TasWater: saving oysters with data
- ChallengePrevent and respond to spills in ecologically sensitive environments
- SolutionUse pump data from the PI System and analytics in Seeq Workbench to identify potential sewage blockages
- BenefitsReduced blockage-response time by up to 13 hours
On August 24, 2017, a customer in Midway Point, a small suburb outside of Hobart, Tasmania, reported a sewage spill on his property. Midway Point stands adjacent to the Pitt Water Nature Reserve, home to rare birds and butterflies and a large oyster farm. To make matters worse, the prized Pitt Water oysters were just coming into peak condition. TasWater, the state’s primary water and sewage service provider, responded immediately. Unfortunately, 6,000 liters of wastewater had already spilled into Pitt Water, shutting down production for three weeks. It was an ecological and economic disaster, and TasWater had to find a way to catch blockages and spills more quickly. In collaboration with Nukon, a PI System Integrator, TasWater began using the PI System as part of a pilot program. This initiative would use sewage pump station (SPS) data to spot and respond to blockages and spills before they turn into ecological disasters.
Sewage networks and sensitive environments don’t mix
Sewage systems and pump stations are, by nature, finicky assets. “Sewage spills are an unavoidable reality of managing a sewage network,” said Matt Jordan, manager of network asset performance at TasWater. “With the volume of our network, keeping it blockage free is impossible.” TasWater, which manages almost 5,000 kilometers of sewer mains, suffers about 2,000 spills and main breaks a year. About 70% of these are due to tree roots damaging sewer pipes, but those can be predicted and prevented through root cutting.
"The one that really gets us is foreign objects,” said Jordan. “These are items like nappies or diapers or hand towels or Coke bottles. You name it, and it somehow ends up in the sewage system. You may have the best maintenance strategy in the world, but the day after you clean it, a foreign object can end up in there. They are the ones that are unpredictable.
SPS monitoring: a simple innovation for the water industry
SPS sites have defined operating characteristics. For example, peaks in pump activity occur at the beginning and end of the day, when people generally use more water. Utilizing advanced analytics from Seeq, some of Nukon’s own tools, and the infrastructure provided by the PI System, Nukon and TasWater devised an innovative method for identifying potential blockages. By analyzing just a single data point—whether a pump was running—TasWater discovered that the time to fill, or duration between pump runs, was the key determinant of a station’s operating profile. When the time to fill extends beyond what is normal, the wet well (which collects the incoming water from the sewer system) may be taking too long to fill, and a blockage may be occurring upstream.
New insights and easy rollout
The system’s low implementation cost and quick deployment were a great benefit to TasWater. Using the templates feature of Asset Framework, the company quickly expanded the pilot project from the initial pump station to all the SPS sites in Midway Point within a month. Currently, TasWater tracks about 50,000 data points, a figure that is expected to quickly increase to over 200,000
Though TasWater hasn’t yet experienced another blockage event, its data model has already given it unexpected insight into its sewage network. For example, an increase in pump activity during a recent storm event revealed leaky segments in the sewer system. This knowledge about the location of leaky sewers has allowed it to target funding to minimize inflow and infiltration and avoid excessive pump usage.
After its pilot program concludes, TasWater hopes to roll out its new system to all its pump stations located in sensitive areas, thus expanding its ability to protect the health and safety of its customers, the environment, and those delicious oysters.
For more information about TasWater and the PI System, watch the full presentation here.