AerationAI™

Standalone, Site-Specific Aeration-Optimization Software

Predictive Aeration Optimization for Reliable Treatment and Lower Energy Use

AerationAI is a standalone wastewater aeration-optimization software application that works through a facility’s existing plant data, PLC/SCADA systems, blowers, variable-frequency drives, valves, and aeration infrastructure.

AerationAI combines site-specific predictive tools, trusted operational information, dynamic airflow-to-DO response modeling, engineering constraints, and governed control to determine how much airflow is supportable as plant conditions change.

Rather than simply reacting to a fixed dissolved-oxygen setpoint, AerationAI predicts baseline aeration requirements, evaluates the direction and dynamics of process response, and adjusts airflow within plant-specific treatment, equipment, information-confidence, and operating boundaries.

The objective is not simply to reduce airflow. It is to provide the airflow needed for reliable biological treatment—no more and no less—as operating conditions change.

Request a Demo
Contact McC AI Group

The Limitation of Fixed DO Setpoints

Dissolved oxygen remains one of the most important measurements for biological wastewater treatment. However, conventional aeration control commonly relies on DO feedback after a process change has already begun affecting the basin.

Changes in organic and ammonia loading, hydraulic conditions, biological activity, temperature, and process operation can alter oxygen demand before the resulting effect becomes fully visible at the DO probe.

The measured DO response may also be delayed by:

A controller that reacts only after DO has moved substantially may therefore respond later than desirable. Conversely, a controller that reacts too frequently to small DO changes may chase measurement noise or repeatedly change airflow before the effect of the previous adjustment has had time to appear.

AerationAI addresses both problems by combining predictive airflow estimation with response-aware DO feedback.

Predict the aeration requirement. Observe the process trajectory. Allow the previous action time to work. Correct only when the evidence supports another adjustment.

What AerationAI Does

AerationAI develops a plant- and basin-specific representation of aeration demand from historical operating data, treatment performance, equipment characteristics, and site operating constraints.

During operation, AerationAI combines the resulting predictive model with available current plant information to determine the supportable airflow for each controlled basin or aeration zone.

The system evaluates not only the current DO value, but also:

This allows AerationAI to distinguish between a condition that requires additional airflow and one in which the process is simply still responding to a previous adjustment.

How AerationAI Works

1. Establish the Site-Specific Operating Model

Historical plant information is used to characterize relationships among loading, influent flow, DO, airflow, treatment performance, basin configuration, equipment operation, and energy use.

These relationships form the basis for plant- and basin-specific predictive airflow models.

2. Validate Operational Information

Available DO, airflow, flow, equipment-status, and related signals are evaluated for plausibility, consistency, reliability, and suitability for the intended decision.

Questionable information can reduce control authority, trigger verification, or cause AerationAI to hold rather than act.

3. Predict Aeration Requirements

AerationAI uses site-specific predictive tools to calculate baseline airflow requirements under the current operating condition.

Historical organic and ammonia loading relationships are incorporated into the site model even when real-time BOD, COD, or ammonia measurements are not available.

4. Evaluate the Dynamic Process Response

AerationAI evaluates:

5. Apply Engineering Guardrails

Proposed airflow adjustments are evaluated against plant-specific operating envelopes, equipment capability, minimum mixing requirements, treatment objectives, sensor confidence, rate limits, deadbands, and approved operating authority.

6. Recommend or Execute

Depending on the authorized deployment stage, AerationAI may:

7. Verify the Response

After an airflow adjustment, AerationAI evaluates whether the DO trajectory is responding within the expected time window before determining whether another adjustment is justified.

8. Continue Learning and Performance Evaluation

Measured performance is compared with expected performance to refine site-specific response relationships and improve future aeration decisions.

SensorAI + AerationAI: Trusted Information Before Optimization

Aeration optimization is only as reliable as the information used to make the decision.

SensorAI supports AerationAI by evaluating whether critical operational measurements remain sufficiently trustworthy for optimization and control.

For DO and airflow measurements, this may include:

AerationAI can then modify its operating authority according to information confidence.

A questionable DO signal should not have the same authority as a validated DO signal.

Advanced Predictive Tools and Airflow Optimization

AerationAI does not depend on a single prediction method.

It combines multiple site-specific predictive and dynamic tools to determine both the expected airflow requirement and whether current process behavior supports an adjustment.

Site-Specific Predictive Airflow

Historical operating information is used to develop predictive relationships among:

These proprietary site-specific models, including Golden Key predictive methods where applicable, establish the expected airflow requirement under changing plant conditions.

DO Trajectory Modeling

AerationAI evaluates the direction, persistence, and rate of DO change, rather than waiting only for DO to cross a fixed control boundary.

A developing decline can therefore be recognized before the absolute DO concentration becomes critically low.

