基于动态知识图谱的新风-空调系统多目标优化调控策略

Multi-objective Optimal Control Strategy for Fresh Air-Air Conditioning System Based on Dynamic Knowledge Graph

摘要

针对同时配备新风系统与空调系统的住宅能耗优化问题,以新风-空调系统动态知识图谱为基础,提出一种新风-空调系统的多目标优化调控策略。首先,融合专家经验知识与实时监测数据建立新风-空调系统动态知识图谱模型,并以此构造新风系统新风量、空调系统设定温度、空调系统设定风速候选集;其次,将候选集作为多目标优化算法约束条件,同时,将降低系统总体能耗与维持室内最佳舒适度设为优化目标来建立多目标优化模型;然后,利用Logistic混沌映射、黄金正弦策略及多项式变异等策略改进多目标鲸鱼优化算法(multi-objective whale optimization algorithm,MOWOA),采用改进后的多目标鲸鱼优化算法(improved multi-objective whale optimization algorithm,IMOWOA)实时优化新风量、空调设定温度,并设定风速收敛至最优解,以有效平衡能耗与舒适度之间的冲突,实现新风系统与空调系统的协同调控;最后,在高度仿真的住宅环境模拟平台中对该策略进行验证,减少了约3.98%的建筑耗电量,并提升了36.8%的求解速度。结果表明,该调控策略在保证室内环境舒适度的同时降低了系统整体能耗,且显著提升了优化算法迭代的效率,较好地实现了降本增效的目标。

Abstract

To address energy optimization for residential buildings equipped with both fresh air systems and air conditioning systems, a multi-objective optimal control strategy for fresh air-air conditioning system is proposed based on the dynamic knowledge graph of fresh air-air conditioning system. First, expert knowledge and real-time monitoring data were fused to establish a dynamic knowledge graph model of fresh air-air conditioning system, and the candidate sets of fresh air volume, air conditioning set temperature, and air conditioning set air speed were constructed based on the model. Second, the candidate sets were used as the constraint condition of the multi-objective optimization algorithm, and the optimization goals of minimizing the overall system energy consumption and maintaining optimal indoor comfort were established. Third, the logistic chaotic mapping, golden sine strategy, and polynomial variation strategies were used to improve the multi-objective whale optimization algorithm(MOWOA), and the improved multi-objective whale optimization algorithm(IMOWOA) was used to optimize the fresh air volume, air conditioning set temperature, and set air speed in real time to converge to the optimal solution, effectively balancing the conflicts between energy consumption and comfort, and achieving coordinated control of the fresh air system and air conditioning system. Finally, the proposed strategy was validated in a highly simulated residential environment simulation platform, which reduces the building power consumption by about 3.98% and improves the solution speed by 36.8%. Results show that the control strategy of this paper ensures indoor environmental comfort while reducing the overall system energy consumption, significantly improves the efficiency of the optimization algorithm iteration, and achieves the goal of reducing costs and increasing efficiency to a good extent.

关键词

新风系统 / 空调系统 / 动态知识图谱 / 多目标优化 / 推荐系统 / 多目标鲸鱼优化算法(multi-objective whale optimization algorithm,MOWOA)

Key words

fresh air system / air conditioning system / dynamic knowledge graph / multi-objective optimization / recommendation system / multi-objective whale optimization algorithm (MOWOA)

参考文献

[1] Cluxton—Keller F, Bruce M L . A pilot study of an educational intervention to increase postpartum medical visit attendance in home—visited mothers[J]. The Journal of Perinatal Education, 2023, 32(2): 83-93.

[2] Salins S S, Kumar S, Chouhan J, et al. Experimental performance of a spray tower system for water desalination and indoor thermal comfort[J]. Process Safety and Environmental Protection, 2023, 180: 122-135.

[3] Zhuang L F, Huang J X, Zhong K . A revised PMV model: from a physiological standpoint[J]. Building Services Engineering Research & Technology, 2023, 44(5): 557-575.

[4] 高宇 . 中央空调多目标优化[D]. 沈阳: 沈阳工业大学, 2019.

[5] Gao Y . Multi—objective optimization of central air—conditioning[D]. Shenyang: Shenyang University of Technology, 2019. (in Chinese)

[6] 孙育英, 赵耀华, 王伟, 等 . 空调系统运行负荷GRNN预测模型的应用研究[J]. 北京工业大学学报, 2013, 39(7): 1084-1091.

[7] Sun Y Y, Zhao Y H, Wang W, et al. Application study of air conditioning system operational load prediction based on GRNN[J]. Journal of Beijing University of Technology, 2013, 39(7): 1084-1091. (in Chinese)

[8] Wang F R, Tan B, Chen Y, et al. A visual knowledge map analysis of mine fire research based on CiteSpace[J]. Environmental Science and Pollution Research, 2022, 29(51): 77609-77624.

