Detection of element content in coal by pulsed neutron method based on an optimized back-propagation neural network

This paper introduces the detection method of principal element content (carbon, hydrogen and oxygen) in coal by using pulsed fast-thermal neutron analysis method (PFTNA). A system for the measurement of coal by PFTNA is also presented. The 14 MeV pulsed neutron generator and a Bi4Ge3O12 (BGO) detector with a 4096 Multi-Channel Analyzer (MCA) were applied in this system. The detection model of element content in coal based on back-propagation (BP) neural network which is optimized by Genetic Algorithms (GAs) was put forward, and the research of verifying the model was made by comparing the practical measured data of coal in power plant. The results show that the detection precision of carbon, hydrogen and oxygen using this model is 0.3%, 0.2% and 0.4%, respectively.