Likewise, a recovering DO trajectory can indicate that a previous airflow increase is beginning to work even while DO is still temporarily declining.

Dynamic Airflow-to-DO Response

Every aeration basin and zone can respond differently after airflow changes.

AerationAI characterizes site-specific response parameters such as:

This enables AerationAI to recognize an important distinction:

DO may still be declining because oxygen demand remains too high—or because the previous airflow adjustment has not yet had sufficient time to become measurable.

The system is designed to distinguish between these conditions.

Response-Aware Airflow Adjustment

Following a significant airflow change, AerationAI can temporarily enter a response-evaluation state.

During this period, calculations continue, but unnecessary additional airflow changes can be withheld while the expected process response develops.

If the DO trajectory deteriorates sufficiently that a treatment boundary may be approached before the expected response occurs, a bounded earlier intervention can still be considered.

This helps prevent repeated control actions from accumulating before the effect of the first action becomes visible.

Influent Flow as Context—not an Automatic Loading Signal

AerationAI uses influent flow as an important hydraulic indicator but does not automatically assume:

Higher Flow = Higher Oxygen Demand

For example, wet-weather infiltration and inflow can increase hydraulic flow while diluting wastewater strength.

AerationAI therefore interprets influent-flow changes together with:

A flow increase accompanied by deteriorating DO behavior may support evidence of increased oxygen demand.

A flow increase accompanied by stable or improving DO may indicate a predominantly hydraulic or dilution event and should not automatically trigger additional aeration.

Designed to Work With Existing Plant Infrastructure

AerationAI is designed to use the instrumentation and control infrastructure already available at wastewater treatment facilities.

When suitable instrumentation, operational data, connectivity, and control interfaces are available, deployment can normally proceed without new process hardware or major physical plant modifications.

Typical existing information includes:

Additional Information When Available

Real-time BOD, COD, or ammonia analyzers are not required for basic AerationAI aeration optimization.

Any instrumentation, data-quality, timing, or connectivity gaps are identified during the initial site assessment.

Engineering Constraints and Governed Control

AerationAI optimization occurs within defined engineering and operational boundaries.

Depending on the facility, these may include:

Optimization therefore does not mean searching for the mathematically lowest airflow without regard to treatment or equipment.

AerationAI seeks the lowest supportable airflow within the plant’s approved engineering and operating envelope.

Better Performance With Less Unnecessary Aeration

AerationAI is designed to reduce unnecessary airflow while maintaining the treatment performance and operating stability required by the facility.

Aeration efficiency is not optimized independently of treatment reliability.

AerationAI and AerationAI-SND Within the Engineering Intelligence Platform

AerationAI applies McC AI Group’s four core Engineering Intelligence modules to wastewater aeration optimization.

OperationsAI™

Defines facility-specific operational objectives, process models, optimization strategies, information requirements, engineering constraints, and control logic. For wastewater treatment applications, OperationsAI provides the facility-specific operational intelligence supporting AerationAI™ and AerationAI-SND™. For other regulated critical facilities, the OperationsAI application can be configured and named according to the facility type, process, or operational function being optimized.

SensorAI™

Validates signal integrity, generates trustworthy virtual sensors where appropriate, evaluates operational consistency, and monitors sensor and equipment health.

WisdomAI™

Applies site-specific engineering knowledge, explainable reasoning, treatment requirements, operating constraints, and contextual guidance.

PilotAI™

Governs recommendations and operational actions within approved operating envelopes, enabling progressive authority from advisory operation to bounded governed control.

Together, these modules transform plant data into trusted aeration decisions that can be explained, constrained, verified, and progressively authorized.

Phased Adoption: Prove Performance Before Expanding Authority

AerationAI does not require a facility to move directly from conventional control to automated AI optimization.

Operational authority is introduced progressively.

Designed for Secure Industrial Deployment

AerationAI is designed for local-first deployment within the facility’s approved computing environment.

Its essential optimization functions can operate locally without requiring plant operating information to be sent to an external cloud service.

AerationAI can be configured so that normal optimization is performed locally using only explicitly approved plant information.

Any external communication pathway, if used, must be separately documented, configured, and authorized according to the facility’s IT/OT cybersecurity requirements.

Control access is also limited by deployment stage. Historical Analysis requires no control connection. Shadow and Advisory operation can remain read-only. Governed Control requires an explicitly approved and bounded control interface.

The plant determines what information AerationAI can access and what level of operational authority, if any, it receives.

Existing Plant Control Is Preserved

AerationAI operates as a supervisory optimization layer. It does not require elimination of the facility’s existing DO-based PLC control strategy.

The original control logic remains available as a fallback.

Authorized operators can:

Loss of AerationAI should not require interruption of biological treatment. The deployment architecture should permit the existing plant-control strategy to resume or continue according to the approved site configuration.

