【作者】 姚克娟；

【导师】 李晋宏；

【作者基本信息】 北方工业大学 ， 计算机应用技术， 2005， 硕士

【摘要】 随着信息技术的不断发展,人们积累了大量的数据,迫切需要将数据转化为信息。基于数据仓库的决策支持技术包括数据仓库、OLAP和数据挖掘,这些技术构成了新一代决策支持技术的三大支柱,解决了企业面临的战略性的问题。 本文主要探讨了数据仓库和OLAP技术的基本理论和实施方法,并探索了基于统计决策的模式识别理论及时间序列预测算法。结合铝用碳素回转窑生产实际,将模式识别中简单距离分类算法改进并运用于回转窑窑炉异常特征提取及检测,将时间序列预测算法应用于窑炉异常预测,开发了铝用碳素回转窑生产数据多维分析系统—CrkAS(carbon rotary kiln production data online analytical system)。 完成的工作主要有:从决策分析需求出发构建铝用碳素回转窑数据仓库CrkDw(carbon rotary kiln data warehouse);提取窑炉温度异常特征,分析异常并进行预警及故障分析;基于Microsoft.NET平台,利用OLAP多维分析技术开发基于Web的多维分析系统,实现对碳素回转窑生产数据进行钻取、切片、切块、旋转等分析;实现生产及管理中各种报表的自动生成及分析统计。 该系统能够有效指导生产、规范碳素生产管理,避免重大事故,为科学化管理提供依据,在实际推广应用中受到用户的好评。更多还原

【Abstract】 With the development of information technique, a large quantity of data are gathered .People urgently want to find a way to transform data into useful information. The following important techniques: data warehouse, OLAP, data mining build up decision support technique based on data warehouse that solves the strategic problems to be faced by enterprises.In this paper, I not only introduce the fundamental theory and technologies of data warehouse and OLAP but also do some research on pattern recognition based on statistical theory and time series pattern and its application of predictive model algorithm. Combined with the practice in production of carbon rotary kiln for aluminum, I has improved the simple distance for classify algorithm and applied it to collecting and detecting in abnormal character of carbon rotary kiln, and besides, I has applied the time series pattern predictive algorithm to forecast abnormal working state of carbon rotary kiln. I put emphasis on the design and implementation of a multidimensional analytical system for carbon rotate kiln production data. In the system, I mainly realize the following functions: to construct suitable data warehouse based on decision and analysis requirement; based on the algorithm of detecting and predicting abnormal state, to analyze abnormal data and give a alarm to users in order to avoid seriously production accident; to make good use of OLAP technique to implement drilling, slicing, rotating so that improve the ability of scientific decision and analysis; to automatic make various reports in production and management and make statistic of these reports.The system is the part of the large scale aluminum electrolysis pot information integration and control system which has been awarded the science and technology creative excellent prize of 2004 NingXia hui Autonomy district. After the system is put into practice, it has played an important role in a series of fields: instructing production, standardizing carbon rotary kiln production, avoiding serious accident, improving the ability of scientific management and received high praise from users.更多还原