How advanced digital technologies are transforming wastewater treatment and biosolids

The wastewater industry faces the unique challenge of managing variable influent flows and loads while ensuring a consistent effluent quality, or at least one that meets all the effluent limits. In addition to this challenge, one can also consider: aging infrastructure, climate change, population migration, deferred maintenance and other environmental and human stressors that make managing wastewater and biosolids a bigger challenge than ever.

As these stressors increase in magnitude, the traditional approaches are struggling to keep up. The result: the industry is turning to digital technologies to make a significant impact. Among the most promising tools are advanced analytics and artificial intelligence (AI), which are helping utilities with data driven strategies to optimize processes, improve sustainability, and recover valuable resources.

These tools can equip utilities to:

• View performance of assets in near real-time to identify risks and vulnerabilities before assets fail and impact customers.
• Make proactive decisions to improve levels of service by offering operational improvement options in place of expensive capital projects.
• Make the most of sensors, smart devices, SCADA, and telemetry already embedded at treatment facilities.

To unlock greater value, utilities need to shift from only monitoring historical performance to proactively utilizing their data. The new array of technologies and tools is now available to assist them to proactively monitor, manage and eventually automate many critical operational processes in near real-time.

Future-ready intelligent wastewater systems

Wastewater systems are complex and keeping them running smoothly is a monumental task. The industry is increasingly adopting the concept of “intelligent wastewater,” which uses AI, machine learning, and advanced analytics to optimize operations and decision-making.

Intelligent wastewater is an evolution of “smart wastewater” which integrates sensors, data, and monitoring to improve the treatment of wastewater but still requires significant human intervention for decision-making. Three of the possibilities becoming a reality include the following.

1. Near real-time problem solving

The water industry has invested in operational technology, sensors, instruments, SCADA and historian, but the data these tools produce has limited inherent value for “intelligent water.” By aggregating this information, utilities can analyze data from different sources as denoted in Figure 1 (p. 38), and provide operators insights for proactive actions to optimize processes. As a result, utilities will have a consolidated view across the entire plant to transform problem solving at every stage of the value chain.

2. From corrective to predictive maintenance

AI analyzes historical and near real time data — like temperatures, pressures, vibration, and flow rates — to detect anomalies and predict potential failures. This means less unexpected downtime, fewer costly repairs, and a longer life for critical infrastructure as well as optimized maintenance schedules.

3. Optimizing every drop

There is an increased emphasis on optimization of process, especially from a resource consumption perspective. AI-based analytics have been deployed to assist facilities in fine-tuning processes like energy use, chemical dosing, pump schedules and aeration control. By adjusting these in near real-time in conjunction with weather patterns and demand fluctuations, utilities achieve significant cost savings and conserve resources while ensuring high levels of performance.

The following discusses specific examples and successful deployment of these technologies.

Example 1: Better dewatering efficiency

Dewatering can be costly and energy-intensive, and operating centrifuges consumes a lot of power and polymer. Meanwhile, a drier cake can significantly reduce transportation costs. Finding the optimum operating condition where excessive power and polymer use is avoided while minimizing transportation costs is an ongoing challenge. AI helps utilities make smarter and more efficient decisions using data that is already collected and may not currently be contributing to improved performance.

Early adjustments using historical data: AI models can be used to monitor and optimize the performance and turning of processes to reduce process variations, compensate for disruptions and ensure the process operations close to optimal to maintain stable operations and maximize cake solids.

Removing noise control systems are only as good as the inputs to which they respond. Bad sensor data can mean chemical dosing issues, asset downtime, compliance issues as well as “noisy” data foundation for continuous improvement programs. AI can assist to continuously monitor and analyze sensor data to target anomalies and minimize their potential impact.

Example 2: Optimizing operations through AI

A groundwater treatment facility in Southern California utilizing reverse osmosis (RO) to purify 3.25 billion gallons of recycled water annually faced inefficiencies in data processing and operational reporting. The challenge included manually normalizing SCADA data and generating more than 100 performance plots monthly, a time-consuming process prone to human error.

Leveraging advanced analytics, an adaptive framework was implemented to automate data normalization, integrate multiple data sources, and streamline key performance indicator (KPI) analysis. This reduced manual effort, increased accuracy, and standardized reporting, while enabling plant managers to optimize maintenance schedules and forecast membrane performance. The solution enhanced operational efficiency (chemical and energy usage), and decision-making capabilities. In addition, an algorithm was created that was applied to learn from previous patterns and predict future fouling indicators of membranes to anticipate maintenance in weeks in advance.

Example 3: Using energy intelligently

One of the most energy-intensive operations is aeration during secondary treatment. Today, AI-driven aeration optimization is transforming wastewater treatment by enhancing efficiency and reducing operational costs. Analytics and machine learning monitor and analyze trends of ammonia-nitrogen concentration, which is used to adjust dissolved oxygen (DO) setpoints in aerated zones in near real-time. Leveraging AI to automate this process achieves lower energy costs, optimizes nitrification, and ensures compliance.

Recent research is showing nutrient removal can be optimized, both in terms of performance and energy use, by consistently reducing DO setpoints. Many operators may be concerned low DO operations could lead to an effluent violation due to loss of nitrification. In this case, AI is used to track system performance to provide early warning if performance starts degrading and to ensure violations are avoided.

To summarize, intelligent wastewater has the potential to improve asset management through enhanced predictive maintenance. By enabling equipment failures to be anticipated and avoided, both downtime and maintenance costs can be reduced. In addition, seamless connectivity and rich visualization along with AI and advanced analytics will enable utility staff to analyze operating scenarios, quantify the operational changes that will have on key performance metrics, and identify causes of performance variations.

By improving access of data, operators are enabled to make informed decisions in real time. This further increases regulatory compliance, which is crucial in maintaining the standards and safety of operations. As the workforce ages and more operators retire, there are fewer people available to manage these systems. Intelligent wastewater can relieve operators from routine troubleshooting tasks, giving them the bandwidth to focus on more impactful work.

In closing, utility staff — engineers, planners, operators, etc. — will need to develop capabilities in AI and analytics to remain competitive — both on an individual professional level as well as to help their respective organization.

Over time, staff can go from small AI and analytic projects to plant-wide optimization with deep applications of AI and analytics. Their deep domain expertise provides a foundation to transform waste water treatment and biosolids management.