By Mike McDonald
A combination of fluctuating and extreme wet weather patterns and aging infrastructure is causing problems with wastewater collection systems across North America like never before. As hundred-, even thousand-year storm events increase in frequency, a holistic understanding of systems and their capacity is needed to plan proactively to protect property and assets.
FlowWorks Infinitii I&I uses machine learning, advanced analytics and statistical analysis to provide ongoing, real-time data on the wet weather response of sewer systems. Using automated tools that are built on proven methods of storm event and infiltration and inflow (I&I) analysis, this web-based software provides time-saving calculations that can help pinpoint areas for immediate investigation after a storm event and provide insight into historical trends that can indicate areas of aging infrastructure in need of upgrade.
It also allows for users to convert their projects and studies into a manual form that allows for the data to be manipulated and exported to a CSV file.
Infiltration happens when groundwater enters a sewer system through holes and cracks in manholes and sewer pipes, and inflow happens when rainfall or snowmelt enters a sewer system from sources like roofs, downspouts and storm drains. Rainfall-derived infiltration and inflow (RDII) into collection systems has long been recognized as a major source of operating problems and poor performance. High levels of I&I reduce pipeline capacity in the collection system that would otherwise be available to transmit sanitary flow.
As a result, during extreme storm events, there may be sewer surcharging, local overflows of untreated sewage, increased incidences and volumes of combined sewage overflows, as well as increased operating costs. RDII is also the main cause of sanitary sewer overflows (SSO) into basements, streets or nearby streams, and can cause operating problems at wastewater treatment facilities.
SSO control plans are necessary that can align with municipalities’ projected annual capital budgets and provide flexibility in future improvements. Currently, there are seven municipalities across Canada and the U.S. piloting FlowWorks Infinitii I&I. In March, the City of North Vancouver became the latest municipal partner to pilot this software.
“I&I management is an important long-term initiative that should be continually supported by reliable and timely monitoring data. Having real-time I&I analysis updates will be valuable for us as a small municipality. Our current team can evaluate our infrastructure and further characterize storm events that are driving I&I, so we can target potential problems and prioritize new projects,” said Dave Matsubara, the City of North Vancouver’s infrastructure engineer.
Infinitii I&I is designed to work as a real-time monitoring platform that provides web-based GIS maps of the locations of sensors and devices. The FlowWorks platform connects to any sensor or data source, and provides tools for historical, real-time and predictive analysis of environmental data. This platform has been used by dozens of North American municipalities over the past 12 years to provide immediate and accessible data.
A network of sensors that includes flow monitoring devices and rain gauges is needed to provide data for the I&I analysis. Quantification begins with accurate, reliable, and repeatable sewer flow data. “Flow data” is defined as continuously recorded, electronic, time-series sewer flow.
A pilot project can be initiated with three months to a year of monitoring data, but large amounts of historical data can be accommodated and actually enhance the performance of the machine learning enabled features of the software.
The dashboard provides a colour-coded map view of the different catchments being monitored. A catchment has at least one flow device and rain gauge associated with it, and the size of the catchment is defined by the municipality’s operations team. Thresholds can be set to indicate when there is an I&I event. The area will be coloured green, yellow, orange or red, with increasing severity.
The storm event finder feature uses the amount of rainfall and duration to automate the process of defining the date and time of a significant storm event. Machine learning is built, using linear regression models to help define the 15 most significant storm events. Time parameters can be set to specify periods of time, such as seasons, where significant storm events relative to that time can be found. This enables fast identification of storms for I&I studies, and is useful for engineers, who must study a variety of storms to ensure accuracy in their studies and reports.
Regression is a method of modelling a target value based on independent predictors. This method is mostly used for forecasting and finding out cause and effect relationship between variables. Regression techniques mostly differ based on the number of independent variables and the type of relationship between the independent and dependent variables. Linear regression is a type of regression analysis where there is a linear relationship between the independent (x) and dependent (y) variables.
The dry weather pattern finder feature predefines an optimized pattern to use when studying a specific storm event. It also allows users to quickly build their own dry weather patterns. Select several dry days, and the system will average them out to determine the pattern for a single dry day.
Infinitii I&I uses the I&I envelope method for analysis in the application since it is relatively easy to apply to collected data and provides a means for normalizing results. It is an economical and practical method of standardizing I&I analysis. This graphical method is based on a summary of rainfall and sewer flow data from flow monitoring. By plotting these results, a relationship between rainfall and rainfall dependent I&I (RDI&I) can be developed. The purpose of the I&I envelope method is to use a collection of recorded storm events to create a correlation between the amount of rain that falls in a catchment and the amount of I&I that shows up at the flow monitoring site.
The FlowWorks RDII features automatically generate a correlation graph for a selected catchment site. For each catchment, users chose the flow channel, dry weather pattern, rainfall site or storm event group RDII charts/correlation graphs. They can toggle between correlation graph view and RDII chart view at any time.
Alan Tse, from the consulting firm, Kerr Wood Leidal, reviewed the Infinitii I&I tool. “This application is very useful after a significant storm event, especially when flooding occurred. After any flooding event, the question anyone asks is: ‘Why did it happen?’ With Infinitii I&I you get a preliminary idea of when, where and how I&I was a factor. And there are already numbers and data to work with instead of rushing to crunch numbers to understand the significance of the storm event in relation to the sanitary system,” said Tse.
Mike McDonald is with FlowWorks Inc. This article appears in ES&E Magazine’s October 2019 issue.