Prediction of Methane Emission Quantity Based on Back- Propagation Neural Network

Yuchen FAN

Research output: ThesisMaster's Thesis

16 Downloads (Pure)

Abstract

The coal industry is an important basic industry that links the country’s economic lifeline and supports the sustained, rapid, and healthy development of the national economy. China is the world’s largest producer and consumer of coal, accounting for 45.7% of the world’s total reserves, ranking first in the world, and producing approximately 38.2% of the world’s total production. China's coal mining conditions are complex. Coupled with the lack of infrastructure for safe production, gas disasters are a major challenge for coal production. Gas disaster prevention and control is the most important task for coal mining and production. At present, the major gas disasters in China's coal mine safety production are methane explosions and coal and methane outbursts. Therefore, it is of great significance to effectively control gas explosions to improve the safety of coal mines in China. In the production of coal mines, gas monitoring and monitoring technologies are used to realize gas warning and management. At present, after years of automation and information construction, the coal mine has initially established a relatively complete backbone network for information collection and information transfer, ground information storage equipment and automatic control systems. However, the systems operate independently from each other, independently analyzed, and simple data processing methods that use threshold alarms are often used. There is no real data value mining. In the event of an accident, the traditional accident analysis is only based on the causal analysis method, and the causal investigation and discussion often has a relatively strong personal habits and experienceoriented, it is difficult to comprehensively explore the many factors that cause the accident, and the accident risk The analysis was not thorough enough. Through the historical accident data correlation analysis, it is possible to find the relationship between the previously hidden data, thus providing decision support for mine gas accident prevention, risk management, and resource allocation. Then use the genetic algorithm to train the data model to achieve real-time data prediction. The application of data mining technology in the field of coal mine safety production supervision and forecasting can provide a new perspective for safety production and a new tool to obtain new levels of management level and reduction of accidents. These are the significance of this research.
Translated title of the contributionPrediction of Methane Emission Quantity Based on Back- Propagation Neural Network
Original languageEnglish
QualificationDipl.-Ing.
Awarding Institution
  • Montanuniversität
Supervisors/Advisors
  • Moser, Peter, Supervisor (internal)
Award date29 Jun 2018
Publication statusPublished - 2018

Bibliographical note

embargoed until null

Keywords

  • Methane Emission Volume
  • Methane Concentration Based
  • Back-Propagation Neural Network

Cite this