基于时空分布的基站异常能耗分析方法

Analysis Method for Abnormal Energy Consumption of Base Stations Based on Spatiotemporal Distribution

文章提出一种基于时空特征的基站异常能耗分析方法。该方法利用基站主设备能耗数据,结合地理空间和时间属性,对能耗数据进行异常标签标识。通过使用支持向量机算法训练标识后的数据,构建异常能耗识别模型。该模型能够输入历史能耗数据,输出基站能耗状态,并统计不同场景的异常率。通过该模型可分析全网能耗数据是否超过标准,从而提高能耗分析精准度,并输出异常基站供维护人员处理。同时,制定了修正机制,以更新标准,适应网络变化。

This paper presents an analysis method of abnormal energy consumption of base stations based on temporal and spatial characteristics.This method uses the energy consumption data of base station main equipment,combined with geographical space and time attributes,to mark the energy consumption data with abnormal labels.By using the support vector machine algorithm to train the identified data,the abnormal energy consumption identification model is constructed.The model inputs historical energy consumption data,outputs the energy consumption status of base stations,and counts the abnormal rates of different scenarios.Through this model,we can analyze whether the energy consumption data of the whole network exceeds the standard,thus improving the accuracy of energy consumption analysis and outputting abnormal base stations for maintenance personnel to deal with.At the same time,a revision mechanism is established to update the standards and adapt to the changes in the network.