SimPilot
DocumentationExamplesAPI Reference
Sign in

Search Documentation

Find pages across the documentation

Getting Started

IntroductionQuickstartHow It Works

Core Concepts

AI AgentSimulation WorkflowEngineering PipelineKnowledge BaseArchitecture

Features

OverviewChat InterfaceCanvas & ArtifactsSharing & CollaborationExports & ReportsVoice InputMulti-Model AIProjectsTemplates & WorkflowsFile UploadsDeep ResearchWeb SearchCode ExecutionImage GenerationMCP ConnectorsInteractive DashboardsParallel ExecutionURL RetrievalFreeCAD CADSimulationsTemplatesConvergence Monitoring

Simulation

OverviewSupported TypesMesh GenerationError RecoveryBatch & SweepsResults Comparison

Studies & Analysis

OverviewDOE & Parametric SweepsOptimizationComparison

Validation & Reviews

OverviewBaselines & VersioningEngineering ReviewsRegulatory Compliance

3D Viewer

OverviewVisualization ToolsKeyboard Shortcuts

Enterprise

OverviewAdmin PanelOrganizationsKnowledge ManagementMethod Packs

Account

Getting StartedSettingsBilling

Examples

OverviewAerodynamicsPipe FlowHeat TransferStructural
  1. Docs
  2. Features
  3. Convergence Monitoring

Convergence Monitoring

Real-time convergence tracking, health status, ETA, and coaching suggestions.

While your simulation runs, SimPilot continuously monitors solver progress and provides real-time feedback on convergence health, estimated completion time, and actionable suggestions when problems arise.

Real-time tracking

During a simulation run, the monitoring system tracks:
MetricDescription
ResidualsPer-field residual values (pressure, velocity components, turbulence quantities) plotted over iterations
Courant numberMaximum and mean Courant number to verify time-step stability
Solver progressCurrent iteration or time step vs. total target
Wall-clock timeElapsed time and estimated time remaining
Residuals and Courant numbers update live in the chat interface as the solver produces new output.

Convergence status

The system classifies your simulation's convergence into one of five states:

Healthy

Residuals are decreasing steadily. The simulation is progressing toward convergence at a normal rate.

Converged

All monitored residuals have dropped below the convergence threshold. The simulation has reached a stable solution.

Stagnating

Residuals have plateaued and are no longer decreasing. The simulation may be stuck at a local minimum or need parameter adjustment.

Oscillating

Residuals are fluctuating without a clear downward trend. Common in unsteady flows or when relaxation factors are too aggressive.

Diverging

Residuals are increasing. The simulation is becoming unstable and may crash if not corrected.

ETA calculation

The estimated time to completion is computed from:
  • Current iteration vs. target iteration count
  • Wall-clock time per iteration (averaged over recent iterations)
  • Convergence rate (for criteria-based stopping rather than fixed iteration count)
The ETA updates as the simulation progresses, becoming more accurate over time.

Health snapshots

Periodic health updates appear in the chat showing the current state:
  • Current iteration and percentage complete
  • Residual levels for all monitored fields
  • Convergence status classification
  • Any warnings or suggestions
These snapshots let you check on a running simulation without interrupting your other work.

Coaching suggestions

When the monitoring system detects an issue, it provides specific, actionable suggestions:

Divergence detection

The system raises an alert when residuals increase by 10x over 5 consecutive iterations. At this point:
  1. The divergence is flagged in the chat with the affected fields
  2. If the simulation has not already crashed, the AI may suggest an immediate corrective action
  3. If the simulation crashes, the error recovery pipeline takes over (see Error Recovery)
Automatic intervention
In some cases, the AI can intervene automatically -- adjusting relaxation factors, reducing the time step, or modifying numerical schemes -- and restart the solver without requiring manual action. This depends on the detected failure mode and the AI's confidence in the fix.

Progress bar

A progress bar at the top of the simulation view shows:
  • Percentage complete: Based on iterations or simulated time vs. target
  • Estimated remaining time: Wall-clock time to completion
  • Status indicator: Color-coded by convergence health (green, yellow, red)
🔧

Error Recovery

What happens when a simulation fails and how the AI recovers.
📊

Interactive Dashboards

Post-simulation dashboards with convergence history and result visualization.
PreviousTemplatesNextOverview

On this page

Real-time trackingConvergence statusETA calculationHealth snapshotsCoaching suggestionsDivergence detectionProgress bar