top of page
Writer's pictureNathan Rudyk

Calgary water main break will not be an isolated infrastructure challenge and offers a perfect application for machine learning technology

The City of Calgary's current infrastructure challenge involving a state of emergency in Alberta's largest city is not and will not be unique, unfortunately. However it offers a perfect application for machine learning technology. 


In an interview published by The Globe and Mail June 16, Professor Tricia Stadnyk made the point that this isn’t merely a Calgary problem. Rather, it’s the canary in the coal mine for cities across the country.


The Canadian Infrastructure Report Card that monitors the state of Canada's core public infrastructure found that 30% of Canada’s water and wastewater infrastructure is at or beyond its service life.


Professor Stadnyk said communities stand to face catastrophic water issues unless they are proactive. She acknowledged that being proactive can cost billions of dollars but said it’s crucial to preserve water supplies.

“We assume with 100-per-cent reliability that we will turn on the tap and water will come out. But that’s not the way that any built structure or system works. There’s always a risk of failure,” she said.
On site camera image displayed alongside sensor data
Machine learning technology applied to site camera image displayed alongside sensor data

Machine Learning software from infinitii ai combines photo data with sensor data to significantly enhance the interpretation and utilization of underground cameras and sensors. Here are three applications:


  • View state of site over time – the photo channel can store years of images for comparison purposes.  This allows you to view images alongside level and velocity data to gauge the operational efficiency of the site over time.  Identify and track cracks and structural damage and prioritize maintenance as required.

  • Investigate unusual site activity – if you have alarms based on data channels, adding a camera to your site and a photo channel to infinitii flowworks monitoring can help easily identify the source of the problem.

  • Identify sensor anomalies and problems with monitoring hardware – adding a photo channel allows you to confirm that a suspect sensor reading is indeed a anomaly. Confirming bad sensor data can assist in making the decision to replace monitoring equipment.


Water utility operators can hover a mouse over a datapoint to see a thumbnail view of the underground photo, then click on the datapoint and open a viewer to see a larger version of the photo. This view can also be opened via the right-click context menu ‘View Photo’ option. In the Photo Viewer window, they can then press the previous or next buttons to scroll through the photos in the time series.


Hover mouse over a datapoint to see a thumbnail view of the photo, then click on the datapoint and open a viewer
Hover mouse over a datapoint to see a thumbnail view of the photo, then click on the datapoint and open a viewer

Combining infinitii flowworks photo channel data with sensor data that takes of advantage of infinitii auto qa/qc and infinitii auto i&i machine learning software ­– included in infinitii flowworks pro or priced separately for existing flowworks customers –­ can significantly enhance the interpretation and utilization of data from underground cameras and sensors in city-wide sewer systems.


With a camera installed and a corresponding photo channel in flowworks, utilities can see the impact of a storm event at a site, then use the flowworks event finder to identify when certain conditions are present then view the image associated with that event.


To learn more about making better sense of data from underground cameras and sensors in city-wide sewer systems, please click here to get in touch with a product specialist or visit www.infinitii.ai.

Comments


Commenting has been turned off.
bottom of page