基于人工神经网络技术的声舒适度评价预测研究  

RESEARCH ON EVALUATION AND PREDICTION OF ACOUSTIC COMFORT BASED ON ARTIFICIAL NEURAL NETWORK TECHNOLOGY DENG

作　　者：邓凌桐 陈曦[2] 康健[1,3] Ling-tong;CHEN Xi;KANG Jian(Harbin Institute of Technology,Key Laboratory of Cold Region Urban and Rural Human Settlement Environment Science and Technology,Ministry of Industry and Information Technology,150001,Harbin,China;Heilongjiang Institute of Technology,150050,Harbin,China;University College London,WC1H 0NN,London,UK)

机构地区：[1]哈尔滨工业大学建筑学院,寒地城乡人居环境科学与技术工业和信息化部重点实验室,哈尔滨150001 [2]黑龙江工程学院,哈尔滨150050 [3]英国伦敦大学学院,英国伦敦WC1H 0NN

出　　处：《建筑技术》2020年第5期627-632,共6页Architecture Technology

基　　金：中国自然科学基金(51778169);政府间科技合作项目(6-19)。

摘　　要：通过对我国3个城市滨海空间进行现场调查和客观实测的方法收集数据,采用人工神经网络技术和多元线性回归模型的方法,以滨海城市开放空间的声环境条件、城市物理环境特征和使用者个体特征来预测使用者对滨海城市开放空间整体声环境的评价.结果表明,构建模型的预测效果均可接受,性别因素和周围人群因素显著影响模型预测结果.预测准确率表明,人工神经网络模型更适合进行声环境的预测.Collecting data through on-site investigation and objective measurement of coastal cities in three cities in China,using artificial neural network technology and multiple linear regression model methods characteristics to predict the user's evaluation of the overall acoustic environment of open spaces in coastal cities.The results show that the prediction effect of the model is acceptable,and gender factors and surrounding crowd factors significantly affect the model prediction results.The prediction accuracy shows that the artificial neural network model is more suitable for the prediction of acoustic environment.

关 键 词：城市滨海区域 声环境 声舒适度 主观评价 预测模型 

分 类 号：TU473[建筑科学—结构工程]