Notes

Notes from the headworks, the basins, the membrane room and the control room.

Short essays on water and wastewater treatment and autonomy, written from the thinking behind the product. No author bylines yet: the team publishes names when people join publicly.

Six notesWater treatment and autonomyNo sales register

run_scn_0418 · agent graph · scenario 10 nodes · 1 approval gate
01 · intake_screen Read influent and forecast ✓ SUCCEEDED 02 · orchestrator Rehearse the next 24 hours ✓ SUCCEEDED 03 · quality_anomaly Fuse cameras and analyzers ✓ SUCCEEDED 04 · dose_coagulate Compute the dose curve ✓ SUCCEEDED 05 · dose_coagulate Trim the coagulant setpoint ✓ SUCCEEDED 06 · aerate_bio Minimum-energy DO schedule ✓ SUCCEEDED 07 · orchestrator Blower move approved ◆ APPROVAL 08 · membrane_disinfect Flux hold and backwash timing ✓ SUCCEEDED 09 · compliance_energy Permit conformance check ✓ SUCCEEDED 10 · conformance Write the shift record ✓ SUCCEEDED

Illustrative scenario: one municipal activated-sludge wastewater plant with a membrane train, planning the next 24 hours after a rain forecast. Not a measured run.

Notes

In the order we wrote them

Origin

Aeration is the power bill

Walk into a wastewater plant and ask where the electricity goes. The answer, more often than not, is the blowers. Aeration can be the majority of a plant's power, and the blowers run high because a nutrient breach is worse than a large bill.

The dissolved-oxygen setpoint that keeps the plant safe at the morning peak is far more than it needs at three in the afternoon, but nobody is standing there re-tuning it. Aerix is built for exactly that gap: a DO schedule that follows the load and the biological state and never thins the nutrient margin.

The baseline is a meter the plant already reads. That is why aeration energy is usually the first conversation.

Why now

The craft is walking out the door

A treatment plant is a living process. Influent changes with the weather and the load, and a handful of experienced operators keep it inside spec by feel. When they retire, the feel goes with them.

The tools most plants have watch and alarm. SCADA shows the drift; a dashboard charts it; a lab result confirms it hours later. None of them closes the loop from what the plant sees to a safe, permit-checked action.

Hydrira is built to capture that craft as corrections, every approval, override and note becoming a labelled example, and to give it back to the plant on every shift.

Simflo

Rehearse the surge

Rain is forecast for the afternoon and the influent will rise by a third. What should the dose do, and when? What happens to the nutrient margin if the DO schedule stays where it is? Operators answer these questions from experience because they cannot answer them any other way.

Simflo is a calibrated, as-run simulation of the plant that evaluates candidate plans over the next 8 to 24 hours before the shift starts. The design target is 50 to 500 plans per run. The plans that thin a permit margin never reach the agents.

A twin that matches the plant as designed is a drawing. One that matches the plant as it is today, aged, fouled and drifting, is a tool.

Turbix

Seeing before the lab result

Experienced operators read water by eye: the blanket in the clarifier, foam on the basin, the colour of the effluent. The online analyzers read numbers, but they drift, and nobody fuses the two.

Turbix puts cameras on the clarifiers, basins and filters and fuses them with the turbidity, pH, chlorine and nutrient analyzers into one calibrated state. The design target is an alert that arrives before the lab result, with the evidence image attached.

It is the perception layer every control agent acts on, which is why it is also the sensing wedge for plants whose first problem is visibility.

Trust

A permit is not a setpoint

It is tempting to treat a permit limit as just another constraint in an optimizer. It is not. A permit limit is a line the plant does not cross, and the margin above it is what keeps a community safe when the model is wrong.

Permia evaluates every planned move against the permit set and blocks anything that would thin a margin. Disinfection residual minimums, integrity limits and discharge release decisions are read, never written. The operator of record stays the operator of record.

Energy and chemicals are optimized inside that margin, not against it. The arbitration rules say so in one line, and the shift record shows it on every run.

Trust

What a shift record is for

Every reading, prediction, limit, approval and write on a Hydrira run is appended to an immutable record the plant owns. It is exportable, it names the model versions that produced it, and it is drafted into the regulatory report with citations.

The point is not only compliance. A record that explains why the plant did what it did is how a process engineer learns to trust a system that acts, and how a regulator learns to accept one.

We build as if someone will read the record. On the plants we want to work with, someone will.

