The following table shows the knowledge sources used
for the design of experiments.
| Name of the source code | Knowledge sources | Action taken | |
| create-space.lisp | 1.initiate | Clears blackboard database | |
| 2.create-space | 1. Defines blackboard spaces viz., element types, objective, parameters, orthogonal-arrays
2.Defines unit classes viz., defined variable, design variable, defined material, design material, obj-fn, type-name etc., | ||
| 3.make-dv-instances | Creates unit instance of defined parameter | ||
| 4.make-mp-instance | Creates unit instanc of defined material | ||
| 5.make-et-instances | Creates unit instance of element types | ||
| read-ansys.lisp | assign-val | Once the ansys file name is input, assign-val executes the following
1.remodify ansys file 2.initiate the balckboard database 3.create space 4.createsorthogonal array instances 5.creates design variables,material variables and element type instances | |
| remodify-ansys-file | reads the ansys input file and 1.retrieves
a. assigned variables b.materials defined c.element types 2.runs ansys to get initial values of variables
| ||
| read-macro | reads the material library and stores the properties | ||
| find-relation | for each assigned variables, identifies all the dependent variables and constants | ||
| variable-value | reads the inital values of all assigned variables from the output file of ANSYS | ||
| ortho-array.lisp | create-oa-instances | The orthogonal tables are stored in the slots. The following information is available for later retrievel:
1.array name 2.no of levels 3.maximum factors 4.level combination | |
| indep.lisp | find-i-variable | finds out whether the selected independent variable is a function of any other variables. | |
| gui2.lisp | gui | 1.Creates a main window and command buttons for :
a. reading ansys file b.selecting objective fn variable c.selecting independent variable d.starting taguchi experiments | |
| parm-menu | Depending on the command button clicked, the parm-menu directs the further action to be done | ||
| create-buttons | Creates list of assigned varible as a command button. When clicked, opens up a window for getting level values from the user. | ||
| scroll-obj-var | Creates a list of parameters that defines the objective function variables. | ||
| user-input | opens up a window for reading the level inputs and initiates the unit creation routine | ||
| create-level-value-
instances | After the user input of level value, unit instances are created and stored in database | ||
| write-obj-fn | Once the user inputs the objective function parameters or read from a file, the same is written as a post processing ANSYS file | ||
| ansys-variables.lisp | ansys-scroll-list | Depending on the element type, the program creates the list of post-processing variables to be used for objective function scrolling | |
| start-taguchi.lisp | start-taguchi-
experiments | 1. Deletes the independent design variables for which all the level values are same.
2.Selects the orthogonal array for the design of experiment 3.initiates the following operations a. writing ansys file b.calculating main effect b.confirmation run c.post processing of results | |
| write2.lisp | write-ansys-file | 1.For each experiment, writes out a new ansys file with the level values changed as per the orthogonal array
2. Runs ANSYS externally. 3. The output file of ANSYS stores the process parameters | |
| taguchi-gui.lisp | taguchi-gui | Plots the layout of the orthogonal array selected.
After experiment is over, the process parameters and optimum level combination also plotted | |
| post-process.lisp | calculate-main-effects | The following parameters are calculated :
1. Variable sum of square 2. Total sum of square 3.Percent contribution 4.Optimum level of each variable | |
| anova-table | calculates the F parameter | ||
| confirmation-run | Conducts a confirmation run to check whether the results are acceptable. | ||
| plot-levels | Plots the mean level values of each design variable | ||
| plot-%-contribution | Plots the percent contribution of each design variable | ||
| write-out-results | Writes out a text file that contains all the information about the experiment conducted |