煤炭是我国的主要能源，煤炭行业是国民经济的重要方面。但是由于我国煤炭资源的赋存与开采技术条件的复杂性和多样性，安全一直是煤炭企业所面临的重要问题。 论文根据我国煤矿企业及其安全生产现状，提出提高矿井安全素质是从根本上保障矿井安全生产的关键，在国内外煤矿及企业安全评价方法发展现状分析的基础上，提出用人工神经网络技术进行矿井安全素质评价，对提高矿井安全素质，促进安全生产有重要作用。 首先，以提高煤矿企业安全生产为目的，结合矿井素质的概念及煤矿行业的特殊性，建立了矿井安全素质评价指标体系，包括矿井形象、矿井能力及矿井作用发挥三个方面的16个指标。然后，建立的矿井安全素质评价的人工神经网络模型，确定了评价模型的网络结构及基于BP算法的训练学习算法。运用MATLAB神经网络工具箱的强大功能，设计了矿井安全素质人工神经网络模型的学习和评价程序，并采用历史数据对人工神经网络进行训练，为矿井安全素质评价网络模型的应用奠定基础。其次，运用该模型对忻州窑煤矿矿井安全素质现状进行评价，得到了与实际比较符合的结果；最后，通过对影响该矿矿井安全素质指数的一些重要指标进行重要度分析，确定了这些指标对矿井安全素质指数的影响，据此提出提高矿井安全素质的建议，对矿井安全管理起到了重要作用。

Coal is the main source of energy in China, and the coal industry is an important aspect of the national economy. However, due to the complexity and diversity of the occurrence and the mining conditions of coal resources in China, safety has always been important issues that the coal enterprises facing with. Paper proposed to improve the coalmine safety quality is a fundamental key to protecting the coalmine safety production on the basis of the China's coal mining enterprises and safety production status. Anglicizing of evolutionary current situation of domestic and international coalmines and enterprises safety evaluation method, the coalmine safety quality was evaluated based on artificial neural network technology. It played an important role to improving the coalmine safety quality and promoting safety production. First, in order to improve the safe production of coal mining enterprises ,the paper combined the concept of the coalmine quality and the particularity of the coal industry, and established the Coalmine Safety Quality Evaluation System which including mine image, mine capability and mine the role of play three aspects of the 16 indicators. Then, the paper established the artificial neural network model of coalmine safety quality evaluation, determined the network structure and training and learning algorithm based on BP algorithm of the evaluation model. Based on the powerful function of MATLAB neural network toolbox, this paper designed the learning and evaluation process of the artificial neural network model of coalmine safety quality evaluation and used historical data for training the artificial neural network. It laid the foundation for the application of the network model of coalmine safety quality evaluation; Finally, this model was put into evaluating the safety quality situation of the Xinzhouyao mine, then reached a conclusion in accordance with actual; through the importance analysis between the index number of coalmine safety quality and several important indicators, it confirmed the influence of these indicators. In view of the above, the paper offered proposals to improving the coalmine safety quality which played an important role in mine safety management.