The Dataset of Quantifying Alignment Deviations for the In-plane Biaxial Test System

For dataset in "alignmentDeviaitons_SP56_54976.csv": The dataset consists of 12 alignment deviations of an in-plane biaxial tensile testing machine as well as 56 strain measurement points on cruciform specimens. A deep learning model is trained on the dataset to quantify 12 alignment deviations using 56 strain values on a shape-optimized cruciform specimen. The design of experiments includes Optimal Latin Hypercube, numerical modelling of Finite Element Methods. Using the Optimal Latin Hypercube, 55000 distinct groups of DOE simulation tests are constructed. Under the boundary conditions of 12 distinct deviations, 56 strain values at the required location on the cruciform specimen are obtained using Python scripts. The illustration of “Code Scripts of Quantifying Alignment Deviations.rar”: To quantify the alignment deviations of the in-plane biaxial testing machine, the AutoML model built-in AutoGluon was used to map relationship between the 12 alignment deviations of the in-plane biaxial testing machine and 56 strain measurement locations on a shape-optimized cruciform specimen. The training data for the AutoML model was obtained by Optimal Latin Hypercube, an algorithm-designed DOE experimental strategy. DOE scheme designed 12 alignment deviations of the in-plane biaxial testing machine. Through the finite element simulation of each group of alignment deviation in the DOE design scheme, 54976 groups of 56 strain values for the cruciform specimen were obtained. AutogluonProgram_ SP56_ 54976 with_ Jupyter. ipynb: the operation procedure of developing the mapping relationship between alignment deviation and strain measurement locations using an automl model is documented in jupyerbook. datasetOutput_ to_ localDir. py: output 56 strain measurement points and related 12 centering deviations to the local data file. measurePointsSets_ in_ AbaqusModel. py: create a set of 56 measuring points for cruciform specimens in ABAQUS.