Supporting data and code for Minimal identity-closed spectral clusters for operator-channel attribution in structural generalized eigenproblems

This dataset contains the supporting data and code for the development and validation of the minimal identity-closed cluster (MICC) framework for operator-channel attribution in structural generalized eigenproblems. It includes analytical and transfer benchmarks, robustness and sampling results, perforated shallow-shell identity and attribution evidence, the registered physical-flux negative control, and an independent 16-endpoint Abaqus verification chain. Model inputs, analysis scripts, extracted modal and thermal fields, processed CSV/JSON results, and SHA-256 integrity records are provided. Large reproducible solver databases, runtime logs, manuscripts, and machine-specific files are excluded.本数据集包含用于开发与验证结构广义特征值问题中算子通道归因的最小身份封闭簇（minimal identity-closed cluster, MICC）框架所需的配套数据与代码。其中涵盖解析基准与迁移基准、鲁棒性与采样实验结果、带孔薄壳相关的身份验证与归因佐证材料、已注册的物理通量阴性对照，以及独立的16终点Abaqus验证链。同时提供模型输入数据、分析脚本、提取得到的模态场与热场数据、经处理的CSV/JSON格式结果文件，以及SHA-256完整性校验记录。本数据集未包含大型可复现求解器数据库、运行时日志、研究手稿以及特定机器适配文件。