An abnormal diagnosis method of monitoring data based on generative confrontation network

Abstract

The invention discloses a monitoring data abnormality diagnosis method based on a generative confrontation network, which is suitable for the abnormality diagnosis of long-term measured data of a structural health monitoring system. By choosing an appropriate basic time interval, the recorded data is divided into several sub-segments. The monitoring time series data of each sub-segment is converted into a grayscale image by using the Gram angle field, and two unsupervised deep artificial neural networks of generative confrontation network (GANs) and autoencoder (AE) are trained accordingly, and according to the test The prediction error of the dataset verifies the training effect of the resulting network. Then, according to the training set and test set samples, the optimal index suitable for data anomaly diagnosis is selected, and the state of the measured data is judged by combining the cumulative sum function. This method can realize rapid and accurate diagnosis of abnormal monitoring data, and provide effective data support for structural state diagnosis and abnormal early warning.

Description

technical field [0001] The invention relates to a data abnormal diagnosis method, in particular to the abnormal diagnosis of long-term monitoring data of a structural health monitoring system, and belongs to the technical field of data abnormal diagnosis. Background technique [0002] The Structural Health Monitoring System (SHMS) includes numerous sensors for monitoring structural responses such as acceleration, velocity, displacement, and strain, as well as environmental factors such as temperature, wind speed, and humidity. Long-term monitoring data can provide reference for structural damage identification and state assessment. Before carrying out damage identification and status assessment, the accuracy of monitoring data must be ensured, otherwise it will lead to wrong assessment results. However, most sensors are located in harsh environments, and it is difficult for sensors to avoid failures. Therefore, the diagnosis and elimination of abnormal data is one of the key...

AI Technical Summary

Problems solved by technology

Unsupervised learning methods can effectively avoid the disadvantages of the above-mentioned supervised learning methods, but the ability to process massive and complex information still needs to be improved

Images