摘要
建模、优化、故障诊断是流程工业CIMS技术中的关键技术。传统的建模、优化、故障诊断方法依赖于数学模型仿真或专家经验规则 ,对于强非线性和非高斯分布噪声的对象存在着知识获取瓶颈。而数据挖掘技术综合运用机器学习、计算智能 (人工神经网、遗传算法 )、模式识别、数理统计等技术 ,从大量数据中挖掘和发现有价值和隐含的知识。本文进一步研究了建模、优化、故障诊断的数据挖掘系统 ,以及规则挖掘、参变量优化、故障诊断建模的算法。
Modeling, optimization and fault diagnosis are key techniques in Computer-Integrated Manufacturing System(CIMS). Traditional approaches for modeling, optimization and fault diagnosis are dependent on simulation of mathematical model or experiential rules, they have limitations of knowledge acquisition to deal with those objects with strong nonlinear and non-Gauss noises. Data mining technique has comprehensively utilized machine learning, computational intelligence (artificial neural network, genetic algorithm), pattern recognition, mathematical statistics to mine and discover valuable and hidden knowledge from databases. The data mining system of modeling, optimization and fault diagnosis has been described. The algorithms of rule generalization, variables optimization and modeling for fault diagnosis have been also studied.
基金
国家 8 6 3 /CIMS主题资助项目!(86 3 -5 11-945 -0 0 5 )
国家自然科学基金资助项目
关键词
数据挖掘
故障诊断
CIMS
建模
优化
data mining
fuzzy control
fault diagnosis
artificial neural network
genetic algorithm
