geolib_plus.shm package

Submodules

geolib_plus.shm.general_utils module

class geolib_plus.shm.general_utils.GeneralUtils

Bases: BaseModel

Contains general static functions.

class Config

Bases: object

arbitrary_types_allowed = True

static linearise_data_over_layer(data_to_linearized: array, depth: array, buffer_zone: int = 0)

Function that returns the linearised data of the layer inputted by the user.

geolib_plus.shm.nkt_utils module

class geolib_plus.shm.nkt_utils.NktMethod(value)

Bases: IntEnum

An enumeration.

REGRESSION = 1

STATISTICS = 2

class geolib_plus.shm.nkt_utils.NktUtils

Bases: BaseModel

This class contains utility functions which are used to determine Nkt and the characteristic value and probabilistic parameters. All methods are according to [11].

static get_characteristic_value_nkt_from_statistics(su: ndarray, q_net: ndarray, std_loc: Optional[float] = None) → float

Gets characteristic value of N_kt through statistics.

The characteristic value of N_kt is calculated by using the following formula:

$$N_{kt,kar}=exp({\mu }_{N_{kt}}+T_{n-1}^{0.05}\cdot {\sigma }_{Ln(N_{kt,gem})\cdot \sqrt{1+\frac{1}{n}}})$$

Parameters:

• su – iterable of undrained shear strength

• q_net – iterable of net cone resistance

• std_loc – local standard deviation of log N_kt, if None: std_loc = 0.5 std_total

Returns:

characteristic value of N_kt

static get_characteristic_value_nkt_from_weighted_regression(su: ndarray, q_net: ndarray, vc_loc: Optional[float] = None) → float

Gets characteristic value of nkt from weighted regression.

The characteristic value of N_kt is calculated by using the following formula:

$$N_{kt,kar}=\frac{{\mu }_{N_{kt}}}{(1-T_{n-1}^{0.05}\cdot V_{\frac{q_{net}}{N_{kt}},gem}\cdot \sqrt{1+\frac{1}{n}})}$$

Parameters:

• su – iterable of undrained shear strength

• q_net – iterable of net cone resistance

• vc_loc – local variation coefficient of q_net/N_kt, if None: vc_loc = 0.5 vc_total

Returns:

characteristic value of N_kt

static get_default_nkt(is_saturated: Optional[Union[ndarray, bool]])

Gets default Nkt values.

For saturated soil: mean Nkt is 20 and variation coefficient is 0.25 For unsaturated soil: mean Nkt is 60 and variation coefficient is 0.25

Parameters:

is_saturated – boolean or numpy array with booleans which indicate of the soil/soils is/are saturated

Returns:

mean of Nkt, std of Nkt

static get_nkt_from_statistics(su: ~numpy.ndarray, q_net: ~numpy.ndarray) -> (<class 'float'>, <class 'float'>)

Calculates log mean of N_kt and total log standard deviation of N_kt through statistics

Parameters:

• su – iterable of undrained shear strength

• q_net – iterable of net cone resistance

Returns:

mean of Ln Nkt, total standard deviation of Ln Nkt

static get_nkt_stats_from_weighted_regression(su: ~numpy.ndarray, q_net: ~numpy.ndarray) -> (<class 'float'>, <class 'float'>)

Gets Nkt statistics from weighted regression. With this method, the mean of Nkt and the variation coefficient of q_net/Nkt are found through weighted regression where the variation coefficient is minimised.

The function to be minimised is:

$$V_{\frac{q_{net}}{N_{kt}}.tot}=\sqrt{\frac{\sum {(\frac{s_{u,i}\cdot {\mu }_{N_{kt}}}{q_{net,i}}-1)}^2}{n-1}}$$

Parameters:

• su – iterable of undrained shear strength [kPa]

• q_net – iterable of net cone resistance [kPa]

Returns:

mean of Nkt, variation coefficient of q_net/Nkt

static get_prob_nkt_parameters_from_statistics(su: ~numpy.ndarray, q_net: ~numpy.ndarray, log_std_loc: ~typing.Optional[float] = None) -> (<class 'float'>, <class 'float'>)

Get Nkt parameters for probabilistic analysis through statistics.

Firstly, statistics are used to get the mean and standard deviation of the log Nkt values. Afterwards, the mean and standard deviation for probabilistic analysis are calculated, where the student T distribution is taken into account.

