Evaluation of Prediction Model for Compressor Performance Using Artificial Neural Network Models and Reduced-Order Models

IF 3 4区工程技术 Q3 ENERGY & FUELS

Energies Pub Date : 2024-07-26 DOI:10.3390/en17153686

Hosik Jeong, Kanghyuk Ko, Junsung Kim, Jongsoo Kim, Seongyong Eom, Sang-Kwon Na, Gyungmin Choi

{"title":"Evaluation of Prediction Model for Compressor Performance Using Artificial Neural Network Models and Reduced-Order Models","authors":"Hosik Jeong, Kanghyuk Ko, Junsung Kim, Jongsoo Kim, Seongyong Eom, Sang-Kwon Na, Gyungmin Choi","doi":"10.3390/en17153686","DOIUrl":null,"url":null,"abstract":"In order to save the time and material costs associated with refrigeration system performance evaluations, a reduced-order model (ROM) using highly accurate numerical analysis results and some experimental values was developed. To solve the shortcomings of these traditional methods in monitoring complex systems, a simplified reduced-order system model was developed. To evaluate the performance of the refrigeration system compressor, the temperature of several points in the system where the compressor actually operates was measured, and the measured values were used as input values for ROM development. A lot of raw data to develop a highly accurate ROM were acquired from a VRF system installed in a building for one year, and in this study, specific operating conditions were selected and used as input values. In this study, the ROM development process can predict the performance of compressors used in air conditioning systems, and the research results on optimizing input data required for ROM generation were observed. The input data are arranged according to the design of experiments (DOE), and the accuracy of ROM according to data arrangement is compared through the experiment results.","PeriodicalId":11557,"journal":{"name":"Energies","volume":null,"pages":null},"PeriodicalIF":3.0000,"publicationDate":"2024-07-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Energies","FirstCategoryId":"5","ListUrlMain":"https://doi.org/10.3390/en17153686","RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"ENERGY & FUELS","Score":null,"Total":0}

引用次数: 0

Abstract

In order to save the time and material costs associated with refrigeration system performance evaluations, a reduced-order model (ROM) using highly accurate numerical analysis results and some experimental values was developed. To solve the shortcomings of these traditional methods in monitoring complex systems, a simplified reduced-order system model was developed. To evaluate the performance of the refrigeration system compressor, the temperature of several points in the system where the compressor actually operates was measured, and the measured values were used as input values for ROM development. A lot of raw data to develop a highly accurate ROM were acquired from a VRF system installed in a building for one year, and in this study, specific operating conditions were selected and used as input values. In this study, the ROM development process can predict the performance of compressors used in air conditioning systems, and the research results on optimizing input data required for ROM generation were observed. The input data are arranged according to the design of experiments (DOE), and the accuracy of ROM according to data arrangement is compared through the experiment results.

查看原文本刊更多论文

使用人工神经网络模型和降序模型评估压缩机性能预测模型

为了节省与制冷系统性能评估相关的时间和材料成本，利用高精度的数值分析结果和一些实验值开发了一种简化阶模型（ROM）。为了解决这些传统方法在监测复杂系统方面的不足，我们开发了一个简化的降阶系统模型。为了评估制冷系统压缩机的性能，对系统中压缩机实际运行的几个点的温度进行了测量，并将测量值作为 ROM 开发的输入值。从安装在一栋建筑中一年的 VRF 系统中获取了大量用于开发高精度 ROM 的原始数据，并在本研究中选择了特定的运行条件作为输入值。在这项研究中，ROM 的开发过程可以预测空调系统中使用的压缩机的性能，并观察了生成 ROM 所需的输入数据的优化研究成果。根据实验设计（DOE）安排输入数据，并通过实验结果比较根据数据安排生成的 ROM 的准确性。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Energies ENERGY & FUELS-

CiteScore

6.20

自引率

21.90%

发文量

8045

审稿时长

1.9 months

期刊介绍： Energies (ISSN 1996-1073) is an open access journal of related scientific research, technology development and policy and management studies. It publishes reviews, regular research papers, and communications. Our aim is to encourage scientists to publish their experimental and theoretical results in as much detail as possible. There is no restriction on the length of the papers. The full experimental details must be provided so that the results can be reproduced.