Research On Coal Mine Safety Situation Assessment Based On Fusion Of Multi-source And Heterogeneous Data

At present,China's economy is developing rapidly.As the pillar industry of my country's economy,the coal industry must adhere to high standards and strict requirements.Therefore,how to strengthen safe production and escort the continuous,stable and healthy development of the coal industry is of great significance.The production environment of coal mines is complex and changeable,and its safety is affected by the combined effects of many factors.Therefore,the study of coal mine safety situation requires comprehensive consideration of various factors.Existing related researches are more based on a single index,while ignoring the study of comprehensive indexes.Therefore,in order to further ensure the safety of coal mines,it is necessary to conduct related research on the overall safety situation of coal mines.Based on this,this article starts with 25 monitoring indicators of six types of accidents that affect the safety situation of coal mines,and conducts the following research work on the safety situation of coal mines,that is,the state of coal mine safety and the trend of coal mine safety:(1)Construction of coal mine safety situation assessment index system.On the basis of reading a large number of relevant domestic and foreign research documents,from the perspective of six major disasters common in coal mines: gas accidents,roof accidents,fire accidents,flood accidents,electrical and mechanical accidents and transportation accidents,the coal mine safety situation indicators have been constructed system.The indicator system consists of 25 quantitative indicators,all of which can be obtained in real time by relevant monitoring equipment,avoiding the unclear division of qualitative and quantitative indicators of the existing indicator system and the shortcomings that cannot be associated with coal mine safety big data,laying a foundation for the assessment of the situation in the future basis.(2)Construction of coal mine safety situation assessment model.Aiming at the shortcomings of traditional coal mine risk assessment methods that fail to consider both the uncertainty and fuzziness of the system,the cloud model can be used to integrate fuzziness and uncertainty,and the set pair analysis can consider the advantages of determination and uncertainty at the same time,and proposed coal mine safety Situation assessment model.The model considers the two aspects of "state" and "potential".The cloud model can well combine the fuzziness and randomness of concepts in statistics and modeling,and realize the characteristics of the conversion between quantitative data and qualitative concepts.The overall security status has been effectively evaluated.Then,in view of the above evaluation results,combined with the advantages of the connection number that can deal with the uncertainty caused by vagueness,randomness and incomplete information,the system under study is evaluated by establishing the connection number between the evaluation object and the ideal solution,so as to achieve The assessment and analysis of coal mine safety trends.(3)Coal mine safety situation assessment based on hidden Markov model.The coal mine safety situation assessment model based on the cloud model and the five-element connection number has a strong dependence on data,and it is difficult to conduct an effective assessment when the data is missing.Based on this,in order to eliminate the impact of data abnormalities in the evaluation process,the article introduces the hidden Markov model and uses its state transition characteristics to realize the evaluation of the coal mine safety situation from a probabilistic perspective,and aims at the existing hidden Markov model.The initial parameter setting of the training model lacks theoretical guidance,and it is easy to cause the training result to fall into the problem of local optimization.The genetic algorithm is introduced to optimize the initial value setting of the above model,and the Baum-Welch algorithm is used to further optimize The model parameters are used to obtain the hidden Markov model under maximum likelihood estimation;finally,the Viterbi algorithm is used to evaluate the safety situation of the coal mine in combination with the observations.This paper has 31 graphs,15 tables,and 84 references.