ass日本风韵熟妇pics男人扒开女人屁屁桶到爽|扒开胸露出奶头亲吻视频|邻居少妇的诱惑|人人妻在线播放|日日摸夜夜摸狠狠摸婷婷|制服 丝袜 人妻|激情熟妇中文字幕|看黄色欧美特一级|日本av人妻系列|高潮对白av,丰满岳妇乱熟妇之荡,日本丰满熟妇乱又伦,日韩欧美一区二区三区在线

In order to judge the accuracy of sensor data in greenhouse environment measurement and control system, a sensor fault identification method was proposed based on the comparison of node information. This method based on the principal component analysis (PCA) was to achieve the sensor system fault detection through the monitoring statistics T2 and SPE changes. When the system detected the fault, the different sensor fault identification by using the comparison of node information based on temporal and spatial characteristics were realized, and to compare the effects with different methods, node information was made a comparison based on temporal scale, spatial scale and temporal-spatial scale, for multi-sensor fault identification. Verification results showed that the sensor fault detection method based on PCA can effectively realize the preliminary detection of the sensor system, and the sensor fault identification method based on the comparison of node information took the time and spatial scale into consideration, which can effectively achieve the specific fault sensor positioning. The value of the sensor nodes fault data average CDR was 98.37%, and the average FAR was 1.72%. Compared with the traditional method for sensor fault identification, the CDR increased by 22.067 percentage points and the FAR reduced by 15.762 percentage points, and it was found that the fault recognition method mentioned can effectively guarantee the efficiency of fault diagnosis improve the accuracy of fault diagnosis, and reduce the false alarm rate with reliability and accuracy.