基于统计学参数与SVM算法的核电风冷式冷水机组高压保护故障研究  

Study on high-pressure protection fault of nuclear power air-cooled chiller based on statistical parameters and SVM algorithm

作　　者：顾明伟 宋海川 Gu Mingwei;Song Haichuan(Shanghai Nuclear Engineering Research and Design Institute;Gree Electric Appliances,Inc.of Zhuhai)

机构地区：[1]上海核工程研究设计院有限公司 [2]珠海格力电器股份有限公司

出　　处：《制冷与空调》2021年第6期42-46,74,共6页Refrigeration and Air-Conditioning

基　　金：国家科技重大专项课题“2018ZX06002003 CAP1400机电一体化关键技术研究”。

摘　　要：采用传统的人工识别方法很难确定核电风冷式冷水机组高压保护故障是源于翅片脏堵还是源于散热不良。本文采用基于统计学参数和机器学习的分析方法对冷水机组高压保护进行分类研究,经数据预处理和模型选择,提出一种基于统计学参数与SVM(支持向量机)算法的核电风冷式冷水机组高压保护故障诊断方法,故障识别率达96%,达到了行业专家分析水平。It is difficult to distinguish the high-pressure protection fault of the nuclear power air-cooled chiller due to dirty fins or poor heat dissipation using traditional manual identification methods.The statistical parameters and machine learning analysis methods are used to classify the high-pressure protection of chiller.After data preprocessing and model selection,a fault diagnose method based on statistical parameters and SVM(support vector machine)algorithm is proposed to deal with the high-pressure protection of nuclear power air-cooled chiller.As result,the fault recognition rate reaches 96%,which achieves the level of industry export analysis.

关 键 词：冷水机组 故障 高压保护 统计学参数 支持向量机 

分 类 号：TB6[一般工业技术—制冷工程]