levelBased ramp + engagement:
- Ramp foot is now max(startLevel, holdLevel) — was max(startLevel,
inflowLevel). inflowLevel is basin geometry, not a control setpoint;
the implicit hold zone it created was causing pumps to "start at
inflowLevel" instead of startLevel.
- New optional `holdLevel` config (defaults to startLevel = no hold band).
When raised, pumps engage at startLevel and hold at 0 % = MGC flow.min
across [startLevel, holdLevel], then ramp 0..100 % to maxLevel.
- Engagement decided in run() (not in `_applyMachineGroupLevelControl`):
rising-edge hysteresis arming gates a clean turnOff early-return.
Once armed, the helper always forwards setDemand(pct, '%') — 0 %
legitimately means "engaged at min flow", no more soft-turnOff at
the boundary.
- Disengagement paths (minLevel hard-stop, stopLevel falling-edge,
pre-arming idle) now all clear the shifted-ramp hysteresis state too.
- Threshold validator drops the startLevel ≤ inflowLevel rule; adds
startLevel ≤ holdLevel < maxLevel (only checked when holdLevel is
explicitly set, so default-null doesn't false-flag).
MGC unit math:
- Replace direct group.handleInput(percent) with group.setDemand(pct, '%')
in _applyMachineGroupLevelControl. The percent → m³/s resolution now
lives in MGC.setDemand (committed separately in the MGC submodule).
FlowAggregator variant picking:
- New _pickFlowSum() helper mirrors selectBestNetFlow's variant
precedence (measured first, then predicted) and resolves each side
independently. Realistic mixed case — real measured upstream sensor +
predicted pump outflow — now feeds the predicted-volume integrator.
Was reading only `flow.predicted.*` so a real upstream sensor
(which writes `flow.measured.*`) never moved the level.
Editor:
- New `holdLevel` and `deadZoneKeepAlivePercent` defaults + side-panel
input rows in the levelbased mode preview.
- Add the missing `ps-mode-line-holdLevel` SVG marker (was declared in
the side-panel coupling but the SVG element didn't exist, so the
dashed line never rendered).
- Relax stopLevel marker gate so it renders for any non-negative typed
value — start/stop ordering is the ribbon's job, not the marker's
(was hiding the line whenever startLevel was momentarily smaller).
- Add holdLevel to the marker loop in mode-preview so changes track.
- Add stopLevel + holdLevel + maxLevel to all three bindRedraw lists
(basin-diagram, mode-preview, bounds.apply) so the SVG, validation
ribbon, and HTML5 min/max attrs update on every edit.
- Initialise stopLevel + holdLevel + deadZoneKeepAlivePercent inputs
in oneditprepare so reopening the editor shows the saved values.
- nodeClass passes holdLevel + deadZoneKeepAlivePercent into the
domain config.
Tests:
- New test/basic/_probe_upstream_emit.test.js: confirms the parent
surfaces flow.measured.upstream.* on Port 0 after a measurement
child write — pins the previously-invisible measured variant flow.
- flowAggregator.basic.test.js: two new regression cases — measured
inflow when predicted side is empty, and the measured-in /
predicted-out mixed case.
- control-levelBased.basic.test.js: new cases for the holdLevel hold
band, the [stopLevel, startLevel] keep-alive, the engagement gate,
and the "0 % at startLevel = setDemand" contract.
- specificClass.test.js: zone tests adjusted to the new ramp foot.
Shifted-ramp tests pin holdLevel = 3 explicitly so their legacy
arithmetic (ramp foot at inflowLevel) stays self-consistent.
- shifted-ramp-end-to-end.test.js: same holdLevel pin for the same
reason.
Packaging:
- Add .gitignore + .npmignore so the published tarball drops the
wiki/, simulations/, test/, tools/, .claude/ etc. The pack went
from 1.5 MB (72 files) to ~57 KB (30 files).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Adds to scalar setters whose payloads are
plain numbers OR {value, unit}. Skipped where payload is compound or
mode-dependent (control-%, {F, C: [...]}, etc.) — documented inline.
Every command gains a description field for wikiGen consumption.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Decision 2026-05-11: 'highVolumeSafetyLevel' is canonical. The legacy
'overfillLevel' name is gone from computeSafetyPoints + the validator
issue tuple. 'overfillVol' parallel alias kept (out of scope for this
task; flagged for follow-up). 130/130 tests pass.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Splits pumpingStation/src/ into focused concern modules. specificClass.js
will be slimmed to an orchestrator in P2.9 (integration); for now both
the inlined logic AND the new modules coexist so tests stay green
throughout.
src/basin/ BasinGeometry + thresholdValidator (pure)
src/measurement/ flowAggregator + measurementRouter + calibration
src/control/ levelBased + flowBased(stub) + manual + index dispatcher
src/safety/ safetyController split into dryRun + overfill rules
src/commands/ registry array + handlers (canonical names from start)
src/editor.js 260 lines of SVG basin-diagram redraw, was inline in .html
examples/standalone-demo.js was if(require.main===module) at bottom of specificClass.js
CONTRACT.md canonical inputs + outputs + emitted events
Modified:
src/specificClass.js removed the 170-line standalone demo block
pumpingStation.html oneditprepare/oneditsave delegate to editor.{init,save}
pumpingStation.js added admin endpoint serving src/editor.js
102 basic tests pass (60 new + 42 existing).
specificClass.js itself is unchanged in behaviour — integration is P2.9.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Volume integrator changes:
- Hard physical floor at 0 added to _updatePredictedVolume. Without
it, a basin seeded below dryRunSafetyVol (calibration / startup
/ low seed) under continued net-outflow drifted volume arbitrarily
negative; the level output looked clamped only because
_calcLevelFromVolume floors at 0, masking the underlying drift.
