摘要
随着计算机、自动化等技术的不断发展 ,在线监测系统已广泛地应用在企业生产中 ,开发研制智能化实时在线诊断系统已成为发展的必然趋势。粗集理论能有效地分析和处理不精确、不一致、不完整的各种信息 ,并以最简单的形式表示属性间相互影响关系 ,它已逐步应用在各类诊断领域中。由于属性的简约为 NP完全问题 ,这就为时效性要求较高的实时在线诊断算法提出了较高的要求。本文提出了一种高效的时态决策表“核”的计算方法 ,该方法不需要遍历所有的对象 (或过程 ) ,提高了求解速度 ,并可利用前一时态决策表“核”的计算结果经过简单推算即可求得下一时态决策表的“核”。应用遗传算法求解决策表的最小简约 ,根据“核”的信息对遗传操作采取了一系列控制策略 ,从而大大提高了求解效率 ,使基于粗集的实时诊断系统应用在实际生产中成为可能。
The real-time monitoring system has gradually applied in manufacturing process Rough set theory is proved to be an effective method for the analysis of the influencing relations between a set of multi-valued attributes extracted from the monitoring information by the attributes reduction and it has been adopted in diagnostic problem. Since the attributes reduction is an NP problem, the traditional method is difficult to be applied in the practical application of rough set, especially in diagnostic problem on line. In this paper, we present an efficient method of calculating relative core of a temporal decision table, we can get the next temporal relative core according to the last one by simple calculating. A genetic algorithm is introduced to attributes reduction, and a series of control strategies is implemented to improve efficiency when the genetic algorithm is operating based on the information from the relative core. The new search method is more efficiet for the temporal decision table and makes real-time diagnosis possible in reality by rough set.
出处
《机械科学与技术》
CSCD
北大核心
2003年第6期938-941,共4页
Mechanical Science and Technology for Aerospace Engineering
基金
广东省科技计划项目 ( c3 190 3 )资助
关键词
实时诊断
时态决策表
遗传算法
粗集
核
最小简约
Real-time diagnosis
Temporal decision table
Genetic algorithm
Rough set
Core
Attribute reduction