Study Of Reservoir Inversion And Prediction On SWJZ Oil Field In Song Liao Basin

The speed and quantity of seismic data gathering is egregious and the cost is expensive. Seismic data includes abundant geological information, which is the main basis of oil and gas prospecting.However, owing to the limit of seismic data processing, the information has not been fully developed and used. Conventional structural interpretation is difficult to meet the needs of describing geological targets in the procedure of oil and gas prospecting. To increase reserves and production, it became one of the basic works of reservoir description that how to trace and identify the thickness of oil and gas sands as well as boundary conformation and distribution, and get more comprehension of its attributes. Therefore, geophysical prospectors looked after more effective methods of seismic data processing, supplied much more and accurate geological information to geologists. Seismic data inversion, as one of seismic technique methods, plays important role to reduce the prospecting risk and increase possibilities of exploration and development day by day.Inversion acoustic impedance is one of the inversion of lithology. It is the core technique of oil reservoir geophysics. The continuity, heterogeneity, lithologies and physical properties among wells could be obtained from this method, supplied basis of reservoir prediction.The aim of the acoustic impedance inversion is to trace and predict reservoir. Using seismic attributes and the results of inversion topredict the parameters of reservoir is an aspect of applying the results. To rebuilt the characteristics of reservoir is the main seismic inversion technique by now.In choosing inversion method, Strata generalized linear inversion has high vertical resolution in the geological condition of simple structure, subdued strata. So that we used well-constrained inversion, i.e. generalized linear inversion for interpreting and processing reservoir. Well-constrained inversion is acceptable for last stage of exploration and the stage of development, suitable to the situation of multi-well area and thin and inter-bedded strata. The properties of technique is that its has high resolution and big reference of model. Well-constrained inversion is one of model based impedance conversion. Based on geological model, the disturbing algorithm of optimized iteration from model. After continuous modification and update of geological model, the best match was reached between model based evolution synthetic seismogram and practical seismic data. The last model data is the result of inversion.With abundant high frequencies and complete low frequencies of well log, the well-constrained inversion technique completes the insufficiency of limited bands width of seismic data. That supplies dependant basis for reservoir depth, thickness and physical properties using known geological information and constrained conditions of well log information to calculate the seismic impedance with high resolution.The essence of well-constrained inversion is of interpolating and extrapolating under controls of seismic data among wells. The results of inversion not only depend upon the geological condition, the amountof drilling wells, the distribution of wells and the quality of seismic data, but also depend on the delicacy of data processing. The keys of processing procedure of well-constrained inversion are of analyzing and correcting well log data, correlating strata, extracting wavelet, and building initial model.Model based well-constrained inversion is that combining the high vertical resolution of well log data with the preferable lateral continuity of seismic data, to obtain the results of inversion with both advantages.The contents of this paper is composed:1. Precise structure interpretation: reservoir structure characteristics,reservoir division,and layers identification.2. Study of reservoir geological and geophysical characteristics.3. The synthetic prediction of reservoir: based on the interpretation of well logging, test data, completion data of drilling well and log data, the attributes of log and 3D seismic attributes were obtained with the inversion of log data and seismic data; At the same time, the spatial parameter variation of reservoir geophysical properties was integrated, to predict the distribution of reservoir, to study the spatial variation of reservoir geophysical properties.4. Overall evaluation of reservoir and disposition of well position .Research method:1. The software used in research are seismic interpretationsoftware of Hampson-Russell, Geoframe, Landmark, Kingdom suite etc, with hardware of stations and micro-computers.2. Appling of seismic inversion: the reservoir layers of SWJZ Field are researched and estimated quantitative and synthetic, synthetically using the information of geology, seismic, log, well logging and test with the method of computer furthest.3. Prediction of reservoir: it was researched that the thickness of sand, various combination of sand and mudstone, the relation among the properties of bearing oil or gas and other seismic attributes of wave-form, amplitude, frequency, etc, using correct correlating position of synthetic seismogram and reservoir layer, to build model of 3D strata. Constraining the strata framework, layer velocity model and log data, the lateral distribution of reservoir is predicted with seismic inversion of HRS system. Synthesizing estimate: combining geophysical features of the reservoir with fine structural interpretation, making the best of the tectonic evolution materials of area main structures and treating well traps, the oil and reservoir storing is analyzed, to identify the prospective traps of oil and gas, and in the basis of these process, the advanced target area and suggestion of exploration wells is being put forward.In the paper, the main solved issues are below:1. Precisely making synthetic seismogram ---correlation of reservoir2. Building initial model of precisely according with geological rules.3. Optimizing analysis, extracting attributes of main features. From the 2D or 3D dataset, the seismic feature information is extracted, and depend upon this result, the variation rule is to be identified.4. Optimizing models. In the composite traces of 2D dataset or 3D families, the initial model is optimized from the similar lithology and geophysics parameters.In this paper, the anticipative target is that the resolution ratio of inversion is wanted to be greater than the conventional seismic resolution, and the contact relation of strata is clear, geophysical properties can reflect the law of litholoy and litho-facies in lateral direction. The results of the inversion include seismic characteristics, i.e. not only the medium frequency information but also the lower frequency information, to characteristic the whole reservoir. Using the results of correlation between log and seismic data, which is accorded with each other, to distinguish the thickness of 10 meters of reservoir sands, to be provided for the variation trends of lateral variation, to be provided for the favorable proofs of sands spatial distribution, to basically meet the needs of lateral prediction of reservoir.