| Acoustic Emission(AE)is a significant method in non-destructive testing and damage assessment which can effectively detect the instantaneous elastic waves emitted by various damage forms such as deformation and crack propagation in the material.Forecast the possible defects and failures of the structure to ensure the safe operation of the equipment and structure.Therefore,acoustic emission has broad development prospects in real-time fault detection.Among them,the acoustic emission source location technology is the key to find the crack source,and its key points is to infer the acoustic emission source based on the received acoustic emission signal,which is the so-called "reverse source".In this paper,the three concrete beams have been made.For generating the acoustic emission sources,the two methods of lead-breaking simulation and trabecular loading is used successively.And the traditional time difference method,correlation coefficient method,two-step method,and joint positioning method are used to solve the problem of acoustic emission source.Meanwhile,evaluate the positioning performance of the four methods.The thesis mainly includes the following aspects of work:(1)This article discussed the existing acoustic emission source time difference positioning method,and analyzed the method of identifying the homology of acoustic emission signal and picking up the first arrival time.At the same time,factor analysis has been made which is about the positioning accuracy of the acoustic emission source.Finally,the analysis confirmed that non-homologous signals and first arrival pick-up accuracy are the main influencing factors in the process of acoustic emission source positioning.(2)In order to study the homology signal identification method based on the acoustic emission impact waveform,this paper starts from the similarity of the acoustic emission waveform by using the method of calculating the waveform correlation coefficient to conduct a waveform similarity study on the waveform of the concrete acoustic emission impact signal.By setting a correlation coefficient threshold of 0.6 as the criterion for identifying homology signals,it intuitively reflects the degree of similarity between the waveforms of concrete acoustic emission impact signals.Based on the similarity model of the impact waveform within the acoustic emission event,a positioning algorithm based on the waveform correlation coefficient is proposed.(3)Aiming at the problem of the picking accuracy of the first arrival time in the acoustic emission positioning process,this paper adopts the transformation time-window energy ratio method.Through counting the picking results under different time window lengths,the suitable time window length for the acoustic emission signal is obtained.And,AIC is introduced at the same time that picking up the preliminary picking results of the energy ratio method accurately.Through this way,the accuracy of the first arrival picking has been improved.What’s more,the energy ratio-AIC two-step method based on the time difference correction also is proposed.At the meantime,the first arrival calculation of the acoustic emission events in the lead-breaking experiment and the trabecular loading experiment were carried out,and the reasons for the unsatisfactory positioning results of the two-step method in the lead-breaking experiment were statistically analyzed.(4)From a qualitative and quantitative perspective,combining the advantages and disadvantages of the correlation coefficient method and the two-step method,and based on the homology signal and using the two-step method to accurately pick the first arrivals,a joint positioning algorithm is proposed.(5)The effectiveness and feasibility of each positioning algorithm was verified by conducting grid-breaking experiments on concrete beams.Furthermore,single-point loading experiments were carried out on beams,and the method of cutting pre-cracks was used to detect cracks in the loading process.The development direction is controlled so that the positioning result and the actual position of the crack can be statistically analyzed.Through the above research work,the following important conclusions are obtained:Time difference is not the most important factor affecting the accuracy of acoustic emission positioning,and homology signal recognition should be fully valued in acoustic emission positioning.A single consideration of correcting time difference or identifying homologous signals is limited in improving the accuracy of acoustic emission source positioning.In future research work,the two should be combined to maximize the accuracy of acoustic emission source positioning. |
