| With the increasing demand of society for sulfide ores,the shallow minerals have been depleted,and the deep development of metallic minerals has become an inevitable trend.With the increasing of mining depth,the rise of ground temperature is obvious.The high environment temperature in deep mines will aggravate spontaneous combustion of sulfide ores.Many metal mines are at risk of internal fire at home and abroad.Once the oxidation and self heating accidents concentrate in the sulfide ore,the disaster will lead to the production of the mine and a series of environmental and safety problems,which result in huge economic losses.The research on the early warning of sulfide ore of spontaneous combustion is of great significance to the sustainable development of enterprises and the safety of personnel.The main research contents of this thesis are as follows:(1)Many accidents of spontaneous combustion of sulfide ores are selected to analyze the cases in metal mines at home and abroad in this paper.The system dynamics is used to sort out the factors of spontaneous combustion,the method of improved AHP based group decision is used to select the early warning index of spontaneous combustion of sulfide ores,and the principal component analysis is used for the second screening to establish a scientific and complete disaster warning index system.Also,the paper divides the four warning and gives the grading standards and the corresponding warning criteria in order to provide the basis and theoretical support for the prevention and control of spontaneous combustion of sulfide ores.The indicators can be optimized form 22 to 18 in 22,which includes personnel,spontaneous combustion tendency of sulfide ores,environmental and management,4 indexes and 18 level two indexes.(2)According to the characteristics of early-warning indicators,the concept of the unascertained rational number method is used to determine the weight of the early-warning system of spontaneous combustion of sulfide ores in this paper.The weigh of the index will help the mine to grasp the key point in early warning work,improving the level of safety management and reducing the disaster loss.The unascertained rationality method can greatly reduce the arbitrariness of the subjective empowerment process and make the conclusion of the evaluation more scientific and reasonable.(3)The early warning models of spontaneous combustion of sulfide ores based on RBF,PSO-SVM and cloud model are established combined with the cases of spontaneous combustion of sulfide ores.The empirical results show that the structure of RBF neural network model is 19-18-4,and the prediction effect of spontaneous combustion disaster level of sulfide ore is better and the performance is relatively stable.Meanwhile,the classification accuracy of PSO-SVM test samples is more than 80%,and it has a good performance on classification.The prediction model based on the unascertained rational number-cloud model has some differences in the prediction of spontaneous combustion of sulfide ore.Therefore,it is necessary to make further research on the problems existing in the cloud model to make cloud parameters more scientific and objective and the results more practical.This paper selects the RBF early warning model to complete the design of sulfide ore early warning system based on the accuracy and reliability comparing three early warning models.(4)Finally,Applying Qt application framework and an early warning system of sulfide ore spontaneous combustion disaster is developed on the Windows platform,which includes user management,spontaneous combustion disaster basic information,spontaneous combustion early warning management and emergency management 4 modules.Thus,the system can realize the early warning function of spontaneous combustion disaster of sulfide ores,and provide intelligent,informative and scientific management platform for the safety production of the enterprise.The Qt can be ported across multiple platforms to improve software development efficiency and reusability. |
PDF Full Text Request