- New cumulative diagnostic: underflowVolume.predicted.atequipment
(m³) + getOutput().predictedUnderflowVolume. Non-zero indicates a
flow-balance error (over-reported outflow / missing inflow).
- The transition-only dryRunSafetyVol clamp is preserved so
startup-from-empty doesn't snap to 2.1 m³ on tick 1.
Spill flow refactor (taxonomic + bug fix):
- Synthetic spill moved from flow.predicted.out.<child='overflow'>
to its own position flow.predicted.overflow.<default>. The spill
is a derived quantity, not a physical sub-source sharing a position
with pumps — .child() was the wrong knob.
- Removes the spillPrev self-subtraction in the integrator (no longer
needed: outflowTotal at ['out','downstream'] cleanly excludes spill).
- Closes a latent fall-through bug exposed during this work:
.child('overflow').getCurrentValue() returned the value of any
available sibling child when overflow itself didn't yet exist.
Hardened separately in generalFunctions@a516c2b.
- _selectBestNetFlow folds the overflow position into the outflow
side so the predicted net-flow balance still reads ~0 while pinned.
Tests: 70/70 pass. 4 new subtests cover the 0-floor, accumulated
underflow tracking, getOutput surface, and refill-from-empty.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Predicted volume is now clamped to [dryRunSafetyVol, maxVolAtOverflow]
in _updatePredictedVolume — the integrator can no longer drift above
the weir crest (only a real measurement can show level > overflow,
e.g. inflow exceeding pump+weir capacity). Excess is recorded as:
- overflowVolume.predicted.atequipment.default — cumulative spill (m3)
- flow.predicted.out.overflow — instantaneous spill rate (m3/s),
registered as a synthetic outflow so net-flow balance reads ~0
while pinned. The integrator subtracts the prior tick's synthetic
flow before integrating so it never feeds back into volume math.
Lower clamp at dryRunSafetyVol fires only on the transition — a low
seed/calibration is left alone; inflow is what brings it back up.
_selectBestNetFlow holds the last non-zero level-rate net flow when
level pins at overflowLevel and dL/dt collapses to 0, so dashboards
keep showing roughly what's coming in. Auto-refreshes once level
drops.
getOutput() exposes predictedOverflowVolume + predictedOverflowRate
as top-level convenience keys; the underlying measurements flow to
InfluxDB via the standard MeasurementContainer flatten path.
9 new test assertions cover the upper-clamp + spill increment, stable
spill across ticks, net-flow ~0 while pinned, spill clearing when
inflow stops, low-seed left alone, drain-across-threshold clamp, and
the new top-level output keys.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Runtime (specificClass.js):
- Replace the "shift left both ramp ends" geometry with a true
hold-then-ramp hysteresis driven by output %, not level:
• Up-curve % crosses shiftArmPercent on the way up → ARM.
• Filling→draining transition while armed → capture the up-curve %
at that moment as _shiftHoldValue.
• Draining + level ≥ shiftLevel → output stays at _shiftHoldValue
(horizontal hold, matching the dashed segment in the SVG).
• Draining + level in [start, shift] → output ramps holdValue → 0 %
along the same curve shape (linear or log) as the up curve.
• Draining + level < startLevel → 0 % AND disarm.
• Returning to filling clears holdValue, stays armed; next drain
transition captures a fresh hold so bouncing fills rearm cleanly.
• Disarm only when level ≤ startLevel.
- New _curveShape(x) helper for shared linear/log shaping.
- Removed legacy _levelBasedRampStart / _levelBasedRampTop /
_updateShiftArmed in favour of the inline state machine.
Adapter (nodeClass.js):
- Pipe shiftArmPercent through to control.levelbased.
Editor (pumpingStation.html + src/editor/):
- Add shiftArmPercent input row (% with unit) to the mode side panel
(only shown when shifted ramp is enabled). Default 95 %.
- Add the horizontal arming-% line + label inside the mode SVG —
this is the "% Threshold triggering shifted ramp down" line from
the original drawing that had been missing.
- Redraw the shifted-down curve to match the SVG geometry literally:
100 % flat from maxLevel → shiftLevel, then ramp shiftLevel →
startLevel down to 0 %, OFF below startLevel. Preview shows the
worst-case envelope (hold = 100 %); runtime hold is captured live.
- Validation extended: 0 < shiftArmPercent ≤ 100; ordering rules
preserved (start < shift ≤ max etc.).
- Auto-default shiftArmPercent to 95 when shift is enabled and the
current value is missing or out of range.
