权益分类	功能权益	普通用户	{{item.name}}会员
{{category.name}}	{{benefitItem.name}}

Fundamental Research on Intelligent Control of Fusion reactors

聚变反应堆智能控制基础研究

基本信息

批准号：
09480111
负责人：
YAMAZAKI Kozo
金额：
$ 4.61万
依托单位：
National Institute for Fusion Science
依托单位国家：
日本
项目类别：
Grant-in-Aid for Scientific Research (B)
财政年份：
1997
资助国家：
日本
起止时间：
1997 至 1999
项目状态：
已结题

来源：
https://kaken.nii.ac.jp/grant/KAKENHI-PROJECT-09480111/
关键词：
Intelligent Control Fusion Plasma Sensor Fusion Large Physics Experimental Control Neural Network Fussy Control センサーフュージヨン

项目摘要

Recently, there is a great progress on intelligent control concepts, and there are many trials of application of Fuzzy logic, neural network, chaos theory to the atomic reactor control and plasma diagnostics/control. Moreover, GA (genetic algorithm) is also being applied to control systems. The authors have being making researches on fuzzy logic control of toroidal plasmas, and now are going to develop wide "sensor fusion" concept for unifying engineering schemes.This research aims at the survey on intelligent control schemes, and checks the possibility of applying them to fusion plasma control. To clarify the issues of application of these intelligent method to fusion power plant reactors is also one of purposes of this research.In JFY 1997, general survey on various intelligent control systems are carried out and are tried to be applied to plasma control and reactor system design. In order to make research wide, the author attended at the International Conference on Accelerators and Large Experimental held in Beijing.In JFY 1998, researches on the software are carried out and the feedback algorithm is checked related to the international conference on plasma control in USA.Moreover, the author presented his research activities in the International Conference PCaPAC' 99 held in Tsukuba.JFY 1999 is the final year of this grand-in-research, and the author presented the control results related to LHD, active control of MHD mode and design work on future fusion reactors in the relevant workshops.

近年来，智能控制的概念有了很大的发展，模糊逻辑、神经网络、混沌理论在原子反应堆控制和等离子体诊断/控制中的应用有了很多尝试。此外，遗传算法（GA）也被应用于控制系统。作者一直在研究环形等离子体的模糊逻辑控制，现在准备发展广泛的“传感器融合”概念，以统一工程方案，本研究旨在综述智能控制方案，并检验它们应用于聚变等离子体控制的可能性。阐明这些智能控制方法在聚变电站反应堆中的应用问题也是本研究的目的之一，1997年日本财政年度对各种智能控制系统进行了综述，并试图将其应用于等离子体控制和反应堆系统设计。为了扩大研究范围，作者参加了在北京召开的国际加速器和大型实验室会议，并于1998年参加了在美国召开的国际等离子体控制会议，对软件进行了研究，对反馈算法进行了校核。作者在筑波召开的PCaPAC' 99国际会议上介绍了他的研究活动。JFY 1999是这个大研究的最后一年，并在有关研讨会上介绍了与LHD有关的控制结果、MHD模式的主动控制以及未来聚变堆的设计工作。