Data-driven based variable geometry design

Conventionally,the method to make up for the missing data of middle-shallow layer in the obstacle area is by variable geometry,for example,deviating physical points and adding sources and receivers.And the missing data of middle-shallow layer is evaluated according to the effective coverage of the target layer.Since the traditional method doesn't consider the actual seismic data,it is impossible to actually predict the gap of section and the imaging effect.The paper proposes the evaluation method of data-driven based variable geometry:Firstly,the obstacle avoidance design is realized according to the coordinate range and safe distance of the obstacle area;Secondly,the local similarity of each common image gather(CIG)is calculated,and the contribution of the sources and receivers to the target area is also calculated;Thirdly,according to the variable geometry design,choose the required trace to perform sorting and stacking according to the contribution of the sources and receivers in the CIG,the stack data volume of the whole work area is generated;finally,evaluate the missing data in the obstacle area by the extracted seismic stacked sections in different direction and guide the designer in the infilling plan.Meanwhile,for area with very low signal to noise ratio(SNR),the new method can be used to evaluate the imaging potential and guide the survey design.The new method has achieved very good effect in the production,and the analysis result is very consistent with the processed result of the actual seismic data.

New approaches to relaxed stabilization conditions and H-infinity control designs for T-S fuzzy systems

This paper focuses on the problem of fuzzy control for a class of continuous-time T-S fuzzy systems. New methods of stabilization design and H-infinity control are derived based on a relaxed approach in which both fuzzy Lyapunov functions and staircase membership functions are used. Through the staircase membership functions approx- imating the continuous membership functions of the given fuzzy model, the membership functions can be brought into the design conditions of fuzzy systems, thereby significantly reducing the conservativeness in the recent fuzzy controller design methods. Unlike some previous fuzzy Lyapunov function approaches reported in the literatures, the proposed design techniques of stabilization and H-infinity control do not depend on the time-derivative of the membership functions that may be pointed out as the main source of conservatism when considering fuzzy Lyapunov functions analysis. Moreover, conditions for the solvability of the controller design given here are written in the form of linear matrix inequalities, but not bilinear matrix inequalities, which are easier to be solved by convex optimization techniques. Simulation examples are given to demonstrate the validity and applicability of the proposed approaches.