 |
PIPER
1.0.1
|
|
PIPER
1.0.1
|
A dedicated XML-based language is used to define the rules needed to parse the Finite Element Model file format. The approach can be used with multiple formats and codes and has been successfully tested against LS-dyna and Pamcrash models. It was selected because it allows the user to implement additional rules for other codes or numerical features without recompiling the software. This is an alternative to a full templates (the only required rules are those needed to build the PIPER model for a given transformation).
Format rule file : in this file (xml based with extension .pfr), general information and rules to parse the finite element format are defined. In the release, rule format files are provided for LS dyna (for both fixed sized and comma separated format) and Pamcrash. The files (with a pfr extension) are stored in the Piper/bin directory. In principle, it is possible to define Format rules file for other finite element formats and use them in the PIPER application as long as:
This section describes how to define rules to add support for additional keywords in the provided files, or to develop new files for another Finite Element format.
Note: this rule file is not specific to a finite element Human Body Model. If a rule file already exists for the finite element format used by your HBM, it is likely not needed to modify it even if you develop new metadata.
The format rule file define rules to parse keywords for a specific Finite Element format. The xml file is organized under the root element "format_description" as follows:
Description of the general structure of the parser file (Text between <– and –> is comment):
The two first lines are mandatory. The first one is requested for the XML format and the second one allows to check the syntax in the parser file.
The formatInformation element has an attribute used to declare a name describing the format (mandatory) and is composed of:
Example of formatInformation element for Pamcrash. In this format, two different characters can be placed at the beginning of comment line.
Example of formatInformation element for LS-Dyna using comma as separator.
The meshComponent element defines relation between keywords of the FE format and a PIPER component. For each keyword defined in this element, a "rule" element must be defined.
Components defined in this element are:
In addition, the element ordering has to be specified in case the one in the considered FE format differs from the one chosen in Piper project (VTK definition adopted VTK documentation (http://www.vtk.org/wp-content/uploads/2015/04/file-formats.pdf)).
elementOrdering element can be used for this purpose. It describes the sequence ordering used to convert FE element definition to the Piper one for the following types:
Example of meshComponent element:
Example of meshComponent element for Pamcrash © format:
Example of meshComponent element for LS-DYNA © format:
The "rule" element is used to define all operations to be performed when a previously registered keyword is found. For each keyword registered in "meshComponent" element, parsing actions are defined and parsed data are stored in variables (Variable manipulation). These variables are used to create and manipulate PIPER components of the model with PIPER Component operation.
The element "variableAssign" is used to assign the result of an operation to the variable identified by name. The element "variable" defines variable as an input of another element.
Usage of "variableAssign" and "variable" elements:
The "append" element allows to add characters defined in value to a variable of type character.
Example of "append" usage:
the "clearVar" element clear variable(s).
Example of "clearVar" usage:
This elements go to the next line of the current line (without considering comment line according to comment character defined in`"formatInformation" element).
This element defines how to (1) parse the current line or variable of type character and (2) assign the result to a variable.
Options of the "parse" element:
format: Type of data to be parsed. It can be C (characters), F (double), UI (integer), UIpos (strictly positive integer).offset: Offset in digit if no separator is defined either in parse element or “formatinformation” element. Number of separator if a separator is defined either in parse element or “formatinformation” element.length: Number of digit to parse. It must be zero if a separator is definedseparator: Defines a local separator character specific to this parse element. Default is no local separator defined. If length equals 0 and a global separator is defined in "formatInformation" element, the global separator is used for parsing if no local separator is defined.repeat: repeat the parse operation considering the offsetrepeat. The result of parsing is accumulated in the variable as a vector.offsetrepeat: Offset in digit between repeated parse operations in case no separator is defined neither in parse element nor in "formatInformation" element . Default value is 0 if length is different from 0 and 1 if length equals 0. separator is defined either in parse element or "formatInformation" element element.
Example 1: Parse in fixed length format.
String to parse (blanks are replaced by dot for visibility and parsed values are in bold)
…….1.5……0.06……..-2………0
As a result, the value 0.06 is assigned to the variable var1 type F.
Example 2: Parse with comma separator.
