Research Of The Forecast Gas Emission Based On Factor Analysis And Kalman Filter

Mine gas disaster threatens the safe production of coal mine,master the prediction method of gas emission quantity,and realizes the accurate prediction of mine gas ermission.It is the basis for studying the law of mine gas emission and the law of gas geology.The mine gas disaster prevention and protection of underground life There's important meaning.In this paper,a gas mine in Yankuang Group,Xinjiang,is taken as the experimental object.The relationship between gas emission and its influencing factors is studied from the aspects of mine geological conditions and mining conditions.The influencing factors of gas emission are affected Many factors,the role of varying degrees;and gas emission between the existence of complex non-linear relationship;with the time,the face of the changes and other characteristics.According to the characteristics of influencing factors of gas emission and the degree of action,this paper proposes a method to select the forecasting index of gas emission based on factor analysis method.By extracting the same common factor method from the original variables of influencing factors of gas emission,The reduction of information between the original variables is reduced,and the optimization of the original variables is reduced and the prediction indexes of the gas emission are obtained.Aiming at the non-linearity of the gas emission index and the gas emission,the BP neural network and the Kalman filter are used to predict the gas emission.The BP neural network not only realizes the nonlinear mapping recognition of the gas emission prediction index,but also provides the state variables for the Kalman filter theory recursive equations.When the prediction index changes with the face of the face,the BP neural network can effectively change the index information to identify the state variables that react to the Kalman filter,and realize the dynamic prediction of the gas emission.Based on the study of factor analysis method and BP neural network Kalman filter theory,this paper uses MATLAB software as the development platform and graphical user interface GUI as software development tools to design and develop based on factor analysis and Karl Mannel Filtering Gas Emission Prediction Software.The software effectively integrates the factor analysis method to select the prediction model module and the BP neural network and the Kalman filter coupling model.The application example shows that the software has the characteristics of convenient operation,friendly interface and high precision,and can meet the actual requirements of mine gas emission prediction.