【目的】高比例新能源发电接入导致电网故障电流幅值降低、方向性改变，传统后备保护的离线整定难以适应环网复杂工况，且新能源低惯量与低电压穿越（low voltage ride through，LVRT）控制加剧了正/负序网络特性变化，故障元件辨识困难，引发保护失配或延时过长。文章旨在解决含新能源电网后备保护的动态适应性问题，突破环网死锁与定值僵化瓶颈。【方法】提出基于广域量测的双判据，针对不对称故障，利用负序电压/电流排序锁定故障关联母线及支路，通过区域集中式架构实现快速辨识；针对对称故障，依托变电站单台行波监测设备，结合全局行波到达时间差异与双端测距算法，实现微秒级故障定位。进一步优化后备保护逻辑，仅对故障线路关联远后备保护进行动态整定，调整阻抗圆范围，并将动作延时固化为2个时间级差，规避传统逐级配合的延时累积。【结果】PSCAD仿真结果表明，不对称故障下负序判据准确率达100%，过渡电阻30 Ω时仍能可靠辨识；对称故障行波测距误差小于100 m，定位时间较传统方法缩短90%；优化后远后备动作延时由4~7个时间级差降至2个，定值覆盖范围提升18.4%，有效避免负荷入侵误动。【结论】所提方法通过负序排序与行波时差判据互补，实现新能源电网多类型故障元件的快速动态辨识，突破环网死锁限制。动态整定策略使远后备保护动作延时大幅缩短，灵敏性与速动性显著提升，且无需依赖高采样设备或复杂通信架构，为高比例新能源电网后备保护在线整定提供了高效、可靠的工程化解决方案。

[Objective] The integration of a high proportion of renewable energy generation has reduced the amplitude of fault currents and changes in their directionality in power grids. Traditional backup protection that relies on offline settings struggles to adapt to the complex conditions of looped networks. Additionally， the low-inertia and low-voltage ride-through （LVRT） control of renewable energy sources exacerbates the changes in the characteristics of positive- and negative-sequence networks， making it difficult to identify faulty components and often resulting in protection mismatch or excessive delay. This study addresses the dynamic adaptability of backup protection in power grids with renewable energy， overcoming the bottlenecks of looped network deadlocks and rigid setting values. [Methods] A dual-criteria approach based on wide-area measurements is proposed. For asymmetrical faults， negative-sequence voltage/current ranking is used to identify fault-associated buses and branches， enabling rapid identification through a regional centralized architecture. For symmetrical faults， a single traveling wave monitoring device at the substation， combined with the global traveling wave arrival time difference and a double-ended ranging algorithm， is utilized to achieve microsecond-level fault location identification. The backup-protection logic is further optimized by dynamically setting only the remote backup protection associated with the fault line， adjusting the impedance circle range， and fixing the action delay to two time intervals， thereby avoiding the cumulative delays of traditional step-by-step coordination. [Results] The PSCAD simulation results indicate that the accuracy rate of the negative-sequence criterion for asymmetrical faults was 100%， with reliable identification possible with a transition resistance of 30 Ω. For symmetrical faults， the traveling wave ranging error is less than 100 m， and the location time is reduced by 90% compared with traditional methods. After optimization， the remote backup-action delay was reduced from 4-7 intervals to 2 intervals， while the setting coverage increased by 18.4%， effectively avoiding misoperations owing to load intrusion. [Conclusions] The proposed method achieved rapid and dynamic identification of multiple types of fault components in renewable energy grids through the complementary use of negative sequence ranking and traveling wave time-difference criteria， overcoming the limitations of looped network deadlocks. The dynamic setting strategy significantly shortens the action delay of remote backup protection， thus considerably enhancing sensitivity and speed. Moreover， this strategy does not rely on high-sampling equipment or complex communication architectures， thus providing an efficient and reliable engineering solution for online backup protection in power grids with a high proportion of renewable energy.