Wavelet analysis applied in measurement of coal mining underground EMI and design of filter

The serious Electro Magnetic Interference (EMI) has caused false alarms in coal mining monitoring systems, which would induce troubles of safety in production. Nearly no measurement has been carried on underground EMI distribution because of the difficulties caused by the equipment explosion-proof enclosure and the limitation of repeated startup and stop during measurement. The improved wavelet analysis was proposed to accomplish the indirect measurement of EMI and the electrical fast transient/burst (EFT/B) was studied to verify the validity. Therefore, the sources, startup time, duration and waveform of EMI were ascertained correctly based on underground running records. Wavelet filter was presented in accordance of the EMI distribution gotten from wavelet analysis. The improved threshold function not only overcomes the deficiencies of the soft-threshold and the hard-threshold methods but also meets the real time demand. Therefore the EFT/B is restrained evidently and the false alarms reduce evidently. The wavelet analysis and the improved wavelet filter are significant to measure underground EMI distribution and enhance the monitoring system immunity.