基于LSTM的高压输电电缆接头异常预警系统设计  

Design of abnormal early warning system for high voltage transmission cable joint based on LSTM

作　　者：贾俊青[1] 思勤[1] 段玮頔 邓凤婷[1] 董文娟[1] JIA Junqing;SI Qin;DUAN Weidi;DENG Fengting;DONG Wenjuan(Inner Mongolia Electric Power Research Institute,Hohhot 011200,China)

机构地区：[1]内蒙古电力科学研究院,内蒙古呼和浩特011200

出　　处：《电子设计工程》2024年第2期78-82,共5页Electronic Design Engineering

基　　金：内蒙古电力公司2022年科技项目(DK-ZXZB-2022-KKX0401-0338)。

摘　　要：高压输电电缆接头异常预警容易受到谐波电流影响,导致预警效果不佳,为此设计了基于LSTM的高压输电电缆接头异常预警系统。检测高压输电电缆接头异常电流谐波,获取电缆接头介质状态。设计CC2430微处理器预警模块,负责处理预警区域内电流数据。构建接地线电流检测模块,负责评估电缆接头异常。构建LSTM神经网络预警模型,对比分析实时状态与模型中检测点状态,获取当前电缆接头状态的预测值。采用移动平均技术分析电流数据的整体变化走向,判断电流预测数据与阈值关系,根据系统指示灯颜色区分异常等级。由实验结果可知,该系统电流数值最大值为125 A,变化曲线与实际曲线一致,说明该系统的预警效果较好。The abnormal early warning of high voltage transmission cable joints is easily affected by harmonic current,resulting in poor early warning effect.Therefore,an abnormal early warning system for high voltage transmission cable joints based on LSTM is designed.Detect abnormal current harmonics of high voltage transmission cable joints,and obtain the medium status of cable joints.Design the CC2430 microprocessor early warning module,which is responsible for processing the current data in the early warning area.Build a ground wire current detection module responsible for evaluating cable splice anomalies.Build an LSTM neural network early warning model,compare and analyze the real⁃time state and the state of the detection points in the model,and obtain the predicted value of the current cable joint state.The moving average technology is used to analyze the overall change trend of the current data,determine the relationship between the current prediction data and the threshold value,and distinguish the abnormal level according to the color of the system indicator light.It can be seen from the experimental results that the maximum current value of the system is 125 A,and the change curve is consistent with the actual curve,indicating that the system has a good early warning effect.

关 键 词：LSTM 输电电缆 接头异常 预警系统 

分 类 号：TN92[电子电信—通信与信息系统]