基于灰色-神经网络联合模型的大型冷却塔风效应预测

基于灰色GM(1,1)模型和BP人工神经网络,建立灰色-神经网络联合的大型冷却塔平均位移和风振系数预测模型。该联合模型增强了预测结果的自适应性和准确性,能解决因气弹模型试验中测点样本数目太少而无法直接建立神经网络预测模型的局限。通过某大型冷却塔气弹模型风洞试验结果的算例分析,表明该组合模型对于平均位移和风振系数的预测结果均与试验结果吻合良好,随后基于已训练的模型给出结构风振反应精细化分析所需的输入参数预测结果。这为冷却塔结构风效应的精细化研究提供了一个新的有效方法。

Prediction on Wind Effects of Large Cooling Towers Based on Grey-Neural Network Joint Model

Based on grey GM(1,1) model and BP artificial neural network, the grey-neural network joint model is established, which is used to predict the displacement and wind induced coefficients for large cooling towers. Using the joint model, the influence caused by little raw data is overcome. Furthermore the self-adaptability and predicting precision for wind-induced responses of large cooling towers are enhanced. Through comparative analysis of the wind-induced responses of domestic large hyperbolic cooling tower in aero-elastic model wind tunnel test, it can be found that the prediction results of wind-induced responses and wind vibration coefficients are in good agreement with the experimental results, which shows good validity and stability of the model, and then input parameters of refined research on wind induced response are predicted. The proposed method provides a new and effective idea for the refined research on wind effects of large cooling towers.