Lithology Auto-identification From Oil Well Logs

This article focuses on the topics of auto-identification of lithology from oil well logs, mainly used to well logging technology. A geologic strata of well identifies the contrast is the part that many wells evaluation, the oil hides to describe the medium indispensability .Passing the geologic strata contrast can understand the variety of the geologic strata layer preface, rock and the geologic strata thickness, understand the geologic strata structure , breaking the layer and relation that integrate, find out the oil stratum of air to distribute and connect the circumstance, for looking for beneficial district that contain oil. During oil field develop, the traditional data explains use the artificial to draw up to match or use the expert's subjective evaluation. Those methods need lots of work, at the same time, the accuracy of the contrast along with the one who explain is well-trained the degree be different. So this affect acquire of right result. In order to avoid the influence of the artificial factor as far as possible, adopt the intelligence of the calculator to identify and then become urgent request on the engineering This article firstly made an in-depth investigation toward the network structure and studying three models of ANN—Bp Network, Hopfield Network and kohonen network, especially discuss bp network and made a improvement toward traditional BP arithmetic, Secondly, it discussed the process of using Visual C++ to realize the network to identify well lithology. At the same time, it applies Aitken arithmetic of numerical analysis to improve convergence speed for BP neural network been explored. ANN( Artificial Neural Network) is a sort of large-scale self-adaptive nonlinear kinetic system, it can realize nonlinear mapping, mode identifying ,function approximating, clustering analyzing ,data compressing and design optimizing etc, and in the mean while, it has many favorable strong characters such as stability, astringency and robustness etc, so it has broad application in data management Bp network is a kind of layer structured front feed neural network with nonlinear mapping function. It is consisted of input layer, concealed layer and output layer; there is no connection between the nerve cells within the layers and no feedback between the layers, signals transfer along a same direction from the input layer through the concealed layer to the output layer. Grades searching technology was adopted by the study arithmetic to minimize the public expense function and adjust the connection power by this, thus acquire knowledge and save into the connection power of the layer network. This endowed the network with the characters of preciseness knowledge structure and efficacious consequence mechanism .It was usually used to make function approaching, model sorting, data compress etc. Hopfield Network is a kind of feedback network also called self associative memory which many favorable characters of nonlinear dynamics system. Its usual structure only has one layer of nerve cell, the feedback between each nerve cell is entire. Its study process is design a network ,save a group of balance point so that when giving a group of initial values, the network finally converge to a certain saved balance point by self execute. This network is used mainly in associative memorizing mode sorting and mode identifying etc Kohonen network is a kind of competitive study network, it is simulate the self-organization feature map function of brain nerve system as its network study method and was hence called SOM network. Its topo-structure is: one input layer, one output layer, input nodes and all nodes of the output layer realize complete interlink age through power value. Its idea is that the network layer nerve cells compete the response chance to the input mode, only one nerve cell become the winner in the end and if adjust the connection power values that have relation with it to the direction which is more favorable to him, this victorious nerve cell express the sort of the input mode. It is usually used to do non-mode sorting, clustering analyzing and design optimizing etc.