Overview of User-Defined Functions
The following table lists the available user-defined functions:
Table 5-17: User-Defined Functions.
Name
Global
Local
Description
Analytic
√
√
Define a function using an analytic expression and its arguments.
Deep Neural Network
√
√
Define a deep neural network with a data source, training method, and validation data.
Elevation (DEM)
√
√
Define a function by importing geospatial elevation data from digital elevation models (on the DEM file format using the USGS standard
External
√
 
Define a function by interfacing with other external functions written in the C language.
Function Switch
√
√
A function switch to switch between different functions.
Gaussian Process
√
√
Define a function that represents a Gaussian process for use with the Uncertainty Quantification Module.
Gaussian Pulse
√
√
Define a function that is a Gaussian pulse, the common bell-shaped curve (Gaussian function).
Image
√
√
Define a function by importing an image and map the image’s RGB data to a scalar function output value.
Interpolation
√
√
Create an interpolation function defined by a table or file containing the values of the function in discrete points.
Least-Squares Fit
√
√
Define a function with coefficients based on a least-square fit.
MATLAB
√
 
Define a MATLAB functions to interface to functions written in MATLAB®. This function requires a LiveLink™ for MATLAB® license.
Normal Distribution
√
√
Define a function that represents a normal distribution.
Partial Fraction Fit
√
√
Create a partial fraction approximation to a function defined by a result table or data from a file.
Piecewise
√
√
Define a function that is created by splicing together several functions, each defined on one interval.
Polynomial Chaos Expansion
√
√
Define a polynomial chaos expansion function. It is intended for use with the Uncertainty Quantification Module.
Ramp
√
√
Define a ramp function — a linear increase.
Random
√
√
Define a function that generates white noise with uniform or normal distribution.
Rectangle
√
√
Define a rectangle function.
Step
√
√
Define a step function — a sharp transition.
Triangle
√
√
Define a triangle function — a linear increase and linear decline.
Waveform
√
√
Define a waveform function: sawtooth, sine, or Gaussian-modulated sine, square, or triangle.
About Local parameters in Functions
Local parameters in functions are added as variables, which means they can be used in variable expressions or in postprocessing. You can access them using <function tag>.<parameter> (for example, an1.p1 or lsq1.a0). Local parameters can be found in Analytic, Piecewise, and Least-Squares Fit functions. Partial Fraction Fit also has local parameters whose names are predefined as Y_inf, R, Q, xi, and zeta.