湿地植被生态质量监测数据异常自适应辨识研究  

Identification of Anomaly Adaption in Wetland Vegetation Ecological Quality Monitoring Data

作　　者：姚雪梅 Yao Xuemei(Industry Development and Planning Institute,NFGA,Beijing 100010,China)

机构地区：[1]国家林业和草原局产业发展规划院,北京100010

出　　处：《环境科学与管理》2023年第2期167-172,共6页Environmental Science and Management

摘　　要：受到环境因素、设备因素的影响,生态监测数据中存在异常数据,影响辨识效果,为此提出监测数据异常自适应辨识方法。统一监测数据量纲,去除噪声数据,依据监测数据之间的相关系数,分层聚类处理监测数据,应用PKNN算法检测并剔除监测数据中的离群数据,依据历史监测数据探究其变化规律,引入自适应因子计算相邻监测数据线段斜率,制定异常数据辨识规则,执行规则,进行监测数据异常自适应辨识。实验数据显示:应用提出方法获得的异常数据比例数值与实际数值保持一致。Affected by environmental factors and equipment factors,there are abnormal data in the ecological monitoring data,which affects the identification effect.Therefore,an adaptive identification method of monitoring data anomaly is proposed.The monitoring data dimension is unified,the noise data is removed,the monitoring data is processed by hierarchical clustering based on the correlation coefficient between the monitoring data,the outlier data in the monitoring data is detected and removed by using PKNN algorithm,the change rule is explored based on the historical monitoring data,the adaptive factor is introduced to calculate the slope of the adjacent monitoring data segment,the abnormal data identification rules are formulated,and the rules are implemented,The monitoring data anomaly adaptive identification is carried out.The experimental data show that the ratio value of abnormal data obtained by the proposed method is consistent with the actual value.

关 键 词：生态质量 数据异常 监测数据 湿地植被 自适应辨识 

分 类 号：X835[环境科学与工程—环境工程]