Solar farms from Amazon

What is the news about?

The solar panels at Baldy Mesa Farm are carefully placed in the western Mojave Desert in southern California. They convert abundant sunlight into emission-free energy and feed it to the grid. All solar power plants eventually shut down after sunset, just like any other power plant. But that doesn’t mean energy stops flowing on this solar and storage farm. In May, a BESS the size of a football field will begin transmitting energy generated by solar panels back to the grid, ensuring access to clean energy at all times of the day and night.

One of the fastest ways to help decarbonize energy grids is to switch to renewable energy sources such as solar and wind, but the amount of this energy can fluctuate when the sun is not shining. One way to solve this problem is to combine solar energy with battery energy storage, which can extend the time that emission-free energy is available. Baldy Mesa is a solar farm built, owned and operated by AES. Machine learning (ML) models powered by AWS help you schedule battery charging and discharging on a project-by-project basis.

The most popular news of the week:10 AI ML trends in supply chain management to watch out for in 2024

Why is this important?

To date, Amazon has enabled ten solar energy projects with battery energy storage systems, totaling nearly 1.5 GW of battery energy storage capacity. These developments include the Baldy Mesa and Bellefield solar and storage megaprojects, the largest such projects in the United States, and the San Bernadino Air Hub, where Amazon placed its first rooftop solar panel and battery storage module. A total of ten projects spread across Arizona and California contribute to the green energy mix that Amazon uses to power its data centers, office buildings and fulfillment centers.
Zero-emission energy owners and operators are increasingly turning to machine learning to increase zero-emission energy production and help stabilize the grid; one such trend is the use of artificial intelligence (AI) to maximize battery performance in Baldy Mesa.

To read: 10 AI ML trends in data storage to watch out for in 2024

Data such as real-time weather forecasts and past records of power grids have become accessible thanks to digitization and the cloud. For example, the IEA estimates that the global wind turbine fleet generates more than 400 billion data points annually, which can be used by AI and ML models to improve the operational efficiency of carbon-free energy projects.
The software at Baldy Mesa, which uses machine learning and was built using Amazon’s AWS SageMaker service, is expected to analyze 33 billion data points annually, according to Fluence, a solutions provider. The goal of the software is to maximize utilization of the Baldy Mesa battery module by adjusting optimal times to purchase, store and sell energy in response to changes in the grid. Last year, the ML solution was deployed at a comparable location in California. This helped predict a statewide heat wave and stabilized the grid by sending stored solar energy at critical times.

Climate change is causing heatwaves to become more frequent and hotter, putting a strain on electricity infrastructure. As a result, this AI invention is gaining importance. As temperatures rise, many homes and businesses turn up their air conditioning to stay cool. This puts a strain on network operators trying to cope with increased demand, and traditional thermal power plants may reduce their output, which could lead to outages. These major weather disasters have highlighted the need to accelerate the world’s transition to clean energy. The 2.5 MW battery energy storage unit and 5.8 MW rooftop solar array are located at Amazon’s San Bernardino Air Hub building, which is an hour away from Baldy Mesa. In nice weather, the Air Hub is powered by solar panels. However, a facility can easily switch to battery use for some of its electricity when the clouds roll in or the sun sets. By deploying solar panels at night, Amazon Air Hub can reduce its dependence on the grid during periods of peak demand and ensure a steady supply of carbon-free electricity.

Teams at Amazon are hard at work on an AI model that will leverage ML capabilities and performance data from Amazon’s rooftop solar panels to help reduce energy consumption in Amazon buildings like the Air Hub.
Teams are laying the groundwork for improved performance tracking and analysis by first collecting data from solar panels on the roof of the Air Hub and other Amazon locations. They will then combine this data with local weather and building information in a central location in the AWS Data Lake. Upon launch, the upcoming AI model is expected to provide predictions about facility performance and efficiency. Due to the time and effort required for humans to examine and monitor each Amazon system and building, this work is currently unfeasible.

Wrapping

To democratize our efforts in the broader energy sector with AWS customers and partners and solve some of the most important sustainability challenges, Amazon is using artificial intelligence in innovative ways. Two examples of these techniques are the San Bernardino and Baldy Mesa projects. The demand for renewable energy is growing, and AWS gives companies the tools they need to streamline their operations and innovate through the use of artificial intelligence. For example, Greenko, a leading renewable energy company in India, has connected all of its wind turbines to AWS, enabling near real-time monitoring and analytics based on artificial intelligence. Additionally, using cutting-edge data analytics and machine learning models, cities like Barcelona can reduce energy consumption by up to 15% with Engie’s “Common Data Hub” built on AWS.

Must Read: What is Experience Management (XM)?

(To share your insights with us through editorial or sponsored content, please write to [email protected])