随着智能变电站新型技术的不断更新，整个智能变电站数据信息实现全站共享，站域后备保护技术将传统继电保护的后备保护由单一电气元件的后备保护延伸到整个变电站的后备保护，为提出新型站域后备保护算法提供了技术支撑。但是目前的站域后备保护算法存在数据传输负担及数据计算压力大的问题；一旦数据信息缺失或者错误，可能导致单一故障识别算法难以正确识别故障。据此，本文提出了智能变电站站域后备保护算法。 为了减轻站域通信网络的数据传输负担和数据计算的压力，提出能量相对熵的站域后备保护故障区域识别算法。利用被保护区域综合动作电流与综合制动电流之间的能量相对熵识别故障区域，若被保护区域的能量相对熵小于阈值，则区域内部故障；若被保护区域的能量相对熵大于阈值，则被保护区域外部故障或者整个系统正常运行。利用PSCAD搭建典型的220kV智能变电站以及相邻电网模型，对本文提出的能量相对熵的站域后备保护故障区域识别算法进行仿真验证，结果表明该算法能够准确识别各种被保护区域的运行状态。 为了进一步识别变电站站内保护区内部的故障元件，提出能量相对熵的站域后备保护故障元件识别算法。利用被保护电气元件综合动作电流与综合制动电流之间的能量相对熵识别故障元件，若被保护元件的能量相对熵接小于阈值，则元件发生短路故障；若被保护元件的能量相对熵大于阈值，则被保护元件正常运行。通过不同短路故障条件下的仿真，验证了能量相对熵站域后备保护故障元件识别算法的正确性与有效性。 为了提高站域后备保护故障元件识别算法的容错性，提出模糊传递闭包站域后备保护故障元件识别算法。首先，选取元件的能量相对熵、元件的故障测量度、元件的方向信息三种优势互补的故障元件识别方法；其次，通过选取的统计指标与被保护元件构建模糊传递闭包矩阵，根据模糊传递闭包矩阵实现故障元件识别；最后，通过不同短路故障条件下的仿真，结果表明该保护算法在信息完整或者缺失的条件下均可以正确地识别故障元件，体现了该保护算法的容错性。

With the continuous updating of new technologies of intelligent substation, the whole intelligent substation data information is shared by the whole station. The station backup protection technology extends the backup protection of traditional relay protection from the backup protection of a single electrical component to the backup protection of the entire substation. A new type of site backup protection algorithm is proposed to provide technical support. However, the current site backup protection algorithm has the problem of data transmission burden and data calculation pressure; once the data information is missing or incorrect, it may be difficult for a single fault identification algorithm to correctly identify the fault. Based on this, this paper proposes a backup protection algorithm for intelligent substation station domain. In order to alleviate the data transmission burden and data compression pressure of the station domain communication network, a station domain backup protection fault area identification algorithm based on energy relative entropy is proposed. The fault area is identified by the energy relative entropy between the integrated action current of the protected area and the integrated braking current. If the relative entropy of the energy of the protected area is less than the threshold, the internal fault of the area; if the relative entropy of the energy of the protected area is greater than the threshold, then The external area of the protected area is faulty or the entire system is operating normally. By using PSCAD to build a typical 220kV intelligent substation and adjacent grid model, the energy-entropy-based station backup protection fault region identification algorithm proposed in this paper is simulated and verified. The results show that the algorithm can accurately identify the operating states of various protected areas. In order to further identify the faulty components inside the protection zone in the substation, a fault identification algorithm for the station backup protection component with energy relative entropy is proposed. The fault element is identified by the energy relative entropy between the integrated operating current of the protected electrical component and the integrated braking current. If the relative entropy of the energy of the protected component is less than the threshold, the component has a short circuit fault; if the relative entropy of the energy of the protected component is greater than The threshold is then operated by the protected component. The correctness and effectiveness of the energy-dependent entropy station domain backup protection fault component identification algorithm are verified by simulation under different short-circuit fault conditions. In order to improve the fault tolerance of the fault identification component identification algorithm in the station domain backup protection, a fuzzy protection closure station domain backup protection fault component identification algorithm is proposed. Firstly, the fault component identification method with the complementary advantages of the energy relative entropy of the component, the fault measurement degree of the component, and the direction information of the component is selected. Secondly, the fuzzy transmission closure matrix is constructed by the selected statistical index and the protected component, according to the fuzzy transmission. The closure matrix realizes the fault component identification. Finally, through the simulation under different short-circuit fault conditions, the results show that the protection algorithm can correctly identify the faulty component under the condition of complete or missing information, which reflects the fault tolerance of the protection algorithm.