矿井涌水量预测的PCA-GA-ELM模型及应用

Research on PCA-GA-ELM Model of Mine Inflow Prediction

为了快速有效地预测矿井涌水量,并进一步提高预测的准确性,在分析矿井涌水量影响因素的基础上,提出一种将主成分分析法(PCA)、遗传算法(GA)与极限学习机(ELM)相结合的矿井涌水量预测新方法。根据矿井涌水实例数据,综合选取9个主要因素作为矿井涌水量的预测指标,通过PCA对数据进行降维预处理,并针对ELM算法的不足,结合GA算法对其进行优化训练,建立矿井涌水量预测的PCA-GA-ELM模型。对模型进行训练及检验,并将PCA-GA-ELM模型与GA-ELM模型、单一ELM模型的预测结果进行对比分析,其预测结果与实际情况更吻合。该模型预测效果优于GA-ELM模型和ELM模型,可对矿井涌水量进行更准确有效的预测,提供科学的参考依据,指导矿山生产。

In order to predict mine inflow more rapidly and effectively, and improve the prediction accuracy, a new method combining principal component analysis （ PCA ）, genetic algorithm （ GA ） and extreme learning machine （ELM） for mine inflow prediction was proposed based on the analyses of mine inflow influence factors. According to the engineering example, 9 main factors were selected as the prediction indexes, and PCA was used to reduce data dimension.Considering the disadvantage of ELM, GA was used to optimize the related parameters of ELM,and then PCA-GA-ELM model of mine inflow prediction was built.Then the model was trained and tested, and the prediction results of PCA-GA-ELM, GA-ELM and ELM model were comparatively analyzed.The prediction results fit better than the other two models with the actual situation.The PCA-GA-ELM model was superior to GA-ELM model and ELM, and it can be effectively applied to mine inflow prediction, which provided scientific references and guidance in mining production.