Machine Learning

Analyze, learn, and adapt.

Machine learning is the foundation upon which the Vigilent Dynamic Cooling Management® System is built. Every minute, data from hundreds or thousands of environmental sensors is collected across the wireless mesh network, where it makes its way to the central artificial intelligence (AI) Engine. There, the data is analyzed and interpreted, and meaning is extracted. Sophisticated dynamic control algorithms then send commands to the facility’s cooling system in real time, making decisions designed to optimize performance.

Trends are analyzed and patterns extracted from days, months, or years worth of historic data, which supply our prescriptive analytics with the information it needs to provide the insight that will inform your decision making.

Our patented approach constantly learns the intricacies of your facility and, with closed-loop control, ensures that it is operating at top performance.

Find out how Vigilent fits into your business Get Started

Our approach to machine learning

How our system works

The speed and depth of our machine learning solution makes our results almost impossible to match. Measuring and monitoring are a good and necessary start, but it’s essential and imperative to apply intelligence and automatic control to manage the complex interactions among heating and cooling sources, and airflow within the facility.

The machine learning technology analyzes historical performance to predict future behavior. It evaluates courses of action to determine the one that yields the optimum result, minimizing energy usage while protecting uptime or tenant comfort. And it provides you the power to make some of your own what-if decisions, with support to evaluate different cooling scenarios by uncovering the intricate and non-obvious relationships between heat loads and cooling resources.

Vigilent artificial intelligence makes predictions

Sensor to measure temperature

Measure and monitor

Hundreds to thousands of wireless sensors throughout your facility gather environmental information every minute, and deliver those readings to the AI engine.

This is what’s known as Big Data – amounts of information that are large relative to the analytical tasks to be solved, and that can provide deep insight into relationships and trends.

Analyze and Learn

Analyze and learn

The AI Engine constantly compares current environmental readings with desired ranges, historic trend data, the current load, and cooling resource availability and efficiency. It determines what changes need to occur to maintain desired temperatures with the least amount of cooling spend, using advanced algorithms fine-tuned to the very specific needs of this task.

Real behavior is compared to expected results, and the system continually learns and improves, as equipment behavior and environmental response are ever-more-accurately characterized.

Over time, the AI Engine evolves to provide increasingly optimized results.

Optimize thermal environment

Adapt and control

Adaptive control allows the system to respond to random and never-ending changes in the environment – whether it be the addition of more server racks, or someone closing a door to a room.

Commands are dispatched by the system to the cooling infrastructure, where they are automatically implemented by turning equipment on or off, or adjusting fan speeds. And this cycle continues over and over, in a closed-loop, with constant adjustments every minute of every day of every year from the moment it is deployed.

Unprecedented insight

Analytics help to surface long tail events.

By collecting Big Data about your facility’s environmental performance every minute, and storing it indefinitely, the Vigilent system builds a data set that enable powerful analysis. The prescriptive analytics used to decipher meaning out of this data can surface trends that might otherwise go unnoticed.

It is finding those rare and rarer events that enable us to reduce costs, deliver improved uptime, identify stranded capacity, and provide the non-intuitive insights that can help you rethink how your facilities are architected.

Prescriptive analytics provides the only way to robustly capture the nuanced approach your facility requires, and deliver it in a scalable, measurable way.

Big Data vs Event Frequency Graph

The Vigilent AI Engine

AI-EngineThe Vigilent AI Engine contains the sophisticated machine learning that makes everything possible. It’s a special-purpose, dedicated hardware appliance that operates the powerful control software which is able to provide such a high level of dynamic control.

The AI Engine serves as the brain of the Vigilent system. It’s where the mountains of Big Data from the network of environmental sensors are aggregated, and stored indefinitely. The engine then analyzes and process that data – back to the day it was first started, if necessary or valuable – to inform decisions on what to do next.

The high-performance hardware engine is specifically designed and tuned to operate the specialized, patented control software at the heart of the Vigilent system. It’s designed to reside on your internal IP network, where it communicates with network gateways, receives monitored data, and issues control commands. It’s designed for high reliability and high availability, and includes significant failsafes to ensure your facility is always protected.