| Passive seismic exploration eliminates the step of artificially exciting source,and has lower cost and better adaptability to complex terrain than the active seismic exploration.According to the distribution of source energy along the time axis,passive sources can be divided into microseismic(transient)source and noise source.The energy of the former is concentrated around the excitation time,while that of the latter is uniformly distributed over a period of time.Passive source exploration is widely used in monitoring various aspects of underground media,among which the spatial coordinate and excitation time of the source are the most noteworthy information.The data used for passive source locating is arrival time(travel time)or record(wavefield),and the methods used is reverse time imaging or inversion.Traveltime inversion has the highest efficiency and the most extensive application at present,but it requires the identification of microseismic event and picking up the arrival time in the record,which is usually not easy to achieve,and cannot be used in locating noise source.Because of the method of inverse fitting,there is a risk of falling into a local minimum.The reverse time imaging based on wavefield does not need to pick up the arrival time,and is effective for both microseismic and noise sources.The imaging result is helpful to grasp the model space as a whole,but the false points in the imaging result will interfere with the identification of the source location.Wavefield based inversion is similar to traveltime inversion,but it does not need to pick up the arrival time and need to know the wavelet information in advance,which is usually difficult to obtain.In addition,due to the unknown number of source points,the inversion process cannot regard the number of source points as the objective parameter,so it is not suitable for simultaneous locating multiple sources.In this paper,three kinds of methods and strategies are proposed to optimize and improve several links of different source locating methods.In the first category,the criterion of source identification is extended from space domain(model domain)to space-time domain to solve the problem of false points in locating microseismic source using the wavefield based reverse time imaging.According to the characteristics of energy concentration of microseismic signal,the condition of identification is established,which can effectively suppress false points in imaging result.In the second category,aiming at the problem of false points in locating noise source using wavefield based reverse time imaging,according to the properties of wavefield propagation,for an imaging point,the wavefield values of several surrounding points are used for imaging.The distribution of the false points is different from that of conventional method.The combination of the two imaging results can effectively suppress false points.The third one combines the reverse time imaging with the traveltime information to solve the problems of local minima and source number in traveltime inversion,which realizes the combination of high efficiency and overall grasp of the model and eliminates the local minima problem.The main methods and achievements presented in this paper can be summarized as follows:(1)For the wavefield based reverse time imaging,the characteristics of the wavefield of microseismic events in the process of forward and backward propagation are studied,and the causes of false points in imaging result are also discussed.The image processing method is used to highlight the imaging value representing the source in the imaging result,which balances the difference between strong and weak sources.It is proposed to incorporate the time axis into the criteria for source identification and judgment,and the flow and formula for quantifying the energy concentration degree of temporal series are established.Combining with threshold,a sifting model is constructed to suppress the false points in imaging result.Because the formula involves the whole temporal series,if time saving is considered,all the wavefield snapshots should be saved and processed uniformly.This paper considers saving space and converts the processing method suitable for data volume into the processing method suitable for data stream.At this time,there is no need to store all wavefield snapshots,but the computing time is increased.(2)The wavefield based reverse time imaging is also effective for noise sources but the problem of false points is more serious than that of microseismic locating.The arrival time in noise source records can not be picked up,nor can the distribution of source wavefield energy on the time axis be used to judge the existence of a source.For this reason,the theoretical basis of reverse time imaging,i.e.time invariance and seismic interference,is studied.The occurrence of interference in the process of backward propagation is discussed.Based on this,a method of imaging by using the temporal series of the backward propagated wavefield around the imaging point is proposed.This method has nothing to do with the nature of the source,so it is also suitable for microseismic locating.Compared with conventional method,this method will suffer from more serious false points,but the distribution range of the false points is different.The combination of the two methods can suppress the false points in the imaging result.It is necessary to calculate the traveltime field around the imaging point in the process of selecting the points around the imaging point.In order to reduce the time consumed of this step,only the extremum points in the conventional imaging result are computed again.The imaging values of the imaging results of different steps are counted.Through the analysis of the statistical results,a method of threshold selection is determined to filter out the non source points.(3)According to the data and methods used in passive source locating,the development law of locating method is studied comprehensively.According to the characteristics of data(traveltime or wavefield)and method(reverse time imaging or inversion)used in passive source locating,the traveltime and reverse time imaging are combined,and both the high efficiency of traveltime and the integrity of reverse time imaging are realized.According to the time invariance of wavefield propagation,a method of reverse calculating the traveltime field is proposed,which can complete the traveltime calculation from all imaging points to all receivers with high efficiency.The imaging conditions of source imaging do not need explicit physical meaning,but only need to indicate the location of source.Based on the relationship between the traveltime of the source wavefield and the arrival time picked up in the record,the imaging conditions for traveltime imaging are established in this paper.According to the characteristics of imaging conditions,a threshold selection method based on statistics is proposed to extract the source location in imaging result.This method is only suitable for microseismic locating because it needs to pick up arrival time,but it does not need to identify the number of microseismic events or the corresponding relationship between the microseismic events and the arrival time,so it is suitable for multi-source location. |
