基于Adaboost集成BP网络原煤生产成本预测研究

为了提高传统BP神经网络预测模型精度,避免BP网络容易陷入局部极值、收敛速度慢等问题,将BP神经网络与Ada-boost算法相结合,提出了一种Adaboost集成BP神经网络模型.结合磁县观台煤矿原煤生产成本相关数据,建立了原煤生产成本预测的Adaboost集成BP神经网络模型,将该模型用于实际的原煤成本预测.结果表明：该模型预测精度高于传统的BP神经网络,收敛速度快,具有较强的鲁棒性,预测精度能满足实际预测需要,为原煤生产成本预测提供了一种新的途径,也为原煤生产成本控制提供了重要依据.

Prediction Research of Raw Coal Production Cost Based on Adaboost Integrated BP Network

In order to improve the precision of the prediction model of traditional BP neural network, and avoid the BP network falling easily into local extremum, slow convergence speed and so on, combing BP neural network and Adaboost algorithm, this paper put forward a kind of Adaboost integrated BP neural network model. Combining with the related data of raw coal production cost of Cixian Guantai coal mine, established a prediction model of Adaboost integrated BP network of raw coal production cost, and applied this model into predicting the actual cost of raw coal. The resuhs showed that the model＇s prediction accuracy was higher than that of the traditional BP neural network, and the convergence speed was quicker, and had the stronger robustness, the prediction accuracy could meet the practical prediction needs, provide a new approach to predict the coal production cost, and provide the important basis for the control of the raw coal production cost.