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.