Improving Collection System Operations with Digital Solutions

Utilities are facing increasing pressures in operating and maintaining their collection and treatment systems, leading to digital solutions to assist with optimization. Smart sewers are collection systems with remote sensors that relay level and flow information, often with cloud-based machine learning to provide insights into the behavior of the sewers. This paper focuses on a solution that provides a dashboard for smart sensors and machine learning to provide insights into the behavior of wastewater collection systems. The solution first puts “eyes on the system,” which includes pulling in data from many sources. The data are analyzed using machine learning to ascertain the behavior of the system and create alerts based on that learning. The alerts are refined over time, including alerting to blockages or impending overflows, which allows operators to proactively respond before an overflow occurs. Additionally, artificial intelligence is used in conjunction with a hydraulic model to continuously calibrate the model and make predictions of system behavior where sensors are not in place. The solution can then use forecasted rainfall to simulate and optimize control setpoints to minimize overflows and maximize catchment while not overloading the downstream wastewater treatment plant. The first case study is at United Utilities in the United Kingdom, where this solution was developed. The case study demonstrates how United Utilities has become more proactive in their response by adding sensors in their collection system, reducing spills to the environment. The second case study is in the eastern United States. The second case study demonstrates how the digital solution tool has enabled the utility to modernize maintenance approaches in their collection system to minimize overflows.