The edge frequencies of each blade of the WT (i.e. the frequencies of the first bending mode in the blade edgewise direction) are estimated from the acceleration signals.

Good morning everyone,

I can’t find the edge frequency of the wind turbine blades.

The available information are:

The edge frequency data of one blade is synthesised using a non-zero mean normal distribution.

The edge frequency data of the second blade is set to be equal to that of the first blade with some additional noise. Finally, the third blade initially has the same edge frequencies of the first blade, but ‘deteriorates’ over time by means of a negative linear gradient. That is,

Fa=N(µa,σ2)

Fb= Fa +N(µa,σ2)

Fc = Fa +dc+ N(µa,σ2)

Dc={0,mx+c}

1-In this work, 360000 data points were generated.

2-Only the first half of the data points of Blade C is considered ‘healthy’.

3-The length of the vector of zeros in equation (29) is therefore 180000.

The rest of the vector dC, which is indexed by the data point number, x (i.e. 180001 to 360000) has a gradient, m, of -1*10 -7 and a y-intercept value, c of 0.018.Whilst not realistic, µa is set to 10 and σ2has a value of 0.01.