pystra.calibration.Calibration#

class Calibration(loadcombobj, target_beta, dict_nom_vals, calib_var, calib_method='optimize', est_method='matrix', print_output=False)[source]#

Bases: object

Class for calibrating partial and combination factors.

The factors are: \(\\phi\), \(\\gamma\), and \(\\psi\) factors for a given load combination instance and target reliability.

betaT#

Target reliability index

Type:: float

calib_method#

The calibration algorithm to use

Type:: str

cvar#

The label of the calibration variable.

Type:: str

df_nom#

A dataframe of nominal values

Type:: DataFrame

df_Xstar#

A dataframe of design point values

Type:: DataFrame

df_phi#

A dataframe of partial factors for resistances

Type:: DataFrame

df_gamma#

A dataframe of partial factors for loads

Type:: DataFrame

df_psi#

A dataframe of load combination factors

Type:: DataFrame

dict_nom#

Dictionary of nominal values

Type:: dict

est_method#

The estimation method

Type:: str

label_comb_vrs#

Labels of combination variables

Type:: str

label_comb_cases#

Labels of combination load variables

Type:: str

label_R#

Labels of resistance variables

Type:: str

label_other#

Labels of other load variables

Type:: str

label_all#

Labels of all variables including design parameter

Type:: str

loadCombObj#

LoadCombination object

Type:: LoadCombination

print_output#

Whether or not to print output to the console

Type:: bool

Initialize class instance.

Parameters:

loadcombobj (Class object) – Class LoadCombination object.
target_beta (Float) – Target reliability index for calibration.
dict_nom_vals (Dictionary) – Dictionary of nominal values.
calib_var (String) – Label of calibration variable in the LSF.
calib_method (String, optional) – Calibration method for the analysis: “optimize” or “alpha”. The default is “optimize”.
est_method (String, optional) – Estimation method for the factors: “matrix” or “coeff”. The default is “matrix”.
print_output (Boolean, optional) – Boolean flag for printing analysis output. The default is False.

Return type:

None.

Methods

`calc_Xst_nom`	Calculate the design point DataFrame divided by the nominal values per load case and adjust for \(\psi\) factors for combination load variables.
`calc_beta_design_param`	Calculate reliabilities based on given design parameter for resistance for all load combination cases.
`calc_design_param_Xst`	Calculate design parameter for resistance from design points.
`calc_df_pgRS`	Calculate the DataFrame of all resistance and load variables nominal values multiplied by their respective factors, \(\phi\), \(\gamma\), and \(\psi\), for all load cases.
`calc_epgS_mat`	Get LHS for matrix estimation method, \(\gamma_j~S_j\).
`calc_gamma`	Calculate Load factors \(\gamma\) from a dataframe of design points factored by the nominal values.
`calc_lsf_eval_df`	Calculate the LSF evaluation of a Dataframe elements.
`calc_pg_coeff`	Calculate \(\phi\), \(\gamma\), and \(\psi\) for the given set of design points and nominals using comparison of design pt coefficients approach.
`calc_pg_matrix`	Calculate \(\phi\), \(\gamma\), and \(\psi\) for the given set of design points and nominals using the Matrix approach.
`calc_phi`	Calculate resistance factors \(\phi\) from a dataframe of design points factored by the nominal values.
`calc_phiRz_egS_vect`	Get RHS for matrix estimation method, \(\phi~R~z-\gamma_g~G-\gamma_i~S_i\)
`get_design_param_factor`	Estimate the resistance design parameter for a given set of safety and combination factors, and nominals.
`get_phi_min`
`get_psi_max`	Get \(\psi\) dataframe corresponding to maximum estimates of dfpsi.
`print_detailed_output`	Print detailed outputs for Pystra Calibration, including calculations from intermediate steps.
`run`	Run calibration analysis to estimate \(\phi\), \(\gamma\), and \(\psi\) factors, and set class dataframe attributes corresponding to each factor.

calc_lsf_eval_df(df, ret_abs=True)[source]#

Calculate the LSF evaluation of a Dataframe elements.

Pass each dataframe value to the LSF and get corresponding LSF evaluation.

The columns of the dataframe must be a subset of LSF arguments.

Parameters:

df (Dataframe) – Dataframe for element-wise LSF evaluation. len(df.index) can be greater-than or equal to 1
ret_abs (Bool, optional) – Return abs of LSF evaluation. The default is True.

Returns:

df_lsf – Dataframe containing element-wise LSF evaluation of passed df.

Return type:

Dataframe

calc_design_param_Xst(dfXst)[source]#

Calculate design parameter for resistance from design points.

Parameters:: dfXst (Dataframe) – Dataframe containing design points. Note: len(dfXst.index) = 1
Returns:: z – design parameter for resistance corresponding to the design pt.
Return type:: Float

run(est_method=None, set_max=False)[source]#

Run calibration analysis to estimate \(\phi\), \(\gamma\), and \(\psi\) factors, and set class dataframe attributes corresponding to each factor.

