一种基于链形混合拓扑的输电线路温度异常检测方法

A Method for Detecting Abnormal Temperature of Transmission Line Based on Chain Hybrid Topology

我国输电线路存在异常检测数据准确性和及时性较低,无线环境恶劣,数据在时空难关联等问题,因此建设一个高效、安全、准确的输电线路异常检测模型迫在眉睫。提出一种基于链形混合拓扑的异常检测方法,将传感器采集到的数据送至基站进行单源和多源多维数据异常检测。该方法首先设计了一种基于时间维度的单源数据异常检测算法(single-source data anomaly detection algorithm,SDADA),对检测时间内的数据进行依次遍历,确定有效和异常数据的个数,然后对异常检测结果进行综合分析。其次,设计了一种在基站端执行的多源多维数据异常检测算法(multi-source and multi-dimensional data anomaly detection algorithm,MDADA),在SDADA的基础上,通过位置相关性定义了不同传感器之间的距离关系,用于确定候选异常检测队列,并对特定时间的异常数据值进行综合分析。实验结果表明,与传统方案相比,该方法具有更高的检测精度和执行效率。

For China’s transmission lines, there are problems such as low accuracy and timeliness of anomaly detection data, harsh wireless environment and data correlation in time and space. It is urgent to build an efficient, safe and accurate anomaly detection model for transmission lines. Therefore, this paper proposes a anomaly detection method for the transmission lines based on a chain hybrid topology which sends the data collected by sensors to the base station for single-source and multi-source multi-dimensional data abnormality detection. This method firstly designs a single-source data anomaly detection algorithm(SDADA) based on time dimension, which traverses the data in detection time sequentially, determines the numbers of valid and abnormal data, and then comprehensively analyzes the anomaly detection results. Secondly, it designs a multi-source and multi-dimensional data anomaly detection algorithm(MDADA) implemented at the base station. On the basis of the SDADA, the distance relationship between different sensors is defined according to position correlation, which is used to determine the candidate anomaly detection queue, and the abnormal data values at specific time are comprehensively analyzed. The experimental results show that compared with the traditional scheme, this method has higher detection accuracy and execution efficiency.