Heat-Balance BasedData Validation and Reconciliation (DVR)

What MapExDVR Does

MapExDVR applies physics first-law discipline to  a system with measured data. Reconciled data is a complete mass and energy balance that matches the mesured data as closely as possible and is more accurate than the sensor values. Measured sensor inputs are weighted by their uncertainties so that the more accurate sensors contribute more to the reconciliation than do less accurate sensors. Any sensor can optiionally be "believed" so that its reconciled value is equal to its measured value. This allows engineers to investigate the imapct of any sensor on the overall system heat balance.

Heat Balance Analysis

MapExDVR applies conservation of mass and energy — plus combustion modeling — to the plant thermodynamic cycle. Every temperature, pressure, flow, and enthalpy in the system is calculated, giving engineers a complete picture of cycle performance that goes far beyond what physical sensors alone can provide. The boiler heat balance model above can determine the fuel flow, moisture content, flue gas composition, and boiler efficiency from the measured data that is typically available.

Data Validation & Reconciliation (DVR)

Plant measured data commonly contains missing values, measurement error, and calibration drift. MapExDVR uses least-squared optimization to detect and resolve inconsistencies, producing reconciled output that is more accurate than the raw sensor data. A reconsiled feedwater flow at a nuclear plant, for example, can unlock a 30 MW power increase.

First-Law Thermodynamics

The reconciled heat-balance quantities every point in the cycle — including locations where no physical sensor exists.  The analysis adds information derived from first-law thermodynamics that is not present in the measured data, enriching the data set available for downstream analysis. For a combined-cycle power plant, MapExDVR will adjust the air flow, fuel flow, and steam/water flows (and their temperatures) so that mass and energy are conserved for each equipment icon as well as the overall plant. That is: the fuel energy into the plant equals the power produced plus the energy out (stack gas energy plus the condenser duty).

Augments Advanced Pattern Recognition (APR)

MapExDVR outputs can be passed to the Advanced Pattern Recognition (APR) engine in SureSense Studio. The richer input dataset enables the APR to detect subtle anomalies and diagnose faults with greater accuracy than measured data alone can support. The figure above shows the detection of an alarm condition in exhaust gas temperatures, and the  improved gas temperature spread after shut-down and maintenance.

Integrated with SureSense Studio

MapExDVR is an analytic component that has been integrated as the heat-balance module within SureSense Studio from Expert Microsystems Inc. SureSense Studio uses the MapExDVR outputs — the complete set of reconciled flows, temperatures, pressures, and enthalpies — as inputs to its APR, fault detection & diagnostics, and equipment health monitoring capabilities.