Skip to main content
To KTH's start page To KTH's start page

Dynamic Thermal Rating for Improved Utilization of Wind Farm Export Systems

A Methodology for Improving Load Profile Estimation of Wind Farm Export Transformers

Time: Mon 2023-05-08 16.00

Location: Sten Velander, Teknikringen 33, Stockholm

Language: English

Subject area: Electrical Engineering

Doctoral student: Zhongtian Li , Elektroteknik

Opponent: Assistant Professor Nguyen Nga, University of Wyoming, Laramie, WY, USA

Supervisor: Associate professor Patrik Hilber, Elektroteknisk teori och konstruktion; Tor Laneryd, Elektroteknik; Professor Stefan Ivanell, Uppsala University

Export to calendar

QC 20230414


The power system components connected to renewable energy sources, such as transformers, are often oversized and conservatively loaded. The design of transformers normally ignores the intermittent nature of the connected renewable energy sources (e.g. solar, wind). Due to the variations in weather conditions and operation states, the transformer load oscillates and the actual hot spot temperature is significantly lower than the designed thermal rating.

For wind farms, the oversized transformer causes extra resourcematerial waste and a higher wind power price. Dynamic thermal rating can be applied to determine the rating of the transformers based on real-time environmental conditions (e.g. ambient temperature, wind speed). However, in order to optimize the operation of the transformers with dynamic thermal rating, the prediction of the load profile of transformers is an obstacle.

The load of wind farm export transformers oscillates due to thechange of load conditions (e.g. turbine availability, power curtailment) and environmental conditions (e.g. wind speed, wind direction and ambient temperature). This thesis proposes a new methodology to improve the utilization of wind farm export transformers by estimating their load profile more accurately and assessing their aging rate. The estimation of the load profile takes the wake effect and turbine availability into account. Specifically, the variation in the wind turbine failure and repair rates, which is influenced by the wind, is considered in the evaluation of turbine availability. Additionally, a correction method is proposed to improve the accuracy of the wake loss computation.

The results demonstrate that the estimation accuracy of the transformer load profile is improved after considering the influence of the wake effect and turbine availability. The wake effect and the turbine availability reduce the generated wind power and to some extent, reduce the load and the aging rate of transformers. However, the wake effect has limited influence when the wind farm reaches peak power production while turbine availability influences the load profile of transformers especially when the load is close to the installed capacity of the wind farm. After considering these two factors, the prediction accuracy of the hot spot temperature in the transformers can be enhanced and dynamic thermal rating can be applied to transformers with improved reliability.