Package neqsim.process.util.reconciliation

package neqsim.process.util.reconciliation

Data reconciliation and steady-state detection for online process optimization.

This package provides two complementary capabilities for online process optimization:

1. Steady-State Detection (SSD) — monitors process variables over a sliding window and determines when the process has reached steady state using the R-statistic (ratio of filtered to unfiltered variance), optional slope and std.dev tests. Only steady-state data should be fed into reconciliation or model calibration.

2. Data Reconciliation — weighted least squares (WLS) adjustment of plant measurements so that mass (and energy) balance constraints are exactly satisfied, with gross error detection via normalized residual tests and iterative elimination of faulty sensors.

Key classes:

• SteadyStateDetector - Monitors variables and evaluates steady-state status using R-statistic, slope, and std.dev criteria
• SteadyStateVariable - A monitored variable with sliding window and computed statistics
• SteadyStateResult - Result of SSD evaluation with per-variable diagnostics
• DataReconciliationEngine - WLS solver with gross error detection and elimination
• ReconciliationVariable - A single measured variable with value, uncertainty, and reconciled result
• ReconciliationResult - Complete reconciliation result with statistics, JSON export, and text report

Typical online workflow:

// 1. Monitor for steady state
SteadyStateDetector ssd = new SteadyStateDetector(30);
ssd.addVariable(new SteadyStateVariable("feed", 30).setUncertainty(20.0));
ssd.addVariable(new SteadyStateVariable("gas", 30).setUncertainty(15.0));
ssd.addVariable(new SteadyStateVariable("liquid", 30).setUncertainty(10.0));

// 2. Push readings and check
ssd.updateVariable("feed", 1000.5);
ssd.updateVariable("gas", 600.2);
ssd.updateVariable("liquid", 399.1);
SteadyStateResult ssResult = ssd.evaluate();

// 3. When steady, reconcile
if (ssResult.isAtSteadyState()) {
  DataReconciliationEngine engine = ssd.createReconciliationEngine();
  engine.addMassBalanceConstraint("Sep", new String[] {"feed"}, new String[] {"gas", "liquid"});
  ReconciliationResult recResult = engine.reconcile();
  System.out.println(recResult.toReport());
}

Version:

1.0

Author:

Process Optimization Team

See Also:

• SteadyStateDetector
• DataReconciliationEngine

Classes

Class	Description
DataReconciliationEngine	Data reconciliation engine using weighted least squares (WLS) with linear constraints.
ReconciliationResult	Result of a data reconciliation run.
ReconciliationVariable	A measured process variable participating in data reconciliation.
SteadyStateDetector	Steady-State Detector (SSD) for process variables.
SteadyStateResult	Result of a steady-state detection evaluation.
SteadyStateVariable	A monitored process variable for steady-state detection.