基于现场数据建模的冷机群控策略研究  

Optimal Control of Chiller Plant based on Models Identificated with Field Data

作　　者：许晓群 王翠灵 侯忠平 库慧益 齐延会 王宝龙[2] XU Xiaoqun;WANG Cuiling;HOU Zhongping;KU Huiyi;QI Yanhui;WANG Baolong(Wuhan Cigarette Factory,China Tobacco Hubei Industrial Co.LTD,Wuhan 430040,China;Department of Building Science,School of Architecture,Tsinghua University,Beijing 100084,China)

机构地区：[1]湖北中烟武汉卷烟厂,武汉430040 [2]清华大学建筑学院建筑科学系,北京100084

出　　处：《建筑节能（中英文）》2023年第4期56-63,共8页Building Energy Efficiency

摘　　要：冷站能耗在中央空调总能耗中占比大。通过优化冷机运行序列和负荷分配,可以在保证末端冷量需求的同时,实现节能。针对冷机性能主要受供冷量、蒸发器出水温度和冷凝器出水温度影响,设计了基于现场数据识别模型优化群控策略,同时以冷机现场可达到的实际最大供冷量和最小供冷量作为约束条件,缩小了冷机序列优化和负荷分配优化的搜索域,加速优化控制计算速度。同时,探讨了不同优化算法对寻优结果的影响。以武汉某烟厂作为案例,对冷机优化群控策略进行验证,结果表明:与传统冷机群控策略相比,采用基于现场数据识别模型的冷机群控策略均可以在整个制冷季实现约20%的节能率。For central air-conditioning systems,the energy consumption of the chiller plant is huge.By optimizing the operation combination of chillers and the load distribution among chillers,it is possible to realize the energy saving while meeting the cooling demand for end users.Considering that the chiller performance is affected by the cooling capacity,supplying chilled water and returned cooling water temperature,an optimized model-based control strategy of chillers is designed based on the actual performances of chillers.In which,the actual maximum cooling capacity is used as constraint to reduce the search domain and accelerate optimization control calculation speed.Meanwhile,the performances of different optimization algorithms on the optimization results are discussed.Taking a cigarette factory in Wuhan as example to verify the optimized cluster control strategy,the result shows that compared to the traditional control strategy of chiller plant,both traversal optimization and genetic algorithm optimization strategies can achieve an energy-saving rate of about 20%in the entire cooling season.

关 键 词：模型控制 冷站 实际最大供冷能力 节能 

分 类 号：TU83[建筑科学—供热、供燃气、通风及空调工程]