遗传算法与神经网络相结合的热带气旋强度预报方法试验
An experiment with methods of forecasting tropical cyclones intensity base on combination of the genetic algorithm and artificial neural network
-
摘要: 以1960~2001年共41 a的7月和8月西行进入南海海域的热带气旋样本为基础,采用遗传算法与神经网络相结合的方法,进行了热带气旋强度预报模型的预报建模研究.并根据相同的热带气旋个例,将这种遗传-神经网络热带气旋强度预报模型与气候持续法热带气旋强度预报方法进行对比分析,试验预报结果表明,遗传-神经网络方法具有更好的预报能力.Abstract: An experiment of establishing a forecasting model of tropical cyclones(TCs)intensity was undertaken by using genetic algorithm and artificial neural network in combination,base on the data of TCs over the South China Sea area in July and August respectively for the period of 1960~2001.Furthermore,by making a comparative analysis of the forecasting result between the genetic algorithm combining with artificial neural network and the Climatology and Persistence(CLIPER)methods base on the same TCs sample.It showed that the former method had better capacity of TCs intensity forecasting than the latter.
-
钮学新,藤卫平.热带气旋强度变化预报的一种动力-统计预报方法[J].科学通报,1986,(13):1 003-1 006. 姚才.0103号台风"榴莲"强度变化特征及暴雨成因的分析[J].热带气象学报,2003,19(增刊):12-17. 马雷鸣,端义宏.利用TRMM资料对热带气旋"威马逊"结构及其降水特征的研究[J].海洋学报,2005,27(1):36-44. 姚才,贺海晏,黄明策.华南西部登陆台风的大尺度条件对比分析[J].中山大学学报,2005,44(6):110-113. 陈联寿,孟智勇.我国热带气旋研究十年进展[J].大气科学,2001,25(3):420-431. 何夏江,曾琮,许永锞.热带气旋路径强度预报——SAPC法[J].应用气象学报,1996,7(1):45-52. RUSSELL L E,GLENN G C,PAUL L K.Statistical forecasts of 24,48 and 72 h typhoon and tropical storm instensity change[J].J Appl Meter,1975,14:445-451. MARK D,JOHN K.A Statistical Model for Predicting Tropical Cyclone Intensity Change[C].19th Conference on Hurricanes and Tropical Meterology,1991.521-526. 薛秋芳,燕方杰,范永强,等.台风强度变化的诊断分析和预报[J].气象,1993,19(2):25-29. 冯利华,骆高远.基于模型叠加法的登陆台风强度预报[J].海洋学报,2001,23(1):127-132. 金龙.神经网络气象预报建模理论方法与应用[M].北京:气象出版社,2004.1-8. JIN Long,LUO Ying,LIN Zhen-shan.Comparison of long-term forecasting of June-August rainfall over Changjiang-Huaihe valley[J].Adv Atmos Sci,1997,14(1):87-92. 陈国良,王熙法,庄镇泉,等.遗传算法及其应用[M].北京:人民邮电出版社,1995.28-127. 李敏强.遗传算法和神经网络的结合[J].系统工程理论与实践,1999,19(2):65-69. 曹永强,张道军.非线性智能组合预测模型及应用[J].中国矿业大学学报,2004,33(4):428-432. 薛福珍,柏洁.基于先验知识神经网络的非线性建模与预测控制[J].系统仿真学报,2004,16(5):1 057-1 059. ZHENG De-ling,LIANG Rui-xin,ZHOU Ying,et al,A chaos genetic algorithm for optimizing an artificial neural network of predicting silicon content in hot metal[J].Journal of University of Science and Technology Beijing,2003,10 (2):68-71. 洪露,马长山.基于遗传算法的神经网络权值优化[J].贵州工业大学学报,2003,32(6):48-51. 谢冰,戴盛,谢科湛.基于遗传神经网络的工业股票指数预测[J].湖南大学学报,2004,18(6):79-64. 李鸿雁,侯光明.提高洪水智能预报中洪峰预报精度方法的研究[J].自然灾害学报,2004,13(4):128-134. BLANCO A,DELGADOM,PEGALAJARMC.A real-coded genetic algorithm for training recurrent neural networks[J].Neural Networks,2001,14:93-105. ARNOLD F S.The merging of neural net-works,fuzzy logic and genetic algorithms[J].Insurance:Mathematics and Economics,2002,31:115-131. 金龙,况雪源,黄海洪,等.人工神经网络预报模型的过拟合研究[J].气象学报,2004,62(1):62-70. SIM D A,CHARLES R S.On the predictability of tropical cyclone tracks in the northwest Pacific basin[J].Monthly Weather Review,2003,131(7):1 491-1 497. SIM D A.Five-day tropical cyclone track forecasts in the north Atlantic basin[J].Weather and Forecasting,1998,13(4):1 005-1 015. 上海台风研究所.2004年热带气旋定位和预报精度评定.第五届全国台风及海洋气象专家工作组第三次会议文件汇编.
点击查看大图
计量
- 文章访问数: 718
- HTML全文浏览量: 4
- PDF下载量: 1383
- 被引次数: 0