Risk factors in wind energy investments and the effect of wind measurement on estimating energy production

Due to increase in global energy need day by day, limited resources of fossil fuels and their negative effects on environment, tendency to use alternative energy sources is increasing. At this point, renewable energy sources including solar energy, wind energy, biomass energy, have became crucial. Furthermore, energy policies of the countries has been shaped to increase the installed capacities in this field. When annual growth rate of the renewable energy is analyzed in 2000's, it is obviously seen that the number and the amount of the wind energy investments have been rising. This requires wind energy investments to be realized through detailed analyses and right strategies. The major aim of this study is to evaluate risk factors in wind energy investments and to investigate the effect of wind measurement data on estimating wind energy production. In the first chapter of the thesis, the goal and the scope of the study are defined. In the following chapter, the development and the current status of the wind energy in world and in Turkey are presented through the statistical information. In the third chapter, the phases of wind energy projects are explained. Wind projects consist of three phases including project development, construction and operation. The process of developing lasts 4-5 years. The construction period is minimum 1 year. The operation of onshore wind farms continues during 20 years. Since winds are less turbulent over the sea, technical lifetime of offshore wind farms is 25 years. After reviewing projects phases in detail, the investment and operating costs of wind plants are investigated in this section. The cost structure of wind projects are more capital intensive compared to fossil energy plants. 75% of the total cost can be composed of equipments. When it comes to the operational phase, it could be said with certainty that the operation and maintenance costs of offshore wind plants are more than the onshore ones. Project risks are defined as uncertainities that are caused by unexpected conditions and have an adverse impact on project parameters such as cost, time and quality. In the fourth chapter of the thesis, the risk factors in wind energy investmens are examined within the scope of wind measurement data, wind resource assessment programmes, site geology, environmental impacts and social acceptance, natural disasters, regulations and equipment procurement. For instance, wind measurement is used to estimate 20-25 years production of the wind farms. This requires that wind speed measurement has to be performed properly to see the exact wind conditions of the potential site. On the other hand, softwares, which are used for wind resource assessment, represent a risk for the feasibility studies as well. WAsP, WindPro, WindSim, WindFarmer, Windographer, Homer are the main softwares which are used in the wind industry and it can be stated that behaviour of the models may differ depending on terrain complexity. Enviromental impacts and social acceptance should definitely be taken into account for wind investments. Such factors can even end up with cancellation of the projects. Another risk factor which needs to be evaluated during project development is natural disasters including earthquake, tsunami, hurricanes, lightning and icing. Long periods of downtimes, which cause a loss of revenue due to these factors, may occur. In the worst case, it is probable to face with major damages on vulnerable equipments of the wind plants. Furthermore, uncertainties pertainining to site geology, regulations and equipment procurement should be taken into consideration. It is significant to follow effective risk management strategies for all specified risks in order to reduce their negative effects for the wind investments. After reviewing literature, a case study is followed to investigate the effect of wind measurement data on estimating wind energy production. An actively operating wind plant has been used for this study. The wind farm consists of 16 wind turbines, a nominal power of 2.5 MW. Total installed capacity of the plant is 40 MW. Since the annual production data has been provided only for 8 turbines, this study has been carried out with 20 MW section of the plant. The average altidude of the project site is approximately 1600 m. 18-month of measurement from 60 m mast was available. Data have been collected between December, 2009 and May, 2011 with 10-min intervals and it covers 6 seasonal periods including 2 winter, 2 spring, 1 summer and 1 autumn. Highest wind speeds have been detected in winter periods. WAsP, which is a wind simulation software based on linear flow model, has been used to estimate power production. In order to provide input for WAsP analyses, six different wind data have been derived by changing the measurement duration length and intervals. For this study, three different scenarios have been created based on the measurement duration length by dividing these data into three periods, namely the first 6-month, the first 12-month and the whole 18-month. By using two different intervals, 10-min and 60-min, for each three periods, six different measurement data have been generated in total. WAsP has been run with six different meteorological data and energy production predictions have been obtained. These values have been compared with the actual energy production of the wind farm. Annual energy produced by 8 turbines is 52881 MWh in 2014. When production data are analyzed, it was seen that there is 3822 MWh energy loss during 2014 due to downtimes. This annual loss value, which had been calculated by using nacelle anemometer measurements and turbine power curve, has been added to actual output (52881 MWh) in order to see the effect of measurement data on energy predictions clearly. Therefore, 56703 MWh has been taken as a final value to make following comparisons. On the basis of annual energy production, it has been found that optimum results are obtained from the cases using the data of 18-month period with relative errors of using the data of first 12-month period have resulted with 12.0% (10-min int.) – 12.8% (60-min int.) relative errors. Since the data of first 6-month period cover only 2 seasons, WAsP predictions are quite different compared to the actual production of the wind farm. When the wind data, which have same period but different averaging intervals, are compared to each other, it is seen that 10-minute interval data have resulted better. On the basis of annual wind energy estimation, the difference between two different intervals is at the rate of 1.2%. It can be stated that the averaging interval does not have a considerable impact on the results of this study. Different results can be obtained in other studies conducted for different regions. Payback period has been calculated as 8.8 years by using plant's energy outputs. It has been found that optimum result has been obtained from the case using 18-month period with 10-min intervals, as payback period of 8.2 year. The case using first 6-month period with 60-min intervals has a payback period of 6.9 year and showed largest deviation. It is crucial to state that in wind energy projects, working with long-term measurement data collected with short intervals will considerably reduce the risks in terms of accurateness of the calculations of energy production estimations.