Renewable energy provides sustainable green power for AI data centers

When we scroll through our phones, use navigation, search for information online, or use voice assistants, image recognition functions, and even ChatGPT for AI chatting, translation, or creation, behind all of this, "data centers" are quietly at work. These data centers operate tens of thousands of servers, functioning like the brain of the digital world, running 24 hours a day, processing countless pieces of information and computational tasks. All of these processes are inseparable from "electricity."

But did you know? With the rapid development of artificial intelligence, AI models require massive computational resources, most of which are concentrated in data centers. The operation of these data centers not only demands enormous amounts of electricity to power the servers, but also requires cooling systems to maintain stable temperatures in the server rooms. As a result, the electricity consumption of data centers has surged dramatically. It is estimated that global data centers now account for 2-3% of total electricity consumption, and this number continues to rise. This leads to an unavoidable issue: while we want AI to keep advancing, how can we reduce the burden that data centers place on the Earth's resources?

To ensure that the development of AI does not become a burden on the planet, more and more tech companies and data centers are adopting renewable energy sources such as solar, wind, and hydro power. These naturally derived energies have minimal environmental impact, reduce carbon emissions, and make data centers more sustainable.

Solar power is a clean, inexhaustible energy source. By installing solar panels on data center rooftops or nearby open spaces, electricity can be directly supplied to servers during the day, reducing reliance on traditional power grids. To enable solar energy to function at night as well, battery storage systems play a critical role. Simply put, the excess solar energy generated during the day is stored in high-capacity batteries, and then released at night or on cloudy days to provide power - allowing AI data centers to operate steadily 24 hours a day.

Arrow Electronics and its partners have long been committed to making intelligent computing more sustainable, fostering a harmonious relationship between technology and the planet, and have launched a wide range of renewable energy solutions. Below, we will introduce the solutions developed by Arrow Electronics and its partners according to different application categories such as energy conversion, battery management systems, system interfaces, and control.

In solar and battery energy storage systems, energy conversion acts like a “power translator,” converting different forms of electricity into usable power for devices. Solar panels generate DC (direct current), but most data center equipment requires AC (alternating current), thus requiring inverters to convert DC to AC. Battery storage systems also store power in DC form, so voltage and current need to be regulated during discharge to ensure stable and safe power supply. Through high-efficiency conversion equipment, the power from solar and battery systems can be stably delivered to AI data centers, enhancing energy efficiency and ensuring uninterrupted operation.

10kW PV Storage Inverter -- This is a hybrid inverter system featuring a fully integrated design with Nexperia’s third-generation SiC (Silicon Carbide) technology. The hardware includes dual interleaved BOOST circuits, a three-phase NPC-T topology, and a CLLLC circuit. It supports bidirectional power conversion and integrates MPPT solar control, three-phase rectification and inversion, bidirectional DCDC control, and charge/discharge management. An external HMI panel can display real-time system status and upload data to the cloud.

A battery management system is like the brain and guardian of the battery, playing a crucial role in solar energy storage. Its primary tasks include monitoring battery voltage, current, and temperature to ensure safe charging and discharging, preventing overcharge/discharge or overheating, balancing each cell module, and extending battery lifespan. It also issues alerts or shuts down systems in case of anomalies. In AI data centers powered by solar energy, BMS ensures batteries operate reliably and efficiently, maximizing green energy use while ensuring uninterrupted 24/7 operation.

In solar and battery storage systems, system interface and control act as the intelligent command center, integrating and coordinating the entire system’s operation. It connects PV panels, batteries, inverters, etc., to work in sync, monitors energy flow and system status, enables smart control and scheduling, and automatically switches between power supply and storage modes based on grid demand. It also communicates with upper-level platforms like EMS or AI data center control system. Robust system interface and control solutions enable stable, efficient, and intelligent support for AI data centers.

HMI for PV inverter and energy storage system -- This reference design is a generic HMI solution that connects with energy storage systems (like PV system with inverter/MPPT and CLLLC battery charger) through CAN bus. Developed by Arrow dashboard, it supports PV storage protocols, includes 5 buttons, an LCD display, RTC feature, and multiple external interfaces: CAN-FD, Ethernet, Wi-Fi, Bluetooth, and dual RS485 interfaces (for connect to KNX or other external devices).

Smart AFCI 2.0 solution -- This solution uses NXP’s MCXN547 MCU, supporting 8-channel sampling and automatic marking. PV systems are prone to arc faults at multiple connection points, and traditional detection often results in false alarms or missed detections in actual applications. This not only brings safety risks but also increases operation and maintenance costs. Based on MCX N series MCUs, this AI-powered industrial arc fault detection greatly improves accuracy and system safety.

As artificial intelligence continues to drive global progress, we must also consider how to ensure that this intelligent power can coexist with our planet. From solar-powered energy to smart battery management and integrated control systems, every step of technological innovation is a step toward sustainability. The AI data centers of the future will not only be the heart of information but also symbols of green transformation. Let us step together into a smarter, cleaner tomorrow - where technology and nature thrive under the sun.