数据中心机房机架温度预测模型研究
|
摘要
为了达到服务器温度需求以实现安全可靠运行,数据中心多采用过量冷却方式来提供冷量。这种冷却方式虽然具有稳定的温度控制,但巨大的空调能耗问题凸显出来。数据中心只能通过采集温度数据,考虑工作量和运行需求对未来的温度变化进行估计,这种温度预测过于经验化,可靠性较低。针对现实情况,机房多台空调共同制冷,位于不同位置的机架制冷情况的不同,该文针对某数据中心机房数据构建温度预测模型,并采用蚁群和粒子群混合算法优化支持向量机加以实现。通过MATLAB仿真和数据中心机房采集数据完成模型的建立与预测。仿真结果表明,与实际测量数据相比,整体呈现出较高的预测精度,温度趋势大致相符。
The excess cooling method is often used by the data center to provide the cooling capacity in order to achieve the temperature requirement for the safe and reliable operation. Although this cooling method has stable temperature control, the energy consumption of air conditioning is huge. Data center can estimate the change of future temperature only by collecting temperature data and taking account of workload and operational requirements, which is quite empirical and has low reliability. Since multiple air conditioners are co-cooled in the equipment room and the cooling situation of the racks at different positions is different, according to the date from a data center computer room, the ant colony optimization mixing with the particle swarm optimization algorithm is used to optimize the support vector machine model and to construct a temperature prediction model in this paper. The establishment of model and the temperature prediction is based on MATLAB software and the true data from computer room. The simulation results show that the overall average prediction has a high degree of accuracy and the temperature trend is roughly consistent compared with the actual measured data.
The excess cooling method is often used by the data center to provide the cooling capacity in order to achieve the temperature requirement for the safe and reliable operation. Although this cooling method has stable temperature control, the energy consumption of air conditioning is huge. Data center can estimate the change of future temperature only by collecting temperature data and taking account of workload and operational requirements, which is quite empirical and has low reliability. Since multiple air conditioners are co-cooled in the equipment room and the cooling situation of the racks at different positions is different, according to the date from a data center computer room, the ant colony optimization mixing with the particle swarm optimization algorithm is used to optimize the support vector machine model and to construct a temperature prediction model in this paper. The establishment of model and the temperature prediction is based on MATLAB software and the true data from computer room. The simulation results show that the overall average prediction has a high degree of accuracy and the temperature trend is roughly consistent compared with the actual measured data.
引文
[1]黄森,潘毅群,Peng XU.数据中心节能研究现状与发展[C].全国暖通空调制冷2010年学术年会资料集.2010.
[2]王恕清,孙立强,张为民.机房专用恒温恒湿空调机组的相关热点问题[J].暖通空调,2008,38(12):84-92.
[3]于洋,孙成伟,张冰冰,等.云模型RBF神经网络的数据中心温度预测[J].沈阳理工大学学报,2013,32(4):9-14.
[4]邹燕.IDC机房温度调控设计研究[D].上海:上海工程技术大学,2016.
[5]安册册.数据中心空调系统预测模型研究[D].北京:北京建筑大学,2017.
[6]DORIGO M,ST譈TZLE T.The ant colony optimization metaheuristic:algorithms,applications,and advances[M].Kluwer Academic Publishers,2002.
[7]KENNEDY J,EBERHART R.Particle swarm optimization[C].IEEE International Conference on Neural Networks,1995:1942-1948.
[8]韩明,张积林,李林.数学建模案例[M].上海:同济大学出版社,2012.
[9]VAPNIK V.Statistical learning theory[M].New York:Wiley,1998.
[10]郭凤仪,郭长娜,王爱军,等.基于粒子群优化支持向量机的煤矿水位预测模型[J].计算机工程与科学,2012,34(7):177-181.
[11]原峰山,朱思铭.利用傅立叶变换筛选核函数的方法[J].中山大学学报(自然科学版),2004,43(3):10-13.
[12]TRELEA I C.The particle swarm optimization algorithm:convergence analysis and parameter selection[J].Information Processing Letters,2003,85(6):3.
[2]王恕清,孙立强,张为民.机房专用恒温恒湿空调机组的相关热点问题[J].暖通空调,2008,38(12):84-92.
[3]于洋,孙成伟,张冰冰,等.云模型RBF神经网络的数据中心温度预测[J].沈阳理工大学学报,2013,32(4):9-14.
[4]邹燕.IDC机房温度调控设计研究[D].上海:上海工程技术大学,2016.
[5]安册册.数据中心空调系统预测模型研究[D].北京:北京建筑大学,2017.
[6]DORIGO M,ST譈TZLE T.The ant colony optimization metaheuristic:algorithms,applications,and advances[M].Kluwer Academic Publishers,2002.
[7]KENNEDY J,EBERHART R.Particle swarm optimization[C].IEEE International Conference on Neural Networks,1995:1942-1948.
[8]韩明,张积林,李林.数学建模案例[M].上海:同济大学出版社,2012.
[9]VAPNIK V.Statistical learning theory[M].New York:Wiley,1998.
[10]郭凤仪,郭长娜,王爱军,等.基于粒子群优化支持向量机的煤矿水位预测模型[J].计算机工程与科学,2012,34(7):177-181.
[11]原峰山,朱思铭.利用傅立叶变换筛选核函数的方法[J].中山大学学报(自然科学版),2004,43(3):10-13.
[12]TRELEA I C.The particle swarm optimization algorithm:convergence analysis and parameter selection[J].Information Processing Letters,2003,85(6):3.
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |