【摘要】 应用基于人工神经网络技术和正交试验法结合的优化方法,本文对座便器虹吸管道结构参数进行了优化设计。作者运用计算流体动力学仿真技术得到样本,在MATLAB环境中建立神经网络模型,并通过正交试验得到新样本,最终建立神经网络模型,代替数值模拟,并对虹吸管道进行了优化。结果表明,将CFD数值模拟、正交试验、神经网络三者结合用于座便器虹吸管道的基本结构参数设计中,可以缩短优化时间,提高设计效率。更多还原

【Abstract】 This paper raises a method of structure optimization, which is based upon the artificial neural network (ANN) and orthogonal design. Based on the numerical model of water closet, a mathematical model of the sipon pipe system was built with backpropagation (BP) neural network in the MATLAB environment. The optimization was carried out using numerical simulation, orthogonal design and ANN. The result showed an agreement with the experiment.更多还原