Analysis to reduce ice-related production losses for wind turbines

Abstract: In the rapidly growing wind energy market, regions with cold climates are currently in the spotlight owing to their abundance of wind resources. However, the operation of wind turbines in cold climate conditions is challenged by serious icing problems. Ice accretion on the rotor blades of a wind turbine results in a decline in power production, an increase in fatigue loads, and raises health and safety concerns. To mitigate these adverse effects, ice protection systems (IPS) are now widely being employed. These systems mainly rely on costly blade heating techniques, yet their efficiency is limited and they cannot effectively prevent or remove ice build-up under all ambient conditions. In this study, the performance of five identical wind turbines, each equipped with an electrothermal heating IPS, is analyzed over several icing events. All data are collected from an undisclosed wind park located in northern Sweden. Historical wind turbine data is studied to explore the extent of icing-induced losses and IPS activities, as well as the dependence of blade icing and IPS efficiency on meteorological conditions. Based on the results from the analysis, suggestions will be provided on how the control settings of the IPS can be modified to increase the de-icing effectiveness and reduce ice-related production losses. For the purposes of better understanding the performance of the wind turbines and their IPS in icing conditions, an analytic dashboard has been internally developed. To derive quantitative information about the IPS efficiency, a set of standardized metrics is utilized. An internal algorithm has been developed that classifies various forms of ice losses and different status codes of wind turbines. These ice losses and turbine status codes are monitored and analyzed using the analytic dashboard. Statistical analysis indicates that the most substantial source of ice losses is the stoppages caused by blade icing, whereas losses during de-icing operations are relatively insignificant. Results from the IPS performance analysis show that the icing-induced losses are further influenced by the inconsistency in the IPS behavior. The systems have shown to be inefficient even when operating under conditions of wind speed and ambient temperature that fall within their specified operational limits, indicating their dependence on external conditions. In the majority of icing events, a delay in IPS activation was observed, particularly when these events coincided with periods of high wind speeds. Moreover, the heating of the blades is not sufficient, as multiple attempts to melt the accreted ice are often required, yet success is not always achieved. The difficulty in validating whether the blades are free of ice stems from the fact that the heat is emitted only from the blade’s leading edge. The author suggests specific immediate measures to improve the control of the IPS, including changing the threshold values for IPS triggering and adjusting the duration and frequency of ice removal cycles. These measures are confined by constraints tied to Intellectual Property Rights, limiting the extent to which elements in the IPS control settings can be modified by the wind operator. Nevertheless, if these constraints are relaxed, there exists significant untapped potential for further optimizing the control of IPS.