Research On Risk Identification Model And Application Of Abnormal Gas Emission Based On Immune Mechanism

From the perspective of China’s energy production and consumption structure,coal’s dominant position will remain unshaken for a long time in the future.In recent years,due to the near exhaustion of shallow coal seams,coal mining has gradually moved to deeper mining,and the abnormal gas emission phenomenon has increased,and the risk of disasters has also been increasing,so the safety production of coal mine has been greatly threatened.Abnormal gas emission,as the source factor of gas accident,has certain risk and uncertainty,which leads to the difficulty of risk identification and factors identification in dynamic changing environment.Therefore,it is of great significance to explore the dynamic identification method of abnormal gas emission risk in complex underground environment,so as to provide evidence for the early prevention and control decision of abnormal gas emission,and to effectively prevent and control abnormal gas emission and reduce the occurrence of gas accidents.In view of the limitations of traditional static risk identification methods and the demand for dynamic identification of abnormal gas emission risk,this paper carried out the research on the identification model and application of abnormal gas emission risk based on immune mechanism by referring to the relevant mechanism of immune recognition of biological immune system.The main research contents and achievements are as follows:(1)On the basis of the overview of biological immune system,the basic principle of immune recognition of biological immune system and the purpose and characteristics of abnormal gas emission risk recognition are specifically analyzed,and the similarity between immune recognition and abnormal gas emission risk recognition was analyzed from the aspects of purpose,action object,action environment,function realization and principle flow by using the principle of bionics.Based on the above analysis,the coupling between them was established and the feasibility of applying the relevant mechanism of immune recognition to the risk identification of abnormal gas emission was established,and completed the design of the framework of abnormal gas emission risk recognition based on immune mechanism.(2)Firstly,on the basis of analyzing the characteristic indexes of abnormal gas emission,the fluctuation rate,variation coefficient and relative change velocity were selected as the characteristic indexes of abnormal degree of abnormal gas emission,and the XGBoost algorithm was used as the weight calculation method of characteristic indexes to complete the construction of the characteristic vector of abnormal degree of abnormal gas emission.Then,according to the division standard of each component index in the characteristic vector,the abnormal grade of abnormal gas emission was divided into four grades.Finally,according to the theory of artificial immune system and abnormal grade division standard,the recognition space of feature vector of abnormal gas emission was constructed,and the ellipse was selected as the boundary to complete the regional division of recognition space.(3)Firstly,on the basis of analyzing the characteristics and application of classical immune recognition algorithm,combined with the identification space division method,the identification model of abnormal degree of abnormal gas emission based on the adaptive recognition algorithm of T-B cell principle was constructed.Then,in order to further identify the types of abnormal gas emission,the DTW algorithm and the related mechanism of immune recognition were used to complete the construction of the type recognition model of abnormal gas emission based on DTW algorithm.Finally,by combining the above two algorithm models,the risk identification model of abnormal gas emission based on immune mechanism was constructed,and the feasibility of the model was analyzed with a practical example.The results showed that the model can identify the abnormal degree and type of abnormal gas emission,which was consistent with the characteristics of biological immune system recognition of "nonself" antigen and its specific type.At the same time,the model has certain adaptability,learning and memory,and the recognition ability could be gradually enhanced with the increase of recognition times.(4)Design of risk identification system for abnormal gas emission based on immune mechanism.In the risk identification of abnormal gas emission model based on immune mechanism as the core,combined with the characteristics of the biological immune system,the overall framework of abnormal gas emission risk identification system based on immune mechanism was designed,and the functional modules of abnormal gas emission risk identification system were designed based on the technologies of Internet of Things,data sensing and system design objectives.Finally,the main functional interface was implemented.