大型建筑空调变过热度压缩制冷系统模型预测控制

Large-Scale Building Air Conditioning System consumes a large amount of energy while providing comfortable indoor environment for us. More than half of the energy consumptions of Large-Scale Building Air Conditioning System is consumed by the Vapor Compression Refrigeration System (VCRS). It is the key task for control system of Large-Scale Building Air Conditioning System to improve the energy efficiency of VCRS and reduce its energy consumptions as soon as possible. However, the close relationships between the superheat degree at the outlet of evaporator and the VCRS’s efficiency are ignored and conventional control strategies are designed based on constant superheat degree, which has become one of the most serious reasons causing low energy efficiency of VCRS. In this proposal, facing to the optimizing control and energy saving problems of VCRS, a variable superheat degree based dynamic model of VCRS is developed based on the hybrid modeling theory with thermodynamic properties and parameters identification; then an optimal selection of controlled variables for the system is proposed. Finally, in order to improve the energy efficiency of VCRS, a novel model predictive control based cascade control strategy with variable superheat degree for VCRS is developed to fulfill the cooling requirements and maximum its energy efficiency as well. The results of the project will provide a reliable theoretical basis and experimental results for comprehensive analysis of the VCRS in Large-Scale Building Air Conditioning System, and the proposed optimizing control strategy has great significance on system energy efficiency improvement and development of large-scale building energy conservation.

大型建筑空调系统在给人们带来舒适室内环境的同时也消耗了大量的能源。压缩制冷系统能耗占大型建筑空调系统能耗一半以上，如何最大限度地提高其能源效率正是大型建筑空调控制系统所面临的核心问题。现有控制方案忽略了过热度与系统能耗间的关系，仅基于恒定过热度设计压缩制冷系统控制器，是造成当前系统能效低下的主要原因之一。本项目拟面向大型建筑空调中压缩制冷系统的优化控制与节能降耗问题，基于过程机理与参数辨识建立考虑系统过热度与能效关系的动态混合模型；并研究基于自寻优的系统最优被控变量选取方法；基于模型预测控制理论提出基于可变过热度的压缩制冷系统串级优化控制策略，满足系统负荷的同时最大化系统能源效率。本项目的开展可为大型建筑压缩制冷系统的综合分析提供可靠的理论基础和实验结果，同时提出的优化控制策略可显著提升系统节能运行效率，对推动我国大型建筑节能具有积极的现实意义。

现代社会对大型建筑空调的制冷品质和节能要求不断提高。作为建筑空调的核心部件，蒸汽压缩制冷系统具有非线性、强耦合、多变量等特点，传统的建模方式以及控制方法很难达到期望的控制效果。本项目针对大型建筑空调压缩制冷循环系统的建模与优化控制问题，首先，基于传热传质理论和能量守恒定律，分别建立了压缩制冷系统中关键组成部件的混合模型，构建了温度、压力等关键变量的动力学函数关系，并结合数据辨识方法建立了系统简化模型；其次，分析了系统关键变量对系统能耗及稳定性的影响，设计了基于能源效率的优化指标，分别提出了自寻优控制策略、改进的差分进化优化策略以及改进的鲸鱼优化策略实现压缩制冷系统变量优化选取及降低能耗的节能优化目标；最后，基于模型预测控制方法，分析了最小稳定过热度与冷负荷的非线性关系，构建了模型预测控制与PI控制相结合的串级控制策略，在变负荷情况下最大限度提高了系统效率；结合模型预测控制与迭代学习控制方法，利用重复迭代特性，设计了迭代学习模型预测控制策略，实现对过热度的实时控制跟踪。本项目的研究可有效提高压缩制冷系统能源效率，为大型建筑空调的实际应用提供理论支撑。