[9] Sangeetha M, Thiagarajan M D . Attribute preserving recommendation system based on graph attention mechanism[J]. Journal of Intelligent & Fuzzy Systems, 2023, 44(6): 9419-9430.

[10] Ahmad H K, Qi C, Wu Z Q, et al. ABiNE—CRS: course recommender system in online education using attributed bipartite network embedding[J]. Applied Intelligence, 2023, 53(4): 4665-4684.

[11] Zitnik M, Agrawal M, Leskovec J . Modeling polypharmacy side effects with graph convolutional networks[J]. Bioinformatics, 2018, 34(13): i457-i466.

[12] Zenkert J, Klahold A, Fathi M . Knowledge discovery in multidimensional knowledge representation framework: an integrative approach for the visualization of text analytics results[J]. Iran Journal of Computer Science, 2018, 1(4): 199-216.

[13] 陈星, 黄志明, 叶心舒, 等 . 智能家居情境感知服务的运行时建模与执行方法[J]. 软件学报, 2019, 30(11): 3297-3312.

[14] Chen X, Huang Z M, Ye X S, et al. Approach to modeling and executing context—aware services of smart home at runtime[J]. Journal of Software, 2019, 30(11): 3297-3312. (in Chinese)

[15] 何坚, 杨洺, 白佳豪, 等 . 多尺度特征与知识图谱融合的景区手写诗词识别[J]. 北京工业大学学报, 2024, 50(3): 316-324.

[16] He J, Yang M, Bai J H, et al. Scenic handwritten poetry recognition based on fusion of multi—scale features and knowledge graph[J]. Journal of Beijing University of Technology, 2024, 50(3): 316-324. (in Chinese)

[17] Li L, Tu B, Xiong Y L, et al. Machine learning—based model for predicting prolonged mechanical ventilation in patients with congestive heart failure[J]. Cardiovascular Drugs and Therapy, 2024, 38(2): 359-369.

[18] Suyambazhahan S, Temesgen T, Nene A A, et al. Energy saving in an air—conditioning system using interdisciplinary energy conversion approach[J]. Smart Science, 2023, 11(1): 54-65.

[19] Ongwandee M, Khianthongkul K, Panyametheekul S, et al. Bangkok school indoor air quality: monitoring and intervention by positive pressure fresh air system[J]. Environmental Science and Pollution Research, 2024, 31(17): 25454-25467.

[20] Mirjalili S, Lewis A . The whale optimization algorithm[J]. Advances in Engineering Software, 2016, 95: 51-67.

[21] 严爱军, 曹付起 . 基于文化鲸鱼优化算法的特征权重优化分配方法[J]. 北京工业大学学报, 2021, 47(11): 1230-1238.

[22] Yan A J, Cao F Q . Method of feature weight optimization allocation based on cultural whale optimization algorithm[J]. Journal of Beijing University of Technology, 2021, 47(11): 1230-1238. (in Chinese)

[23] GB 50176—2016 民用建筑热工设计规范[S].

[24] Kyere E B, Jen T C, Tartibu L . Analysis of the influence of chilled water temperature setpoint on thermal comfort and energy consumption[J]. Arabian Journal for Science and Engineering, 2024, 49(8): 10409-10429.

[25] Ge L F, Zhang D P, Huang J L, et al. A novel sensorless control method for SRMs based on gain—optimized SMO[J]. Electrical Engineering, 2024, 106(2): 2011-2019.

[26] Ishibuchi H, Imada R, Setoguchi Y, et al. How to specify a reference point in hypervolume calculation for fair performance comparison[J]. Evolutionary Computation, 2018, 26(3): 411-440.

[27] 陈靖文, 何治斌, 张洪富 . 变风量空调系统的能耗建模与优化[J]. 船舶工程, 2022, 44(1): 27-31, 58.

[28] Chen J W, He Z B, Zhang H F . Energy consumption modeling and optimization of VAV central air conditioning system[J]. Ship Engineering, 2022, 44(1): 27-31, 58. (in Chinese)

[29] 刘忠宝, 韦自妍, 刘梓宸 . 基于太阳能的新风双级处理模块化节能墙[J]. 北京工业大学学报, 2023, 49(7): 800-809.

[30] Liu Z B, Wei Z Y, Liu Z C . Modular energy—saving wall with two—stage treatment of fresh air based on solar energy[J]. Journal of Beijing University of Technology, 2023, 49(7): 800-809. (in Chinese)

基金资助

国家自然科学基金资助项目(62373012)

佛山市科技创新专项资金资助项目(BK22BF003)

中央高校优秀青年团队培育项目(FRF-EYIT-23-06)