Run timeline

The scenario most notes refer to

Every Hydrira run is an ordered, inspectable sequence. This is run_scn_0418, an illustrative scenario: one municipal activated-sludge wastewater plant with a membrane train, planning the next 24 hours after a rain forecast. Ten steps, nine automatic, one human approval on the blower setpoint move, every action written to the shift record.

run_scn_0418 scenario · not measured SUCCEEDED
  1. 01 Read influent and forecast SUCCEEDED 1.2 s

    Intake-and-Screen · Flow, turbidity, ammonia and alkalinity at the headworks, and a 24-hour rain forecast that raises the expected influent by mid-afternoon.

  2. 02 Rehearse the next 24 hours SUCCEEDED 4 min 10 s

    Plant orchestrator · Simflo evaluates candidate dose, DO and flux plans over the 24-hour horizon and keeps the plans that hold permit margin through the surge.

  3. 03 Fuse cameras and analyzers SUCCEEDED 180 ms

    Quality-and-Anomaly · Clarifier blanket, foam and colour from the basin cameras fused with turbidity, pH, nutrient and residual analyzers into one calibrated plant state.

  4. 04 Compute the dose curve SUCCEEDED 2.4 s

    Dose-and-Coagulate · Coagulant and polymer dose ramp for the surge window under turbidity, phosphorus, sludge and chemical-cost constraints, cited to the plant's dosing SOP.

  5. 05 Trim the coagulant setpoint SUCCEEDED 0.6 s

    Dose-and-Coagulate · A bounded setpoint move inside the declared rate and magnitude limits for the dosing pump, with the rollback value recorded first.

  6. 06 Minimum-energy DO schedule SUCCEEDED 3.1 s

    Aerate-and-Bio · Blower and DO setpoint schedule across the basins that meets the nitrogen and phosphorus targets at the lowest kilowatt-hours the biological state allows.

  7. 07 Blower move approved APPROVAL 6 min 02 s

    Plant orchestrator · The DO setpoint change is above the plant's threshold for this loop. The process engineer on shift reviews the predicted nutrient margin and approves.

  8. 08 Flux hold and backwash timing SUCCEEDED 1.8 s

    Membrane-and-Disinfect · Flux held through the surge and the next backwash moved earlier, inside the integrity envelope; the disinfection residual minimum is untouched.

  9. 09 Permit conformance check SUCCEEDED 0.9 s

    Compliance-and-Energy · Every planned setpoint checked against the discharge permit, the nutrient limits and the residual minimum, with excursion risk forecast for the horizon.

  10. 10 Write the shift record SUCCEEDED 0.3 s

    Quality-and-Conformance · Inputs, model versions, predictions, limits, the approval and every write appended to the immutable shift record the plant owns.

Tool calls

The agent shows its work

Every tool invocation, argument and result is written to an immutable, human-readable log, and every reasoning step is expandable. Nothing about a run is hidden from the plant. The log below is the scenario run, not a measured one.

tool-call stream · run_scn_0418
  1. 05:40:02scada.read_influent({ plant: "WWTP-A", tags: ["FLOW", "NTU", "NH4", "ALK"] })
  2. 05:40:03ok flow 412 L/s · turbidity 118 NTU · ammonia 31 mg/L · alkalinity 184 mg/L
  3. 05:40:03weather.read_forecast({ horizon_h: 24 })
  4. 05:40:04ok rain from 14:00 · expected influent +38% by 17:00
  5. 05:40:04twin.simulate({ horizon_h: 24, candidates: 240, objective: ["permit_margin", "kwh", "chemical"] })
  6. 05:44:14ok 240 plans evaluated · 31 hold permit margin through the surge · best plan selected
  7. 05:44:15vision.fuse_state({ cameras: ["CLAR-1", "CLAR-2", "AER-3"], analyzers: "all" })
  8. 05:44:15ok blanket 1.1 m · foam none · effluent turbidity 1.8 NTU · residual 0.9 mg/L
  9. 05:44:16cuopt.solve_dose({ window_h: 12, constraints: ["NTU", "P", "sludge", "cost"] })
  10. 05:44:18ok coagulant ramp 38 → 46 mg/L from 13:30 · polymer unchanged · cite sop:DOSE-07 §3
  11. 05:44:19policy.evaluate({ loop: "dose", stage: "bounded", tag: "DOS1.SP" })
  12. 05:44:19ok permitted · max_step 4 mg/L per 15 min · rollback 38 mg/L
  13. 05:44:19plc.write_setpoint({ tag: "DOS1.SP", value: 42, unit: "mg/L" })
  14. 05:44:20ok written · ack from PLC · audit id 0418-05
  15. 05:44:20cuopt.solve_blowers({ basins: 4, constraints: ["NH4", "NO3", "P", "DO_min"] })
  16. 05:44:23ok DO 2.0 → 1.6 mg/L basins 1–3 from 06:00 · predicted −9% kWh · nutrient margin held
  17. 05:44:24policy.evaluate({ loop: "aeration", stage: "assist", tag: "BLW.DO_SP" })
  18. 05:44:24above threshold for this loop · routed to approve.human · engineer on shift notified
  19. 05:50:26approved by the process engineer on shift · reason: nutrient margin 0.6 mg/L above limit
  20. 05:50:27membrane.schedule_backwash({ train: "MBR-2", surge_at: "14:00" })
  21. 05:50:29ok flux held 24 LMH · backwash moved 16:00 → 13:15 · residual minimum untouched
  22. 05:50:29permit.evaluate({ permit: "NPDES-WWTP-A", horizon_h: 24 })
  23. 05:50:30ok all parameters inside limit · excursion risk 24 h: low · cite permit §II.A
  24. 05:50:30audit.commit({ run: "run_scn_0418", steps: 10, approvals: 1 })
  25. 05:50:30ok shift record written · immutable · exportable

