Training fuzzy logic based software components for reuse

Proceedings 1997 27th International Symposium on Multiple- Valued Logic Pub Date : 1997-05-28 DOI:10.1109/ISMVL.1997.601396

Junda Chen, D. Rine

{"title":"Training fuzzy logic based software components for reuse","authors":"Junda Chen, D. Rine","doi":"10.1109/ISMVL.1997.601396","DOIUrl":null,"url":null,"abstract":"A training framework of an effective method for off-line training of a class of control software components (for example, for first-order nonlinear feedback control systems) using combinations of three kinds of adaptation algorithms is presented. Each control software component is represented at the abstract level by means of a set of adaptive fuzzy logic (FL) rules and at the concrete level by means of fuzzy membership functions (MBFs). At the concrete representation level adaptation algorithms specified for use in adapting MBFs are: genetic algorithms, neural net algorithms, and Monte Carlo algorithms. We specify effective combinations of these three existing adaptation algorithms to train an erroneous FL rule-based software component in the tracker problem. In the framework, training consists of two phases: testing and adapting. In this paper, only the adapting phase is addressed. For each fault scenario adaptation algorithms and their combinations are used to modify the MBFs of the component. Effectiveness of the adapting phase is determined in terms of flexibility, adaptability, and stability. We perform experiments using a genetic algorithm. Simulation results are discussed.","PeriodicalId":206024,"journal":{"name":"Proceedings 1997 27th International Symposium on Multiple- Valued Logic","volume":"15 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"1997-05-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"4","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proceedings 1997 27th International Symposium on Multiple- Valued Logic","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ISMVL.1997.601396","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}

引用次数: 4

Abstract

A training framework of an effective method for off-line training of a class of control software components (for example, for first-order nonlinear feedback control systems) using combinations of three kinds of adaptation algorithms is presented. Each control software component is represented at the abstract level by means of a set of adaptive fuzzy logic (FL) rules and at the concrete level by means of fuzzy membership functions (MBFs). At the concrete representation level adaptation algorithms specified for use in adapting MBFs are: genetic algorithms, neural net algorithms, and Monte Carlo algorithms. We specify effective combinations of these three existing adaptation algorithms to train an erroneous FL rule-based software component in the tracker problem. In the framework, training consists of two phases: testing and adapting. In this paper, only the adapting phase is addressed. For each fault scenario adaptation algorithms and their combinations are used to modify the MBFs of the component. Effectiveness of the adapting phase is determined in terms of flexibility, adaptability, and stability. We perform experiments using a genetic algorithm. Simulation results are discussed.

查看原文本刊更多论文

训练基于模糊逻辑的软件组件以实现重用

提出了一种结合三种自适应算法对一类控制软件组件(例如一阶非线性反馈控制系统)进行离线训练的有效方法的训练框架。每个控制软件组件在抽象层次上用一组自适应模糊逻辑规则表示，在具体层次上用模糊隶属函数表示。在具体的表示级别上，用于自适应mbf的自适应算法有:遗传算法、神经网络算法和蒙特卡罗算法。我们指定了这三种现有自适应算法的有效组合，以训练跟踪器问题中基于错误FL规则的软件组件。在这个框架中，培训包括两个阶段:测试和适应。在本文中，只讨论了适应阶段。针对每个故障场景，采用自适应算法及其组合来修改组件的mbf。适应阶段的有效性取决于其灵活性、适应性和稳定性。我们用遗传算法做实验。对仿真结果进行了讨论。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Proceedings 1997 27th International Symposium on Multiple- Valued Logic

自引率

0.00%

发文量