矿井通风风速的准确预测对提高矿井安全性具有重要意义.为提高矿井通风风速预测的准确度,提出一种基于改进遗传算法和BP神经网络的矿井通风风速预测方法.该方法采用前向神经网络建立矿井通风风速预测模型,将排序选择策略和概率生存策略相结合,从而代替传统选择算子,得到一种改进遗传算法.采用改进遗传算法对网络最优权值和阈值进行全局搜索,在此基础上,再利用BP算法进行局部寻优,从而得到网络的权值和阈值.采用矿井工作面的数据作为实验数据进行仿真预测,并与已有的几种模型进行比较,仿真结果表明,该模型提高了矿井通风风速的预测精度.%Accurate wind velocity forecasting plays an important role in guaranteeing safety of the mine.In order to improve the accuracy of wind velocity forecasting of mine ventilation,a method of mine ventilation prediction based on improved genetic algorithm and BP neural network was presented.Meanwhile,by using the forward neural network,the prediction model of mine ventilation was established.In the model,the ranking selection strategy combines with probability survival method is used instead of the traditional selection operator,which obtained an improved genetic algorithm.The improved genetic algorithm is used to search the optimal weights and thresholds of the network.Based on this,BP algorithm was used to find the local optimization and then obtained the weights and threshold value of network.Meanwhile,the data of mine are used as the experimental data to conduct simulation and prediction,and the prediction results are compared with the several models.The forecasting result shows that the model improves the prediction accuracy of mine ventilation velocity.
展开▼
