Dataset and Source Code for A weighted multi-response optimization method of clamping position in shaft turning: A case study of parting-off

Project

Context & Methodology:</p><p dir="ltr">The dataset encompasses machining responses—specifically cutting time, energy consumption, and surface roughness—collected from parting-off experiments under flood cooling conditions. The materials include Q235, 45 Steel, 40Cr, and 42CrMo.</p><p dir="ltr">2. Included Files:</p><p dir="ltr">Supplementary_Table_S1.csv: The complete dataset including geometric parameters, material properties, and experimental results.</p><p dir="ltr">BPNN_groupkfold.py: A training script for the BP neural network utilizing a 5-fold GroupKFold cross-validation strategy.</p><p dir="ltr">MC-SA.py: The Membrane Computing Simulated Annealing algorithm script for parameter optimization.</p><p dir="ltr">WC-LHS.py: The script utilized for generating geometric parameters via Weighted-Constraint Latin Hypercube Sampling.</p><p dir="ltr">README.txt: A detailed data dictionary explaining all variables, units, and experimental conditions.</p><p dir="ltr">For detailed instructions on running the code and interpreting the dataset variables, please refer to the included README.txt file.</p>