The Study On Occupational Hazards Monitoring And Warning Technology Of Coal Mine Dust Based On Internet Of Things

With the increasing scale of coal mine operations in China,the safety risks and occupational health problems of coal mine dust are becoming more serious.Coal mine dust,especially respiratory dust,is extremely harmful to workers.Monitoring of coal mine dust is an important means to control and reduce its damage.In recent years,the Internet of Things and dust separation technology have developed rapidly.Coal mine dust concentration detection technologies and monitoring equipment based on the Internet of Things are gradually emerging.The coal mine dust detection device provides important technical means for real-time data collection and monitoring of dust.On this basis,how to effectively collect dust concentration,manage and apply the collected dust data has become the focus of coal mine dust controlling.Coal miners are prone to suffer from pneumoconiosis when working in environments with high concentrations of dust for a long time.Pneumoconiosis is a type of disease which causes serious health hazards to workers in occupational diseases.Therefore,the establishment of a prediction model with high accuracy has important practical significance for the formulation of pneumoconiosis prevention and control policies.This paper uses the statistical number of people affected pneumoconiosis in China published by the Ministry of Health from 2000 to 2016 as the research object,and studies the veracity of the combination forecasting models.Firstly,this paper constructs a coal mine dust occupational hazards monitoring platform based on the Internet of Things and proposes a cloud service model for coal mine dust monitoring.Secondly,this paper improves the existing method for estimating individual accumulated dust,and proposes the new method combined with personnel positioning technology.Finally,in order to accurately predict the number of people affected pneumoconiosis in China,this paper combines grey theory and neural networks to study the prediction effect of various combinations of different models.The final experimental results prove the effectiveness of multidimensional input gray-generalized regression neural network prediction model proposed in this paper,which comprehensively considers multiple factors affecting the number of people affected pneumoconiosis,achieved the ideal state of prediction.The main research content of this article can be summarized as the following aspects:(1)This paper aims at the problems of coal mine dust's existing detection methods,such as poor timeliness,manual sampling and analysis,and the inability to continuously monitor.Introducing the Internet of Things and cloud computing technologies,and constructing a coal mine dust occupational hazards monitoring platform based on the Internet of Things.The platform has formed a coal mine dust monitoring cloud service model by integrating integrated dust monitoring intelligent terminals,real-time dust data collection and transmission,big data storage and management,and application services.Through this platform,multi-level supervisory departments,enterprises,and workers can monitor the status of dust in real time and achieve flat and off-site continuous supervision.This will increase the level of occupational hazards supervision services in the industry.(2)The calculation of individual cumulative dust exposure was studied.In this paper,the existing methods for estimating total accumulated dust amount have been improved,and the personal cumulative dust exposure calculation method combined with the personnel positioning technology has been proposed.(3)In order to master the trend of pneumoconiosis in China more accurately,different combinations of gray neural network and Grey-generalized regression neural network prediction models have been used for predicting the number of people suffering from pneumoconiosis in China in the future.Considering that the number of people suffering from pneumoconiosis would be affected by many factors,the number of miners in the mining industry,the number of mining companies of different types,the number of enterprises above designated scale and raw coal production were selected from the national statistical database,used as assisting decision factors for the prediction.The comparative analysis and model evaluation have verified the feasibility and effectiveness of the gray-generalized regression neural network prediction model in forecasting the number of people suffering from pneumoconiosis.