Automatic Generation Model for Coal Mine Alonormal Monitoring Data Based on Characteristic Point Compression

Diagnosis and inspection on the coal mine ventilation system suffers the lack of effective and timely debugging data. In this paper, the time series of abnormal data from many mine ventilation monitoring systems are analyzed, and several types of abnormal monitoring curves are deeply investigated. By combining the data compression theory and the random function, we propose a characteristic point compression-based model （CPC） that can automatically generate abnormal coal mine monitoring data via free sampling and DP global sampling. Based on .Net and C#, a platform is proposed to simulate coal mine ventilation anomaly in real-time. Simulations are performed on several typical anomalies. The obtained anomaly curves are quite consistent with the actual abnormal data, and the proposed model can generate abnormal ventilation monitoring data for relevant researches.