Parameters:

• su – iterable of undrained shear strength

• q_net – iterable of net cone resistance

• log_std_loc – local standard deviation of log N_kt, if None: std_loc = 0.5 std_total

Returns:

mean of N_kt and variation coefficient of q_net/N_kt for probabilistic analysis

static get_prob_nkt_parameters_from_weighted_regression(su: ~numpy.ndarray, q_net: ~numpy.ndarray, vc_loc: ~typing.Optional[float] = None) -> (<class 'float'>, <class 'float'>)

Get Nkt parameters for probabilistic analysis through weighted regression.

Firstly, weighted regression is used to get the mean of the Nkt values and the variation coefficient of q_net / N_kt. Afterwards, the variation coefficient is adjusted for the number of samples with the student T distribution

Parameters:

• su – iterable of undrained shear strength

• q_net – iterable of net cone resistance

• vc_loc – local variation coefficient of q_net/N_kt, if None: vc_loc = 0.5 vc_total

Returns:

mean of N_kt and variation coefficient of q_net/N_kt for probabilistic analysis

geolib_plus.shm.prob_utils module

class geolib_plus.shm.prob_utils.ProbUtils

Bases: BaseModel

Class contains probabilistic utilities for parameter determination following the methodology as described in [11].

class Config

Bases: object

arbitrary_types_allowed = True

static calculate_characteristic_value_from_dataset(data: ndarray, is_local: bool, is_low: bool, is_log_normal: bool = True, char_quantile: float = 0.05)

Calculates the characteristic value of the dataset. A normal distribution or a lognormal distribution can be assumed for the dataset. The student-t distribution is taken into account. And the spread reduction factor is taken into account.

The characteristic value is calculated as follows:

$$x_{kar}=exp({\mu }_{ln(x)}\pm T_{n-1}^{0.05}\cdot {\sigma }_{ln(x)}\cdot \sqrt{(1-a)+\frac{1}{n}}$$

where:

$x_{kar}$ is the characteristic value

$T_{n-1}^{0.05}$ is the student t factor at 5 $\mathrm{\%}$ quantile with n-1 degrees of freedom

$a$ is the spread reduction factor; 0.75 if data collection is regional, 1.0 if data collection is local

Parameters:

• data – dataset, X

• is_local – true if data collection is local, false if data collection is regional

• is_low – true if low characteristic value is to be calculated, false if high characteristic value desired

• is_log_normal – True if a log normal distribution is assumed, false for normal distribution

• char_quantile – Quantile which is considered for the characteristic value

Returns:

characteristic value of the dataset

static calculate_characteristic_value_from_prob_parameters(mean: float, std: float, n: int, char_quantile=0.05, a=0)

Calculates characteristic values from probabilistic parameters.

Parameters:

• mean – mean of x

• std – standard deviation of x

• n – number of data points

• char_quantile – quantile of characteristic value (default = 0.05)

• a – spread reduction factor (default = 0)

Returns:

characteristic value of X

static calculate_log_stats(data: ndarray)

Calculates mean and std of LN(X)

Parameters:

data – dataset, X

Returns:

mean and std of LN(X)

static calculate_normal_stats(data: ndarray)

Calculates mean and std of X

Parameters:

data – dataset, X

Returns:

mean and std of X

static calculate_prob_parameters_from_lognormal(data: ndarray, is_local: bool, quantile=0.05)

Calculates probabilistic parameters mu and sigma from a lognormal dataset, as required in D-stability. This function takes into account spread reduction factor and the student t factor.

Firstly the standard deviation of LN(X) is corrected for the number of test samples and type of samples (local or regional), secondly, the mean and standard deviation of X are calculated from the mean and corrected standard deviation of LN(X)

Parameters:

• data – dataset, X

• is_local – true if data collection is local, false if data collection is regional

• quantile – quantile where the student t factor should be calculated

static calculate_std_from_vc(mean: float, vc: float)

Calculates standard deviation from variation coefficient

Parameters:

• mean – mean of distribution

• vc – variation coefficient of distribution

static calculate_student_t_factor(ndof: int, quantile: float) → float

Gets student t factor from student t distribution at a specific quantile.

