基于SSA-SVM的继电保护装置状态评估方法研究

运维检修人员对保护装置检修时需制定检修计划。为了对保护装置准确评估和精准检修，提出了一种基于麻雀搜索算法优化SVM的继电保护装置状态评估方法。根据建立的继电保护装置指标体系对保护装置健康状态进行分类，实现继电保护装置的状态评估。首先，建立继电保护装置状态评价指标体系，根据指标体系采集保护装置的相关数据。其次，建立SSA-SVM继电保护装置状态评估模型，采用该模型将保护装置健康状态按正常、注意、异常和严重进行分类。最后，分别采用SSA-SVM、BA-SVM、GWO-SVM、WOA-SVM算法评估继电保护装置的健康状态，并比较各算法评估保护装置健康状态的准确性和实时性。仿真结果表明，SSA-SVM算法评估保护装置健康状态的准确性最高，保护装置状态评估结果可为运维检修人员提供检修依据。

A state evaluation method for a relay protection device based on SSA-SVM

Operation and maintenance personnel formulate a maintenance plan when maintaining a protection device. To accurately evaluate and repair a protection device, a state evaluation method of relay protection device based on the sparrow search algorithm optimized SVM is proposed. Using the established relay protection device index system, the health state of the device is classified to realize its state evaluation. First, this paper establishes the state evaluation index system of relay protection devices, and collects relevant data on the protection device according to that system. Secondly, the SSA-SVM relay protection device state evaluation model is established. This is used to classify the health state of the protection device in terms of normal, attention, abnormality and severity. Finally, SSA-SVM, BA-SVM, GWO-SVM and WOA-SVM algorithms are used to evaluate the health status of relay protection devices, and the accuracy and real-time of each algorithm is compared. The simulation results show that the SSA-SVM algorithm has the highest accuracy in evaluating the health state of the protection device, and the state evaluation results can provide a basis for operation and maintenance personnel. This work is supported by the Science and Technology Project of State Grid Corporation of China (No. 5212002000AR).