Reasoning trace

Goal, plan, thought, action, observation, exactly as the orchestrator would record it.

GoalHold effluent turbidity, nutrients and disinfection residual i…

Hold effluent turbidity, nutrients and disinfection residual inside the discharge permit for the next 24 hours while taking aeration energy and coagulant use to the minimum the permit margin allows.

PlanRead influent and the forecast; rehearse the horizon on Simflo…

Read influent and the forecast; rehearse the horizon on Simflo; fuse the plant state; compute dose and DO schedules; write the dose inside bounds; route the DO move for approval; hold flux and move the backwash; check the permit; commit the record.

ThoughtRain arrives mid-afternoon and the influent will rise by about…

Rain arrives mid-afternoon and the influent will rise by about a third. Dosing has to ramp before the surge, not after, and the DO schedule has to keep nutrient margin through it.

Actioncuopt.solve_dose over a 12-hour window under turbidity, phosph…

cuopt.solve_dose over a 12-hour window under turbidity, phosphorus, sludge and cost constraints; plc.write_setpoint on the coagulant pump inside the declared rate and magnitude limits.

ObservationThe dose ramp holds predicted effluent turbidity under 2 NTU t…

The dose ramp holds predicted effluent turbidity under 2 NTU through the surge. The DO schedule saves energy but is above the threshold set for the aeration loop.

ThoughtEnergy never outranks permit margin. The DO change goes to the…

Energy never outranks permit margin. The DO change goes to the engineer with the predicted nutrient margin attached, and nothing is written until it is approved.

Actionapprove.request to the process engineer on shift; membrane.sch…

approve.request to the process engineer on shift; membrane.schedule_backwash to hold flux and move the next backwash ahead of the surge; permit.evaluate across the horizon.

OutcomeTen steps, one human approval, all parameters inside limit on …

Ten steps, one human approval, all parameters inside limit on the forecast, and the shift record written. Illustrative scenario; nothing here is a measured plant result.

Topics

What we write about

Aeration and energy

Dissolved oxygen, blowers, nutrient margin and the plant's power bill.

Dosing and clarity

Coagulant, polymer, jar tests, weather-driven influent and dose response.

Membranes and twins

Fouling, flux, cleaning, as-run simulation and rehearsing the surge.

Trust

Permits, audit logs, autonomy stages and what never leaves the operator of record.

Autonomy

Four stages, set per loop

A plant does not go from manual to bounded control in one step. Hydrira makes the stage explicit per loop, auditable and reversible at any time, and some decisions never leave the person.

Autonomy stages and the human role at each
StageWhat the agent doesWhat the person doesWhen
ShadowPredicts water quality, fouling, excursion risk and energy; writes nothingRuns the plant as today; compares predictions to lab, analyzer and SCADA outcomesFirst phase with every design partner
AssistRecommends dose, DO, flux and cleaning moves with predicted quality, energy and permit impactApproves, corrects or rejects each move; corrections train the modelsAfter accuracy is proven against the plant's own baseline
BoundedWrites to an allow-list of low-risk loops inside declared rate and magnitude limitsSets the limits, watches the record, pulls the loop back at any timePer loop, after the assist-mode gate and compliance sign-off
ExceptionsEscalates anything above the threshold, every permit-adjacent move and every releaseDecides; the operator of record is never removed from compliance decisionsAlways
Design targets

Figures we cite most

The numbers Hydrira is engineered against, from the GPU utilization plan.

