摘要
为了实时、准确地判断净油储罐运行状态,本文提出了一种基于聚类分析和专家系统相结合的状态模式识别方法,有效地解决了传统判别方法存在的误判问题,并且利用离散Hopfield网络对联合站所有净油储罐的运行过程故障状态进行了正确识别,确保了原油集输系统的稳定、安全、高效运行,提高了整个原油集输系统的工作效率。实践表明,本文提出的状态识别与故障诊断方法具有很好实用价值。
In order to accurately discriminate the real-time operating status of net crude-oil tanks, a new method based on cluster analysis associated with expert system is proposed, which efficiently solves the traditional misjudgement problems in production. Discrete Hopfield neural network (DHNN) is used to accurately identify the fault status in operating process of all the net crude-oil tanks in a combined station, which ensures the crude oil production and transportation system to operate stably, safely, and effectively. Therefore, the operating efficiency of the whole crude oil production and transportation system is improved. Production practice shows that the proposed status dis- crimination and fault diagnosis method has excellent practical value.
出处
《仪器仪表学报》
EI
CAS
CSCD
北大核心
2007年第4期703-707,共5页
Chinese Journal of Scientific Instrument