华中科技大学学报（自然科学版）2016，Vol.44Issue(7)：1-4,4.DOI:10.13245/j.hust.160701
基于 PCA-DT的多联机制冷剂充注量故障诊断
PCA-DT based refrigerant charge fault diagnosis for variable refrigerant flow system
摘要
Abstract
In order to solve the problem that variable refrigerant flow refrigerant charge breaks dow n , a fault diagnosis method based on principal component analysis‐decision tree (PCA‐DT ) algorithm was put forward .The raw data was preprocessed at first .Then the PCA method was employed to reduce the dimension of the data .Further ,the data was divided into two sets ,i .e .training set and test set . The decision tree model was developed based on training set to achieve fault detection and diagnosis . Through detection and diagnosis result ,the practicability of the decision trees model was confirmed . Other further test was taken out to check its feasibility ,and practical running data from 3 typical re‐gions was collected .The results show that the proposed method performs better at detection and diag‐nosis to VRF refrigerant charge problem compared to the decision tree model .关键词
多联机/故障/检测与诊断/主成分分析/决策树/制冷剂/充注量Key words
variable refrigerant flow/fault/detection and diagnosis/principal component analysis/decision trees/refrigerant/charge分类
通用工业技术引用本文复制引用
王江宇,陈焕新,刘江岩,李冠男..基于 PCA-DT的多联机制冷剂充注量故障诊断[J].华中科技大学学报（自然科学版）,2016,44(7):1-4,4.基金项目
国家自然科学基金资助项目（51576074，51328602）． （）