<100 ms per camera inference at the plant edge, on Jetson Orin
15–30 FPS foam, clarity and anomaly vision per camera
1–10 Hz sensor fusion across SCADA tags and online analyzers
8–24 h simulated on the as-run twin before each shift starts
50–500 candidate control plans compared per twin run
10k–100k synthetic surge, fouling and upset scenario variants per plant archetype

Design targets from the GPU utilization plan, all marked aspirational in the architecture. Nothing on this page is a measured plant result.

Guardrails

An agent that can move a setpoint needs a leash

Hydrira writes to control equipment in a plant that keeps a community's water safe. Every capability is scoped, every write is policy-checked, every output is cited, and every action is written to an append-only audit log the plant owns.

  • Bounded action space. Each agent can only write to an explicit tag allow-list, inside per-tag rate and magnitude limits, at the autonomy stage the plant has set.
  • Permit limits are read, never written. Permit-limit setpoints, disinfection residual minimums, integrity limits and discharge release decisions stay with the permit and the operator of record.
  • Policy engine before every write. Autonomy stage, interlock state, permit margin, excursion forecast and operator presence are evaluated before a setpoint moves.
  • Grounded and cited. Every recommendation carries a citation to the permit, SOP or regulation it relied on. Retrieval is tenant-isolated and permission-aware.
  • Immutable audit log. Append-only, exportable and retained on the plant's own storage, aligned to Safe Drinking Water Act, discharge-permit and EU Water Framework record-keeping.
  • Hard fallback. If the edge node, the network or a model stops, the plant keeps running on its existing SCADA, PLC and DCS control and interlocks.
  • Tenant isolation and critical-infrastructure posture. Telemetry, permits, SOPs and models never cross a customer boundary. On-prem and edge-only deployment available for utilities.

Assurance posture

Assurance and certification status
ItemScopeStatus
SOC 2 Type I + critical-infrastructure cyber reviewCloud control plane, edge QUEUED Roadmap, first 6 months
SOC 2 Type IICloud control plane QUEUED Roadmap, 6 to 12 months
SDWA · discharge permits · EU WFD · nutrient and PFAS rulesGrounding and record-keeping RUNNING Design basis for retrieval and the audit log
Per-tenant isolationData, memory, models RUNNING Design rule
On-prem or edge-only deploymentUtilities, critical infrastructure RUNNING Design option

Hydrira supports your compliance records. It is not a certification, it does not replace your permit and it never removes the operator of record from a compliance decision.

Read the security overview

Integrations

It speaks plant, not cloud

Hydrira reads and writes through the systems already in the plant. No parallel historian, no new HMI, no rip-and-replace. Connectors are built against each design partner's stack first; long-tail integrations are out of scope for the first release.

SCADA and historians

Plant SCADA, historians, PLC and DCS interfaces

Tag reads at 1 to 10 Hz, guarded setpoint writes, interlocks untouched

Dosing and chemical skids

Coagulant, polymer, pH-adjust and disinfectant dosing controllers

Dose reads, bounded setpoint writes inside declared limits

Aeration and blowers

Blower controls, DO and ammonia-based aeration control

DO and blower reads, bounded setpoint writes after approval

Membrane skids

Membrane and RO skid PLCs, TMP, flux, conductivity, integrity monitors

Reads; flux and backwash writes inside the integrity envelope

Online analyzers

Turbidity, pH, chlorine residual, nutrient, TOC and flow analyzers

Reads; drift and fault flags

Lab and LIMS

Lab information systems, jar-test and grab-sample records

Reads for labels and calibration

Cameras and robots

Fixed cameras on clarifiers, basins, filters and membrane rooms; inspection robots

Streams fused at the edge; supervised routes

Utility MES, CMMS, permits

Utility MES, maintenance systems, permit and reporting systems, weather

Reads; shift record and report drafts written back

See all integrations

FAQ

Straight answers

The questions operations managers, process engineers and compliance leads ask in the first meeting.

Only at the autonomy stage your plant has set, and only to an explicit allow-list of low-risk loops inside documented limits. Every design partner starts in shadow mode, where Hydrira predicts and writes nothing. Assist mode follows, where a process engineer approves each move. Bounded write-back comes last, per loop, after the accuracy gate and compliance sign-off, and can be pulled back at any time.

Get started

Get new notes

Ask to be added to the list. Notes only, no drip campaign.