摘要
提出了一种基于粗集约简的粒子群储层识别方法,即应用粗糙集进行属性约简,应用粒子群(PSO)聚类算法对约简和正规化后的数据进行处理。实验表明,约简后的PSO聚类较约简前在识别率上有明显的提高。
The unconventional reservoir identification is presented based on rough set attribute reduction and particle swarm optimization(PSO), which means utilizing the rough set attribute reduction approach to reduce data space and using PSO clustering algorithm to deal with processing normalized data. Experiment shows that identification rate of the unconventional reservoir with reduced attributes is much higher than all feature attributes in PSO clustering algorithm.
出处
《湖南工业大学学报》
2008年第5期46-48,共3页
Journal of Hunan University of Technology
基金
国家自然科学基金资助项目(70573101)
关键词
粒子群算法
属性约简
粗糙集
聚类
particle swarm optimization
attribution reduction
rough set
clustering