Generation of long term wind speed histories
EWEC 2007, (2007)Abstract
In wind turbine fatigue load calculations, it is customary to consider
separate 10-minute periods, characterised by mean wind speed, mean
wind direction and turbulence intensity. This approach misses large
low frequency load cycles (with periods of a few hours or days) that
may influence total fatigue loading. To remedy this, a Markov chain
method is proposed that generates artificial long term wind and load
histories. Comparisons of generated and measured wind speed histories
are done. Using artificial wind speed histories, the expected amount
of starts and stops is established, and the fatigue damage equivalent
loads including large load cycles over the life of a standard turbine
operating in an IEC class II wind regime. It is shown that the influence
of low frequency load cycles on fatigue is significant.