How Artificial Intelligence Optimizes Water Treatment and Predicts Operational Challenges
Artificial Intelligence (AI) enables smarter, more efficient water treatment by continuously analyzing process data to optimize performance and anticipate operational issues before they escalate. Rather than relying on static settings or reactive interventions, AI systems adapt in real time to changing conditions—improving energy efficiency, membrane life, and water quality.
By processing flow rates, pressure, conductivity, turbidity, and cleaning logs, AI can: Predict ultrafiltration (UF) membrane fouling and trigger early intervention, Optimize cleaning-in-place (CIP) schedules based on actual performance decline, not rigid time intervals, Detect scaling trends in reverse osmosis (RO) membranes before irreversible damage occurs, Balance flow between treatment units to extend overall plant lifespan and minimize downtime. These insights allow operators to make better decisions faster, reduce chemical consumption, and prevent unplanned outages.
Examples:
Predictive Monitoring of UF and RO Membrane Fouling
Membrane fouling is a major operational challenge in both Ultrafiltration (UF) and Reverse Osmosis (RO) systems. Fouling can be caused by particulates, organics, biofilms, or scale, leading to increased transmembrane pressure (TMP), reduced permeate flow, higher energy consumption, and shorter membrane lifespan. SensQuant’s AI-driven platform continuously monitors key parameters such as differential pressure, flow decline, conductivity, turbidity, and cleaning frequency. By learning system-specific patterns, the AI can accurately detect early signs of fouling—before they affect performance.
PSIORI’s model for automated classification of cells is based on deep neural networks for image classification. It additionally uses some methods from classic computer vision. The model classifies cells based on microscopic images from the Single-Cell-Printer.
Our model recognizes non-viable cells by evaluating image information. Deep neural networks are especially suited to this kind of task as they are able to learn precise classification features through small differences in images. To facilitate this task for the neural network the images are preprocessed by methods of classical computer vision.
AI-Driven Cleaning-in-Place (CIP) Optimization
CIP operations are essential but often overused or delayed. SensQuant’s AI dynamically schedules CIP based on actual membrane performance degradation, chemical cleaning history, and operational load—saving water, chemicals, and labor.
A deep neural network classifies individual PCR curves. The network learns the typical form of exponential rise as a positive and can recognize its variations. The PSIORI model classifies curves into positive or negative but can distinguish examples with high uncertainty into their own class. In practice these cases would be given to a human processor to decide. The model achieves results comparable to state-of-the-art solutions after training on a low five digit number of PCR curves without being fine tuned and can be adjusted to new assays through simple retraining.
The model is based on PSIORI-STaR, a specific architecture of deep neural networks for processing multivariate time series. Since the deep neural network does not only learn the characteristic global form of a positive or negative curve, but also local features, a simple retraining on little data should be enough to transfer it to a new assay.
Scaling Detection in RO Membranes
Scaling caused by mineral precipitation is a leading cause of RO membrane failure. Our algorithms detect early signs of scaling through conductivity drift, pressure increase, and permeate quality degradation, enabling targeted antiscalant dosing or timely intervention.
The deep learning model is trained on thousands of ECGs. It detects time intervals that might contain atrial fibrillation. The software presents those to the cardiologist for evaluation. Our model works on 3-channel-ECGs as recorded by small devices that can be worn in everyday life. The model is tuned to not produce any false negatives and present any potentially suspect time intervals to the doctor.
The model is based on PSIORI-STaR, a specific architecture of deep neural networks for processing multivariate time series. It jointly analyzes the three channels of the ECG and is therefore robust against small irregularities produced by, for example, not attaching the device tight enough while wearing.
Dynamic Flow Balancing in Multi-Stage RO Systems
In complex RO installations with multiple stages or parallel trains, maintaining balanced flow is essential for efficiency, membrane longevity, and stable operation. Uneven load distribution can lead to premature fouling, energy waste, or suboptimal recovery. SensQuant’s AI platform dynamically adjusts flow rates between RO units based on real-time monitoring of key parameters such as feedwater quality, permeate flow, pressure differentials, and energy consumption. Instead of using fixed control logic, the system learns from historical and live data to optimize flow balance in changing conditions.
Process Anomaly Detection in Pretreatment and Chemical Dosing
Chemical dosing in pretreatment must be precise. AI tracks real-time sensor data and lab results to detect process drifts—such as under- or overdosing—and automatically adjusts feed-forward controls to stabilize water quality before membranes are affected.
The data is pulled from a data warehouse that can contain data from several studies and registers. Based on this, the backend of the data science tool does not store data. It instead stores filter and calculation rules for populations along with configurations of predefined analyses.
The tool enables various standard analyses as well as the creation of tables and specific analyses such as time-to-event analyses and Cox regressions. For approaches not included in the standard analyses, the tool provides a Jupyter notebook server with R and Python connected to the backend and data warehouse databases.
