Influence of data preprocessing on neural network performance for reproducing CFD simulations of non-isothermal indoor airflow distribution
Influence of data preprocessing on neural network performance for reproducing CFD simulations of non-isothermal indoor airflow distribution
复制标题
DOI：
10.1016/j.enbuild.2020.110525
复制
发表时间：
2021-01-01
影响因子：
6.7
通讯作者：
Ooka, Ryozo
中科院分区：
文献类型：
--
作者：
Zhou, Qi;Ooka, Ryozo
室内环境对人类的日常生活至关重要。在耦合模拟、逆向设计和系统控制等实际应用方面，对室内环境进行快速且准确的预测是非常必要的。神经网络（NN）是一种流行的机器学习模型，用于在具有非线性关系的目标变量之间建立联系。为了证实神经网络用于快速准确预测室内环境（包括速度和温度分布）的可行性，本研究设定了非等温的有因次情况，并构建了一个神经网络模型，其中输入为边界条件（即入口速度、温度和窗户表面温度），输出为速度和温度分布。采用了各种数据预处理方法，并对结果进行比较，以揭示数据预处理对神经网络性能的影响。结果表明，在大多数情况下，不同的预处理方法可使神经网络具有相似的性能，每种情况的预测时间约为350秒，最大值的预测误差小于10%，平均值的预测误差小于5%。如果不进行数据预处理，可能会出现误差淹没现象，梯度下降算法可能无法在训练过程中降低数量级较小的变量的误差。在不进行数据预处理的情况下对多个变量分别进行预测，其准确性与进行数据预处理后同时预测的准确性相当；然而，应该考虑为分别预测而训练多个神经网络的计算成本。© 2020爱思唯尔有限公司。保留所有权利。
The indoor environment is important to the daily lives of humans. Fast and accurate prediction of indoor environments is desirable with regard to practical applications, such as coupled simulation, inverse design, and system control. Neural network (NN) is a popular machine learning model used to build pings between target variables with nonlinear relations. To confirm the feasibility of an NN for fast accurate prediction of indoor environments (including both velocity and temperature distributions), dimensional non-isothermal cases are set and an NN model is constructed in this study, where the inputs are boundary conditions (i.e. inlet velocity, temperature and window surface temperature) and outputs are velocity and temperature distributions. Various data preprocessing methods are utilized, and results are compared to reveal the impact of data preprocessing on NN performance. The results show that, for most cases, different preprocessing methods can lead to similar NN performances with a prediction time of approximately 350 its for each case and a prediction error of less than 10% for the maximum value and 5% for the mean value. Without data preprocessing, error submergence is likely to occur, the gradient descent algorithm may fail to reduce errors of variables with smaller orders of magnitude during the training process. Separate prediction of multiple variables without data preprocessing achieve accurate predictions as simultaneous prediction with data preprocessing; however, the computation cost for training multiple NNs for separate predictions should be considered. (C) 2020 Elsevier B.V. All rights reserved.