Parameters:

• ndof – number of degrees of freedom

• quantile – quantile where the student t factor should be calculated

Returns:

Student t factor

static calculate_vc_from_std(mean: float, std: float)

Calculates variation coefficient from standard deviation

Parameters:

• mean – mean of distribution

• std – standard deviation of distribution

static correct_std_with_student_t(ndof: int, quantile: float, std: float, a: float = 0) → float

Calculates corrected standard deviation at a quantile with the student-t factor. This includes an optional spread reduction factor. The corrected standard deviation is calculated as follows:

$${\sigma }_{ln(x).prob}\approx \frac{T_{n-1}^{0.05}}{u^{0.05}}\cdot {\sigma }_{ln(x)}\cdot \sqrt{(1-a)+\frac{1}{n}}$$

where:

$T_{n-1}^{0.05}$ is the student t factor at 5 $\mathrm{\%}$ quantile with n-1 degrees of freedom

$u^{0.05}$ is the value of the standard normal distribution at the 5 $\mathrm{\%}$ quantile

$a$ is the spread reduction factor; 0.75 if data collection is regional, 1.0 if data collection is local

Parameters:

• ndof – number of degrees of freedom

• quantile – quantile where the student t factor should be calculated

• std – standard deviation

• a – spread reduction factor, 0.75 for regional sample collection; 1.0 for local sample collection

Returns:

corrected standard deviation

static get_log_mean_std_from_normal(mean: float, std: float)

Calculates mean and standard deviation of LN(X) from the mean and std of X

Parameters:

• mean – mean of X

• std – std X

Returns:

mean and std of LN(X)

static get_mean_std_from_lognormal(log_mean: float, log_std: float)

Calculates normal mean and standard deviation from the mean and std of LN(X)

Parameters:

• log_mean – mean of LN(X)

• log_std – std of LN(X)

Returns:

mean and std of X

geolib_plus.shm.shansep_utils module

class geolib_plus.shm.shansep_utils.ShansepUtils

Bases: BaseModel

Class contains shansep utilities for parameter determination following the methodology as described in [11].

class Config

Bases: object

arbitrary_types_allowed = True

static calculate_characteristic_shansep_parameters_with_linear_regression(OCR: ~typing.Union[float, ~numpy.array], su: ~typing.Union[float, ~numpy.array], sigma_effective: ~typing.Union[float, ~numpy.array], S: ~typing.Optional[float] = None, m: ~typing.Optional[float] = None) -> (<class 'float'>, <class 'float'>)

Calculates characteristic values of parameters s and m, according to [11]. This methodology assumes a log normal distribution of S and a normal distribution of m Optionally with a given S or m.

Parameters:

• OCR – Union[float, np.array], over consolidation ratio [-]

• su – Undrained shear strength [kPa]

• sigma_effective – Effective stress [kPa]

• S – S value, shear strength ratio [-]

• m – m value, strength increase component [-]

Returns:

characteristic value of S and m

static get_shansep_prob_parameters_with_linear_regression(OCR: ~typing.Union[float, ~numpy.array], su: ~typing.Union[float, ~numpy.array], sigma_effective: ~typing.Union[float, ~numpy.array], S: ~typing.Optional[float] = None, m: ~typing.Optional[float] = None) -> ((<class 'float'>, <class 'float'>), (<class 'float'>, <class 'float'>), <class 'numpy.ndarray'>)

Determines shansep parameters s and m with linear regression according to [11]. This methodology assumes a log normal distribution of S and a normal distribution of m. Parameter S or m can also be given as an input.

Parameters:

• OCR – Union[float, np.array], over consolidation ratio [-]

• su – Undrained shear strength [kPa]

• sigma_effective – Effective stress [kPa]

• S – S value, shear strength ratio [-]

• m – m value, strength increase component [-]

Returns:

(mean and std of S), (mean and std of m), covariance matrix

geolib_plus.shm.shm_tables module

class geolib_plus.shm.shm_tables.ShmTables(*, soils: List = [])

Bases: BaseModel

“Schematiseringshandleiding macrostability” tables: 7.1, 7.2, 7.3, 7.4.

load_shm_tables(path_table: Path = PosixPath('resources'), filename: str = 'shm_tables.json') → None

Function that reads the shm tables json file.

Parameters:

• path_table – Path to the tables file

• filename – Name of the tables file

Returns:

Dictionary with the NEN data structure

soils: List

geolib_plus.shm.soil module

class geolib_plus.shm.soil.DistributionType(value)

Bases: IntEnum

An enumeration.

