Automating the Edge with Robotics

Mobile microgrids: Tapping into the rugged edge to ensure power reliability anywhere

Categories Edge Computing News  |  Guest Posts

By Dustin Seetoo, director of product marketing for Premio

Real-time decision-making and predictive analytics are an increasing imperative among infrastructure operations — powered by rapid digital transformation and an intensifying demand for automation upgrades across the board.

Rugged edge computing is critical in this landscape, speeding data processing based on myriad sensor input data and supporting access and analytics close to the data source. At the same time, engineers often find themselves pressed to design compute solutions for environments that are unstable, uncontrolled, and operating in stark contrast to a data center structure.

Industries face an increasing need for specialized hardware that can run intelligent software algorithms efficiently and consolidate workloads that matter most — machine learning and intelligent decision-making capabilities that improve automation. Specifically developed to endure the rigors of harsh usage conditions with built-in durability and ruggedized features, rugged edge computers are primed to tackle the challenges inherent to the current array of infrastructure applications.

The nation’s power grids provide a prime example, illustrating the risks being faced by infrastructure computing and the innovation that can be unlocked by rugged edge systems.

Powering up with the rugged edge

Overwhelming demand, extreme weather events, cyber terrorism strikes, and more have put today’s power grid in serious jeopardy. Electrical outages are on the rise and such failures pose significant challenges for businesses and individuals alike. One power solutions company specializing in remote and grid-paralleling applications is working to remedy the situation with a new breed of power solution —– small-scale, self-contained microgrid trailers. Operating independently or in conjunction with other microgrids, these units ensure a reliable power supply, enhance resilience and mitigate the impact of grid disturbances.

Microgrids, as defined by the National Renewable Energy Laboratory (NREL), are interconnected loads and distributed energy resources that function as a single controllable entity with respect to the grid. They can connect to the main grid or operate in isolation (island mode), improving reliability for customers. The implementation of microgrids enhances power quality, integrates on-site generation resources, reduces peak demand charges, and provides standby power generation.

To manage its microgrid operations effectively, the company employs an advanced energy management and generation platform, facilitated by an intuitive software portal. The platform enables grid operators to access power services, track emissions, utilize energy storage, and streamline back-office management. By incorporating artificial intelligence (AI) and machine learning, the platform offers real-time insights. This data can be used to monitor equipment conditions, analyze energy consumption rates, generate accurate cost projections, and support environmental reporting.

GPUs bring more power to the rugged edge

As technological advancements and growth objectives emerged, the team came to realize GPU processing would be necessary and identified Nvidia A2 GPUs as the ideal choice due to their suitability for AI, machine learning, and deep learning applications. Unfortunately, the systems’ existing consumer-grade computers were incompatible with the GPUs; significant hardware improvements would be required to optimize microgrid operation. Compute power, ruggedness, component availability, and scalability were all crucial considerations.

By leveraging the rugged edge and embedded computing technology, the company found a GPU-compatible solution to its hardware challenges. The latest AI edge inference computer offered high-performance socket-type processor design, ruggedness, and modular capabilities. With improved reliability in extreme temperatures, wider input voltages, and resistance to shock and vibrations, this new system integrated the Nvidia A2 GPU into its rugged PC architecture, facilitating efficient edge computing. This empowers the platform to execute complex algorithms for edge compute applications which in turn enables organizations utilizing mobile microgrids to access mission-critical business insights in near-real-time, enhancing their ability to respond to situational data effectively.

To combat potential heat generation resulting from GPU usage, a custom airflow chamber was designed into the solution, ensuring optimal thermal management. By effectively mitigating excessive heat and related problems, the overall reliability of the system was enhanced.

Right away, the power solutions company achieved significant savings related to time, resources, and financial cost. With a consistent hardware configuration, they eliminated the need for re-engineering and testing new platforms. The reliability of the systems resulted in zero defects reported to date, enabling the company to focus its resources on customer service, platform growth, and innovation.

This approach has had an enormous impact on the power solutions company and has great potential for the microgrid industry at large. By combining rugged edge computing with powerful GPUs, the company has developed a scalable and efficient hardware solution that offers reliable and resilient power solutions in the face of frequent grid outages. With the assurance of quality products and on-time delivery, they can confidently focus on expanding their expertise, engaging customers, and developing new products. Such success can be readily replicated for other critical infrastructure applications to vastly improve systems’ performance and reliability — and keep life as we know it moving ever forward.

About the author

Dustin Seetoo is the director of product marketing for Premio. Premio is a global solutions provider specializing in the design and manufacturing of computing technology from the edge to the cloud.

DISCLAIMER: Guest posts are submitted content. The views expressed in this post are that of the author, and don’t necessarily reflect the views of Edge Industry Review (EdgeIR.com).

Article Topics

 |   |   |   |   |   |   |   | 

Comments

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Featured Edge Computing Company

Edge Ecosystem Videos

Automating the Edge

“Automating

Deploying AI Models at the Edge

“Deploying

Latest News