Amazon’s High Desert solar farm with fully functioning artificial intelligence

Amazon officials recently announced that its first energy storage project, including a solar farm and batteries, has become fully operational in the High Desert.

Located in Baldy Mesa near Adelanto, the project is considered one of the largest solar and DC storage facilities in California, the company said Tuesday.

Additionally, artificial intelligence helps new solar and storage project provide energy even when the sun is not shining

Amazon officials recently announced that its first AI-powered solar and battery energy storage project has become fully operational in the High Desert.

Amazon uses solar energy

Solar panels mounted in rows in San Bernardino County’s western Mojave Desert at the Baldy Mesa solar farm convert enough sunlight into carbon-free energy and send it to the grid.

When the sun stops shining, a football field-sized energy storage system (BESS) next to the solar panels sends electricity collected during the day back to the grid, providing access to carbon-free energy at night.

Switching to solar and wind power is one of the fastest ways to help decarbonize the power grid, but the amount of renewable energy available may vary when the sun isn’t shining, Amazon officials said.

Combining a solar project with a battery storage system can help solve the problem of lack of sunlight, providing a constant supply of emission-free energy for more hours of the day.

At Baldy Mesa, a solar farm launched by Amazon, machine learning models powered by Amazon Web Services help predict when and how the project’s battery should charge and discharge energy back to the grid.

“Artificial intelligence is an important tool that is already helping our society transition to carbon-free energy and address climate change at scale,” said Kara Hurst, Amazon’s vice president of global sustainability. “Combining Amazon’s solar projects with AI technologies powered by AWS ensures that the grid and the customers it serves receive a consistent supply of carbon-free energy for more hours every day, while also helping Amazon make progress on our commitment to be more sustainable business.”

Amazon has so far enabled the development of 10 solar projects in conjunction with battery energy storage systems, representing nearly 1.5 gigawatts (GW) of battery energy storage capacity, company officials say.

Projects include Baldy Mesa as well as Bellefield near Mojave, the largest planned solar and storage project in the US. Additionally, Amazon also has its first rooftop solar panel combined with a battery storage module, which was installed at the Amazon Flight Center in San Bernardino.

All 10 projects are located in California and Arizona and support Amazon’s operations, including fulfillment centers, office buildings and Amazon data centers, by matching the electricity they use with 100% renewable energy.

Battery energy storage

At Baldy Mesa, the use of artificial intelligence to optimize battery performance is an emerging trend among carbon-free energy owners and operators, who are increasingly turning to machine learning to enhance carbon-free energy production and help stabilize the grid.

Digitization and the cloud have enabled the growth of data such as real-time weather data and historical network data. For example, according to the International Energy Agency, the world’s wind turbine fleet is estimated to generate more than 400 billion data points annually, from which artificial intelligence and machine learning (ML) models can learn to improve the operational efficiency of carbon-free energy design.

Baldy Mesa’s machine learning software, built using Amazon’s AWS SageMaker service, is expected to analyze up to 33 billion data points each year, according to solution provider Fluence. The software is designed to help extract more value from the Baldy Mesa battery by dynamically optimizing when to buy, store and sell power based on grid conditions.

The same ML solution deployed at a similar site in California helped predict last year’s statewide heat wave and helped stabilize the grid by delivering stored solar energy when it was needed most, Amazon officials say.

This AI-powered innovation is becoming increasingly important as climate change makes heatwaves hotter and more frequent, stretching power grids to their limits. As temperatures rise in homes and businesses, the capacity of ventilation units often increases to stay cool, putting pressure on network operators to meet the high demand, and at the same time the heat can lead to a reduction in the output of traditional thermal power plants, risking outages.

“We are very proud of our long-standing partnership with Amazon, supporting the company’s journey to match the electricity used by its data centers, fulfillment centers and other operations to 100% renewable energy by 2025.” – said Kleber Costa, Chief Commercial Officer at AES Corporation “Projects like Baldy Mesa and Bellefield operate around the clock, and battery energy storage combined with AI/ML-based trading is essential to ensure that renewable energy deployment is optimized and streamlined. energy to meet your needs.”

Amazon Aviation Center

An hour from Baldy Mesa at Amazon’s San Bernardino Air Center, a 5.8-MW rooftop solar array is connected to a 2.5-MW energy storage module.

When the sun is shining, solar panels power the Air Hub. However, when clouds roll in or the sun sets, the facility can instead seamlessly draw some of its power from the batteries. This enables the use of solar power in the middle of the night, providing the Amazon Air Hub with a stable supply of emission-free energy while reducing dependence on the grid during periods of high demand.

Amazon teams are currently developing an artificial intelligence model that will leverage ML capabilities as well as performance data from Amazon’s rooftop solar panels to help Air Hub and other Amazon buildings minimize energy consumption, the company said.

Teams first aggregate performance information from solar panels on the roof of the Air Hub and other Amazon locations, then combine it with local weather and building data in a central location in the AWS Data Lake, creating the foundation for advanced performance tracking and analysis.

Once operational, the future AI model is expected to offer predictive information on facility performance and energy generation. This task is currently impossible to perform by humans as it requires manual analysis and monitoring of each Amazon system and building.

“Battery storage projects enable increased use of renewable energy by helping ensure continued access to the grid for clean energy from solar and wind projects,” said Kevin Hagen, interim CEO of the Clean Energy Buyers Association. “Energy storage, smarter and more interactive load management tools, and artificial intelligence are among the emerging technologies that hold significant potential for a cheaper transition to zero-emission energy.”

Energy sector

Amazon’s work on innovation in San Bernardino and Baldy Mesa is one of many ways the company is using artificial intelligence to solve some of the most pressing sustainability issues and democratize our work in the broader energy sector with AWS customers and partners.

For example, Greenko, one of India’s leading renewable energy companies, has digitally deployed its entire wind turbine fleet to AWS, enabling near real-time monitoring and analytics using artificial intelligence. Similarly, Engie’s “Common Data Hub” built on AWS enables cities like Barcelona to reduce energy consumption by up to 15% through advanced data analytics and ML models.