针对空天地一体化网络中卫星网络易受单粒子效应影响，导致链路中断和服务降级等问题，对面向单粒子效应的卫星网络弹性组网及重构方法进行了研究，旨在提升卫星网络在单粒子效应干扰下的抗毁性与服务可靠性。提出包含天基、空基、地面三层网络的协同弹性架构，及包含备份资源池和备份节点池的弹性资源池，并通过风险感知模块使卫星网络尽早发现故障节点。基于 Walker 网络架构与时间片机制，在 Dijkstra 算法基础上融入路径单粒子效应故障率与链路时延作为权重，提出最小时延及风险的路由规划算法。同时，将单粒子效应引发的故障分为 SEU 导致的 1 级可逆故障与 SEB 导致的 2 级不可逆故障，构建本地检测、邻居校验与地面确认相结合的故障分级检测机制，并提出局部路由修复算法与节点协同补位修复算法，形成分级弹性重构体系。基于铱星星座的STK和NS3联合仿真表明，针对故障节点的检测和重构的时间开销均在系统可接受范围。所提方法在网络出现故障节点后，能够及时对网络进行重构，并将数据包交付率由82.4%提升至97.1%。

To address the issues of satellite networks in integrated space-air-ground networks being susceptible to single-event effects (SEEs), which can lead to link disruptions and service degradation, this study investigates SEE-oriented re-silient networking and reconfiguration methods for satellite networks. The aim is to enhance the survivability and service reliability of satellite networks under SEEs. A collaborative resilient architecture comprising three layers of space-based, air-based, and ground-based networks is proposed, along with a resilient resource pool that includes a backup resource pool and a backup node pool. A risk awareness module is integrated to enable the satellite network to detect faulty nodes at an early stage. Based on the Walker network architecture and time-slice mechanism, a routing planning algorithm with the minimum delay and risk is proposed by incorporating the path SEE failure rate and link delay as weights into the Dijkstra algorithm. Meanwhile, faults caused by SEEs are classified into two levels: Level 1 reversible faults induced by single event upsets (SEUs) and Level 2 irreversible faults caused by single event burnouts (SEBs). A fault detection mechanism combining local detection, neighbor verification, and ground confirmation is established. Additionally, a local route repair algorithm and a node collaborative replacement repair algorithm are proposed, forming a hierarchical elastic reconfiguration system. STK-NS3 joint simulations based on the Iridium constellation show that the time overhead for detecting and reconfiguring faulty nodes is within the acceptable range of the system. After faulty nodes appear in the network, the proposed method can reconfigure the network in a timely manner, increasing the data packet delivery rate from 82.4% to 97.1%. The research results pro-vide key technical support for the stable operation of satellite networks in high-radiation environments.