By Amir Cahn
Growing urban populations, aging infrastructure, rising customer expectations, limited budgets, and climate change are increasingly putting a strain on water and wastewater utility management. While global water utilities have long operated SCADA systems and GIS to monitor critical functions across their network, the use of Big Data analytics, artificial intelligence (AI), and machine learning are now becoming more widespread. To appreciate the impact of these solutions, one must first identify some of the main challenges involved in water and wastewater network monitoring, as well as the benefits of real-time, proactive versus traditional, reactive approaches. This impact is intertwined by economic, social, and environmental dimensions, which are best enabled through industry collaboration.
On The Water Side
A key role for any water utility is to effectively measure its customers’ consumption patterns. Traditional water metering systems must be manually read, and customer water usage volume is only recorded on a periodic basis. In contrast, smart meter devices connect a conventional water meter to an online data logger enabling continuous monitoring of water consumption. Smart meters also identify leaks so that customers can act quickly to save water, which otherwise might not be identified for several months. Through AI, virtual agents and chatbots can now fully automate customer service.
Water loss from aging, leaky infrastructure is a major contributing factor to water stress. It is estimated that 30 percent of all global drinking water is lost to “non-revenue water” (NRW). The most widely deployed leak-control strategies involve regular network sweeps by field detection teams using such methods as noise logging and step-testing. However, these techniques can be time-consuming and costly. Advanced leak-detection solutions apply fixed network sensors or analytic software to remotely alert system operators about various network problems, which prevents water loss, large bursts that can cause significant property damage. Drones built with AI can even be trained to automatically identify asset defects and predict failures without interrupting operations.
Another primary challenge for water operators is obtaining a reliable assessment of water quality over time. Water quality monitoring is typically conducted by manually collecting discrete samples sent to a laboratory for analysis, which only provide a limited snapshot of current conditions. Today, water quality sensors have evolved from traditional lab-based sensors to “in situ” sensors capable of real-time measurement of water-quality parameters on site. These devices can locally process and transmit the measured data, enabling decision-makers to receive data from multiple remote sensor devices in a timely manner.
On The Wastewater Side
The limited capacity of current stormwater infrastructure faces mounting pressure due to changes in urban density and unpredictable weather patterns. During rainfall, wastewater systems can overflow when collection system capacity is exceeded, blocked, or there is a mechanical failure, which can lead to large volumes of untreated wastewater, toxic materials, and other debris being discharged directly into nearby water bodies. Combined sewage overflows (CSOs) are a significant source of elevated contamination in many regions of the world and are also particularly difficult to quantify and regulate due to their abrupt nature.
The past decade has witnessed significant advances and cost reduction in novel stormwater sensors, wireless communications, and data platforms such as “wireless sensor networks” (WSNs). WSNs are ideal for low-power and low maintenance applications, making them well-suited for the monitoring of large water systems like rivers and watersheds. AI can also now assist stormwater operators with reducing hours of CCTV pipeline inspection footage to a few minutes through automatic analysis.
Industrial wastewater poses an even more considerable threat of environmental damage since it contains a broad range of organic and inorganic pollutants, often in high concentrations depending on the specific industrial process. These effluents are highly variable, which can result in shock loads, toxic effects on biological treatments, and the contamination of nearby soil and groundwater. Instead of a utility relying on infrequent, regulated site visits, technologies are now available to indicate sewage quality parameters in real-time. For example, samples can be automatically analyzed and then sent to an authorized laboratory to make an accurate analysis, enabling 24/7 control on suspected contaminating factories.
Gaining A Global Perspective
If you are interested in further exploring the connection between smart water innovation and its impact, we invite you to join the upcoming, SWAN 12th Annual Conference to be held through a hybrid format May 24-26, 2022, in Washington, DC. SWAN’s flagship event will bring together senior utility managers, technology providers, researchers, consultants, regulators, and academics at the leading global smart-water event of the year. Last year’s Conference attracted over 650 industry professionals and 90 unique utilities from over 40 countries
There is also now an opportunity to share your unique insights through the Conference Call for Abstracts (due January 14th).
Learn more about the Conference and register here: www.swan-2022.com
Amir Cahn is the executive director of the Smart Water Networks Forum (SWAN), the leading global hub for the smart water and wastewater sectors.