As an environmental-benign fuel, methane (CH₄) has received considerable interest for developing high-capacity energy storage systems. Herein, we aim to rapidly discover covalent–organic frameworks (COFs) for ultrahigh CH₄ storage among 530,000+ COFs, including one experimental (Curated) and two hypothetical (Berkeley and Genomic) databases. First, the feature space of all the three COF databases is projected by t-Distributed Stochastic Neighbor Embedding (t-SNE) technique, which reveals a potential but unexplored regime in Genomic COFs. Subsequently, an active learning (AL) approach is developed by integrating parallel acquisition with molecular simulation to efficiently explore Genomic COFs. The parallel AL model demonstrates remarkable screening efficiency and shortlists top COFs by evaluating only 50 out of 445,845 Genomic COFs. A record-breaking Genomic COF is identified with CH₄ deliverable capacity of 222.2 v/v, surpassing the current world record (208.0 v/v from experiment and 217.9 v/v from simulation). Our AL approach is significantly faster than brute-force simulation and conventional machine learning, it would accelerate the discovery of advanced porous materials for broad applications.

中文翻译：

---

主动学习促进了用于超高 CH4 存储的共价有机框架的计算发现

作为一种对环境有益的燃料，甲烷（CH ₄）在开发高容量储能系统方面受到了极大的关注。在此，我们的目标是快速发现用于超高 CH _{4的共价有机框架 (COF)}存储在 530,000 多个 COF 中，包括一个实验性（策划）和两个假设性（伯克利和基因组）数据库。首先，所有三个 COF 数据库的特征空间都通过 t 分布随机邻域嵌入 (t-SNE) 技术进行投影，这揭示了基因组 COF 中潜在但尚未探索的机制。随后，通过将并行采集与分子模拟相结合，开发了一种主动学习 (AL) 方法，以有效地探索基因组 COF。并行 AL 模型通过仅评估 445,845 个基因组 COF 中的 50 个，展示了卓越的筛选效率并入围了顶级 COF。_{用 CH 4}鉴定出破纪录的基因组 COF可交付容量为 222.2 v/v，超过了当前的世界纪录（实验结果为 208.0 v/v，模拟结果为 217.9 v/v）。我们的 AL 方法比蛮力模拟和传统机器学习要快得多，它将加速先进多孔材料的发现，以用于广泛的应用。