摘要
针对GRNN神经网络预测RC柱耗能能力的方法存在计算复杂度高和空间复杂度高等问题,文中研究得出一种新的评价RC柱耗能能力的指标,提出了INFO-GRNN神经网络的RC柱耗能能力预测方法。研究结果选取212组矩形柱,计算RC柱的滞回环面积和面周系数,结果表明面周系数曲线的预测值与真实值接近,具有较高的应用价值。
Aiming at the problems of high computational complexity and high spatial complexity of GRNN neural network,this paper proposes the prediction method of INFO-CRNN neural network and a new index to evaluate the energy consumption capacity of RC column.212 sets of rectangular columns are selected to calculate the stagnant ring area and circumferential coefficient of RC columns.The results show that the predicted value of the circumferen-tial coefficient curve is close to the true value and has high application value.
作者
徐常文
曾森
周祥
XU Changwen;ZENG Sen;ZHOU Xiang(School of Civil Engineering,Qingdao University of Technology,Shandong Qingdao 266525,China)
出处
《低温建筑技术》
2023年第11期76-80,共5页
Low Temperature Architecture Technology
关键词
钢筋混凝土柱
INFO-GRNN神经网络
面周系数
耗能能力
reinforced concrete column
INFO-GRNN neural network
circumferential coefficients
energy consump-tion capacity
