The Comprehensive Safety Evaluation Study Of The Ventilation System In Coal Mine Based On The Artificial Neural Network (ANN) Measure

The ANN measure theory was fetched out and applied into the safety evaluation in ventilation system of coal mine. It was put in practice that the safety evaluation based on the ANN measure theory in ventilation system in colliery, the important problem is that establishing an index system of the safety evaluation and ascertaining its weight. The index system of the safety evaluation for the ventilation system in coal mine was established according to any practical experience and 4M method after the principia of establishing an index system and the relation of the index levels were analyzed in detail, then the safety grade of all indexes were plot out and the standard about them was given out in the paper.Due to the fundamentality about the influence of the index weight for the evaluation result, it is certain for bringing on different project of making weight, which affects the evaluation result. A count means with message entropy theory was introduced in the paper, which can ensure the index weight of ventilation safety evaluation in colliery. Two kinds of ANN, BP neural network and radial basis function neural network, are respectively used in the paper. Two kinds of BP algorithm, improved self-adaptive learning algorithm with momentum, and L-M algorithm, are discussed and compared in the course of analysis. And RBFNN is also introduced and used in this paper, the application of which has not been published in the field of safety evaluation yet. And from the research, it is concluded that RBFNN whose running-speed is 102-103 times faster than traditional BP algorithm's is more efficient than BP neural network. By using RBFNN can solve the problem on ventilation system in coal mine efficiently and correctly. It makes count of weight value become more scientific and reasonable. These weight value ensured by it can show the important extent of each index.