This thesis aims to analyze how different temperature deviations affect the components of wind turbines. A study of the temperatures in several different components was carried out to get a clearer picture of how these affects operational reliability. By trying to link, for example, extreme values of temperatures to possible alarms and downtime in the wind power plants, the hope is that this study will be able to contribute to safer and more maintenance-free wind turbines in the future. The 12 wind turbines that have been evaluated in this study are localized in southern Sweden. The time period for the collected data extends over 6 years.

In the thesis, an analysis has been carried out with the help of the computer program Matlab to examine different types of data collected from the components in the wind turbines. The data was plotted with the help of various functions in Matlab and with the help of mathematical equations. The data was then observed ocularly by the authors, which led to the discovery of different connections between temperature variations in the components and different error codes. The generator bearing was the component that was found to have the highest correlation between temperature deviations and specific error codes related to various yaw-errors, which can be explained as error codes that occur when the wind turbine fails to adjust itself to the wind.

An analysis has also been made to determine whether there was a temperature-related upward trend in any of the components in the wind turbines. This was carried out with the help of various standard deviation calculations and calculation of the mean temperature value of different components. The standard deviation in this case describes how much the temperature of each component varies from the mean value. It turned out that there was such an upward trend for the Generator Bearing and Generator Phase 3 components. Some turbines also turned out to have a more distinct upward trend than others