基于动态温度调控的空调系统能耗预测

doi:10.16182/j.issn1004731x.joss.21-0741

摘要/Abstract

摘要：

针对动态温度调控的空调系统能耗预测问题，设计了动态温度调控策略并通过EnergyPlus仿真得到空调系统逐时能耗数据集。在采用集成方法分析能耗的基础上，建立改进PSO算法优化BP神经网络（improved particle swarm optimization-back propagation neural network，IPSO-BPNN）预测模型。集成聚类、分类和关联度分析方法挖掘空调系统能耗模式、确定预测模型的输入变量；设计非线性变化策略调整PSO算法的惯性权重和加速度因子，提高训练速度和优化效果；建立IPSO-BPNN模型对空调系统逐时能耗进行预测。结果表明，收敛速度明显提高，且平均预测精度提高了3.4%。

关键词: 动态温度调控, 能耗仿真, 集成方法, 预测模型, IPSO-BPNN

Abstract:

To solve the problem of energy consumption prediction for air-conditioning systems implementing dynamic temperature control, we designed a dynamic temperature control strategy and obtained a dataset on the hourly energy consumption of the air-conditioning system through EnergyPlus simulation. An improved particle swarm optimization-back propagation neural network (IPSO-BPNN) prediction model was built on the basis of energy consumption analysis by an integrated method. Clustering, classification, and correlation analysis methods were integrated to mine the energy consumption pattern of the air-conditioning system and determine the input variables for the prediction model. A nonlinear change strategy was designed to adjust the inertia weight and acceleration factor of the PSO algorithm and thereby improve the training speed and optimization effect. An IPSO-BPNN model was constructed to predict the hourly energy consumption of the air-conditioning system. The results show that the convergence speed is significantly improved and that the average prediction accuracy is enhanced by 3.4%.

Key words: dynamic temperature control, energy consumption simulation, integrated method, prediction model, IPSO-BPNN

中图分类号:

TU831

白燕, 武璐璐, 贺引娥, 王玉英. 基于动态温度调控的空调系统能耗预测[J]. 系统仿真学报, 2022, 34(2): 366-375.

Yan Bai, Lulu Wu, Yin'e He, Yuying Wang. Energy Consumption Prediction for Air-conditioning System Based on Dynamic Temperature Control[J]. Journal of System Simulation, 2022, 34(2): 366-375.

图/表 14

图1

表1

图2

表2

图3

图4

图5

表3

图6

表4

图7

图8

图9

表5

参考文献 20

1	Thach T Q, Mahirah D, Dunleavy G, et al. Prevalence of Sick Building Syndrome and Its Association with Perceived Indoor Environmental Quality in an Asian Multi-ethnic Working Population [J]. Building and Environment(S0360-1323), 2019, 166(c): 106420.
2	Vellei M, Le Dréau J. A Novel Model for Evaluating Dynamic Thermal Comfort under Demand Response Events [J]. Building and Environment(S0360-1323), 2019, 160(8): 106215.
3	白燕, 冯壮壮, 张玮. 基于动态模糊推理的舒适温度在线预测 [J]. 计算机测量与控制, 2020, 28(7): 74-80.
	Bai Yan, Feng Zhuangzhuang, Zhang wei. Online Prediction of Comfort Temperature based on Dynamic Fuzzy Reasoning[J]. Computer Measurement and Control,2020, 28(7): 74-80.
4	孙淑凤, 赵荣义, 许为全, 等.动态空调策略研究 [J]. 制冷与空调, 2003, 3(6): 27-32.
	Sun Shufeng, Zhao Rongyi, Xu Weiquan, et al. Research on Dynamic Air Conditioning Strategies [J].Refrigeration and Conditioning,2003, 3(6): 27-32.
5	Zhang, Zhao. Relationship Between Thermal Sensation and Comfort in Non-uniform and Dynamic Environments [J]. Building and Environment(S0360-1323), 2009, 44(7): 1386-1391.
6	Zhu Y, Ouyang Q, Cao B, et al. Dynamic Thermal Environment and Thermal Comfort [J]. Indoor Air(S0905-6947), 2016, 26(1): 125-137.
7	刘争敏. 集中空调动态温度调控策略及能耗分析研究 [D]. 西安: 西安建筑科技大学, 2019.
	Liu Zhengmin. Dynamic Temperature Control Strategy and Energy Consumption Analysis of central Air-conditioning [D]. Xi 'an: Xi 'an University of Architecture and Technology, 2019.
8	Yang S, Wan M P, Chen W, et al. Model Predictive Control with Adaptive Machine-learning-based Model for Building Energy Efficiency and Comfort Optimization [J]. Applied Energy(S0306-2619), 2020, 271(c): 115147.
9	Amasyali K, El-Gohary N M. A Review of Data-Driven Building Energy Consumption Prediction Studies[J]. Renewable and Sustainable Energy Reviews(S1364-0321),2018, 8(1): 1192-1205.
10	Shao M, Wang X, Bu Z, et al. Prediction of Energy Consumption in Hotel Buildings via Support Vector Machines[J]. Sustainable Cities and Society(S2210-6707), 2020, 57(4): 102128.
11	Zhang B, Ren J, Cheng Y, et al. Health Data Driven on Continuous Blood Pressure Prediction based on Gradient Boosting Decision Tree Algorithm [J]. IEEE Access(S2169-3536), 2019, 7(9): 32423-32433.
12	杨熊, 于军琪, 郭晨露, 等. 基于改进PSO-BP神经网络的冰蓄冷空调冷负荷动态预测模型 [J]. 土木与环境工程学报, 2019, 41(01): 168-174.
	Yang Xiong, Yu Junqi, Guo Chenlu, et al. Dynamic Prediction Model of Ice Storage Air Conditioning Cooling Load based on Improved PSO-BP Neural Network [J]. Journal of Civil and Environmental Engineering, 2019, 41(1): 168-174.
13	Xu X, Liu W, Lian Z. Dynamic Indoor Comfort Temperature Settings based on the Variation in Clothing Insulation and its Energy-saving Potential for an Air-conditioning System [J]. Energy and Buildings(S0378-7788), 2020,220(c): 110086.
14	Ryu J, Kim J, Hong W, et al. Quantifying Householder Tolerance of Thermal Discomfort before Turning on Air-conditioner[J]. Energy and Buildings(S0378-7788), 2020, 211(15): 109797.
15	罗茂辉, 余娟, 杨月婷, 等. 等温工况下分体空调送风动态化与人体热舒适实验研究 [J]. 暖通空调,2014, 44(5): 130-134.
	Luo Maohui, Yu Juan, Yang Yueting, et al. Experimental Study on dynamic State of Separable Air Supply and Human Thermal Comfort under Isothermal Condition[J]. HVAC,2014, 44(5): 130-134.
16	李慧, 张庆范, 段培永. 基于用户学习的智能动态热舒适控制系统 [J]. 四川大学学报(工程科学版),2011, 43(2): 128-135.
	Li Hui, Zhang Qingfan, Duan Peiyong. Intelligent Dynamic Thermal comfort Control System based on User Learning [J]. Journal of Sichuan University (Engineering Science Edition),2011, 43(2): 128-135.
17	Fanger P O. Thermal Comfort. Analysis and Applications in Environmental Engineering [J]. Thermal Comfort. Analysis and Applications in Environmental Engineering(S0003-6870), 1970, 3(3):181.
18	西安市城乡建设委员会. 公共建筑节能设计标准: [S]. 陕西省住房和城乡建设厅, 陕西省质量技术监督局, 2011: 7-35.
	Xi'an Urban-Rural Construction Committee. Energy-saving Design Standards for Public Buildings: [S]. Shaanxi Provincial Department of Housing and Urban-Rural Construction, Shaanxi Provincial Bureau of Quality and Technical Supervision,2011: 7-35.
19	Ma Z, Song J, Zhang J. Energy Consumption Prediction of Air-Conditioning Systems in Buildings by Selecting Similar Days based on Combined Weights [J]. Energy and Buildings(S0378-7788), 2017, 151(9): 157-166.
20	Jiang Q, Huang R, Huang Y, et al. Application of BP Neural Network Based on Genetic Algorithm Optimization in Evaluation of Power Grid Investment Risk [J]. IEEE Access(S2169-3536), 2019, 7(9): 154827-154835.

