In the process of exploration and development of oil and gas, using geophysical well logging data logging interpretation identification of hydrocarbon reservoir is one of the important tasks.So accurately identify reservoirs in logging interpretation is very important. Hydrocarbon reservoir recognition is an important content of the oil well logging evaluation, It will have a direct impact on the efficiency of oil and gas exploration and success rates.The signal to be collected and analyzed is often a noise signal synthesis that generated by system true signal and a number of noise signal sources in complex signal processing and application of practical engineering. To access useful composition of signals which reflect its features, in signal analysis and processing, the noise is removed according to the spectrum and time-scale characteristics of different signal characteristics. This paper combines theory of wavelet analysis with ANN, integrating the merit of wavelet transform with that of ANN, on one hand, using wavelet transform-frequency characteristics of local, and prominent useful signal characteristics and denoising. On the other hand, using the abilities of self-learning and adaptive for ANN data processing, new model and methods are provided in signal processing.The paper studied wavelet transform modulus maxima denoising theory, And provided the wavelet transform modulus maxima of denoising algorithm. Studied the theoretical basis theory and mathematical characteristics of wavelet transform,the relationship of Wavelet transform and signal singularity,the relationship of signal singularity detection and wavelet transform modulus maxima, wavelet transform modulus maxima of scale propagation characteristics in signal singularities and white noise. Therefore according to the above results ,constructs the denoising algorithm based on wavelet transform modulus maxima , and the practical application.This paper builds up three kinds of neural network model, and the three kinds of neural network model of the structure and properties, the algorithm is studied. Respectively as follows: multi-layer proceeding feedback network model, continuous wavelet process neural networks model,and based on the wavelet transform and the process of neural network integrated model.In multi-layer proceeding feedback network model, BP algorithm is studied, and the BP algorithm was improved, the network structure parameters were analyzed.In continuous wavelet process neural networks model, using the the nature of wavelet function to signal properties of fitting, thus expresses the mapping relational about system input/output signal, and realize approach of the nonlinear function. based on the wavelet transform and the process of neural network integrated model, Through wavelet transform modulus maxima of denoising algorithm for well logging curves signal denoising, Then reuse process neural networks for processing after the signal for hydrocarbon reservoir automatic identification.The paper establishes three the model structures, the high efficiency of the method is confirmed through examples. In view of the third kind of model structure, and it is applied hydrocarbon reservoir recognition in 8 oil wells of Daqing Oilfield. According to the actual data processing the test results are good. |