Dam safety monitoring data anomaly detection method based on unsupervised learning

Abstract

The invention provides a dam safety monitoring data anomaly detection method based on unsupervised learning, and the method comprises the following steps: (1), obtaining to-be-detected time series data of a monitoring amount during the operation of a dam, carrying out the normalization processing of the collected to-be-detected time series data, performing rolling sampling on the normalized time sequence data to be detected by adopting a moving sliding window, and establishing a training sample data set and a test sample data set; (2) based on a training sample data set and a test sample data set long-short memory (LSTM) recurrent neural network regression prediction model, performing regression prediction on the to-be-detected time series data, and calculating a residual sequence of the to-be-detected time series data and the reconstructed sequence data; and (3) establishing an anomaly detection model based on an isolated forest (iForest) algorithm, and inputting the residual sequence into the anomaly detection model to complete real-time detection of the abnormal value of the dam monitoring data. According to the method, the problem of online intelligent identification of the abnormal value of the monitoring data in the dam safety monitoring real-time acquisition process can be solved, the method has high generalization ability and wide application range, the data types acquired by different sensors can be detected, and a large amount of data can be quickly processed.

Description

technical field [0001] The invention relates to the technical field of dam safety monitoring and data processing, in particular to a method for detecting abnormalities in dam safety monitoring data based on unsupervised learning. Background technique [0002] The long-term safety monitoring of the dam structure is the key to ensure its safe operation during its entire life cycle. The dam safety monitoring system provides reliable data for the safety evaluation, management and maintenance of the dam structure. On the one hand, due to the large number and variety of sensors contained in the dam safety monitoring system, the monitoring data is massive and diverse, and with the increase in service time of the dam and the continuous updating of monitoring methods, the amount of monitoring data is also gradually exploding. On the other hand, in the long-term safety monitoring process of the dam, due to the influence of many factors such as the complex environment on the site, the ...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to overcome the deficiencies of traditional methods to solve the prob

Images