首页 | 本学科首页   官方微博 | 高级检索  
     检索      

预报指标自动归纳和实时自动推理预报系统
引用本文:李开乐.预报指标自动归纳和实时自动推理预报系统[J].海洋预报,1992,9(3):74-79.
作者姓名:李开乐
作者单位:广州中东气象台 广州市
摘    要:本文结合实际应用介绍一种人工智能系统。系统的重要特点是,根据预报指标经验知识的产生过程的机理性知识所设计成功的系统主程序之一的“预报指标自动归纳程序”,能够通过对多因子历史资料的分析归纳,自动生成含有较全面知识规则的规则库。有关专家只须凭感性经验提供可能因子和原始资料,系统便能帮助升华出理性的经验规则,自动建成某一专项业务的应用预报系统。对这种“母系统”的反复使用,就可以根据不同类别的资料生成相应的多种应用系统。既适用于气象上风雨暑寒各类天气预报,也适用于其他某些专业,如水文、地震等的定性预报或分析。 本文除简单介绍系统的构造、功能、设计原理之外,还给出一个应用实例,介绍了如何使用本系统去产生应用性系统的具体方法步骤。

关 键 词:天气预报  指标  归纳  推理预报系统

THE FORECASTING SYSTEM WITH FACTOR INDUCTING AND REASONING FUNCTIONS
Li Kaile.THE FORECASTING SYSTEM WITH FACTOR INDUCTING AND REASONING FUNCTIONS[J].Marine Forecasts,1992,9(3):74-79.
Authors:Li Kaile
Institution:Guangzhou Central Meteorlogical Observatory
Abstract:A forecast system with artificial intelligence'function has been shown in this paper. Its important feature is that the system has a function of automatic induction of forecast factors, simulating the mechanism of men's experience collection. This function makes the system can produce some rule libraries by automatically analysing and inducting historical factors.Different forecast systems can be made by using the "mother system" to different kinds of data, and can be applied in different fields such as weather forecast, hydrological and earthquake qualitative analysis and predictions. The paper describes the structure, function and designed principle of this system simply, and also gives an example how to use this system
Keywords:forecast  regulation  induction  rule-library  inference  
本文献已被 CNKI 维普 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号