论文部分内容阅读
为提高煤与瓦斯突出强度的预测精度及预测速度,用最大最小蚂蚁系统和BP神经网络相结合的方法进行预测模型设计。根据煤与瓦斯突出强度及其主要影响因素之间的关系数据,建立其神经网络的预测模型。以网络的权值和阈值为自变量,网络误差为目标函数,通过蚁群算法的迭代运算,搜索出误差的全局最小值,以实现BP神经网络的初始权值、阈值优化,并用优化后的网络进行瓦斯突出强度的预测。实例结果表明,MMAS-BP算法的预测值均方差为0.089,约为BP神经网络的0.1倍,且输出稳定性好,适用于煤与瓦斯突出强度的预测。
In order to improve prediction accuracy and prediction speed of coal and gas outburst intensity, a prediction model design is carried out by a combination of maximum and minimum ant systems and BP neural network. According to the relationship between the intensity of coal and gas outburst and its main influencing factors, a prediction model of its neural network is established. Taking the weight of the network and the threshold as the independent variable and the network error as the objective function, the global minimum of the error is searched through the iterative operation of the ant colony algorithm, so as to realize the initial weight and threshold optimization of the BP neural network. Network to predict the intensity of gas outburst. The case study shows that the mean square error of prediction of MMAS-BP algorithm is 0.089, which is about 0.1 times that of BP neural network. The output stability of MMAS-BP algorithm is good and it is suitable for the prediction of coal and gas outburst intensity.