基于改进遗传算法和Apriori算法的气旋强度预测方法
Cyclone Intensity Prediction Method Based on Improved Genetic Algorithm and Aprior Algorithm
刘健 1江天乐2
作者信息
- 1. 南方海洋科学与工程广东省实验室(珠海) 珠海 519080
- 2. 复旦大学 上海 200433
- 折叠
摘要
为了提高气旋强度预测的准确度,在遗传算法和多支持度阈值的Apriori算法基础上提出了气旋预测算法GA-MSApriori.利用滑动窗口对历史气旋监测数据进行处理,经过处理后的数据能够描述各监测要素值所处范围以及要素值在过去时间内的变化情况,并以此作为输入数据,最终输出预测准确度较高的气旋强度预测规则.相较于传统的预测方法,气旋强度的预测准确度提高了6%～10%.
Abstract
In order to improve the accuracy of cyclone intensity prediction,GA-MSApriori algorithm is proposed based on the genetic algorithm and Apriori algorithm with multiple minimum supports.The sliding window is used to process the historical cy-clone monitoring data.Taking the processed data that can describe the range and change of the element value in the past time as the input data,and output the cyclone intensity prediction rules with high prediction accuracy finally.Compared with the traditional pre-diction method,the prediction accuracy is improved by 6%～10%.
关键词
气旋强度/遗传算法/Apriori/GA-MSAprioriKey words
cyclone intensity/genetic algorithm/Apriori/GA-MSApriori引用本文复制引用
基金项目
国家科技重大专项(2016YFC1400304)
出版年
2024
国家科技期刊平台
NSTL
万方数据