| With the developments of the National Petroleum and chemical industry, China hasbecome the second largest oil consumer in the world at present. It will be exceeding Americaand becoming the first oil consuming country in the future. The crude oil reserve plan hadbeen started from1993in order to cope with the international oil price fluctuation and the statesafety. With the third stage of the project complete, China oil reserves will reach eightymillion cubic meters. Taking into account the business reserves, the reserve scale is evenmore than the National Petroleum reserve. China has a huge oil reserve system. The storagetank is the key equipment in the petroleum reserve system. Corrosion is a serious threat to thestorage tank. Therefore, an effective storage tank corrosion detection technology is essential,especially on-line detection technology.Acoustic emission detection technology is an online testing technology. It detects the tankbottom corrosion with several sensors that are placed on the tank wall. It is quite differentfrom other tank bottom corrosion detection technology which needs open the tank to detect thetank bottom corrosion. But acoustic emission corrosion testing technology has the followingproblem: First, the engineer's experience and the experiment system is essential and muchmore needed; second, the mechanism of corrosion acoustic emission and the characteristic ofacoustic emission signal is not understood to people; third, the theory is lack when theengineers extract the characteristic of acoustic emission signal and select the key parameters ofacoustic emission signal, so the reliability and availability of acoustic emission detectionshould be improved; forth, the relationship between acoustic emission and corrosion processhas been not found, so the physical meaning of corrosion acoustic emission signal can't beunderstood and the accurate test results are hard to be made by engineers. To resolve theseproblems, the mechanism of corrosion acoustic emission, the validity of corrosion acousticemission detection technology, the relationship between acoustic emission signal and thecorrosion process, the extraction of acoustic emission signal characters and the identificationof acoustic emission sources were studied in theory and experiment. The following contentshave been studied:(1) Based on the foundation of corrosion, the reaction of cathode and anode in steelcorrosion process, the possible relationship between the reaction of steel corrosion and acoustic emission signal, the possible relationship between the reaction products of steelcorrosion and acoustic emission signal and the change of stress in corrosion process werestudied. The cause of hydrogen gas bubble growth is presented. The atmospheric pressure,liquid level, and bubble surface tension are the sources of bubble burst acoustic emission stresspulse. And atmospheric pressure, liquid level, and residual stress may be the potential sourcesof steel crack acoustic emission stress pulse. Based on the theory of knopoff point stress pulse,the parameters of atmospheric pressure, liquid level and bubble surface tension weresubstituted the relationship between the response of elastic sheet displacement and the pointforce pulse, and then the displacement was substituted the relationship between the responseof acoustic emission sensor and the elastic sheet displacement. The relation between acousticemission signal and corrosion process is constructed in theory. Based on the numberconclusion and the steel corrosion acoustic emission detection experiment, the effectiveness ofcorrosion acoustic emission detection techniques is directly provided by theory andexperiment. To some extent, the mechanism of corrosion acoustic emission detection wasrevealed.(2) Considering the phenomenon of bubble burst, the crack of steel, the crack of steelcorrosion product and the fluctuation of corrosion potential in corrosion process, therelationship between the gas bubble scales, corrosion cracking scale, cracking speed andcorrosion potential fluctuation with AE signal time-frequency characteristics were studied intheory. The characteristics of acoustic emission of corrosion and the relationship betweencorrosion acoustic emission and the bubble diameter and the liquid level were directly derivedin theory. It is that the amplitude of AE signal is proportional to the bubble radius square andthe liquid level?and its frequency is inversely proportional to the bubble diameter. The AEsignal strength of steel corrosion cracking is proportional to the local stress strength and theAE signal frequency is proportional to the crack propagation distance and inverselyproportional to the crack propagation velocity. The characteristics of corrosion acousticemission were revealed. That is helpful to the engineers who detect the corrosion with acousticemission technology to determine the height of liquid level, select the sensor bandwidth,extract the feature of the acoustic emission signal and evaluate the detection conclusion.(3) The steel corrosion experiment with the same area immersed in the etchant solutionwas designed under sealed condition. The AE systems with high frequency and low frequency were used to detect AE signal during the whole corrosion process at the same distance. TheAE signal parameters were studied. The difference and relationship between AE signal andcorrosion process, corrosion types were studied. The spectrum of bubble breaking out acousticemission signal and corrosion cracking acoustic emission signal that is different was proofedin theoretical and experimental. The frequency range of bubble burst acoustic emission is from10?80kHz. The frequency range of steel cracking acoustic emission is between the80?1000KHz. These two styles acoustic emission signals could be distinguished by thedistribution of power spectrum.(4) The acoustic emission signal characteristics extraction and noise reduction methodswere studied and the rule of choosing the best wavelet function for acoustic emission signalprocessing and noise threshold were proposed The AE signal in time and frequency domaincharacteristics was studied by power spectrum and the S transform spectrum. Thecharacteristic of corrosion acoustic emission signal and the noise was analyzed db8wavelet isone of the best wavelet for the acoustic emission signals noise processing and featureextraction. S transform method can effectively extract the local features of acoustic emissionsignals. |
PDF Full Text Request