Dashboard example (examples/basic-dashboard.flow.json):
- Parser now reads `level.predicted.atequipment.default` etc. The
MeasurementContainer flatten format includes the implicit 'default'
childId; consumers must include it. Comment in the parser points
at the documenting source in generalFunctions.
Tests:
- test/basic: replace old level-armed-shift tests with two new ones
that exercise the hold-then-ramp arming, capture, hold, ramp-down,
disarm, and the bounce case (filling→draining→filling→draining
captures a fresh hold each time).
- test/integration/shifted-ramp-end-to-end.test.js: new file. Drives
Q_IN/Q_OUT through the full runtime tick with a controllable clock,
asserting the same hysteresis path the dashboard exercises.
- test/integration/basic-dashboard-flow.test.js: fixture keys updated
to the .default-suffixed form so they match the real flatten output.
56/56 tests pass.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Runtime (specificClass.js):
- Replace direction-based hysteresis with level-armed _shiftArmed state.
Arms when level rises past shiftLevel; disarms when level drops below
startLevel. While armed, ramp foot moves to startLevel and ramp top
to shiftLevel — both ends shift left, then saturate at 100 % up to
maxLevel.
- _scaleLevelToFlowPercent now takes (rampStartLevel, rampTopLevel) so
the saturation point follows the shift state.
- New setManualOutflow mirroring setManualInflow.
Adapter (nodeClass.js):
- Pipe enableShiftedRamp / shiftLevel through to control.levelbased.
- New q_out topic handler.
Editor (pumpingStation.html + new src/editor/ modules):
- Split monolithic <script> into modules: index.js (helpers),
basin-diagram.js, mode-preview.js, hover-couple.js, oneditprepare.js,
oneditsave.js — served via /pumpingStation/editor/:file.
- Mode preview redrawn per the SVG diagrams: OFF tier below 0 %, 0 %
flat from start→inlet, ramp inlet→max, optional shifted-down curve
start→shift with 100 % saturation past shift.
- Mode preview gains zone bands (dryRun / safetyLow / safe / safetyHigh /
overflow), level markers (dryRun derived, start, inlet, max, shift,
overflow), validation ribbon that blocks save on bad ordering.
- Auto-default shiftLevel to 0.9 × maxLevel on enable so the marker is
always visible.
- All level inputs moved to a side panel left of each diagram, color-
coded to match line strokes; hover-couple highlights the paired SVG
line on input focus / mouseover.
- Removed UI for non-static parameters: minHeightBasedOn,
pipelineLength, maxDischargeHead, staticHead, defaultFluid,
maxInflowRate, temperatureReferenceDegC,
timeleftToFullOrEmptyThresholdSeconds, inletPipeDiameter,
outletPipeDiameter, minLevel (now derived = dryRunLevel).
- foreignObject inputs in basin SVG removed (single source of truth in
side panel).
Dashboard example (examples/basic-dashboard.flow.json):
- Add manual Q_OUT slider + q_out builder mirroring the existing q_in
trio so the basin can be exercised end-to-end without a connected
rotating-machine downstream.
Tests (test/basic/specificClass.test.js):
- Replace direction-shift test with two new cases covering shift-disabled
hold-zone behaviour and shift-armed/disarmed transitions through
shiftLevel and startLevel boundaries. 53/53 tests pass.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
### Guardrails (specificClass.js)
New _validateThresholdOrdering() runs in the constructor. Checks every
ordered pair of basin + control + derived-safety levels and logs a
warning for each violation; returns the list as this.thresholdIssues
so tests and the eval harness can inspect. Non-fatal — we prefer a
running-but-warned station to a refusal-to-start (availability-first).
Strict invariants (bottom → top):
0 < outflowLevel < inflowLevel < overflowLevel ≤ basinHeight
dryRunLevel ≤ minLevel ≤ startLevel < maxLevel ≤ overfillLevel
Uses a list-of-checks pattern rather than a switch — easier to add new
invariants without reflowing cases, and the list itself is readable
documentation.
### Bug fix (specificClass.js)
calibratePredictedLevel was writing the volume value into the LEVEL
slot. Root cause: MeasurementContainer is stateful — its type()/
variant()/position() calls mutate the container's own cursor, so
caching chain references (const levelChain = ...; const volumeChain
= ...) doesn't isolate them. The second cached chain ended up sharing
the state of the last type() call. Rebuilt chains fresh each time,
matching the calibratePredictedVolume pattern that already worked.
### Tests (test/basic/specificClass.test.js)
Ported from Jest to node:test + node:assert — the project's standard
per .claude/rules/testing.md. Deleted the stale test/specificClass.test.js
(tests referenced methods that no longer exist post-rename).
New coverage, 42 passing subtests:
- Basin geometry derivations + minHeightBasedOn
- Level/volume roundtrip
- Threshold guardrails (5 violation cases)
- Direction derivation
- Mode change accept/reject
- Calibration (volume and level paths — catches the bug above)
- Levelbased control zones (STOP / DEAD ZONE / RAMP / saturate)
- getOutput flattening
- setManualInflow
Run with: node --test test/basic/*.test.js
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
