Ordovician grey water inrush early warning method, system and device

Abstract

The invention provides an Ordovician grey water inrush early warning method, system and device. The method comprises the following steps: A, collecting a plurality of water sample data and aquifer labels for each aquifer; b, inputting the data into a random forest algorithm, training by a ten-fold cross validation method, and outputting a confusion matrix and the contribution rate of parameters to a classification result; c, calculating the percentage of each aquifer which is wrongly divided into Ordovician grey water as similarity, and determining a water inrush risk threshold value; d, if the similarity reaches a water inrush risk threshold value, the water inrush risk exists in the aquifer; e, taking the first n parameters with the highest contribution rate as sensitive indexes, taking the sensitive indexes of the water sample in the step A as a training set, and inputting a random forest algorithm to obtain a recognition model; and F, collecting data of a to-be-identified water sample, inputting the data into the identification model, and if the similarity exceeds a water inrush risk threshold, determining that the water inrush risk exists. The method has the advantages that risk early warning is carried out through the properties of the water samples, judgment is integrally carried out through the multiple water samples, and the accuracy is high.

Description

technical field [0001] The invention relates to the technical field of mine water disaster prevention and control, in particular to an early warning method, system and device for Austrian ash water inrush. Background technique [0002] With the improvement of mining technology and the decline of storage capacity, most coal fields are gradually developing towards deep and lower group coal mining. The site conditions of mines are becoming more and more complex, and the mechanism and types of water inrush are complex and changeable. The existing shallow aquifers are no longer suitable. For the prevention and control of deep floor water hazards, especially Austrian ash water disasters, the traditional method of prevention and control of Austrian ash water disasters is mainly based on water level and water pressure data for risk warning, such as the floor water inrush early warning method disclosed in the invention patent with the publication number CN112432727A, However, shallow...

AI Technical Summary

Problems solved by technology

[0002] With the improvement of mining technology and the decline of storage capacity, most coal fields are gradually developing towards deep and lower group coal mining. The site conditions of mines are becoming more and more complex, and the mechanism and types of water inrush are complex and changeable. The existing shallow aquifers are no longer suitable. For the prevention and control of deep floor water hazards, especially Austrian ash water disasters, the traditional method of prevention and control of Austrian ash water disasters is mainly based on water level and water pressure data for risk warning, such as the floor water inrush early warning method disclosed in the invention patent with the publication number CN112432727A, However, shallow aquifers will also affect water level and water pressure data. The current early warning methods are too simple and the monitoring results are inaccurate

Images