首页|期刊导航|常州大学学报(自然科学版)|Mobile gas pipeline leakage detection based on machine learning method

Mobile gas pipeline leakage detection based on machine learning methodOA

中文摘要

Mobile detection of gas pipelines through mobile sensors and automatic detection of gas pipeline leaks relying on data obtained by mobile sensors is an important method for gas leak detection.This article uses data collected from mobile methane sensors at Colorado State University to conduct an analysis of gas pipeline leak detection.First,a preliminary exploration of the data set was conducted,and then the data was cleaned,mainly to deal with missing values and outliers.On this basis,the machine learning model was constructed.This article mainly used Decision Tree,Logistic Regression model,KNN model and Naive Bayes model for comparison.By analyzing the performance gaps of various models,the models were then optimized.The important parameters are adjusted and combined through grid search and cross-validation,and each model is evaluated.Through comparison,it is found that the decision tree and K Nearest Neighbors model have the best effect and have a high accuracy rate.It can provide an effective method for mobile gas pipeline leakage detection to ensure the safe operation of the gas pipeline system.

XU Ke;TANG Zongrui;SHI Lin;YAN Xiaopeng;JIANG Zhongyi

School of Computer Science and Artificial Intelligence,Changzhou University,Changzhou 213164,ChinaSchool of Computer Science and Engineering,Northeastern University,Shenyang 110819,ChinaSchool of Computer Science and Artificial Intelligence,Changzhou University,Changzhou 213164,ChinaSchool of Petroleum and Natural Gas Engineering,Changzhou University,Changzhou 213164,ChinaSchool of Computer Science and Artificial Intelligence,Changzhou University,Changzhou 213164,China

信息技术与安全科学

machine learningmobile sensorsgas leaksautomatic detection

《常州大学学报(自然科学版)》 2026 (5)

P.62-72,11

中国石油-常州大学创新联合体资助项目(2021DQ06)江苏省研究生科研创新计划资助项目(YPC24020224)。

10.3969/j.issn.2095-0411.2026.05.007

评论