Setting up Model of Forecasting Core Reservoir Parameters by Fusion of Soft Computing Methods

The paper utilizes fusion of soft computing methods to distinguish the key attributes of reservoir oil-bearing formation and establishes model with fusion of soft computing methods to forecast these key attributes. The steps as follows: Firstly, use genetic algorithm (GA) and fuzzy c-means nesting algorithm (GA-FCM) to reduce the log attributes of oil-bearing formation and obtain the key attributes that can describe oil-bearing formation of reservoir. Secondly, fuses genetic algorithm and BP neural networks (GA-BP) to construct the fusion model that forecasts the key attributes ,which searches the log attributes and the best number nodes of hidden layer of BP through GA for determining the optimal structure of BP forecasting model. Judge the forecasting model by the error of testing sample. Finally, take oilsk81, oilsk83 and oilsk85 wells of some oil field in China done research and obtain the available results.