Application Of Water-flooded Zone Identifying Based On Wavelet-SVM

At present, oilfield is stepping into water-injection period, some even into secondary recovery or tertiary recovery. The composite water-cut increases, the average single well oil production decreases and the unit oil production costs goes up, it is therefore necessary and important to study the features of water-flooded reservoirs, to improve the interpretation conformity of water-flooded zones in order to stabilize the oil production and improve the integrated profit of the oil field.Wavelet transform can fully demonstrate the characters of the target problem and it enjoys the outstanding capability of obtaining comparatively less traits to the logging curve in the feature detection. Support vector machine (SVM) is specially devised to solve the small sample problem. SVM selects the optimal separate hyperplane that either correctly separates the sample set or gets the biggest margin between two classes. The separate problem can be formulated as a quadratic optimization problem satisfied simple restriction. By introducing the kernel function, the nonlinear separate samples are projected into a high dimension space (so call"feature space"), thus separate problem is solved in the linear separate feature space. SVM exhibits some particular superiority for the classification of small samples and has been the preferred classifier at the international level.This paper proposes the approach in integration of combining wavelet transform and support vector machines.The main work of the paper:1. Introduce the status quo of water-flooded identification;2. Introduce the support vector machine(SVM),its training algorithm and SVM which using in multi-classification;3. Introduce the basis knowledge of wavelet transform, strictly expatiate on the wavelet basis and it's selection;4. On the base of these theories, we do wavelet transformation for logging signal, distilling its feature, and then using SVM for training and classification.Experiments prove that this can gain the upper identification ability in the water-flooded identifying.