基于TrAdaBoost的冷水机组故障迁移诊断
摘要：冷水机组故障可以通过机器学习进行诊断，但需大量训练数据，而获取有效故障数据难度大、成本高。传统故障诊断主要针对单台机组已有数据，很难覆盖全部工况，新工况下诊断性能恶化。本文提出了用对数据进行空间挤压的多重数据处理方法缩减不同分布间的差异，并利用TrAdaBoost算法对不同数据分布的信息迁移能力，结合不同基分类器搭建了冷水机组故障诊断模型，实现了新工况故障的有效诊断，有望缩减实验成本。对冷水机组7类典型故障的诊断结果显示：在新工况数据仅为20组时，相比于未进行迁移诊断的情况，总体正确率分别提升了22.00%、2.50%和32.33%。通过增补2个工况数据验证了不同模式下迁移诊断对冷水机组故障诊断的有效性：单模式下迁移诊断性能较常规诊断提高18.39%～22.43%,全模式下提高1.21%～2.55%;参数寻优对单模式迁移诊断有辅助提升效果(3.06%),对全模式则因过拟合导致性能下降(-4.23%)。可见，基于源工况知识与目标工况少量数据的迁移诊断模型是解决新工况数据缺乏问题的有效途径。
关键词：冷水机组机器学习信息迁移TrAdaBoost算法基分类器故障诊断
尊敬的用户，本篇文章需要2元，点击支付交费后阅读
限时优惠福利：领取VIP会员
全年期刊、VIP视频免费!
全年期刊、VIP视频免费!
参考文献[1] 崔治国，曹勇，魏景姝，等.数据挖掘技术在建筑暖通空调领域的研究应用进展[J].建筑科学，2018,34(4):85- 97.
[2] 吴孔瑞，韩华，任正雄，等.基于XGBoost-RF的制冷剂泄漏故障检测与诊断[J].暖通空调，2023,53(1):105- 111,130.
[3] 周玉焰，李博，孙晓峰，等.持续调适在绿色建筑运营中的应用[J].暖通空调，2021,51(5):111- 118,78.
[4] PINTO G,WANG Z,ROY A,et al.Transfer learning for smart buildings:a critical review of algorithms,applications,and future perspectives[J].Advances in applied energy,2022,5:100084.
[5] CHEN Y J,TONG Z M,ZHENG Y Z,et al.Transfer learning with deep neural networks for model predictive control of HVAC and natural ventilation in smart buildings[J].Journal of cleaner production,2020,254:119866.
[6] FAN C,SUN Y J,XIAO F,et al.Statistical investigations of transfer learning-based methodology for short-term building energy predictions[J].Applied energy,2020,262:114499.
[7] QIAN F Y,GAO W J,YANG Y W,et al.Potential analysis of the transfer learning model in short and medium-term forecasting of building HVAC energy consumption[J].Energy,2019,193:116724.
[8] ZHU X,CHEN K,ANDUV B,et al.Transfer learning based methodology for migration and application of fault detection and diagnosis between building chillers for improving energy efficiency[J].Building and environment,2021,200:107957.
[9] 范雨强，崔晓钰，韩华，等.不平衡数据技术在冷水机组故障诊断中的应用[J].工程热物理学报，2019,40(6):1219- 1228.
[10] 徐晶.基于小波包和迁移学习的飞机燃油泵故障诊断[J].液压与气动，2020(6):183- 188.
[11] FREUND Y,SCHAPIRE R E.A decision-theoretic generalization of on-line learning and an application to boosting[J].Journal of computer and system sciences,1997,55(1):119- 139.
[12] HASTIE T.Ridge regularization:an essential concept in data science[EB/OL].(2020-05-30)[2022-12-15].http://doi.org/10.48550/arXiv.2006.00371.
[13] 董红瑶，王弈丹，李丽红.随机森林优化算法综述[J].信息与电脑(理论版),2021,33(17):34- 37.
[14] 窦海波，陈琼环，杨宜楠，等.深圳某数据中心节能运营优化实践总结[J].暖通空调，2023,53(1):112- 116.
[15] COMSTOCK M C,BRAUN J E.Development of analysis tools for the evaluation of fault detection and diagnostics in chillers[R].Atlanta:ASHRAE Inc.,1999:26- 53.
[16] 李祺尧.TrAdaBoost算法在结构化数据异常点的改进[D].成都：西南财经大学，2020:17- 21.
[17] 郭亚军，易平涛.线性无量纲化方法的性质分析[J].统计研究，2008,25(2):93- 100.
[18] 徐艳平.基于改进的随机森林算法的城市空气质量预测模型：以重庆市为例[D].重庆：重庆工商大学，2021:51- 54.
