Century Engineering turns to the cloud to keep our waterways clean

Solving challenges to deliver higher performance

A key driver for Century was the development of this innovative stormwater management solution into a product ready for commercial-level applications. As a result, the performance of the system had to be resilient under highly variable conditions, reliable so that it effectively improves water quality and reduces runoff, and secure to prevent unwanted intruders into the system.

These objectives were met thanks to Rackspace's recommendations. Rackspace's team worked on a solution for Century in two phases. First, it focused on refining the prior proof-of-concept hardware solution by integrating useful, disparate functions to read environmental data and control equipment. A key design consideration included overcoming a prior challenge with the analog-to-digital circuitry interface, verifying that the existing loop conversion did not destroy the analog signals.

Filling in the gaps with AWS Cloud and IoT expertise

In phase two, the Rackspace team focused on enhancing the AWS environment and making improvements to the hardware by defining requirements for tamper detection and alerting, managing the signal-to-noise ratio of the existing pressure transducer, and providing component guidance, such as with the cellular modem.

“The sensor that measures the water depth is an analog device,” said Bathurst. “The signal it sends looks like an EKG report. Obviously, water surfaces don’t modulate like EKGs. Rackspace helped us to solve tedious issues like this.”

“There were many little fixes like that along the way through which we were able to leverage Rackspace's experience,” Bathurst said. “They were able to listen to our objectives, review our existing system, and apply their in-depth knowledge of IoT and automation to help us implement industry best practices into the SmartSWM™ product. The system is a lot more reliable and secure thanks to Rackspace's support.”

Delivering high performance via real-time remote access

“We can extract hyper-local weather forecast data for any location,” Bathurst said. “We can also monitor our solar generation and consumption, and water temperature. That feeds into our backend decision support system, which sits in the cloud. It computes the optimal water surface elevations of the ponds, along with the discharge rate at which they should be releasing stormwater.

“All of that information is then transferred to our on-site micro-controller. Then it dictates whether the release valves are opened or closed, and by how much. Everything is programmed. And we can go in remotely and perform actions, such as change the stormwater discharge rate if we need to.”

Making land use more efficient

“A major benefit to smart stormwater management is that we’re able to use land more efficiently,” said Bathurst. “Stormwater management facilities are different sizes and hold different volumes of water. With the SmartSWM™ technology, we can actively manage stormwater facility volumes and, as a result, each stormwater facility can occupy a smaller land ‘footprint.’ This efficiency yields more land for development or alternate uses.

“In addition, our experience has shown that public entities appreciate the higher pollutant removal performance, land efficiency and minimization of flooding and land erosion,” Bathurst said. “It’s just a new era in stormwater management.

“In my mind, the future of stormwater management involves connected, intelligent and adaptive infrastructure. We are already working on making interrelated facilities in a geographic region work as a system to further improve how we manage an entire watershed. This presents us with the ability to improve water quality throughout larger regions, not just discrete facilities, resulting in even greater positive impacts.”

Keeping an eye on the future of innovation

As far as the SmartSWM™ team has come in developing this innovative approach to stormwater management, Bathurst already sees an opportunity to take it to the next level. “Traditional stormwater modeling techniques consider the hydrologic characteristics of watersheds to estimate runoff volume,” he said. “But we’ve come to recognize those estimates to be somewhat inaccurate. It can vary wildly. This is likely due to the industry-accepted practice of assuming average conditions for parameters like antecedent soil moisture condition and mischaracterization of the soil’s ability to absorb water. However, once we have accumulated enough precipitation and runoff data, we can then apply machine learning to create predictive models that tell us how each stormwater facility will respond to weather events based on how the facilities have responded in the past.

“Machine learning is where we think we’re going to make our next big improvement,” Bathurst added. “We’ve already started developing algorithms to begin training. This is going to be another paradigm shift.”

Staying out in front of the market is critical for smart technologies. Bathurst said, “Working with Rackspace has allowed us to stay on top of the available technologies, and they have supported our project in a way that reinforces we are building one of the most reliable IoT and automated approaches to stormwater management.”

*Photos provided courtesy of Century Engineering, Inc.