煤与瓦斯突出危险性预测的SαS-PNN模型及应用

发布时间：2018-02-23 19:35

本文关键词： Alpha稳定分布(SαS) 高斯分布概率神经网络煤与瓦斯突出预测　出处：《传感技术学报》2017年07期 　论文类型：期刊论文

【摘要】：较高精度的煤与瓦斯突出预测是煤矿安全生产的必要前提和保证。为了实现对煤与瓦斯突出危险性快速、准确和动态预测,考虑煤与瓦斯突出多种影响因素。提出一种改进的概率神经网络(PNN)煤与瓦斯突出预测模型。首先,引进一种对称Alpha稳定分布(SαS),SαS有更广泛的数学表达,其径向对称特性可充当PNN样本层中的高斯分布。在SαS的基础上,建立煤与瓦斯突出危险性预测的SαS-PNN模型。将SαS-PNN模型应用于国内26个典型矿井的煤与瓦斯突出危险性等级预测。预测结果表明:在3种不同的训练和测试下SαS-PNN模型仍具有良好的预测效果,其误判率分别为7.69%、11.54%和15.38%。说明该模型可为煤矿开采中煤与瓦斯突出危险性预测提供了一种可能的思路。
[Abstract]:High precision prediction of coal and gas outburst is the necessary premise and guarantee of coal mine safety production. In order to realize fast, accurate and dynamic prediction of coal and gas outburst, Considering the influence factors of coal and gas outburst, an improved probabilistic neural network (PNN) model for predicting coal and gas outburst is proposed. The radial symmetry can act as Gao Si distribution in PNN sample layer. The S 伪 S-PNN model of coal and gas outburst risk prediction was established. The S 伪 S-PNN model was applied to the prediction of coal and gas outburst risk grade in 26 typical mines in China. The prediction results showed that S 伪 S-PNN model was used in 3 different training and testing conditions. Still have good prediction results, The misjudgment rates are 7.69% and 15.38%, respectively. It shows that the model can provide a possible way to predict the risk of coal and gas outburst in coal mining.
【作者单位】：昆明理工大学国土资源工程学院;淮阴工学院建筑工程学院;中国铝业遵义氧化铝有限公司;
【基金】：国家自然科学基金项目(51264018,51064012)
【分类号】：TD713;TP183

【参考文献】