基于NSGA-Ⅱ的冷源机房设备运行参数多目标优化

Multi-objective Optimization of Operating Parameters of Central Air Conditioning Cold Source Equipment Room Based on NSGA-Ⅱ

中央空调冷源机房设备系统运行参数优化是提高空调运行性能的重要途径之一,但由于机房设备种类繁多,运行工况复杂,运行参数诸多,参数之间耦合作用较强,其优化问题是学者们广泛关注的重点和难点。提出了一种基于非支配排序遗传算法Ⅱ(non-dominated sorting genetic algorithmⅡ, NSGA-Ⅱ)的冷源机房设备运行参数多目标优化方法。以制冷量最大和能耗最低为目标函数,以冷冻出水温度、冷冻回水温度、冷却回水温度、冷冻水流量和冷却水流量5个变量为决策变量,根据实际运行工况设定其约束条件,建立系统运行参数优化模型。此外,以负荷率、室外环境温度和室外环境湿度作为划分工况的依据,采用等宽离散化方法并结合K-means聚类方法共划分出32种运行工况。最后仿真结果表明,在4种典型运行工况下,多目标优化方法相对于普通优化方式,能得到更高的循环性能系数(coefficient of performance, COP);同时,多目标优化方法能够使系统能效提升14.35%,能耗降低12.52%。因此,本文方法适用于空调领域,并能为工程应用时的参数设置提供一定的指导作用。

The optimization of operating parameters of the equipment system of central air-conditioning cold source computer room is one of the important ways to improve the operating performance of air conditioner. However, due to the variety of equipment in the computer room, complicated operating conditions, numerous operating parameters and the strong coupling between parameters, its optimization is one of the key points and difficulties that draws wide attention of scholars. To this end, a multi-objective optimization method for operating parameters of cold source equipment room was proposed. It was based on non-dominated sorting genetic algorithm Ⅱ(NSGA-Ⅱ). Both the maximum cooling capacity and the lowest energy consumption were set as objective functions, while the frozen outlet water temperature, frozen return water temperature, cooling return water temperature, frozen water flow rate and cooling water flow rate were set as decision variables. A system optimization model of operating parameters was established, which constraints were set accor-ding to the actual conditions. In addition, based on the load rate, outdoor ambient temperature and outdoor ambient humidity, 32 kinds of conditions according to the method of constant width discretization and the K-means clustering were divided. The final simulation result shows that the multi-objective optimization method can obtain a higher coefficient of performance(COP) which compares to the ordinary optimization method under four typical conditions. At the same time, the energy efficiency can increase by 14.35% while the energy consumption decrease by 12.52%. In conclusion, the proposed optimization method is applicable to the field of air conditioning. It can also provide certain guidance for the parameter setting in engineering applications.