Journal of Computer Applications ›› 2016, Vol. 36 ›› Issue (9): 2555-2559.DOI: 10.11772/j.issn.1001-9081.2016.09.2555

Previous Articles     Next Articles

Short-term lightning prediction based on multi-machine learning competitive strategy

SUN LiHua1, YAN Junfeng1, XU Jianfeng1,2   

  1. 1. College of software, Nanchang University, Nanchang Jiangxi 330047, China;
    2. College of Electronics and Information Engineering, Tongji University, Shanghai 201804, China
  • Received:2016-04-12 Revised:2016-05-21 Online:2016-09-10 Published:2016-09-08
  • Supported by:
    This work is partially supported by the National Natural Science Foundation of China(61070139, 61273304); the Science and Technology Supported Programme (nckjj2014) of Nanchang City; the Graduate Student Creative Research Funds for the Nanchang University (cx2015097).

基于多机器学习竞争策略的短时雷电预报

孙丽华1, 严军峰1, 徐健锋1,2   

  1. 1. 南昌大学 软件学院, 南昌 330047;
    2. 同济大学 电子与信息工程学院, 上海 201804
  • 通讯作者: 徐健锋
  • 作者简介:孙丽华(1955-),女,江西南昌人,教授,主要研究方向:信息处理、人工智能、粒计算;严军峰(1987-),男,陕西宝鸡人,硕士研究生,主要研究方向:数据挖掘、机器学习;徐健锋(1973-),男,江西南昌人,副教授,博士研究生,主要研究方向:智能信息处理、数据挖掘、粒计算、机器学习。
  • 基金资助:
    国家自然科学基金资助项目(61070139,61273304);南昌市科技支撑计划项目(nckjj2014),南昌大学研究生创新资金资助项目(cx2015097)。

Abstract: The traditional lightning data forecasting methods often use single optimal machine learning algorithm to forecast, not considering the spatial and temporal variations of meteorological data. For this phenomenon, an ensemble learning based multi-machine learning model was put forward. Firstly, attribute reduction was conducted for meteorological data to reduce dimension; secondly, multiple heterogeneous machine learning classifiers were trained on data set and optimal base classifier was screened based on predictive quality; finally, the final classifier was generated after weighted training for optimal base classifier by using ensemble strategy. The experimental results show that, compared with the traditional single optimal algorithm, the prediction accuracy of the proposed model is increased by 9.5% on average.

Key words: lightning forecast, attribute reduction, ensemble learning, machine learning

摘要: 传统的雷电数据预测方法往往采用单一最优机器学习算法,较少考虑气象数据的时空变化等现象。针对该现象,提出一种基于集成策略的多机器学习短时雷电预报算法。首先,对气象数据进行属性约简,降低数据维度;其次,在数据集上训练多种异构机器学习分类器,并基于预测质量筛选最优基分类器;最后,通过对最优基分类器训练权重,并结合集成策略产生最终分类器。实验表明,该方法优于传统单最优方法,其平均预测准确率提高了9.5%。

关键词: 雷电预报, 属性约简, 集成学习, 机器学习

CLC Number: