[1]王 可,卢焕达,郑军红,等.基于混合深度学习模型的企业中央空调蒸汽预测[J].南京师范大学学报(工程技术版),2022,22(03):053-62.[doi:10.3969/j.issn.1672-1292.2022.03.008]
　Wang Ke,Lu Huanda,Zheng Junhong,et al.Steam Prediction of Central Air Conditioning Based on Hybrid Deep-learning Model[J].Journal of Nanjing Normal University(Engineering and Technology),2022,22(03):053-62.[doi:10.3969/j.issn.1672-1292.2022.03.008]
点击复制

基于混合深度学习模型的企业中央空调蒸汽预测
分享到：

南京师范大学学报（工程技术版）[ISSN:1006-6977/CN:61-1281/TN]

卷:
22卷
期数:
2022年03期
页码:
053-62
栏目:
计算机科学与技术
出版日期:
2022-09-15

文章信息/Info

Title:
Steam Prediction of Central Air Conditioning Based on Hybrid Deep-learning Model
文章编号:
1672-1292(2022)03-0053-10
作者:
王 可1; 卢焕达2; 郑军红1; 何利力1
(1.浙江理工大学信息学院,浙江 杭州 310018)(2.浙大宁波理工学院计算机与数据工程学院,浙江 宁波 315100)
Author(s):
Wang Ke1; Lu Huanda2; Zheng Junhong1; He Lili1
(1.School of Information Science and Technology,Zhejiang Sci-Tech University,Hangzhou 310018,China)(2.School of Computer and Data Engineering,NingboTech University,Ningbo 315100,China)
关键词:
空调蒸汽消耗; 三维卷积; PredRNN++; 门控循环单元; 空间特征
Keywords:
air conditioning steam consumption; convolutions over volumes; PredRNN++; GRU; spatial feature
分类号:
TP181
DOI:
10.3969/j.issn.1672-1292.2022.03.008
文献标志码:
A
摘要:
为解决企业生产车间多空调能耗与生产任务、气候环境匹配的精准供能问题,实现多台大型中央空调机组蒸汽消耗预测,提出一种基于GRU和3DConv-PredRNN++的混合深度学习预测模型. 针对多台空调机组动态联动关系,使用三维卷积和PredRNN++方法提取机组间蒸汽损耗关系作为空间因素特征参与模型预测; 为捕捉蒸汽消耗量序列的总体趋势和局部变化,数据集采用平滑过程模式、趋势性模式和周期性模式作为模型输入; 为提高模型预测性能,基于门控循环单元(GRU)耦合外部因素特征并捕捉时间因素特征; 最后通过参数矩阵融合方式来构建模型. 通过与多种预测模型的对比实验,证明混合深度学习模型预测精度的优越性和空间因素特征参与模型预测的必要性. 与现有模型相比,所提模型平均能耗折标(ASEC)降低了60.09%.
Abstract:
In order to solve the steam consumption prediction problem of several large central air-conditioning units,a hybrid deep-learning prediction model based on GRU and 3DConv-PredRNN++ is proposed. Aiming at the dynamic linkage relationship of multiple air conditioning units,the relationship between steam consumption of units is extracted by convolutions over volumes and PredRNN++method,which is used as a feature of spatial factors to participate in model prediction. In order to capture the overall trend and local variation of the steam consumption series, smooth process model,trend model and periodic model are used as model inputs in the data set. In order to improve the prediction performance of the model,a GRU is used to combine the external factor features and capture the temporal factor features. Finally,the model is constructed by parameter matrix fusion. Through comparison experiments with various prediction models,it is proved that the prediction accuracy of the model is superior and that it is necessary to extract spatial factors to participate in model prediction. Compared with the current model,the average standard energy consumption of this model is reduced by 60.09%.

参考文献/References:

[1]CHOWDHURY H,CHOWDHURY T,HOSSAIN N,et al. Exergetic sustainability analysis of industrial furnace:a case study[J]. Environmental Science and Pollution Research,2021,28:12881-12888.
[2]杨裔. 中央空调节能技术综述[J]. 现代信息科技,2019(13):193-194.
[3]CHEN X,SHUAI C Y,WU Y,et al. Understanding the sustainable consumption of energy resources in global industrial sector:Evidences from 114 countries[J]. Environmental Impact Assessment Review,2021,90:106609.
[4]刘亚东. 造纸厂蒸汽消耗需求预测研究及预测系统开发[D]. 广州:华南理工大学,2013.
[5]刘文华. 基于机器学习的火力发电蒸汽量预测方法研究[D]. 太原:太原科技大学,2019.
[6]MOGHADASI M,OZGOLI H A,FARHANI F. Steam consumption prediction of a gas sweetening process with methyldiethanolamine solvent using machine learning approaches[J]. International Journal of Energy Research,2021,45(1):879-893.
[7]王梦柯. 大型制造企业蒸汽智能供应策略优化的研究与应用[D]. 杭州:浙江理工大学,2020.
[8]彭桐歆,韩勇,王程,等. 面向短时地铁客流量预测的混合深度学习模型[J]. 计算机工程,2022,48(5):297-305.
[9]ZHANG J B,ZHENG Y,QI D K. Deep Spatio-temporal residual networks for citywide crowd flows prediction[C]//AAAI'17:Proceedings of the Thirty-First AAAI Conference on Artificial Intelligence. San Francisco,USA:AAAI Press,2016:1655-1661.
[10]CHO K,MERRIENBOER B V,GULCEHRE C,et al. Learning phrase representations using RNN encoder-decoder for statistical machine translation[C]//Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing(EMNLP). Doha,Qatar:ACL,2014.
[11]查玉坤,张其林,赵永标,等. 基于三维卷积和CLSTM神经网络的水产养殖溶解氧预测[J]. 应用科学学报,2021,39(4):615-626.
[12]DIBA A,FAYYAZ M,SHARMA V,et al. Temporal 3D ConvNets:new architecture and transfer learning for Video classification[J/OL]. arXiv preprint arXiv:1711.08200,2017.
[13]WANG Y B,GAO Z F,LONG M S,et al. PredRNN++:towards a resolution of the deep-in-time dilemma in spatiotemporal predictive learning[J/OL]. arXiv preprint arXiv:1804.06300,2018.
[14]国家市场监督管理总局,国家标准化管理委员会. 综合能耗计算通则:GB/T 2589-2020[S]. 北京:中国标准出版社,2020.
[15]简毅文,江亿. 住宅供暖空调能耗计算模式的研究[J]. 暖通空调,2005,35(2):11-14.
[16]JALAL M,JALAL H. Retracted:behavior assessment,regression analysis and support vector machine(SVM)modeling of waste tire rubberized concrete[J]. Journal of Cleaner Production,2020,273:122960.
[17]XU L,HOU L,ZHU Z Y,et al. Midterm prediction of electrical energy consumption for crude oil pipelines using a hybrid algorithm of support vector machine and genetic algorithm[J]. Energy,2021,222(1):119955.
[18]EAPEN J,VERMA A,BEIN D. Improved big data stock index prediction using deep learning with CNN and GRU[J]. International Journal of Big Data Intelligence,2021,7(4):202-210.
[19]陈聪,候磊,李乐乐,等. 基于GRU改进RNN神经网络的飞机燃油流量预测[J]. 科学技术与工程,2021,21(27):11663-11673.
[20]党建武,从筱卿. 基于CNN和GRU的混合股指预测模型研究[J]. 计算机工程与应用,2021,57(16):167-174.
[21]LIU L J,WANG L,YU Z. Remaining useful life estimation of aircraft engines based on deep convolution neural network and LightGBM combination model[J]. International Journal of Computational Intelligence Systems,2021,14:165.
[22]MUSLIM M A,DASRIL Y,ALAMSYAH A,et al. Bank predictions for prospective long-term deposit investors using machine learning LightGBM and SMOTE[J]. Journal of Physics:Conference Series,2021,1918(4):042143.
[23]WEI J,LI Z Q,PINKER R T,et al. Himawari-8-derived diurnal variations in ground-level PM2.5 pollution across China using the fast space-time Light Gradient Boosting Machine(LightGBM)[J]. Atmospheric Chemistry and Physics,2021,21(10):7863-7880.
[24]FARID M. Data-driven method for real-time prediction and uncertainty quantification of fatigue failure under stochastic loading using artificial neural networks and Gaussian process regression[J]. International Journal of Fatigue,2022,155:106415.
[25]REN C H,YANG Y X,DONG X,et al. Prediction of the maximum temperature of sulfur-containing oil using gaussian process regression for hazards prevention[J]. International Journal of Performability Engineering,2018,14(12):2951-2959.

备注/Memo

备注/Memo:
收稿日期:2020-10-15.
基金项目:浙江省重点研发计划项目(2021C01110).
通讯作者:卢焕达,博士,副教授,研究方向:机器学习、数据分析. E-mail:huandalu@163.com
更新日期/Last Update: 2022-09-15