Method for automatically detecting signal abnormality of seismic instrument by applying machine learning

Abstract

The invention discloses a method for automatically detecting signal abnormality of a seismic instrument by applying machine learning. The method comprises the following steps: S1, searching a previousdata set of the same type; s2, taking continuous records of each channel of each station in the data set in a fixed time period as a sample; s3, extracting each characteristic value capable of representing a signal state from each sample; s4, carrying out normalization processing on each characteristic value; s5, making a training set, a cross validation set and a test set; s6, constructing a probability density function model; selecting a threshold value epsilon for judging the boundary; s7, checking the probability density function model by adopting data in the test set; s8, checking and analyzing the sample which is wrongly calculated and judged, and increasing a new characteristic value of the abnormal characteristic of the sample; carrying out s4 to S7 again, and training an optimization model; and S9, processing the real-time data of the seismic station according to the S2-S4, and then detecting the real-time data by using the optimization model.

Description

technical field [0001] The invention relates to the field of earthquake monitoring, in particular to a method for automatically detecting abnormal signals of seismic instruments by applying machine learning. Background technique [0002] At present, in the field of earthquake monitoring, the seismic instruments and equipment in the network system can obtain and view data in real time, and from the real-time data waveforms, it is possible to manually distinguish the abnormality of some station signals. However, with the acceleration of the construction of seismic stations, the total number of stations in a province has increased from dozens to hundreds to thousands. After the station data is sent back to the system, it is difficult to distinguish among the huge amount of waveform data by manpower alone. There are abnormal signals, which brings inconvenience to the earthquake monitoring work. Contents of the invention [0003] The object of the present invention is to solve...

AI Technical Summary

Problems solved by technology

However, with the acceleration of the construction of seismic stations, the total number of stations in a province has increased from dozens to hundreds to thousands. After the station data is sent back to the system, it is difficult to distinguish among the huge amount of waveform data by manpower alone. There are abnormal signals, which brings inconvenience to the earthquake monitoring work

Images