The Wavelet Packet Analysis Method Of Underwater Blasting Vibration Signal

With the rapid development of the national economy construction, underwater blastingtechnology is playing a more and more important role in all construction projects. Particularlynoteworthy Among them is the damage that caused by blasting vibration to aquatic organismsand building structures, this leads to more and more attention being paid to blasting vibrationtest technology field, but to the actual collected signals, many problems need to be solvedamong the processing of data. Blasting vibration is a stable random process from essentially,this is abhorrent with the original Fourier transformation theory foundation personality. Withthe spread of blasting vibration wave go through by the complex water medium and thedifferent rock qualitative interference, through the different rock qualitative interference it willoften mixed with various frequency components of the different interference wave, based on theabove reasons, it’s very difficult for us to analysis the blasting vibration signal frequencycharacteristics, for the reason, is almost impossible to get the accurate time-frequency analysisof blasting vibration signals, and it makes us to look for a new and more accurate analysismethod, with the deep research of wavelet analysis theory, especially the application in recentyears in the geological aspects, make people realized that the advantage of wavelettime-frequency analysis can be good to fill the shortcomings of blasting vibration signal in timeand frequency analysis. By theoretical analysis, discuss the measured signal combined with thefield test. Get the conclusions as follow:(1) The millisecond waveform is superimposed by a single segment waveform.(2) Along with the increase of segment explosive dosage, every detail signal ofwavelet packet analysis in the main vibration frequency band peak particle vibration velocityhas the increasing trend, but its growth rate to vary within the scope of the main vibration band.(3) When the time of single-segment waveform vibration is less than millisecond, the mainvibration band corresponding to a greater value-frequency components of the back section ofthe waveform will show a single-segment waveform independent features, there is no amplitudesuperposition phenomenon.The wavelet analysis has its own unique advantages in the field of time-frequencydomain, it should have even broader prospects in the vibration signal analysis.