ABB Jiang Ying: Accurate matching of sources and loads is the key to promoting low-carbonization of data centers

admin

Byadmin

Nov 29, 2024

With the development of artificial intelligence such as big models, the demand for intelligent computing has increased rapidly, the industrial scale of data centers has expanded rapidly, and energy consumption is huge and still growing. Under the “dual carbon” goal, accelerating energy conservation and carbon reduction has become an inevitable direction for data centers, and green development has become the focus of discussion.

  Jiang Ying, head of digital business of ABB Electric China’s power distribution system

  At the China Intelligent Computing Ecological Development Conference hosted by China IDC Circle, Jiang Ying, head of digital business of ABB Electric China’s power distribution system, pointed out that there are three main ways to achieve green and low-carbon development of data centers: one is green and clean energy; the second is to achieve accurate matching between sources and loads; the third is low-carbon operation and sustainable carbon reduction.

  Data centers belong to high-energy-consuming industries. In the face of energy transformation and change, how can data centers achieve green and low-carbon development?

  On the energy side, data centers consider building distributed photovoltaics to increase the access ratio of renewable energy. However, the construction area of ​​​​data center buildings that can be used for photovoltaics is usually very limited, which cannot effectively promote the transformation and development of green data centers. “Mutual aid and sharing” is used to achieve lean transformation of data center sources and loads, and through the “diesel-generator mutual aid” method, the phased deployment capability of backup power supplies is optimized, effectively reducing the overall investment cost of diesel-generator units.

  Through the full-site digital IEC61850/GOOSE redundant networking communication architecture, rapid switching of key power supplies within 1 minute can be achieved, and precise matching between sources and loads can be achieved through digital control technology, so as to effectively simplify the configuration of high-redundancy heavy assets in the data center power distribution system, build a lightweight data center power distribution system structure, and improve the efficient utilization of assets.
  The following is the content of the interview:

  What is the path for data centers to achieve green and low-carbon?

  Data centers have gone through about four stages of transformation.

  In 2012, the first stage of data centers was to pursue power supply safety and reliability;

  In 2016, they began to consider the improvement of power usage efficiency (PUE);

  In 2021, the construction of the East Data West Computing cluster promoted the coordinated optimization between energy and computing power;

  So far, with the progress of sustainability, building green and low-carbon data centers has become the main goal.

  There are three main ways to achieve green and low-carbon data centers:

  First, adjust the energy structure, increase the proportion of renewable energy input, and achieve green and clean energy;

  Second, achieve precise matching between data center sources and loads through “mutual assistance and sharing”;

  Third, low-carbon operation and sustainable carbon reduction.

  Can the construction of renewable energy effectively help data centers achieve green and low-carbon transformation?

  Data centers are asset-heavy industries, and the power generation of photovoltaic power generation built on rooftops accounts for less than 3% of the power consumption of data centers, which is extremely limited. If we build several hundred megawatts of photovoltaic power to achieve the greening of data centers, the entire investment is high, and even if we achieve clean energy, it is not replicable. And if there is no professional operation and maintenance of photovoltaic power, the power generation efficiency of photovoltaic power cannot be guaranteed; without a smart energy management system to maximize the local consumption of photovoltaic power, it cannot help the data center achieve the transformation of green and low-carbon.

  Does ABB have any innovative solutions for the coordinated optimization of computing power and power?

  If the realization of green and low-carbonization at the source is limited, we need to explore the feasibility of the matching relationship between source and load. Diesel generators are important backup power sources for data centers. Their diesel-generator co-existence: “co-existence and sharing” promotes the lean transformation of data center sources and loads, optimizes the phased deployment capabilities of backup power sources, and effectively reduces the overall investment cost of diesel-generator units.

  On the other hand, the green, low-carbon and high-quality development of data centers deeply cooperates with computing power and energy, digs deep into the energy space based on the load characteristics of data centers, and determines the optimal diesel-generator input capacity and load through optimization mechanism algorithms to achieve precise matching, achieve dynamic balance between source and load, and further co-existence and sharing to optimize asset allocation. At the same time, the computing load and cooling system are jointly optimized to further improve the energy efficiency of the data center, while establishing the flexible adjustment capability of the data center load and exploring the practical path of the “production and consumption type” data center.

  What strategic plans does ABB have for the future low-carbon transformation of data centers?

  As a global technology leader, ABB has always been adhering to the attitude of continuous innovation, empowering the data center industry to achieve green and low-carbon transformation and development with digital technology and a century of accumulated industry experience.

  We will promote the application of innovative technologies in data center customers:

  First, through IEC61850/GOOSE digital technology to optimize power supply and distribution assets, reduce high-redundancy backup configuration, and still ensure the continuous reliability of power supply in data centers.

  Second, expand the other application values ​​of existing backup power supplies; in the future new power system, the secondary and tertiary load backup capacity of the data center will be replaced with energy storage, and energy storage will be used as a flexible energy source to achieve rapid response on the demand side and build a virtual power plant.

  How to realize the power security supply system in the future energy system architecture, high redundant backup capacity of data centers: such as diesel generation and energy storage can be used as emergency power backup capacity of the power grid to improve the efficient utilization rate of power assets in data centers

  ”China IDC Circle Enterprise Club”

  Belonging to the media business sector of Kezhi Group, initiated by China IDC Circle, it is an industry exchange and cooperation platform jointly formed by the government, computing power centers, upstream and downstream enterprises of the cloud computing industry chain, investment and financing, etc. While covering the entire ecology of the data center, the enterprise club relies on a complete service process and project mechanism to provide enterprises with technology promotion and exchange, resource cooperation, brand marketing, supply and demand docking and capital services. It is committed to the development and application promotion of the data center industry, helping enterprises to achieve complementary advantages, resource sharing, win-win cooperation, and build a digital industry basic platform.
Related product recommendations:
IMCAM07B0000
3BSE092980R1
3BSE069296R1
PP885R
3BSE069296R1
3BHE023784R2530
PPD113B01-25-111000
PPD113B01-25-111000 3BHE023784R2530
3BSE082896R
PM858K02 3BSE082896R1
more……

Leave a Reply

Your email address will not be published. Required fields are marked *