String to parse (parsed values are in bold)
1.5,0.06,-2,0
Example 3: Parse in fixed length format with repetition.
String to parse (blanks are replaced by dot for visiblity and parsed values are in bold)
.7020553……..0.546867……-93.736504………14.2672
The result is stored in var1 and var1 is a vector of 3 double numbers (0.546867,-93.736504,14.2672).
Example 4: Same as example 3 with comma separator.
String to parse (parsed values are in bold)
7020553,0.546867,-93.736504,14.2672
The result is stored in var1 and var1 is a vector of 3 double numbers (0.546867,-93.736504,14.2672).
This element is used to describe "parse"" element to be performed on the current line or variable of type character.
Example of a set of two parse actions on the current line:
line to parse (blanks are replaced by dot for visiblity)
.7020553……..0.546867……-93.736504………14.2672
As a result, value 7020553 is assigned to variable var1 and var2 is a vector with two elements (-93.736504,14.2672).
Example of a set of two parse actions on variable varstr (this variable must be a variable of type C):
Content of variable varstr to parse (blanks are replaced by dot for visiblity)
.string_1……..string_2……-93.736504………14.2672
As a result, string_2 is assigned to variable var1 and var2 is a vector with two elements (-93.736504,14.2672).
The “if” statement is used for conditional execution:
The “while” statement is used to repeat execution as long as the condition is true:
The “forEach” statement is used to iterate over elements of a vector variable:
The following set of operations return a Boolean depending on the result of the evaluation.
This operation returns true if the set of characters defined in “value” is found either in the variable (variable must be a variable of type C) or in the current line being parsed. The option pos is used to specify if “value” can be at any position (pos="any") in the variable or only at the beginning (pos=”start”).
This operation returns true if the set of characters defined in “value” is not found either in the variable nor in the current line being parsed. Syntax and options are the same as for the “find” operation.
This operation returns true if the set of characters "str" defined in value is equal to the current line being parsed or to the variable "varname" of type character.
Utilities operations correspond to operations that can help to manipulate data stored in variables.
the "generateId" element generates an id comprised between a start and an end value stored in variables of type UI and store the result in a variable (vector of UI).
This section is related to the operations that can be performed on a PIPER component. Each PIPER component created during the parsing of files can be identified by its Id (or name) and the keyword defined in keyword defined in the "rule" element element used to create it. For each PIPER component, methods are available to create it and assign to it parsed data stored in variables.
The "objectModelFile" element defines a FE model file to be parsed. It includes only one methode element: "setFile".
The "objectNode" element adds a node in the Piper model. It includes two methods:
Example of "objectNode" element usage: line that defined node in Pam-Crash format (blanks are replaced by dot for visiblity) NODE../..7020551.......-0.869206......-92.601196.........9.03719
"objectElement1D", "objectElement2D" and "objectElement3D" elements respectively add one-dimensional, two-dimensional and three-dimensional elements in the Piper model. It includes two methods:
One-dimensional element:
Two-dimensional element:
Three-dimensional element:
This element defines the group in PIPER model and adds content to it. type: defines type of group content (Node, Element3D, Element2D, Element1D, Group)
Example of group of three-dimensional elements:
In this example, Id stored in variable "varnameforelementId" corresponds to Id associated with one of the keywords associated in "meshComponent" element with Piper component declared in type of the "objectGroup" element (Element3D in the example). If Id added is associated with another keyword(s), these keywords can be specified in the "sourceId" element as in the following example. In the example, Id of groups are added in a Group of group PIPER component and Id of group(s) to be added are associated with keywords defined in the "sourceId" element.
Example of group of group using sourceId:
The "objectFrame" element adds a frame definition in the PIPER model. Methods of the "objectFrame" element are:
Example of "objectFrame" element usage:
The "objectJoint" element adds a joint definition in the PIPER model. Revolute and Spherical joints are the allowed type of joint. Methods of the "objectJoint" element are:
Example of "objectJoint" element usage (REVOLUTE joint):
The "objectParameter" element adds a parameter definition in the Piper model. A parameter allows to associate a parameter name with a value parsed in the file. Methods of the "objectParameter" element are:
Example of "objectParameter" element usage:
1.8.10