基于深度学习的大质量黑洞双星并合前探测
Premerger detection of massive black hole binaries using deep learning
AI总结：
针对传统大质量黑洞双星引力波信号搜寻计算成本高、难以为电磁后续观测提供并合前预警的问题，本研究提出深度学习模型，可快速处理模拟数据并在并合前数小时至数天探测信号，为引力波-电磁协同观测提供可能。
摘要：
并合中的大质量黑洞双星（MBHBs）是空间引力波（GW）观测的主要源之一。这类双星的并合预计会产生可探测的电磁（EM）辐射，且时间窗口很窄。引力波信号的并合前探测对后续电磁观测至关重要。传统的引力波信号搜寻方法计算成本很高。本研究中，我们提出一种深度学习模型，用于搜寻来自MBHBs的引力波信号。我们的模型能在0.01秒内处理4.7天的模拟数据，并可在最终并合前数小时到数天探测到引力波信号。该模型为MBHBs的引力波与电磁协同探测提供了可能。
英文摘要：
Coalescing massive black hole binaries (MBHBs) are one of primary sources for space-based gravitational wave (GW) observations. The mergers of these binaries are expected to give rise to detectable electromagnetic (EM) emissions with a narrow time window. The premerger detection of GW signals is vital for follow-up EM observations. The conventional approach for searching GW signals involves high computational costs. In this study, we present a deep learning model to search for GW signals from MBHBs. Our model is able to process 4.7 days of simulated data within 0.01 seconds and detect GW signals several hours to days before the final merger. The model provides the possibility of the coincident GW and EM detection of MBHBs.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
