Flat-panel composite curing distortion dataset: curated tabular data and representative deformed surfaces
收藏资源简介：
This record presents a curated dataset of process-induced distortion in flat laminated composite panels generated within the DIDEAROT project (DIgital DEsign strategies to certify and mAnufacture Robust cOmposite sTructures). The dataset has been produced using the Process Induced Distortion (PID) synthetic data generator, a high-fidelity numerical workflow based on Alya’s simulation framework. Each sample corresponds to a finite-element simulation of a flat rectangular composite laminate subjected to a cure cycle, with the objective of capturing the distortion that emerges after tool release due to residual stresses developed during manufacturing. The simulations account for the coupled effects of thermoelastic residual stresses and chemical shrinkage, providing a physically grounded description of post-cure out-of-plane deformation. The released dataset includes design parameters, laminate descriptors, derived geometric quantities, scalar distortion metrics, and selected numerical and runtime metadata. It is intended to support exploratory data analysis, benchmarking, surrogate modelling, and machine learning workflows focused on composite manufacturing and structural response prediction. The dataset is released through two complementary Zenodo records. One provides the raw simulation outputs, preserving the original case-level results generated by the PID workflow and intended for full traceability, detailed inspection, and reprocessing. The present record provides the curated tabular data and representative deformed surfaces, offering the user-facing, analysis-ready version of the dataset. This dataset is one of the synthetic data assets developed within the DIDEAROT framework to enable data-driven methodologies for robust composite design and certification. Additional details on the simulation framework, modelling assumptions, and workflow implementation can be found in Deliverable D3.7 of DIDEAROT and in the work entitled “A High-Fidelity HPC Workflow for Predicting Process-Induced Distortions in Composites Using Surrogate Models” [1]. [1] M. Teixidor-Vilarrasa, A. Quintanas-Corominas, A. Ortega, I. Zárate, E. Marquinez, I. Otero and G. Guillamet, A High-Fidelity HPC Workflow for Predicting Process-Induced Distortions in Composites Using Surrogate Models, Materiales Compuestos (Online first). URL https://www.scipedia.com/public/Vilarrasa_et_al_2025a This project has received funding from the European Union’s Horizon Europe Framework Programme under grant agreement No. 101056682 for the project “DIgital DEsign strategies to certify and mAnufacture Robust cOmposite sTructures (DIDEAROT)”. The contents of this publication are the sole responsibility of the participants and do not necessarily reflect the opinion of the European Union. Neither the European Union nor the granting authority can be held responsible for them.
本记录呈现了DIDEAROT项目（全称：用于认证与制造稳健复合结构的数字化设计策略，Digital Design strategies to certify and manufacture Robust composite sTructures）所生成的平面层压复合板材工艺诱导变形精选数据集。该数据集由工艺诱导变形（Process Induced Distortion, PID）合成数据生成器构建而成，这是一套基于Alya仿真框架的高保真数值工作流。每个样本对应一次对平面矩形复合层合板进行固化循环的有限元模拟，旨在捕捉脱模后因制造过程中产生的残余应力所引发的变形。 本次模拟考虑了热弹性残余应力与化学收缩的耦合效应，为固化后的面外变形提供了基于物理机制的描述。本次发布的数据集包含设计参数、层合板特征描述、派生几何量、标量变形指标，以及精选的数值与运行时元数据。本数据集旨在支撑面向复合材料制造与结构响应预测的探索性数据分析、基准测试、代理建模以及机器学习工作流。 本数据集通过两条互补的Zenodo记录发布。其中一条记录提供原始仿真输出，保留PID工作流生成的原始工况级结果，用于全溯源、详细检查与二次处理。本记录则提供经过整理的表格数据与代表性变形曲面，为用户提供可直接用于分析的数据集版本。本数据集是DIDEAROT框架下开发的合成数据资产之一，旨在支持采用数据驱动方法开展稳健复合结构设计与认证。有关仿真框架、建模假设与工作流实现的更多细节，可参阅DIDEAROT项目的可交付成果D3.7，以及题为《基于代理模型的复合材料工艺诱导变形预测高保真高性能计算工作流》的文献[1]。 [1] M. Teixidor-Vilarrasa、A. Quintanas-Corominas、A. Ortega、I. Zárate、E. Marquinez、I. Otero与G. Guillamet. 基于代理模型的复合材料工艺诱导变形预测高保真高性能计算工作流[J]. Materiales Compuestos（在线优先出版）. 链接：https://www.scipedia.com/public/Vilarrasa_et_al_2025a 本项目获得欧盟地平线欧洲框架计划（资助协议号101056682）资助，项目全称：用于认证与制造稳健复合结构的数字化设计策略（DIDEAROT）。本出版物的内容仅由参与方负责，未必代表欧盟的观点。欧盟与资助机构均不对本内容承担责任。
