mdof.markov module

mdof.markov.okid(inputs, outputs, **options)

Identify Markov parameters, or discrete impulse response data, for a given set of input and output data. Observer Kalman Identification Algorithm (OKID) [1].

Parameters:

• inputs (array) – input time history. dimensions: \((q,nt)\), where \(q\) = number of inputs, and \(nt\) = number of timesteps

• outputs (array) – output response history. dimensions: \((p,nt)\), where \(p\) = number of outputs, and \(nt\) = number of timesteps

• m (float) – (optional) number of Markov parameters to compute, excluding timestep zero. default: \(\min(300, nt)\)

• rcond – (optional) cut-off ratio for small singular values in pseudoinverse computation. default: 1e-15

Returns:

the Markov parameters, with dimensions \((p,q,m+1)\)

Return type:

array

References

[1]

Juang, J. N., Phan, M., Horta, L. G., & Longman, R. W. (1993). Identification of observer/Kalman filter Markov parameters-Theory and experiments. Journal of Guidance, Control, and Dynamics, 16(2), 320-329. (https://doi.org/10.2514/3.21006)