Deterministic = 4

LogNormal = 3

Normal = 2

Undefined = 0

class geolib_plus.shm.soil.Soil(*, name: ~typing.Optional[str] = None, unsaturated_weight: ~typing.Optional[~typing.Union[float, ~geolib_plus.shm.soil.StochasticParameter]] = StochasticParameter(is_probabilistic=False, mean=None, standard_deviation=0, distribution_type=<DistributionType.Normal: 2>, correlation_coefficient=None, low_characteristic_value=None, high_characteristic_value=None, low_design_value=None, high_design_value=None, limits=None), saturated_weight: ~typing.Optional[~typing.Union[float, ~geolib_plus.shm.soil.StochasticParameter]] = StochasticParameter(is_probabilistic=False, mean=None, standard_deviation=0, distribution_type=<DistributionType.Normal: 2>, correlation_coefficient=None, low_characteristic_value=None, high_characteristic_value=None, low_design_value=None, high_design_value=None, limits=None), shear_strength_ratio: ~typing.Optional[~typing.Union[float, ~geolib_plus.shm.soil.StochasticParameter]] = StochasticParameter(is_probabilistic=False, mean=None, standard_deviation=0, distribution_type=<DistributionType.Normal: 2>, correlation_coefficient=None, low_characteristic_value=None, high_characteristic_value=None, low_design_value=None, high_design_value=None, limits=None), strength_increase_exponent: ~typing.Optional[~typing.Union[float, ~geolib_plus.shm.soil.StochasticParameter]] = StochasticParameter(is_probabilistic=False, mean=None, standard_deviation=0, distribution_type=<DistributionType.Normal: 2>, correlation_coefficient=None, low_characteristic_value=None, high_characteristic_value=None, low_design_value=None, high_design_value=None, limits=None), cohesion: ~typing.Optional[~typing.Union[float, ~geolib_plus.shm.soil.StochasticParameter]] = StochasticParameter(is_probabilistic=False, mean=None, standard_deviation=0, distribution_type=<DistributionType.Normal: 2>, correlation_coefficient=None, low_characteristic_value=None, high_characteristic_value=None, low_design_value=None, high_design_value=None, limits=None), dilatancy_angle: ~typing.Optional[~typing.Union[float, ~geolib_plus.shm.soil.StochasticParameter]] = StochasticParameter(is_probabilistic=False, mean=None, standard_deviation=0, distribution_type=<DistributionType.Normal: 2>, correlation_coefficient=None, low_characteristic_value=None, high_characteristic_value=None, low_design_value=None, high_design_value=None, limits=None), friction_angle: ~typing.Optional[~typing.Union[float, ~geolib_plus.shm.soil.StochasticParameter]] = StochasticParameter(is_probabilistic=False, mean=None, standard_deviation=0, distribution_type=<DistributionType.Normal: 2>, correlation_coefficient=None, low_characteristic_value=None, high_characteristic_value=None, low_design_value=None, high_design_value=None, limits=None), pop_layer: ~typing.Optional[~typing.Union[float, ~geolib_plus.shm.soil.StochasticParameter]] = StochasticParameter(is_probabilistic=False, mean=None, standard_deviation=0, distribution_type=<DistributionType.Normal: 2>, correlation_coefficient=None, low_characteristic_value=None, high_characteristic_value=None, low_design_value=None, high_design_value=None, limits=None))

Bases: SoilBaseModel

Schematisation manual macrostability soil class

cohesion: Optional[Union[float, StochasticParameter]]

dilatancy_angle: Optional[Union[float, StochasticParameter]]

friction_angle: Optional[Union[float, StochasticParameter]]

name: Optional[str]

pop_layer: Optional[Union[float, StochasticParameter]]

saturated_weight: Optional[Union[float, StochasticParameter]]

shear_strength_ratio: Optional[Union[float, StochasticParameter]]

strength_increase_exponent: Optional[Union[float, StochasticParameter]]

static transfer_soil_dict_to_class(soil_dict, soil)

Transfers items from dictionary to soil class if the item is not None :param soil_dict: soil dictionary :param soil: geolib_plus soil class soil in model

Returns:

unsaturated_weight: Optional[Union[float, StochasticParameter]]

class geolib_plus.shm.soil.SoilBaseModel

Bases: BaseModel

class Config

Bases: object

extra = 'forbid'

classmethod fail_on_infinite(v, values, field)

class geolib_plus.shm.soil.StochasticParameter(*, is_probabilistic: bool = False, mean: Optional[float] = None, standard_deviation: Optional[float] = 0, distribution_type: Optional[DistributionType] = DistributionType.Normal, correlation_coefficient: Optional[float] = None, low_characteristic_value: Optional[float] = None, high_characteristic_value: Optional[float] = None, low_design_value: Optional[float] = None, high_design_value: Optional[float] = None, limits: Optional[List] = None)

Bases: SoilBaseModel

Stochastic parameters class

correlation_coefficient: Optional[float]

distribution_type: Optional[DistributionType]

high_characteristic_value: Optional[float]

high_design_value: Optional[float]

is_probabilistic: bool

limits: Optional[List]

low_characteristic_value: Optional[float]

low_design_value: Optional[float]

mean: Optional[float]

standard_deviation: Optional[float]

Module contents