Abstract—To solve the technical problems of identification
for the dangerous degree of coal spontaneous combustion, the
coal spontaneous combustion process is divided into three
stages: slow oxidation, accelerating oxidation and intense
oxidation in the paper. The prediction method of fusion
identification for characteristic information of coal spontaneous
combustion is proposed and the sensitivity index of coal
spontaneous combustion degree is determined. The quantitative
relationship of the characteristic temperature of coal
spontaneous combustion and the gas concentration of each
index is determined by polynomial least squares fitting method
according to the sample test data. The different feature states
are classified by SVM and PSO-SVM algorithm. The criterion
of the diagnosis and early warning of coal spontaneous
combustion is given according to the results of the data level and
feature level. The proposed method can effectively solve the
problem of low recognition rate. The experiment shows that the
prediction classification accuracy of SVM is 80%, the
prediction classification accuracy of PSO - SVM is
approximately 100%. The PSO-SVM algorithm can
significantly improve the prediction accuracy compared with
the traditional method, which provides criterion for the
diagnosis and early warning of coal spontaneous combustion.
The classification identification of the dangerous degree of coal
spontaneous combustion is implemented. It is of great
significance and practical application value for improving the
level of prevention and control technology of coal spontaneous
combustion early hazards.
Index Terms—Coal spontaneous combustion, characteristic
information, fusion recognition, prediction method.
Wei-Feng Wang and Jun Deng are with the School of Safety Science and
Engineering, Xi’an University of Science and Technology, Xi’an, Shaanxi
710054, China (corresponding author: Jun DENG; tel.: +86-29-85583749;
e-mail: 251044098@qq.com, 693167478@qq.com).
Yuan-Bin Hou is with the School of Electrical and Control Engineering,
Xi’an University of Science and Technology, Xi’an, Shaanxi 710054, China
(e-mail: 739053026 @qq.com).
Nai-Guo Wang is with the Shandong Xinjulong Energy Limited Liability
Company, Heze, Shandong 274918, China (e-mail: 81864920@qq.com).
[PDF]
Cite: Wei-Feng Wang, Jun Deng, Yuan-Bin Hou, and Nai-Guo Wang, "Study on the Prediction Method of Fusion Recognition for Characteristic Information of Coal Spontaneous Combustion," International Journal of Future Computer and Communication vol. 6, no. 2, pp. 67-71, 2017.
