pygmm.kishida_2017.calc_cond_mean_spectrum_vector

pygmm.kishida_2017.calc_cond_mean_spectrum_vector(periods: ~typing.Union[~typing.List[float], ~numpy.ndarray], ln_psas: ~typing.Union[~typing.List[float], ~numpy.ndarray], ln_stds: ~typing.Union[~typing.List[float], ~numpy.ndarray], ln_psas_cond: ~typing.Union[~typing.List[float], ~numpy.ndarray]) -> (<class 'numpy.ndarray'>, <class 'numpy.ndarray'>)[source]

Kishida (2017, []) conditional spectrum.

Conditional mean spectrum vector (CMSV) by Kishida (2017, []) is specifying the target spectral acceleration at multiple periods, rather than the single conditioning period by Cornell and Baker (2008). If this approach is used for a single period, then the resulting spectrum is the same as computed by Cornell and Baker (2008) – implemented by calc_cond_mean_spectrum().

Parameters

  • periods (array_like) – Spectral periods of the response spectrum [sec]. This array must be increasing.

  • ln_psas (array_like) – Natural logarithm of the spectral acceleration. Same length as periods.

  • ln_stds (array_like) – Logarithmic standard deviation of the spectral acceleration. Same length as periods.

  • ln_psas_cond (np.ma.masked_array) – The vector of conditioning spectral accelerations. This is a masked array with the same length as periods. Masked values are not used for defining the CMSV.

Returns

  • ln_psas_cmsv (np.ndarray) – Natural logarithm of the conditional mean spectral accelerations.

  • ln_stds_cmsv (np.ndarray) – Logarithmic standard deviation of the conditional mean spectral acceleration.

Raises

ValueError – If periods are monotonically increasing.