火灾工况下公路隧道通风智能控制模型研究

Research on Intelligent Ventilation Control Model of Highway Tunnel under Fire Condition

为研究公路隧道火灾工况下的智能通风控制方法,以某隧道右线为研究对象,进行如下研究:1)探索RBF神经网络模型的适用性、网络组成与训练过程,并结合工程实例通过经验公式得到火灾规模与临界风速的关系;2)计算获取51组随机工况下不同火灾发生规模、火灾发生位置、隧道交通量下的风机开启台数数据;3)利用径向基RBF(Radial Basis Function)神经网络算法进行模型训练,共设置30组训练数据与21组检验数据进行预测精度验证。工程实例预测结果表明:1)RBF神经网络算法训练组与检验组的相对误差分别为4.6%与8.7%,预测精度在预设阈值范围内;2)构建的火灾工况下,公路隧道风机智能控制模型有助于提升公路隧道营运期安全水平。

In order to study the intelligent ventilation control method under the fire condition of highway tunnel,the right line of a tunnel is taken as the research object,and the following research is carried out.The applicability,network composition and training process of RBF neural network model are explored,and the relationship between fire scale and critical wind speed is obtained through empirical formula combining with engineering examples.51 groups of data on the number of fans opened are calculated and obtained under different fire scale,fire location and tunnel traffic volume under random conditions.The radial basis function(RBF)neural network algorithm is used to train the model.A total of 30 groups of training data and 21 groups of test data are set to verify the prediction accuracy.The prediction results of engineering examples show that the relative errors of the training group and the inspection group of the RBF neural network algorithm are 4.6%and 8.7%respectively,and the prediction accuracy is within the preset threshold range.The intelligent control model of highway tunnel fan under fire conditions is helpful to improve the safety level of highway tunnel during the operation period.