基于自适应捕食遗传算法的改进BISQ模型多参数反演

发布时间：2018-03-07 22:08

本文选题：改进BISQ模型　切入点：遗传算法　出处：《东华理工大学》2016年硕士论文　论文类型：学位论文

【摘要】：基于双相介质模型的多参数反演在油气储层预测、海洋底质探测等领域中广泛应用,理想的介质模型与高效的反演方法是其两个重要组成部分。改进BISQ模型在同时考虑Biot和喷射流机制的基础上,成功简化了流体压力表达式,较BISQ模型更加简易适用。为进一步探索研究改进BISQ模型的优越性,以改进BISQ模型为基础,进行了改进BISQ模型的多参数反演。反演算法方面,在传统遗传算法的基础上引入捕食搜索策略、自适应等思想,设计完成了自适应捕食遗传算法。通过数值反演试算,自适应捕食遗传算法较传统遗传算法、小生境遗传算法表现出明显优势,无论是反演效率还是反演精度上都优于后两者;同时得出,改进BISQ模型的多参数反演在保持与BISQ模型反演精度相同的情况下,具有更高的反演效率,这也正是改进BISQ模型较BISQ模型简化的充分体现。为检验自适应捕食遗传算法在改进BISQ模型反演中的抗噪性,对算法进行了抗噪性检测,通过与小生境遗传算法抗噪性对比,显示出较强的抗干扰能力。为了进一步检验算法的有效性和实用性,最后将算法实用于海底原位声学实测数据的反演中,反演参数由三参数逐渐增至六参数,逐步完成高维参数反演。反演结果与海洋实测数据、实验室测量结果及前人的研究成果相统一,反演与实测速度拟合效果良好,进一步显示出自适应捕食遗传算法的实用性和可靠性。
[Abstract]:Multi-parameter inversion based on two-phase media model is widely used in oil and gas reservoir prediction, ocean sediment exploration and other fields. The ideal medium model and the efficient inversion method are two important parts of the model. The improved BISQ model simplifies the fluid pressure expression on the basis of considering the Biot and jet mechanism simultaneously. In order to further explore the superiority of improved BISQ model and based on the improved BISQ model, multi-parameter inversion of improved BISQ model is carried out. Based on the traditional genetic algorithm (GA), the predation search strategy and adaptive algorithm are introduced, and the adaptive predation genetic algorithm (AGA) is designed and implemented. By numerical inversion, the adaptive predation genetic algorithm (AGA) is compared with the traditional genetic algorithm (GA). Niche genetic algorithm is superior to the latter two in inversion efficiency and precision. It is also concluded that the multi-parameter inversion of improved BISQ model has the same accuracy as that of BISQ model. The improved BISQ model is more simplified than the BISQ model. In order to test the anti-noise performance of the adaptive prey-prey genetic algorithm in the improved BISQ model inversion, the anti-noise detection of the algorithm is carried out. Compared with niche genetic algorithm, the anti-noise ability of the algorithm is shown to be strong. In order to further test the validity and practicability of the algorithm, the algorithm is applied to the inversion of in-situ acoustic measured data. The inversion parameters are gradually increased from three parameters to six parameters, and the inversion results of high-dimensional parameters are gradually completed. The inversion results are consistent with the ocean measured data, the results of laboratory measurements and the previous research results, and the fitting results of inversion and measured velocities are good. The practicability and reliability of adaptive predation genetic algorithm are further demonstrated.
【学位授予单位】：东华理工大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：P631.4

【相似文献】