果蝇算法融合SVM的开采沉陷预测模型

    Mine Subsided Prediction Model Based on SVM Combined With Fruit Fly Optimization Algorithm

    • 摘要: 针对目前开采沉陷预计方法的种种缺陷,提出了一种新的预计方法。将果蝇优化算法(FOA)与支持向量机(SVM)相结合,建立FOA-SVM预测模型。选取煤层倾角、采厚、平均采深等参数作为模型的输入参数,最大下沉量作为模型的输出参数。选取训练集样本,应用FOA对SVM的参数进行寻优,确定最佳的SVM参数。采用预测集样本对该预测模型进行检验,同时将该模型预测性能与其他预测模型进行对比分析。结果表明:与GA-SVM模型、PSO-SVM模型和神经网络预测模型相比,该模型具有更高的预测能力和泛化能力,可以较好地实现对开采沉陷的预测。

       

      Abstract: As to the shortcomings of current mine subsided prediction methods, a new method was proposed. Fruit Fly Optimization Algorithm (FOA) was combined with Support Vector Machine (SVM), and FOA-SVM prediction model was built. Seam dip, mining thickness, average mining depth was selected as input parameters of the model; the maximum subsidence was output parameter. Selecting the training set sample, the optimal parameters of SVM was determined with applied FOA. The prediction model was tested by prediction set sample and the prediction performance was compared with other models. The results showed that the model has higher predictive ability and generalization ability compared with GA-SVM model, PSO-SVM model, BP neural network. The model could better realize mining subsidence prediction.

       

    /

    返回文章
    返回