Process Model Lifecycle Management

This module provides infrastructure for managing the lifecycle of process models (digital twins) from concept through operation to archival.

Overview

As process simulation models evolve from early feasibility studies to live digital twins, they require:

Classes

ProcessSystemState

Serializable state snapshot of a ProcessSystem for checkpointing and version control.

Key Features

Equipment-Specific Properties Captured

Equipment Type Properties Captured
Compressor outletPressure, polytropicEfficiency, isentropicEfficiency, power, speed
Pump outletPressure, power
Valve percentValveOpening, outletPressure, Cv
Heater duty, outletTemperature
Cooler duty, outletTemperature
Separator pressure, temperature, liquidLevel
Stream temperature, pressure, flowRate

Usage Example

// Create a process system
ProcessSystem process = new ProcessSystem();
process.setName("GasProcessingPlant");
// ... configure process ...
process.run();

// Export state to file
ProcessSystemState state = ProcessSystemState.fromProcessSystem(process);
state.setVersion("1.2.3");
state.setDescription("Post-commissioning tuned model");
state.setCreatedBy("john.doe@company.com");
state.saveToFile("model_v1.2.3.json");

// Later: restore state
ProcessSystemState loaded = ProcessSystemState.loadFromFile("model_v1.2.3.json");
if (loaded.validateIntegrity()) {
    ProcessSystem restored = loaded.toProcessSystem();
}

Convenience Methods on ProcessSystem

// Quick state export
ProcessSystemState state = process.exportState();

// Save directly to file
process.exportStateToFile("checkpoint.json");

// Load and apply state
process.loadStateFromFile("checkpoint.json");

ModelMetadata

Tracks lifecycle phases, validation history, and calibration status.

Lifecycle Phases

Phase Description
CONCEPT Early screening and feasibility studies
DESIGN Detailed engineering (FEED, Detailed Design)
COMMISSIONING Construction and commissioning
OPERATION Live digital twin, actively used
LATE_LIFE Decommissioning planning
ARCHIVED No longer in active use

Calibration Status

Status Description
UNCALIBRATED Never validated against real data
CALIBRATED Validated but may be outdated
IN_PROGRESS Calibration underway
FRESHLY_CALIBRATED Recent validation, high confidence
NEEDS_RECALIBRATION Model diverged from plant data

Usage Example

ModelMetadata metadata = new ModelMetadata();
metadata.setAssetId("PLATFORM-A-TRAIN-1");
metadata.setAssetName("Gas Processing Train 1");
metadata.setLifecyclePhase(LifecyclePhase.OPERATION);
metadata.setResponsibleEngineer("jane.doe@company.com");

// Record validation against field data
metadata.recordValidation(
    "Matched well test data within 2%", 
    "WELL-TEST-2024-01"
);

// Record model modifications
metadata.recordModification(
    "Tuned compressor curves based on vendor data",
    "john.smith@company.com"
);

// Update calibration status
metadata.updateCalibration(CalibrationStatus.FRESHLY_CALIBRATED, 0.02);

// Check if revalidation is needed (after 90 days)
if (metadata.needsRevalidation(90)) {
    System.out.println("Model needs revalidation");
}

Integration with Git-Based Workflows

The JSON export format is designed for version control:

# Commit model checkpoints
git add model_v1.2.3.json
git commit -m "feat: Updated compressor curves after field test"

# Compare model versions
git diff model_v1.2.2.json model_v1.2.3.json

# Track model evolution over time
git log --oneline -- model_*.json

Best Practices

  1. Regular Checkpoints: Save state after major modifications
  2. Semantic Versioning: Use MAJOR.MINOR.PATCH for model versions
  3. Validation Records: Document every validation against field data
  4. Modification History: Record all changes with author attribution
  5. Lifecycle Transitions: Update phase when model role changes