Chaotic Dynamics And Seismic Oil Reserves Information Detection Method

Chaos theory is a new subject aimed at nonlinear issue in the last thirty years. Deterministic chaos offers a striking explanation for irregular behaviour and anomalies in system which do not seem to be inherently stochastic. This thesis was supported by Chinese Natural Science Fund Program(No.49874030). By studying on chaos theory and writing program, a series of methods have been brought forward which are used to detect reservior and oil information from seismic signal in the thesis. Main contents as follows:A series of concepts and theorem involved in the paper on chaotic dynamical systems are discussed, such as dynamical systems, optimization arithmetic, phase space & orbit, Li-York law, chaos control and so on. In order to understand chaos theory, chaotic evolving law of Logistic equation is addressed in detail. Logistic equation is a simple nonlinear dissipative system.Chaos theory offers completely new concepts and algorithms for time series. Fractal dimension and Lyapunov exponents are striking feature of nonlinear time series. The key of calculating fractal dimension and Lyapunov exponents is to establish a good phase space. Moreover, catastrophe parameter is an important feature of nonlinear time series. Methods of calculating fractal dimension, Lyapunov exponents and catastrophe parameter from seismic trace are exhausted. The results of data simulating are satisfied.To find out characteristic of lithology and liquid under surface, seismic trace inversion is very important in geophysics. Based on chaos theory, chaotic feature of inversion dynamical system is analyzed, and targeting chaos, controlling chaos and algorithm of Lyapunov exponent in high dimension dynamical system are discussed, then a new method of seismic trace nonlinear inversion controlled by Lyapunov exponent is presented. The successful emulation is realized.Artificial neural network can be used to detect oil & gas information from seismic data. Self-organizing feature map and Back-Propagation Network are discussed firstly. Aimed at issue of local optimization in Back-Propagation Network and based on chaotic feature in Logistic equation, a kind of chaos optimized artificial neural network is presented. Using seismic attributes of fractal dimension, Lyapunov exponents and catastrophe parameter as the input of system, the good stimulating results are achieved.To check up performance of above methods in processing actual seismic data, one seismic section of a certain sediment basin is utilized. Compared with traditional methods, the methods proposed in this paper possessed higher resolution and stability.