dc.contributor.advisor | Güler, Önder | |
dc.contributor.author | Şen, Emre | |
dc.date.accessioned | 2020-12-07T08:47:06Z | |
dc.date.available | 2020-12-07T08:47:06Z | |
dc.date.submitted | 2019 | |
dc.date.issued | 2019-09-25 | |
dc.identifier.uri | https://acikbilim.yok.gov.tr/handle/20.500.12812/117138 | |
dc.description.abstract | Günümüzde hızla artan bulut bilişim teknolojilerinin ve mobil cihazların kullanımı sebebiyle, veri merkezi kavramı her geçen gün hayatımızdaki önemini arttırmakta ve teknoloji dünyasının vazgeçilmezleri arasında yükselmektedir. Her geçen gün veri merkezlerinin önemini arttırması; dünya üzerindeki veri merkezi sayısınının ve veri depolanmasını hızlı bir şekilde arttırmaktadır. Veri merkezleri, elektrik tüketiminin en yüksek olduğu tesislerin en başında gelmektedir. Dünya üzerinde sayıları çok hızlı bir şekilde artış gösteren veri merkezleri çok büyük miktarda elektrik tüketimine neden olmakta, şehirlerin elektrik şebekelerinin altyapısı için büyük sorunlar teşkil etmektedir. Enerjinin insanlık için büyük bir önem kazandığı günümüz dünyasında, veri merkezleri tarafından harcanan yüksek miktardaki enerji: araştırma merkezlerinin, teknoloji üreticilerinin ve çevrecilerin önem verdiği bir konu haline gelmiştir. Bu durum; veri merkezi verimliliği kavramını ortaya çıkarmış ve veri merkezi projelerinin en verimli hale getirilmesini zorunlu kılmıştır. Günümüzde, hizmet ettiği amaç doğrultusunda en değerli binalar olarak gösterilen veri merkezleri için, dikkat edilmesi gereken diğer bir unsur ise veri merkezi sürekliliğidir. İnternet üzerinden gerçekleşen tüm işlemler (bankacılık, sağlık, turizm, telekominikasyon vb.) veri merkezleri içerisinde bulunan IT ekipmanlar üzerinden gerçekleştirilmektedir. IT ekipmanlar üzerinde meydana gelebilecek yarım saniyelik bir enerji kesintisi bile, büyük çapta para ve bilgi kayıplarına neden olabilir. Veri merkezlerinde ön görülen bu riskler; veri merkezi tasarımcılarını, yedekli ve sürekli veri merkezi tasarlamalarına zorlamaktadır.Bu tez çalışması kapsamında; veri merkezlerinin genel yapısı açıklanarak, veri merkezlerinde enerji verimliliğini arttıracak yötemler; UPS, soğutma sistemleri, PDU açısından ve sürekliliği artıracak yöntemler ise kullanılacak UPS sayısı ile elektriksel tasarım açısından incelenmiştir. Tez kapsamında; bir çok farklı üreticiye ait teknolojiler incelenerek, gereklilikleri ve kullanım amaçları açıklanmış, verimliliği en yüksek tasarımlar gerçek test düzeneklerinde test edilmiştir. Tez kapsamında; ülkemiz sınırları içerisinde yapılması planlanan gerçek bir proje için vaka analizi gerçekleştirilmiştir. Vaka analizi kapsamında; UPS, soğutma ve PDU sistemleri incelenmiş, fiyat ve enerji verimliliği açısında en optimum çözüm önerisi sunulmuştur. | |
dc.description.abstract | In today world, data center concept and importance of data center has been enhancing quickly due to increasing usage rate of cloud and mobile services and internet. In the past decades, internet access could be performed via only computers with slow connection speed. But in today world, every single electronic device can connect to internet with fascinating connection speed compared to last decades. For that reason; each day, number of date center all around the world is increasing in order to cover huge increasing demand.Data centers are called as high consumption places whose electricity consuptiom level is one of the highest among all facilities. Data centers whose numbers is increasing rapidly have potential risk to cause high electricity consumptions for data center managers and huge amount of load for electricity grid. Especially data centers located in city center or city areas, may have risk to affect stabilization of electricity grid in negative way. In addition to negative effect for electrical grid, datacenters as major consumers of power which may cause low effinciency in grid or may enable highly efficient operation in electrical grid. In these days when energy becomes one of the important topic for globe, high energy consumed by data centers are pointed by research labs, techonology companies and environmentalist in order to reduce consumption. This awareness has ascertained new term which is called Data Center Efficiency. Today, data centers are called most valuable buildings and facilities in terms of their intended purposes. For these valuable buildings, there is one more important term which must be taken into account which is called Data Center Sustainability. All online process and data storage are done through servers located in data centers. Even half second electricity cut in data centers may cause millions of dolars damage and loss of huge data. These potential risks force to data center designers to design redundant and sustainable data centers.In this thesis; general structure of data centers, methods for increasing data center efficiency and sustainability has been explained by examing energy efficiency in terms of UPS (Uniterruptible Power Supply), cooling systems, PDU (Power Distrubition units) and data center structer in terms of electricity structure diagram of data centers and number of UPS which is used to supply electrical load of data center. To make better analysis, necessity and intended purposes of different technologies belong to different manufacturer has been examined in detail. Moreover, high