Gross Error Identification of Reservoir and Dam Safety Monitoring Data Based on Outlier Characteristics

Based on safety monitoring results to evaluate the overall safety and stability of reservoirs and dams, meanwhile to analyze the reason of abnormal condition and take corresponding measures when necessary, all of those are important means to ensure reservoirs and dams safety from accidents. At present, reservoir and dam safety monitoring is developing towards online monitoring. In order to improve the accuracy and timeliness of analysis results, it is necessary to effectively eliminate the gross error from the prototype monitoring data. Based on the outlier characteristics of gross error, this paper processed a new “group” by using effect variable value and its measured interval, so that the corresponding relationship between the real gross error in the original data and the outlier value in the new “group” could be established, and then the gross error was identified in combination with outlier reasons. The case analysis verifies that this approach is capable of identifying gross error outstandingly.