| 黄朝强,廖基定,尹邦华.优化灰色GM(1,N)-加权Markov模型在道路交通噪声预测中的精度研究[J].南华大学学报(自然科学版),2019,33(1):35~43.[HUANG Chaoqiang,LIAO Jiding,YIN Banghua.Study on Forecasting Accuracy of Road Traffic Noise by Optimizing Grey GM (1,N)-Weighted Markov Model[J].Journal of University of South China(Science and Technology),2019,33(1):35~43.] |
| 优化灰色GM(1,N)-加权Markov模型在道路交通噪声预测中的精度研究 |
| Study on Forecasting Accuracy of Road Traffic Noise by Optimizing Grey GM (1,N)-Weighted Markov Model |
| 投稿时间：2018-08-11 |
| DOI： |
| 中文关键词: 噪声预测 GM(1,N)预测模型 加权Markov模型 关联度分析 背景值优化 |
| 英文关键词:noise prediction GM(1,N) prediction model weighted Markov model correlation analysis background value optimization |
| 基金项目:湖南省科技厅软科学课题基金资助项目(2010zk3052) |
|
| 摘要点击次数: 557 |
| 全文下载次数: 448 |
| 中文摘要: |
| 通过对北京市2007年至2016年城市道路交通噪声及相关影响因素数据分析,以GM(1,N)模型为基础,建立了优化灰色-加权Markov模型,为有效控制交通噪声污染提供理论依据和决策意见。首先,利用平滑公式对原始数据进行预处理,用数值积分中的Simpson公式改变背景值来提高传统多因素GM(1,N)模型精度。其次,用加权Markov模型对得到的模拟值中的异常值进行了修正,将其应用到城市交通噪声的预测上,实证计算表明优化灰色GM(1,N)模型的模拟值与实际值拟合效果很好,比传统的GM(1,N)模型精度有较大提高。最后,用该模型对北京市2017年和2018年城市交通噪声进行预测,基本符合噪声数据实际变化趋势。 |
| 英文摘要: |
| Based on the analysis of urban road traffic noise and related influencing factors from 2007 to 2016 in Beijing,the study establishes an optimized grey-weighted Markov model based on GM(1,N) model,which provides a theoretical basis and decisive suggestions for effective control of traffic noise pollution.Firstly,it uses the smoothing formula to preprocess the original data,and uses the Simpson formula in numerical integration to change the background value to improve the accuracy of the traditional multi-factor GM(1,N) model.Secondly,the outliers in the simulated values obtained by the weighted Markov model are modified and applied to the prediction of urban traffic noise.Empirical calculation shows that the simulated value of the Optimized Grey GM(1,N) model fits the actual value very well,and the accuracy of the model is much higher than that of the traditional GM(1,N).Finally,the model is used to predict the urban traffic noise in Beijing in 2017 and 2018,and it basically accords with the actual trend of noise data. |
| 查看全文 查看/发表评论 下载PDF阅读器 |
| 关闭 |
|
|
|