Research On Production Classification Of Coalbed Methane Well Based On MATLAB

Abstract

China’s coalbed methane resources are very rich, but the coalbed methane mining effect is not good. Fracturing is an important means of coalbed methane mining. It is very important to select coalbed methane wells with mining value before fracturing. For most fracturing wells in China, CBM wells are selected according to the experience of the site and workers. It has blindness and increases the cost of coalbed methane mining. In order to reduce the loss, this paper proposes a prediction model based on support vector machine coal fracturing effect. First, pretreatment studies of fracturing well data are performed to ensure the quality of fracturing well data. Secondly, the gray correlation analysis is carried out on the pre-processed fracturing well data, and eight main controlling factors affecting the coal cracking effect are extracted. Finally, using the support vector machine classifier, the coalbed methane wells are divided into three categories: low-yield wells, middle-production wells and high-yield wells. At the same time, the support vector machine nonlinear regression model is used to predict the stable production of coalbed methane wells. Using the support vector machine model to predict, predict the stable production value of coalbed methane wells is basically consistent with the field data, which has important guiding significance for on-site coalbed methane mining.