基于关键特征的制冷剂泄漏故障软测量研究  

Soft Measurement of Refrigerant Leakage Based on Key Features

作　　者：凌敏彬 杨钰婷 韩华[1] 徐玲 崔晓钰[1] Ling Minbin;Yang Yuting;Han Hua;Xu Ling;Cui Xiaoyu(School of Energy and Power Engineering,University of Shanghai for Science and Technology,Shanghai,200093,China;Carrier Air Conditioning&Refrigeration R&D Management(Shanghai)Co.,Ltd.,Shanghai,200436,China)

机构地区：[1]上海理工大学能源与动力工程学院,上海200093 [2]开利空调冷冻研发管理(上海)有限公司,上海200436

出　　处：《制冷学报》2025年第2期145-154,共10页Journal of Refrigeration

基　　金：国家自然科学基金(51506125)资助项目。

摘　　要：针对制冷剂泄漏难以直接测量的问题,建立基于数据挖掘和关键特征的制冷剂泄漏故障软测量研究。通过随机森林重要性排序和距离相关系数对制冷剂泄漏故障的表征特征进行筛选,建立支持向量回归(SVR)软测量模型对泄漏进行定量测量。经一台额定制冷量为1440 kW、充注量为330 kg螺杆式冷水机组泄漏实验验证,基于3个表征特征建立的SVR软测量模型在测试集上的均方根误差(RMSE)和平均绝对误差(MAE)分别为0.844 kg和0.734 kg,软测量性能较其它3个特征子集显著提升。Refrigerant leakage is a frequent and costly fault that deteriorates the normal operation of a chiller;however,it is difficult to measure directly.This study proposes a data mining-and key-feature-based approach for the soft measurement of refrigerant leakage.Random forest importance ranking and distance correlation coefficients were used to select the characteristic features,and a support vector regression(SVR)soft measurement model was established to measure leakage quantitatively.The proposed model was validated through a leakage experiment conducted on a screw chiller with a rated cooling capacity of 1440 kW and a refrigerant charge of 330 kg.The results showed that the SVR soft measurement model established on the three selected key features achieved significantly improved performance.The model had a root mean square error(RMSE)of 0.844 kg and a mean absolute error(MAE)of 0.734 kg,outperforming the other three feature subsets.

关 键 词：制冷剂泄漏 特征选择 软测量 随机森林 支持向量回归 

分 类 号：TB612[一般工业技术—制冷工程] TB657