Specifying Ekster & Associates control systems in your designs
DO/Nmaster, SRTmaster, BLANKETmaster and CLmaster are machine-learning based real-time control systems that replace the conventional PID control of aeration, sludge wasting, return sludge and chlorine dosing. They are supplied independently of the blowers, instruments and automation platform, so they can be specified in any upgrade without restricting the choice of equipment.
System architecture
- Runs alongside the existing SCADA system. Each controller exchanges data with the plant SCADA system using the standard OPC protocol. The existing PLCs or DCS continue to operate the valves, blowers and pumps. If the controller or its communication link stops, the plant's own control system responds according to the fallback logic programmed for that plant.
- Data validation first. All incoming data passes through a filtration stage. Faulty readings (spikes, frozen values, values out of range, physically impossible rates of change) are detected and not acted on, and operators are alerted.
- DO/Nmaster. Machine-learning modules calculate the dissolved oxygen, airflow and valve position setpoints for each aeration grid, with ammonia-based aeration control (ABAC) where ammonia analyzers are installed.
- SRTmaster. Model predictive control (MPC) calculates the waste sludge flow setpoint from on-line suspended solids and flow measurements, to hold the sludge age (SRT) target.
- BLANKETmaster. Return sludge flow is set clarifier by clarifier from the measured sludge blanket depth. Commissioning begins in advisory mode, in which setpoints are calculated but not applied.
- CLmaster. Sodium hypochlorite and aqueous ammonia doses are controlled together to hold the target CT (total chlorine residual multiplied by modal contact time).
- Operator authority. Operators set the setpoints and limits within which each controller operates, and can override it at any time.
Measured control accuracy
Pomona Water Reclamation Plant, Los Angeles County Sanitation Districts (U.S. Department of Energy final report, DE-EE0009509):
| Controlled variable | Result |
|---|---|
| Airflow, DO/Nmaster | 0.75% average error, 0.25% median (24-hour test, setpoint changed every 3 seconds) |
| Dissolved oxygen, DO/Nmaster | 0.06 mg/L average error, against 0.20 mg/L with tuned PID control (setpoints 0.5–2 mg/L) |
| Sludge age, SRTmaster | 0.13 days average absolute error (less than 2%) |
Over the twenty years to 2016, at SRTmaster installations, sludge age has been held within 5% of setpoint, and most of the time within 2.5% (WEFTEC 2016).
Instrument requirements
Control accuracy depends on the instruments. Ekster & Associates sells no instruments and provides recommended performance specifications, so that the instruments can be bought from any supplier.
- DO/Nmaster: fast-responding airflow meters, modulating actuators designed for process control, dissolved oxygen analyzers and, for ammonia-based control, ammonia analyzers.
- SRTmaster: two suspended solids analyzers (mixed liquor and return or waste sludge), a waste flow meter and a flow control element (a valve or a variable-speed pump).
- BLANKETmaster: a sludge blanket depth measurement in each clarifier and return sludge flow control.
- CLmaster: chlorine and ammonia analyzers, flow measurement and the existing chemical feed pumps.
The published record
- U.S. Department of Energy final report: Young, M. et al. (2025). Transforming Aeration Energy in Water Resource Recovery Facilities through Suboxic Nitrogen Removal. DE-EE0009509, DOI 10.2172/2587564.
- PG&E post-installation verification report: QuEST (2018). California Wastewater Process Optimization Program, City of Chico WWTF.
- FAQs and white papers for DO/Nmaster, SRTmaster, BLANKETmaster and CLmaster (PDF).
- Nine conference and journal papers from 1999 to 2016, including "Twentieth Anniversary of the Real Time Sludge Age Control" (WEFTEC 2016) and the Sacramento Regional SRTmaster paper (WE&T 2013).
Independence
Ekster & Associates sells no hardware (no blowers, analyzers, actuators or control valves), no automation platform, and no service contract that places a vendor inside plant operations. Specifying the control system does not tie the owner to any equipment supplier.