新息累积GM(1,N)模型在交通噪声预测中的应用  

Application of New-information Accumulative GM (1, N) Model to Traffic Noise Prediction

作　　者：沈艳[1] 余冬华[1] 李丽萍[1] 

机构地区：[1]哈尔滨工程大学理学院,哈尔滨150001

出　　处：《噪声与振动控制》2013年第3期184-187,共4页Noise and Vibration Control

基　　金：国家自然科学基金(基金号:10632040);理学院自由探索计划(基金号:GK2110260103)

摘　　要：从传统的灰色GM(1,N)模型出发,利用灰色关联分析法确定相关因素的关联度,引入累积法相关理论,对GM(1,N)模型进行参数辨识,建立起多因素的累积GM(1,N)模拟模型,在此基础上,充分利用最新信息,用新息思想建立新息累积GM(1,N)预测模型。将该模型分别应用到南方某城市及北京市道路交通噪声的模拟和预测上,结果表明,所建立的新息累积GM(1,N)模型的模拟精度高,预测结果平均相对误差比GM(1,1)模型还低,预测效果好,预测值还表明,接下来两年内,噪声值基本维持稳定。Based on the traditional grey GM （1, N） model, the grey correlation analysis was used to determine the relation of the related factors. With the theory related to accumulative methodology, the parameter identification of the grey GM （1, N） model was carded out. The multifactor grey GM （1, N） simulation model was built up. On this basis, using the latest information and new information idea, the grey GM（1,N） forecast model was established. This model was then applied to the simulation and forecast of the traffic noise of Beijing and a city in south China, respectively. The result shows that the simulation precision of the new-information accumulative grey GM （1, N） model is high. Its mean relative error of prediction is lower than that of the traditional grey GM （1, N）. The prediction of traffic noise of the two cities shows that the noise value will basically remain stable in the next two years.

关 键 词：声学 累积法 GM(1 N)模型 噪声预测 

分 类 号：X121[环境科学与工程—环境科学] N941.5[自然科学总论—系统科学]