项目		详情
冷水机组	效率	5.0
风机	静压值/ Pa	600
	效率/%	70
	工作时间	工作日08:00—20:00基于动态温度调控策略运行，其余时间不运行
制冷温度		工作日08:00—20:00进行动态温度设定，其余时间不工作
末端布置		每层有9个空调系统，即9个AHU和9个ATU

日期	时刻	时间类别			能耗/ (kW·h)	能耗模式标签
日期	时刻	工作日	加班	上班	能耗/ (kW·h)	能耗模式标签
06.01	1:00	否	否	否	0	零能耗
06.01	2:00	否	否	否	0	零能耗
06.01	3:00	否	否	否	0	零能耗
……
07.24	8:00	是	否	是	32.27	中能耗
07.24	9:00	是	否	是	38.72	高能耗
07.24	10:00	是	否	是	34.19	中能耗
……
08.31	18:00	是	否	是	32.87	中能耗
08.31	19:00	是	是	否	23.31	低能耗
08.31	20:00	是	是	否	23.46	低能耗

变量	关联度
R时刻室外温度	0.923 21
R-1时刻室外温度	0.917 59
R-1时刻室外相对湿度	0.893 89
R-1时刻室外风速	0.803 01
R-1时刻太阳直射辐射	0.648 06
R-1时刻太阳散射辐射	0.765 04
R时刻空调温度设定值	0.936 18
R时刻空调系统能耗模式	0.897 12

分析-模型	MAPE/%	MAE/(kW·h)	RMSE/(kW·h)
BPNN	8.24	2.816 9	4.179 5
集成-BPNN	6.13	2.110 9	2.715 9
集成-PSO-BPNN	5.59	1.910 8	2.561 7
集成-IPSO-BPNN	4.87	1.684 1	2.299 2

[1]	董海, 高秀秀, 魏铭琦. 基于ACS-DBN的电弧增材制造焊道尺寸预测[J]. 系统仿真学报, 2021, 33(12): 2828-2837.
[2]	陶文华, 袁正波. 焦炭质量的DE-BP神经网络预测模型研究[J]. 系统仿真学报, 2018, 30(5): 1650-1656.
[3]	方永锋, 陈建军. 可修复k/n表决系统随时间的生命状态分析与仿真[J]. 系统仿真学报, 2016, 28(8): 1846-1853.
[4]	张国强, 刘小荷, 蒋方婷, 李志平. 反导体系试训一体化环境开发集成方法研究[J]. 系统仿真学报, 2016, 28(4): 898-906.
[5]	唐旭清, 李建林. 气候变化对多针茅竞争分布预测与影响分析[J]. 系统仿真学报, 2016, 28(4): 956-965.
[6]	李元, 燕亚运, 唐晓初. 基于局部模型的多阶段在线产品质量预测[J]. 系统仿真学报, 2016, 28(4): 966-971.
[7]	李小喜, 陈浩光, 李大喜, 陈疆萍. 基于Elman神经网络的作战效能预测模型研究[J]. 系统仿真学报, 2015, 27(1): 43-49.