Research On Key Technologies For High Density Seismic Acquisition In Mountainous Areas

In China,the strategy of oil and gas exploration and development on shore is gradually changing to "balancing attention to oil and gas,conventional,and unconventional reservoirs",and seismic exploration in mountainous areas is necessary for discovering strategic replacement areas for oil and gas resources in "double complex" areas,thus to activate new impetus for increasing reserves and production,and guarantee national energy security.However,the characteristics of complex surface structure and complex underground geological structure in mountainous terrain,make the signal-to-noise ratio and resolution of seismic data low,and bring unprecedented problems and challenges to seismic acquisition.The main problems can be summarized as: poor imaging quality and high acquisition cost in mountainous areas,difficulties in the optimal design of economically effective observation system,high working risks and troubles in deploying sources and receivers safely and efficiently due to massive complex surface obstacles,the difficulty of selecting shot and receiver methods to improve spectrum of seismic data and to record signal with high fidelity and precision resulted from low signal to noise ratio and resolution of seismic data,and the hardship of rapid quality control and objective evaluation of high-density data at the10,000-trace level.Aiming to solve above problems in seismic acquisition in mountainous areas,this paper studies several key aspects of high-density seismic acquisition in mountainous areas,and the main research content and novelties include:(1)In order to design a more cost-effective high-density 3D observation system scheme and improve the imaging quality of complex targets,this paper systematically studies the optimization of observation system based on analysis of seismic illumination energy with key parameters evaluation of target imaging effect.Through the 3D seismic illumination analysis of the complex tectonic area,particular shot-adding scheme that is more cost-effective was formulated,which improved the illumination of the structural body,and the seismic imaging quality was significantly improved.Due to the high cost of high-density acquisition in mountainous areas,a swath observation method is proposed,which greatly improves the imaging quality of complex structures with reasonable acquisition cost,and effectively guides the design of observation system schemes for the technical and economic integration of "double complex" areas..(2)In order to reduce the risk of mountain seismic acquisition with dynamic sources and improve the deployment efficiency of physical source points,this paper studies the shot point optimization based on terrain risk assessment and the automatic obstacle avoidance technology of shot point based on electronic fence.A terrain risk automatic identification technology of multi-directional slope and undulation constraint is proposed,and a four-level terrain risk evaluation standard of the high-difficulty mountain construction is established.Physical locations of source points are optimized according to the automatic obstacle avoidance technology along with the risk evaluation standard,which greatly reduces safety risks while maintaining the criteria of acquisition design.Profile tracing method and Unet-based deep learning are studied to automatically extract buildings,rivers,reservoirs and roads.The electronic fence is used to realize the automatic obstacle avoidance of the physical locations of source points,and the efficiency of deploying source points is effectively improved.The calculation method of the uniformity of the physical locations of source points is improved,and the uniformity of the observation system is improved after practical application.(3)In order to broaden the frequency bandwidth of seismic data with high signal-to-noise ratio in the mountainous areas,a high-precision surface lithology model was established by applying comprehensive surface survey methods such as lithological core retrieval and lithological logging.In order to increase the energy of down-going wave in the limestone outcrop area,the single well cavity excitation method was studied and a new single well excitation method with no in-filling water was innovatively applied which achieved obviously better results.As dynamic sources cannot be deployed in large composite barrier areas,which may lead to gaps in shallow seismic data,a grid design method of source points for vibrators based on road classification have been established,and the quality of seismic data has been greatly improved by applying dynamic sources and portable vibrator sources for joint acquisition.Based on objective analysis of the advantages and disadvantages of single-point and group receivers,the Fold number of observation system needed for single-point receiver is obtained through field experiments.The application of mountainous smart nodes in the high-difficulty mountain areas ensures the quality of seismic data.At the same time,the safety risk is effectively reduced,and the work efficiency is increased.(4)By quickly accessing the data at 10,000-trace scale,automatically identifying noise and detecting the shot-receiver relationship in real time,the real-time monitoring of seismic data with such scale is realized.According to the difference in characteristics of effective signal and noise in the time domain,spatial domain and frequency domain,the intensity and influence range of noise are quantitatively analyzed,and the multi-domain joint identification technology of noise is investigated.In this paper,the various attributes of the raw seismic data are spatially expanded,and the influencing factors of the quality of shot gather data are analyzed from multiple aspects.Through quantitative extraction of monitoring parameters such as energy,surface wave,sound wave,and shot-receiver relationship,two threshold values are set for the degradation factors,and the results of the shot gather analysis are compared with the threshold values,and the three-level automatic evaluation of the raw seismic data in mountainous areas is finally realized.