Kernel principal component analysis (KPCA) algorithm based method for classifying coal mine well water sources

Abstract

The invention discloses a kernel principal component analysis (KPCA) algorithm based method for classifying coal mine well water sources. The method is performed as follows. A laser induced fluorescence spectrum sensor is adopted to acquire the fluorescence spectrum of to-be-detected water. A laser induced fluorescence spectrometer acquires the fluorescence spectrum data of to-be-detected water under a coal mine through optical fiber transmission. The obtained fluorescence spectrum data of to-be-detected water are uploaded through a CAN bus. Considering the depth and the complex environment under the coal mine, the method converts the CAN bus read fluorescence spectrum data into RS 485 signals through a CAN / RS 485 bus converter. After long distance transmission, the RS485 signals are transmitted to a RS485 / RJ45 bus converter on the ground where they are converted into RJ45 network port signals that PCs could pick up. In this method, long distance transmission is made possible and data are finally transmitted to PCs where they undergo a spectrum pretreatment of wavelet de-noising through MATLAB and are then sent to a KPCA algorithm model to classify water resources. The classifying results are displayed on supervisory control and data acquisition software. With the method, it is possible to rapidly and accurately classify the water sources of water under a coal mine.

Description

technical field [0001] The invention relates to the technical field of coal mining, in particular to a KPCA algorithm classification method for underground water sources in coal mines. Background technique [0002] Mine water damage is one of the major disasters in the process of coal mining, and its occurrence probability and destructive power are second only to underground fire. The phenomenon of mine water inrush is a kind of water disaster. The water inrush event will occur in a short period of time and can cause serious disasters to the underground mining of coal mines. Submerged, causing casualties to underground workers who are performing mining. It can be seen that the harm caused by mine water inrush is enormous. [0003] Nowadays, there are different methods to classify mine water inrush sources, the common methods are water level, water temperature discrimination method and water chemistry method. The water chemistry method is to analyze and model the ion conce...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to provide a kind of KPCA algorithm classification method of underground water body water source of coal mine, to solve the problems of time-consuming long and data having redundancy and noise in the prior art method

Images