为更好地表达降雨量的空间分布,将粒子群算法(PSO)优化后的反向传输(BP)神经网络分别运用于三峡区间流域日、月和年降雨量的空间插值中,并与单纯BP神经网络和克里金的插值效果作对比。研究结果表明:在日和年的时间尺度上,PSO-BP插值性能较BP有明显改善,且优于克里金的插值效果;在月时间尺度上,PSO-BP插值效果与BP接近且优于克里金。因此,PSO-BP能较好地揭示降雨量在空间的分布规律,也具备在不同时间尺度上对降雨量进行空间插值的能力,是一种较优的降雨空间插值方法。
Abstract
To better describe the spatial distribution of rainfall, we applied BP neural network optimized by particle swarm optimization to the daily, monthly and yearly rainfall spatial interpolation of the Three Gorges reservoir area, and compared the performance with those of simple BP and Kriging interpolation. We found that in daily and yearly time-scale, PSO-BP neural network performs better than BP and Kriging; while in terms of monthly time-cale, PSO-BP result is close to BP and better than Kriging. We conclude that BP neural network optimized by particle swarm optimization could better reveal the law of spatial distribution of rainfall and has the ability of spatial interpolation in different timescales, and therefore is an excellent method for rainfall spatial interpolation.
关键词
粒子群算法 /
BP神经网络 /
优化 /
克里金插值 /
降雨插值
Key words
particle swarm optimization /
BP neural network /
optimization /
Kriging interpolation /
rainfall interpolation
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] 胡广义.分布式降雨量估算模型与方法研究[D].武汉:华中科技大学,2009.[2] 朱 蕾,黄敬峰.山区县域尺度降水量空间插值方法比较[J].农业工程学报,2007,23(7):80-85.[3] 张余庆,陈昌春,姚 鑫,等.江西省信江流域极端降水时空变化特征[J].水土保持研究,2015,22(4):189-194,200.[4] 汤国安,杨 昕,等.地理信息系统空间分析实验教程[M].2版.北京:科学出版社,2012.[5] PIAZZA A D, CONTI F L, NOTO L V, et al.Comparative Analysis of Different Techniques for Spatial Interpolation of Rainfall Data to Create a Serially Complete Monthly Time Series of Precipitation for Sicily, Italy[J]. International Journal of Applied Earth Observation and Geoinformation, 2011, 13(3): 396-408.[6] SEO Y M, KIM S W, VIJAY P. Estimating Spatial Precipitation Using Regression Kriging and Artificial Neural Network Residual Kriging(RKNNRK)Hybrid Approach[J].Water Resources Management, 2015, 29(7): 2189-2204.[7] 莫 林,张秋文.人工神经网络在降水量空间插值中的应用研究[J].计算机与数字工程,2007,35(9):9-12.[8] 胡广义,张秋文,张勇传.基于BP人工神经网络的分布式降雨量插值估算[J].华中科技大学学报(自然科学版),2009,37(4):107-110.[9] 胡广义,张秋文,张勇传.GA优化的BPNN模型在分布式降雨量插值中的应用[J].武汉大学学报(工学版),2009,42(4):466-469.[10]SETIONO R. On the Solution of the Parity Problem by a Single Hidden Layer Feedforward Neural Network[J]. Neurocomputing, 1997, 16(3): 225-235.[11]JOU C, YOU S S, CHANG L W.Analysis of Hidden Nodes for Multi-layer Perceptron Neural Networks[J]. Pattern Recognition, 1994, 27(6): 859-864.[12]王小川,史 峰,郁 磊. MATLAB神经网络43个案例分析[M] .北京:北京航空航天大学出版社,2013.[13]KENNEDY J, EBERHART R. Particle Swarm Optimization[C]∥Proceedings of the 1995 IEEE International Conference on Neural Networks. Perth, Australia, November 27-December 1, 1995: 1942-1948.[14]SHI Yu-hui, EBERHART R C. Parameter Selection in Particle Swarm Optimization[C]∥Lecture Notes in Computer Science: International Conference on Evolutionary Programming. California, USA, March 25-27, 1998: 591-600.[15]SHI Y,EBETHART R.C. A modified Particle Swarm Optimizer[C]∥Proceedings of the IEEE Congresson Evolutionary Computation.Piscataway:IEEE,1998:69-73.[16]IMIRE C E, DURUCAN S, KORRE A. River Flow Prediction Using Artificial Neural Networks: Generalization Beyond the Calibration Range[J]. Journal of Hydrology,2000, 233(1/2/3/4): 138-153.
PDF(1737 KB)
