Study Of Physics Simulation Experiment And Calibration Methods About The Loss Gas In CBM USBM Gas Content Testing Process

Coalbed gas content is one of the key parameters in calculating CBM resources and geological constituency evaluation.It is quite significant to obtain accurate coalbed gas content in coalbed methane exploration and exploitation.This paper expanded the views from three aspects:the determination of the coal gas content,the prediction of deep coalbed gas content and the physical simulation of on-site lost gas,in order to improve test accuracy,we proposed a comprehensive test method of coal gas content which is applicable to reservoir characteristics of the Chinese system.This paper described coalbed gas content measurement method such as the USBM direct method,improved direct method,Smith-Williams curve fitting method and pressure coring method,it described specific steps of each method in detail,and analyzed the advantages and disadvantages of each method and its applicability.This paper predicts the deep gas content by establishing mathematical models.We specific studied the coal gas content prediction model based on multiple regression analysis containing,BP neural network prediction model and Gray correlation prediction model.Accorrding to the coalbed known parameters and gas content data,we analyzed the relationship between factors which affected gas content,in order to build a mathematical model to predict deep gas content.Aiming at the phenomenon of low estimation in loss gas content,we designed and implemented physical experiment to simulate high pressure homogenizer quick buck adsorption process to measure coalbed loss gas actually.Studying about the factors affecting the field coring process:the size and speed of drilling,and analyze influence of different factors of the loss gas,and verified low volume estimate by the United States direct method,and finally proposed correction factors for different ways.