Research On The Application Of Information Fusion Technology In Mine Ventilation System Safety Estimation

Mine ventilation system is a complex and dynamic system, it can be affected by several inside and outside factors, and it is a multi-factor problem to estimate the safety and reliability of mine ventilation system. The results showed that we can not find a single item to estimate the security of the mine ventilation system exactly. With the diversification of ventilation system monitoring methods, the classification of monitored data become more and more, and also become more and more complicated. There are possibly redundancy and mutual complementarities in these data and they are also conflicted with others. Adding the complex inspect circumstance and the inaccuracy of the sensor, the inspect data become illegible, inaccurate and bushtailed. Aimed at effectively disposing of the inspect data of estimation of mine ventilation system, this paper applies the information fusion technology in the estimation of mine ventilation system security for the first time. The main work is as follows:(1) Applying the D-S evidence theory and the BP neural network to dispose of the mine ventilation monitoring data, and comparing the two methods, the result indicates that the two methods are viable; yet because it is difficult to assign the basic reliability of the D-S evidence theory ,as well as the output of the single BP neural network is instable, this paper proposes a fusion method combining the D-S evidence theory with the BP neural network, viz. using the BP neural network to assign the basic reliability for the D-S evidence theory. The experiment result indicated that this method is more efficient.(2) This paper applies the fusion method combining the fuzzy neural network with the D-S evidence theory in estimation of mine ventilation system stability, and compares the estimation result with the fusion result without fuzzy information. The experiment result indicates that the information fusion methods based on the FNN and the D-S evidence theory have many virtues such as improving the anti-jamming feature of the system and the calculation speed, boosting up the allow fault ability of the system, reducing the uncertainty of the system etc.This paper applies the multi-sensor management technique in the inspection of the mine ventilation, in order to improve the veracity of the inspection system capturing information, and to consummate the fusion and decision-making of the system .It managed the soft and hard resource in mine ventilation inspection multi-sensor system. The sorting of multi-target's threat and the managing of the fusion algorithm in the multi-sensor management system is analyzed in detail.