The Wu-Xia Deep Targets Predicting And Evaluating Methods Reasearch Of Junggar Basin

At present, people pay more and more attention to the exploration of deep layers, and deep exploration has been to the new hotspot of oil and gas exploration and development. Deep seismic data has more problems than shallow and middle layers, for example: weaker energy, lower frequency, narrower effective bandwidth, lower signal-to-noise, harder description, and so on. It has take lots of difficulties to structure interpretation and the seismic reservoir prediction in deep seismic data.The paper aims at the deep layers of Wu-Xia zone. Its target is to enhance the efficiency of oil and gas exploration and development. Aiming at the difficulties facing in practice, we take the following method researching in the subtle structure interpretation: integrate horizon calibration, the technique of seismic coherence cube, 3-D seismic volume visualization method, the technology of variable-velocity mapping, and so on. Basing on the method of traditional"physiognomy", we use the technology of Kohonen neural networks to identify the seismic facies of deep layers in Wu-Xia zone and transfer the seismic facies into sedimentary facies through seismic interpretation. The applicability of seismic attributes have been analyzed by using forward model. The fracture development zones have been predicted by integrating the technique of seismic coherence cube and structural attribute analysis along horizon (the analysis of dip/azimuth,edge detection, and so on) and the analysis of absorption coefficient. The reservoir parameters (porosity, permeability, the percentage of sandstones) in the deep layers have been predicted by integrating the technique of the analysis of seismic attributes and the method of predicting seismic reservoir parameters. We take the method of the log rebuilding of the log constrained inversion to predict the thickness and the transverse range of deep volcanic rocks in Wu-Xia zone. The porosity of volcanic rocks has been predicted by integrating the method of Radial Basis Function Neural Network and the log rebuilding of the log constrained inversion. Basing on the analysis above, integrating the geologic and geophysical characteristic of researchful zones, we have a good comprehensive evaluation of deep targets in Wu-Xia zones. We give advice of well deployment, which builds the foundation of the thorough research of deep targets in this area in the next stage.Basing on the researching above, we summarize a series of geophysical methods which adapt to the prediction and evaluation of deep target in Wu-Xia zones. The methods have been proved to gain better results.