage or additional power sources. Hence, it is not possible to make a firm conclusion of which turbine is most suitable without further investigations.