Package Layout

This document will describe the layout of SDynPy to help users find the functions they are looking for.

SDynPy Subpackages¶

SDynPy is split up into several subpackages to help organize its content. While these subpackages are convenient to organize the content, they are not convenient to use, as typical function calls might require descending multiple package levels, resulting in inconveniently long code. Therefore, SDynPy heavily relies on aliases to commonly used functions and classes.

The following sections describe the major subpackages that exist in SDynPy.

SDynPy Core Objects¶

The sdynpy.core subpackage contains Python modules that define the core SDynPy objects and other core functionality. These are designed to represent common data used in structural dynamics, including degrees of freedom, geometry, data, or dynamic systems. Because of their ubiquitous nature, almost all of the modules, classes, and their corresponding helper functions are aliased to the top-level SDynPy namespace. The core modules are:

In addition to the modules being aliased to the SDynPy namespace, many of the classes and functions from the core subpackage are also aliased to the top-level namespace, as these are the most often called functions and classes in SDynPy.

File Input and Output¶

Structural dynamics data often comes from external sources, whether it is a modal, vibration, or shock test or an equivalent finite element simulation. Therefore, being able to quickly and easily bring external data into SDynPy is a priority. While users could strip data from their external files and manually construct SDynPy objects from that data, this risks bookkeeping and other translation errors. Therefore if a file type is commonly read into or written from SDynPy, it is useful to add a dedicated reader or writer into SDynPy to handle this translation correctly. The sdynpy.fileio subpackage contains much of the code to handle these conversions. For convenience, the modules are aliased to the top-level namespace. The fileio modules are:

Finite Elements and Similar Numerical Functionality¶

SDynPy has a limited set of finite element and other numerical functionality in the sdynpy.fem subpackage. This includes simple beam finite elements, electrodynamic shaker models, and sensor optimization routines to select optimal sensors for a test from finite element results. For convenience, the modules are aliased to the top-level namespace. The fem modules are:

Because the Exodus file format is used often at Sandia National Laboratories where SDynPy was originally developed, key classes from the sdynpy_exodus module are also aliased to the top-level namespace.

Modal Analysis¶

SDynPy has capabilities for performing experimental modal analysis which entails fitting modes to frequency response functions measured on the test article. These and similar capabilities exist in the sdynpy.modal subpackage. Many of the modules in this subpackage provide both code-based and GUI-based tools to fit modes to data. In the case of the modal subpackage, much of the useful content boils down to a handful of classes. Therefore, instead of aliasing the modules in the subpackage to the top-level namespace, SDynPy aliases these classes to the top-level namespace.

Signal Processing¶

The sdynpy.signal_processing subpackage includes a wide variety of tools that are related to structural dynamics. Often, the code in this package is at a “lower level” than the rest of the SDynPy code, operating on raw data rather than SDynPy objects. Many times, SDynPy objects will often call these lower-level functions as part of their own operations. signal_processing functions can be faster in that they remove the bookkeeping overhead that a lot of SDynPy objects implement, but the user must then be aware that their data must be correctly sorted prior to using the techniques. Additionally, some of the content in signal_processing does not fit well within the existing SDynPy core objects, so the subpackage serves as a kind of catch-all for tools that can be used in structural dynamics, but perhaps are not commonly used. The modules of signal_processing are generally aliased to the top-level namespace.

Documentation¶

SDynPy has limited functionality to create documentation for test or analysis results. This functionality is currently a work-in-progress, but capability exists to dump common formats into PowerPoint or LaTeX documents.

Demonstration Objects¶

Sometimes when a user has a new idea that they would like to try out, they simply want to quickly create a somewhat “interesting” system (e.g. something more than a simple beam or spring-mass-damper system), and generate data to explore. The sdynpy.demo subpackage provides two demonstration objects: a flat plate and a simple airplane-like model made out of beam elements. Note that there is some overhead in creating these demonstration objects, therefore the sdynpy.demo subpackage is not imported automatically with SDynPy, it must be imported separately using import sdynpy.demo. Because these are imported separately, there is no aliasing of the modules to the top-level namespace.