Abstract:

Coal-bed methane reservoirs are characterized by low porosity, low permeability and low pressure, making their industrial exploitation primarily reliant on techniques like hydraulic fracturing. Currently, more than 50 percent of the gas wells in the Shizhuang block in the Qinshui Basin currently produce less than 500 m³/d of coal bed methane. However, the increase in production after gas well retrofitting has not been ideal and the main factors affecting gas well productivity remain unclear, directly impacting the overall improvement. To address this, the degree of influence of geological and engineering factors on fracturing productivity in coal-bed gas wells is described using the gray correlation method, and the main factors controlling gas well productivity after fracturing are analyzed. A correlation mathematical model between the main control factors and gas well production is established using the Pearson correlation analysis method to predict gas well productivity. The reliability of the prediction model is verified through gas well data validation. Furthermore, a classification decision tree is established using the chi-square automatic interactive detection decision tree method, in conjunction with gas well productivity data, to understand the impact of geological and engineering factors on gas well productivity in fractured wells. Under high gas content conditions, engineering factors have a relatively small impact on the productivity improvement of gas wells. However, as the gas content decreases, the impact of different engineering factors on the gas well productivity gradually increases, which helps optimize the main design parameters such as displacement, sand volume, and total liquid volume, enriching the evaluation methods for post fracturing productivity of coal seam pressure.

Key words: coal-bed methane, grey correlation method, fracturing effect, K-means clustering algorithm, optimization of fracturing parameters