基于正负相位和奇异点偶识别规则的煤矸识别技术对比研究

采用数字信号处理技术对煤和矸石撞击刮板运输机产生的振动信号进行适当处理,可以完成煤矸实时识别,即时做出对液压支架的控制。文章对所采集到的放煤阶段产生的5组振动信号,分别采用时域一阶差分分析和小波域内模系数极大法分析。在时域内定义了振动信号的正负相位概念,在小波域内定义了正向、负向奇异点偶概念。由所定义的正负相位和奇异点偶概念分别导出煤和矸石识别规则,以实现动态的煤矸自动识别操作。从可靠性和算法的高效性看,正负相位法识别规则优于奇异点偶法识别规则。从深层数据规律看,奇异点偶法识别规则优于正负相位法识别规则。

Conclusions compared using Singularity -point Couple JR and Positive -negative Phase JR for distinguishing rock from coal

Using digital signal processing technology to process the vibration signals of coal and stone bumping the conveyer, we can control the hydraulic support instantly according to the identification results of real - time distinguishing rock from coal. In this paper, five groups of vibration signals sampled with the speed 250k samples/sec during caving are analyzed by the wavelet transform modulus - maxima method and the first order forward difference（FOFD） method respectively. We define the concept of Singularity - point Couple in wavelet domain and the concept of Positive - negative Phase in time domain based upon the characteristics of vibration signals in different analyzed domain. Based upon the concept of Singularity - point Couple and Positive - negative Phase, the Judgement Rule（JR） for distinguishing rock from coal is proposed individually in two domain to identify rock from coal automatically. From the efficiency and reliability of JR, the Positive - negative Phase JR is superior to the Singularity - point Couple JR. From the perspective laws of the analyzed data, the Singularity - point Couple JR is better than the Positive - negative Phase JR.