高效采集中的随机地震观测系统设计及数据重建

在压缩感知采样理论下开展“两宽一高”高效地震数据采集,在可控的投资成本下能获得更高质量的成像结果。提出了随机采样观测系统设计的基本原则,即在高密度地震数据采集(或常规地震数据采集)观测系统建立的基础网格的基础上,按照高斯随机采样的理论要求,将规则欠采样的网格作为高斯随机采样位置的期望,分别在炮集范围进行空间随机检波点位置的设计和在整个工区进行空间随机炮点位置的设计。以特征波场(初至波场或标志性的反射波场)为随机观测系统感知的对象,用频率域地震数据Hankel矩阵的低秩特性作为稀疏性的度量标准,通过生成符合高斯分布的随机观测系统,测试随机采样加稀疏提升算法对于恢复无假频的地震数据能力及其影响因素。数值实验结果表明,随机采样加压缩感知数据重建后,可以较好地恢复密网格采样数据。同时,相对于随机采样和规则采样的地震数据,重建数据的偏移成像质量都有所提升。实际应用中,若考虑近地表散射噪声和静校正量等因素,则对随机观测系统的设计及数据重建算法提出了更高的要求。

Observer-Based Fuzzy Control Design for Discrete-Time T-S Fuzzy Bilinear Stochastic,Systems with Infinite-Distributed Delays

This paper is concerned with the problem of observer-based fuzzy control design for discrete-time T-S fuzzy bilinear stochastic systems with infinite-distributed delays. Based on the piecewise quadratic Lyapunov functional （PQLF）, the fuzzy observer-basedcontrollers are designed for T-S fuzzy bilinear stochastic systems. It is shown that the stability in the mean square for discrete T-S fuzzy bilinear stochastic systems can be established if there exists a set of PQLF can be constructed and the fuzzy observer-based controller can be obtained by solving a set of nonlinear minimization problem involving linear matrix inequalities （LMIs） constraints. An iterative algorithm making use of sequential linear programming matrix method （SLPMM） to derive a single-step LMI condition for fuzzy observer-based control design. Finally, an illustrative example is provided to demonstrate the effectiveness of the results proposed in this paper.