efficiency technologies which is guaranteed by manufacturer has been tested with real test equipment to prove high efficient operations. In the section of analyzing UPS system and topology for increasing efficiency of data centers; chosing suitable UPS type in terms design has been emphasized as first step. According to design and operation benefits, both moduler and monoblock UPS types were compared. As second step for suitable and efficienct UPS operations for data centers, how important suitable UPS capacity choice has been emphasized by performing real UPS load test with 120kVA UPS. Withing scope of applied regarding test, 120kVA UPS was tested in order to calculate UPS efficiency at different load level and efficiency was calculated for each load level. As a result of UPS load test, it is seen that UPS operation with higher load without excessing UPS rated capacity enables higher UPS efficient operation. In addition to choice for suitable type and capacity of UPS, making UPS to work with suitable UPS operation has been also emphasized. To see each characteristic of each UPS operation mode, 1200kVA UPS was tested with real load in double conversion, eco mode and enhanced eco mode in terms of output voltage waveform, transfer time to battery mode efficiency in case of power cut and. Each test outputs were examined to obtain maximum sustained and efficient UPS operation. Besides UPS choice for data center project, importance of UPS system topology and general term regarding topology are explained and exemplified.In the section of analyzing cooling methods for data centers; basic structure and working principle of of traditional cooling methods used in data center for decades has been presented, then their disadvantages and inefficient operations are explained. In the light of given information related to traditional data center cooling methods, new methods which is containment cooling are explained by eliminating root reasons that cause inefficient operations. Basic structures and working principle of containment cooling methods are presented in details and their advantages against to traditional methods are given. In addition to theoretical part of thesis, case study for new data center project has been performed by examining cooling system, UPS system and PDU system in order to propose optimum solution in terms of energy efficiency and cost optimization. In the scope of case study, UPS system is analyzed to determine whether moduler or monoblock UPS systems is appropriate for intended data centers. Regarding analyze is performed by comparing capital investment, 10 years maintenance cost and rent cost for area in facility where UPS system is installed for both UPS type. Within analysis for UPS system, how many battery is needed and capital cost of batteries for each system are calculated and compared.In the section of case study related to cooling system, how to optimum solution can be chosen according to customer request (which determines that containment cooling system is only option to be used in intended project) is explained. By determining optimum solution for regarding project, each parameter and system requirement to satisfy customer requirement are calculated and explained step by step for both temperature and pressure controller containment cooling systems. According to obtained outputs of calculations and required heat capacity for cooling machines, once appropriate CRAC unit is proposed, then yearly energy consumtion of different installed CRAC units, that can cover at least IT room heat capacity, are compared for both temperature and pressure control containment cooling systems. As a result of performed comparison; it is seen that temperature control containment cooling system is more efficient than pressure control containment cooling system. In addition to comparison between temperature and pressure control systems, it is seen that making cooling systems' EC fans to operate about 30% of full speed much more efficient that making them to operate at full speed for both temperature and pressure control cooling systems.As a final section of case study; in order to evaluate how much data center is efficient, PUE calculations for intended data center project is performed after optimum UPS system and cooling system are chosen. All necessary parameters except UPS losses, cooling system energy consumption and PDU losses are provided by customer in order to perform calculation of PUE. | en_US |
dc.language | Turkish | |
dc.language.iso | tr | |
dc.rights | info:eu-repo/semantics/openAccess | |
dc.rights | Attribution 4.0 United States | tr_TR |
dc.rights.uri | https://creativecommons.org/licenses/by/4.0/ | |
dc.subject | Elektrik ve Elektronik Mühendisliği | tr_TR |
dc.subject | Electrical and Electronics Engineering | en_US |
dc.title | Veri merkezlerinde enerji verimliliğini arttırıcı yöntemler | |
dc.title.alternative | Methods for improving data centers energy efficiency | |
dc.type | masterThesis | |
dc.date.updated | 2019-09-25 | |
dc.contributor.department | Enerji Bilim ve Teknoloji Anabilim Dalı | |
dc.identifier.yokid | 10264987 | |
dc.publisher.institute | Enerji Enstitüsü | |
dc.publisher.university | İSTANBUL TEKNİK ÜNİVERSİTESİ | |
dc.identifier.thesisid | 562046 | |
dc.description.pages | 111 | |
dc.publisher.discipline | Enerji Bilim ve Teknoloji Bilim Dalı | |