Operator authority and existing plant safeguards remain independent of AerationAI optimization authority.

Instrumentation and Pilot Readiness

AerationAI begins with the information already available at the facility.

Minimum Information for Initial Historical Analysis

Typical minimum information includes:

Preferred Additional Information

Where available:

The initial assessment determines which available signals are usable, which require validation, and whether any additional information is necessary for the desired deployment stage.

AerationAI is designed around the instrumentation commonly available at existing wastewater treatment facilities—not around an assumed requirement for extensive new instrumentation.

Start With Historical Analysis

A facility does not need to install active control software to determine whether AerationAI may provide value.

The first step can be completed with exported historical plant data.

McC AI Group can evaluate:

Evaluate the opportunity first. Connect to the plant only when the facility is ready.

See AerationAI Using Real Plant Operation

A demonstration can replay historical plant operation at accelerated speed so that plant personnel can see how AerationAI responds as conditions change.

The demonstration can show:

This allows operators, engineers, administrators, and IT personnel to evaluate how AerationAI would behave before expanding its operational authority.

Request a 30-Minute Demonstration

AerationAI™ and AerationAI-SND™

AerationAI provides site-specific predictive aeration optimization for municipal and industrial wastewater treatment facilities.

AerationAI™

Optimizes aeration across activated-sludge and water-recycling treatment systems while balancing treatment performance, energy use, sensor reliability, equipment constraints, and operator authority.

Explore AerationAI
AerationAI-SND™

AerationAI-SND™ extends the AerationAI architecture to controlled low-DO operation for simultaneous nitrification and denitrification, using site-specific low-DO operating envelopes, nutrient-removal models, process-response verification, and governed control.

Explore AerationAI-SND

Evaluate AerationAI for Your Facility

Every wastewater treatment facility has different:

AerationAI is therefore developed and validated site by site.

The process begins with the plant information already available and progresses only as technical performance, information reliability, engineering constraints, cybersecurity requirements, operator confidence, and facility authorization support the next stage.

Predict the Requirement. Validate the Information. Understand the Response. Govern the Action.

Integration With Existing Plant Systems

AerationAI operates as a supervisory intelligence layer above existing plant instrumentation, automation, and aeration equipment. It uses available operational data and communicates through established PLC and SCADA infrastructure.

Operational Inputs may include:

Recommendations and Control Outputs may include:

Exact data requirements are plant-specific. McC AI Group begins with the information already available and identifies any additional data needed during the initial assessment.

Evaluate AerationAI for Your Facility

AerationAI is introduced progressively so that data quality, model performance, engineering constraints, and operator confidence can be evaluated before operational authority expands.

Stage 1 — Historical Analysis & Site-Specific Predictive Algorithm Development
Typical Duration: 1–2 Weeks

Historical and available real-time operating information is used to:

Stage 2 — Shadow-Mode Verification
Typical Duraton: 2–4 Weeks

AerationAI operates in parallel with existing control systems without changing plant operation.

During shadow operation:

Stage 3 — Advisory Operation
Typical Duration: 2–6 Weeks

Following successful Shadow Mode verification and facility approval, AerationAI may advance to Advisory Operation.

During this stage, AerationAI evaluates plant conditions in real time and provides site- and zone-specific airflow recommendations for operator review. AerationAI does not independently execute control actions during Advisory Operation.

AerationAI recommendations may consider:

Operators can review, accept, reject, defer, or modify each recommendation based on current plant conditions and operating judgment.

During Advisory Operation:

A key objective of Advisory Operation is to demonstrate that AerationAI recommendations are technically sound, understandable, operationally practical, and consistent with treatment requirements before any automated control authority is considered.

During Advisory Operation, AerationAI recommends and explains. The operator decides and acts.

Stage 4 — Governed Control
Ongoing After Authorization

Following successful Shadow Mode and Advisory Operation, and only after engineering review, cybersecurity approval, operator acceptance, and explicit facility authorization, AerationAI may advance to Governed Control.

During Governed Control, AerationAI may execute specifically authorized supervisory airflow adjustments through an approved plant-control interface.

Any active control authority remains bounded by site-specific requirements, including:

AerationAI continues to evaluate the result of each implemented airflow adjustment before determining whether another change is justified.

For example, if airflow has been increased but the affected aeration zone normally requires several minutes before a DO response becomes detectable, AerationAI can recognize that continued DO decline during the expected response period does not necessarily justify another immediate airflow increase.

This response-aware control helps prevent repeated adjustments from accumulating before the effect of the previous action becomes observable.

The facility’s existing control strategy remains available as the approved fallback, and authorized operators retain the ability to limit, pause, or discontinue AerationAI control.

The facility determines the authority AerationAI receives. AerationAI does not determine its own authority.

