基于高阶累积量的煤矿主通风机振动故障检测研究

Research on Vibration Fault Detection of Coal Mine Main Ventilation Fan Based on High-Order Cumulative Quantity

受煤矿主通风机运行过程中自身转子等多种因素的影响,在设计煤矿主通风机振动故障检测方法时,经常会出现振动故障类别检测不准确的情况,导致方法的检测精度差。对此,现提出基于高阶累积量的煤矿主通风机振动故障检测方法。对采集到的振动信号进行降噪处理,再计算其多尺度函数,对振动信号进行时频域分析,通过计算振动信号的协方差,计算其振动信号的高阶累积量,由此提取出振动信号的偏度特征和峰度特征,构建振动信号的特征向量。通过计算振动信号特征向量的熵值,识别出其中的故障特征,再计算信号的故障值,检测出不同类型的振动故障。实验结果表明,设计的检测方法在实际应用中误报率均值为5.6%,检测精度高。

Due to various factors such as the rotor during the operation of the main ventilation fan in coal mines,inaccurate detection of vibration fault categories often occurs in the design of vibration fault detection methods for coal mine main ventilation fans,resulting in poor detection accuracy of the methods.A method for detecting vibration faults in coal mine main ventilation fans based on highorder cumulative quantities is proposed.Denoising the collected vibration signals,calculating their multiscale functions,conducting time-frequency domain analysis on the vibration signals,calculating the covariance of the vibration signals,and computing their higher-order cumulants to extract the skewness and kurtosis features of the vibration signals,and constructing the feature vectors of the vibration signals.By calculating the entropy value of the vibration signal feature vector,identifying the fault characteristics,and then calculating the fault value of the signal,different types of vibration faults can be detected.The experimental results show that the designed detection method has an average false alarm rate of 5.6%and high detection accuracy in practical applications.