2020Journal of Physics Conference SeriesOpen access

Research On Production Classification Of Coalbed Methane Well Based On MATLAB

Yifang Tang, Tianxiang Zhang, Zixi Guo, Weiping Zhu, Pengbo He

Open full text 0 citations

Abstract

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.

Open-access reader

About this research paper

What this paper is about

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.

Why it matters

A significance statement is not available in the OpenAlex record.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

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.

Key concepts: Coalbed methane, Petroleum engineering, Coal, Methane, Coal mining, Hydraulic fracturing, Mining engineering, Support vector machine

Related papers

Back to paper searchBrowse research topicsOriginal source
Research On Production Classification Of Coalbed Methane Well Based On MATLAB — Research Paper | ScholarLens