A Method for Recognition of Abnormal Data in Mine Gas Monitoring

Abstract

The invention discloses a mine gas monitoring abnormal data identification method, which considers the impact of production factors on gas gushing, reconstructs real-time monitoring data samples, sets initial centroid vectors for clustering, and analyzes the discrimination samples in the clustering , if there is a sudden change and it is outside the 95% confidence interval of historical data, it is a small probability event, and it is determined that the monitoring is abnormal, otherwise it is normal detection data. This method can effectively identify the abnormal data of underground gas monitoring in coal mines. Starting from the characteristics of gas gushing, flow and accumulation in different positions underground, comprehensively considering ventilation factors, fluid flow laws in the ventilation network and gas accumulation sources in the upper corner, and other factors, based on the history Based on the statistical analysis of monitoring data, through online analysis with the safety monitoring and monitoring system, the abnormal data of gas monitoring is screened, and the problem of false signals in the gas monitoring information affecting the processing and calculation accuracy of monitoring data is not high and the problem of monitoring false alarms is solved.

Description

【Technical field】 [0001] The invention belongs to the technical field of coal mine safety monitoring and monitoring, and in particular relates to a mine gas abnormal data monitoring and identification method, big data processing and safety early warning application based on real-time data of a safety monitoring and monitoring system. 【Background technique】 [0002] Gas disaster is one of the most harmful disasters to mine production, which seriously threatens the safe production of coal mines. Mine gas monitoring is an important means of preventing gas disasters. Important locations and areas in coal mines are key locations for gas monitoring, and monitoring data comes from detection Component methane sensor, the sensor transmits the monitoring data to the monitoring host through the underground substation, the monitoring host can be connected to the data acquisition program, and the real-time monitoring data is applied to the safety early warning analysis, that is, the gas g...

AI Technical Summary

Problems solved by technology

[0003] As far as gas concentration detection is concerned, the reliability and accuracy of the methane sensor itself is controlled by on-site quality access and adjustment by professional technicians. During daily operation, due to complex underground conditions, temperature, water vapor, mine dust, etc. Environmental factors will affect the accuracy of the methane sensor; factors such as electromagnetic interference during signal transmission and high-voltage shocks on communication lines will cause false signals, so there will inevitably be abnormal data in the monitoring data transmitted to the monitoring host

[0004] At present, no systematic research results have been submitted for field application on the identification of abnormal gas monitoring data, and data processing and application based on gas monitoring data must be based on more accurate and real data. Identifying and appearing critical

Images