[2] 吴孔瑞，韩华，任正雄，等.基于XGBoost-RF的制冷剂泄漏故障检测与诊断[J].暖通空调，2023,53(1):105- 111,130.
[3] 周玉焰，李博，孙晓峰，等.持续调适在绿色建筑运营中的应用[J].暖通空调，2021,51(5):111- 118,78.
[4] PINTO G,WANG Z,ROY A,et al.Transfer learning for smart buildings:a critical review of algorithms,applications,and future perspectives[J].Advances in applied energy,2022,5:100084.
[5] CHEN Y J,TONG Z M,ZHENG Y Z,et al.Transfer learning with deep neural networks for model predictive control of HVAC and natural ventilation in smart buildings[J].Journal of cleaner production,2020,254:119866.
[6] FAN C,SUN Y J,XIAO F,et al.Statistical investigations of transfer learning-based methodology for short-term building energy predictions[J].Applied energy,2020,262:114499.
[7] QIAN F Y,GAO W J,YANG Y W,et al.Potential analysis of the transfer learning model in short and medium-term forecasting of building HVAC energy consumption[J].Energy,2019,193:116724.
[8] ZHU X,CHEN K,ANDUV B,et al.Transfer learning based methodology for migration and application of fault detection and diagnosis between building chillers for improving energy efficiency[J].Building and environment,2021,200:107957.
[9] 范雨强，崔晓钰，韩华，等.不平衡数据技术在冷水机组故障诊断中的应用[J].工程热物理学报，2019,40(6):1219- 1228.
[10] 徐晶.基于小波包和迁移学习的飞机燃油泵故障诊断[J].液压与气动，2020(6):183- 188.
[11] FREUND Y,SCHAPIRE R E.A decision-theoretic generalization of on-line learning and an application to boosting[J].Journal of computer and system sciences,1997,55(1):119- 139.
[12] HASTIE T.Ridge regularization:an essential concept in data science[EB/OL].(2020-05-30)[2022-12-15].http://doi.org/10.48550/arXiv.2006.00371.
[13] 董红瑶，王弈丹，李丽红.随机森林优化算法综述[J].信息与电脑(理论版),2021,33(17):34- 37.
[14] 窦海波，陈琼环，杨宜楠，等.深圳某数据中心节能运营优化实践总结[J].暖通空调，2023,53(1):112- 116.
[15] COMSTOCK M C,BRAUN J E.Development of analysis tools for the evaluation of fault detection and diagnostics in chillers[R].Atlanta:ASHRAE Inc.,1999:26- 53.
[16] 李祺尧.TrAdaBoost算法在结构化数据异常点的改进[D].成都：西南财经大学，2020:17- 21.
[17] 郭亚军，易平涛.线性无量纲化方法的性质分析[J].统计研究，2008,25(2):93- 100.
[18] 徐艳平.基于改进的随机森林算法的城市空气质量预测模型：以重庆市为例[D].重庆：重庆工商大学，2021:51- 54.
Chiller fault transfer diagnosis based on TrAdaBoost
Abstract: Chiller faults can be diagnosed by machine learning, but it needs a lot of training data, and it is difficult and costly to obtain effective fault data. Traditional fault diagnosis is mainly based on the existing data of a single unit, which is difficult to cover all working conditions, and the diagnostic performance deteriorates under new working conditions. In this paper, a multiple data processing method of data space extrusion is proposed to reduce the differences between different distributions. Moreover, the information transfer ability of the TrAdaBoost algorithm for different data distributions is utilized to build a chiller fault diagnosis model combined with different base classifiers, which realizes effective fault diagnosis in new working conditions and is expected to reduce the experiment cost. The diagnosis results of seven typical faults of chillers show that when the data of new working condition is only 20 groups, the overall accuracy increases by 22.00%, 2.50% and 32.33%, respectively, compared with the case without transfer diagnosis. By adding two working conditions data to verify the effectiveness of migration diagnosis for chiller fault diagnosis under different modes, the performance of transfer diagnosis under single mode is improved by 18.39% to 22.43% compared with conventional diagnosis, and the performance under full mode is improved by 1.21% to 2.55%. Parameter optimization can help improve single-mode migration diagnosis(3.06%), but can decrease the performance of full-mode migration diagnosis(-4.23%) due to overfitting. It can be seen that the transfer diagnosis model based on the source working condition knowledge and a small amount of target working condition data is an effective way to solve the lacking problem of the new working condition data.
Keywords: chiller; machine learning; information transfer; TrAdaBoost algorithm; base classifier; fault diagnosis;
311
0
0