In this work, a high-throughput computation (HTC) and machine learning (ML) combined method was applied to identify the determining factors of the detonation velocity ($v_{\mathrm{d}}$) and detonation pressure ($p_{\mathrm{d}}$) of energetic molecules and screen potential high-energy molecules with acceptable stability in a high-throughput way. The HTC was performed based on 1725 sample molecules abstracted from a dataset of over 10⁶ linear nitroaliphatics with 1- to 6-membered C backbones and three types of substituents, namely single nitro group (-NO₂), nitroamine (-NNO₂), and nitrate ester (-ONO₂). ML models were established based on the HTC results to screen high-energy molecules and to identify the determining factors of $v_{\mathrm{d}}$ and $p_{\mathrm{d}}$. Compared with quantum chemistry calculation results, the absolute relative errors of $v_{\mathrm{d}}$ and $p_{\mathrm{d}}$ obtained using the ML models were less than 3.63% and 5%, respectively. Furthermore, eight molecules with high energy and acceptable stability were selected as potential candidates. This study shows the high efficiency of the combination of HTC and ML in high-throughput screening.

在这项工作中，采用高通量计算（HTC）和机器学习（ML）相结合的方法来确定爆速的决定因素（$v_{\mathrm{d}}$) 和爆轰压力 ($p_{\mathrm{d}}$) 的高能分子，并以高通量的方式筛选具有可接受稳定性的潜在高能分子。^{HTC 是基于从超过 10 6 个}具有 1 至 6 元 C 主链和三种取代基的线性硝基脂肪族数据集中提取的 1725 个样本分子进行的，即单硝基 (-NO ₂ )、硝基胺 (-NNO ₂ ) , 和硝酸酯 (-ONO ₂ )。基于 HTC 结果建立 ML 模型，筛选高能分子，识别决定因素$v_{\mathrm{d}}$和$p_{\mathrm{d}}$. 与量子化学计算结果相比，绝对相对误差$v_{\mathrm{d}}$和$p_{\mathrm{d}}$使用 ML 模型获得的误差分别小于 3.63% 和 5%。此外，八个具有高能量和可接受稳定性的分子被选为潜在候选者。这项研究表明了 HTC 和 ML 的结合在高通量筛选中的高效性。