基于LS-SVM的回采工作面瓦斯涌出量预测

Prediction for gas emission quantity of the working face based on LS-SVM

提出了一种基于最小二乘支持向量机(LS-SVM)的回采工作面瓦斯涌出量预测新方法.一方面,该方法基于结构风险最小化,能较好地解决小样本学习问题,避免了人工神经网络等智能方法在对回采工作面瓦斯涌出量进行预测时所表现出来的过学习、泛化能力弱等缺点;另一方面,该方法用等式约束代替不等式约束,降低了计算的复杂性,使得预测容易实现.实验表明,该方法具有预测精度高、速度快、容易实现等优点,适合对回采工作面瓦斯涌出量的预测.

Put forward a new prediction method for gas emission quantity of working face based on least squares support vector machine （LS -SVM）. It can solve the small-batch learning better and avoid such disadvantages as over-training, weak normalization capability, ect. , which artificial neural networks prediction has, because the latter is based on structure risk minimization. This proposed method is simpler and can be realized easily, because it uses equality restriction instead of inequality restriction. Experiments prove that it offers an effective method for predicting gas emission quantity of working face.