Progression Between Stages

Progression from one AerationAI deployment stage to the next is based on demonstrated performance and facility readiness—not simply elapsed time.

Advancement depends on factors such as:

The stated durations are typical planning ranges rather than guaranteed schedules. A facility may remain in Historical Analysis, Shadow Mode, or Advisory Operation for as long as desired.

Progression is not automatic, and Governed Control is not required for a facility to obtain value from AerationAI.

Historical Analysis → Shadow Mode → Advisory Operation → Governed Control
Evaluate → Validate → Recommend → Authorize

Continuous Performance Evaluation

AerationAI monitors:

Energy savings may be estimated against an agreed historical baseline adjusted for flow, organic loading, treatment requirements, seasonal conditions, and other relevant operating factors.

Performance results depend on the facility’s existing equipment, instrumentation, control configuration, treatment requirements, operating variability, data quality, baseline methodology, and authorized deployment stage.

Designed for Secure Industrial Deployment

Cybersecurity is a critical consideration for any software introduced into a wastewater treatment facility or other regulated industrial environment.

AerationAI is designed for local-first deployment within the facility’s approved computing environment. Its essential monitoring, prediction, optimization, and governed-control functions can operate locally without requiring plant operating information to be transmitted to an external cloud service.

No Required Internet Connection

AerationAI can perform its essential optimization functions without a continuous Internet connection.

The software can operate within the facility’s approved local computing environment using designated plant information and locally executed predictive and optimization algorithms.

No Cloud Processing Required

AerationAI does not require cloud computing to perform its essential aeration-optimization functions.

Plant operating information can remain within the facility’s approved network environment while AerationAI performs prediction, engineering evaluation, and optimization locally.

No Automatic External Data Transfer

AerationAI does not require automatic transmission of plant operating data outside the facility network for normal optimization operation.

Any external communication pathway, if desired for an approved application such as software maintenance or authorized support, should be separately documented, configured, and approved according to the facility’s IT/OT cybersecurity requirements.

No Required Remote Access

Routine AerationAI operation does not require external personnel to maintain a persistent remote connection to the plant control environment.

Remote access, if ever permitted for an approved purpose, remains subject to the facility’s cybersecurity policies, authentication requirements, access controls, and explicit authorization.

No Unrestricted PLC Access

AerationAI does not require unrestricted access to PLCs or other plant-control systems.

Access can be limited to explicitly approved process information and, where Governed Control is authorized, to specifically approved supervisory control variables.

AerationAI is not intended to:

During Historical Analysis, no plant-control connection is required.

During Shadow Mode, AerationAI can operate using read-only information and has no active control authority.

During Advisory Operation, AerationAI provides recommendations for operator review and does not independently execute plant-control actions.

Only during Governed Control, and only after engineering review, cybersecurity approval, facility authorization, and operator acceptance, may AerationAI communicate specifically authorized bounded supervisory adjustments through an approved control interface.

Plant-Controlled Access and Authority

The facility determines:

AerationAI does not determine or expand its own access privileges.

The facility determines what information AerationAI can access and what level of operational authority, if any, it receives.

Existing Plant Control Is Preserved

AerationAI is designed as a supervisory optimization layer above the facility’s existing instrumentation, PLC/SCADA systems, airflow controllers, blower controls, and other established process-control functions.

It does not require elimination of the facility’s existing DO-based or other approved control strategy.

Existing plant-control logic remains available as the approved fallback according to the facility’s control architecture and operating procedures.

Existing Safeguards Remain Authoritative

AerationAI does not replace or bypass existing equipment-protection and safety functions.

Existing PLC and equipment safeguards may continue to enforce requirements such as:

AerationAI recommendations and authorized supervisory adjustments remain subject to these constraints.

Operator Authority Is Preserved

Authorized operators retain the ability to:

Human oversight, operational visibility, intervention authority, and organizational accountability remain integral throughout deployment.

Failure Does Not Require Loss of Plant Control

AerationAI should be configured so that loss or disablement of the optimization software does not require interruption of biological treatment.

If AerationAI becomes unavailable because of a software, computer, communication, or data problem, the facility’s approved fallback control strategy remains available according to the site-specific implementation.

Conceptually:

AerationAI Available
→ Site-specific predictive optimization and governed supervisory control

AerationAI Unavailable
→ Existing approved PLC/operator control remains available

AerationAI therefore supplements the existing control system rather than becoming the sole means by which the aeration process can operate.

Governed Authority Remains Bounded

Even after Governed Control is authorized, AerationAI control authority remains constrained by:

Operator authority and existing plant safeguards remain independent of AerationAI optimization authority.

AerationAI is designed to optimize within the plant’s approved control environment—not to replace the plant’s control authority.

Discuss Your Facility
Contact McC AI Group
Return to Home