Parameters:: est_method (String, optional) – Calibration method override. The default is “matrix”.
Return type:: None.

get_psi_max(dfpsi)[source]#

Get \(\psi\) dataframe corresponding to maximum estimates of dfpsi.

Parameters:: dfpsi (DataFrame) – Dataframe of \(\psi\) per load case.
Returns:: df_psi_max – Dataframe of \(\psi\) corresponding to maximum of each load effect.
Return type:: DataFrame

calc_pg_coeff(dfXst, print_output=False)[source]#

Calculate \(\phi\), \(\gamma\), and \(\psi\) for the given set of design points and nominals using comparison of design pt coefficients approach.

Parameters:

dfXst (Dataframe) – Dataframe containing all design points at target reliability.
print_output (Boolean, optional) – print_output flag for displaying intermediate and final output of function. The default is False.

Returns:

df_phi (Dataframe) – Dataframe containing \(\phi\) estimates for resistance variables per load case.
df_gamma (Dataframe) – Dataframe containing \(\gamma\) estimates for all static and combination load variables per load case.
df_psi (Dataframe) – Dataframe containing \(\psi\) estimates for all static and combination load variables per load case.

calc_Xst_nom(dfXstar)[source]#

Calculate the design point DataFrame divided by the nominal values per load case and adjust for \(\psi\) factors for combination load variables.

Parameters:: dfXstar (DataFrame) – DataFrame containing all design points at target reliability for all load cases.
Returns:: df_Xst_nom – Design point DataFrame factored by the nominal values per load case.
Return type:: DataFrame

calc_phi(dfXstnom)[source]#

Calculate resistance factors \(\phi\) from a dataframe of design points factored by the nominal values.

Parameters:

dfXstnom (DataFrame) – Design point DataFrame factored by the nominal values per load case.
set_max (Boolean, optional) – Set \(\phi\) estimates to their corresponding max values. The default is False.

Returns:

df_phi – Resistance factors \(\phi\) for resistance variables per load case.

Return type:

DataFrame

calc_gamma(dfXstnom)[source]#

Calculate Load factors \(\gamma\) from a dataframe of design points factored by the nominal values.

Parameters:

dfXstnom (DataFrame) – Design point DataFrame factored by the nominal values per load case.

Returns:

dfgamma_static (DataFrame) – Load factors for static variables per load case.
dfgamma_comb (DataFrame) – Load factors for combination variables per load case.

calc_pg_matrix(dfXst, print_output=False)[source]#

Calculate \(\phi\), \(\gamma\), and \(\psi\) for the given set of design points and nominals using the Matrix approach.

Parameters:

dfXst (Dataframe) – Dataframe containing all design points at target reliability.
print_output (Boolean, optional) – print_output flag for displaying intermediate and final output of function. The default is False.

Returns:

df_phi (Dataframe) – Dataframe containing \(\phi\) estimates for resistance variables per load case.
df_gamma (Dataframe) – Dataframe containing \(\gamma\) estimates for all static and combination load variables per load case.
df_psi (Dataframe) – Dataframe containing \(\psi\) estimates for all static and combination load variables per load case.

calc_phiRz_egS_vect(dfXstar)[source]#

Get RHS for matrix estimation method, \(\phi~R~z-\gamma_g~G-\gamma_i~S_i\)

Parameters:: dfXstar (Dataframe) – Dataframe containing all design points at target reliability.
Returns:: phiRz_egS_vect – RHS for matrix estimation method as 1D Array.
Return type:: Array

calc_epgS_mat(dfgammanom)[source]#

Get LHS for matrix estimation method, \(\gamma_j~S_j\). The LHS is evaluated by evaluating the LSF with appropriate random variables to account for any constant multipliers. The implementation works for both, linear and non-linear LSFs. For more algorithmic details, ref to Appendix A, Caprani and Khan, Structural Safety, 2023.

Parameters:: dfgammanom (Dataframe) – Dataframe containing product of nominal values and safety factors, along with calibrated z values.
Returns:: epgS_mat – LHS for matrix estimation method as 2D Array.
Return type:: Array

calc_beta_design_param(design_z)[source]#

Calculate reliabilities based on given design parameter for resistance for all load combination cases.

Parameters:: design_z (Float) – design parameter for resistance
Returns:: arr_beta – Array containing reliability indices corresponding to design_z for each load combination case.
Return type:: Array

calc_df_pgRS(min_phi, max_psi)[source]#

Calculate the DataFrame of all resistance and load variables nominal values multiplied by their respective factors, \(\phi\), \(\gamma\), and \(\psi\), for all load cases.

Returns:: df_pgRS
Return type:: DataFrame

get_design_param_factor(min_phi=True, max_psi=True)[source]#

Estimate the resistance design parameter for a given set of safety and combination factors, and nominals.

Returns:: array_z – Array containing design parameters for all load combination cases.
Return type:: Array

print_detailed_output(precision=2)[source]#

Print detailed outputs for Pystra Calibration, including calculations from intermediate steps.

Parameters:: precision (float, optional) – Decimal precision for roundingo off output. The default is 2.
Return type:: None.