CHEMCAD simulations can match your measured process data to create a digital twin

In the world of process simulation, we often only need to run simulations to get an answer that is “close enough” to reality so that we can make engineering decisions. But what if you need to match your plant results exactly to create a digital twin of a real-world process? You can do that in CHEMCAD too. You can fine-tune your simulations to exactly match measured process data though a powerful feature known as Data Reconciliation.

What is Data Reconciliation?

Data Reconciliation is a tool in CHEMCAD that allows you to modify a steady-state simulation to match actual process data. This tool will perform simultaneous parameter regression to determine the best fit for your measured data. The result is a converged steady-state simulation that closely mirrors the real conditions of your process.

Getting Started with Data Reconciliation

Before you begin the reconciliation process, you should start with a converged steady-state model based on your process. This model should produce results that are quantitatively similar to your actual process data. CHEMCAD Data Reconciliation will then modify this simulation to fit the real observed data. It's also recommended to save a separate copy of the original steady-state simulation for future reference.

Identifying Target Data

The real-world (measured) data from your process serves as the target for Data Reconciliation. While you need to collect data from your process, you don’t have to use all the sensor data you may have available. It is important to keep in mind that Data Reconciliation will adjust the steady-state simulation to provide the best overall fit to the selected variables. Too much data can create noise, so focus on what you believe to be most critical to match with your reconciled simulation.

It is also important to try and collect observed data during stable, pseudo steady-state operation. If the process is unstable when you collect data, the reconciled model may not accurately reflect the process.

Example Scenario: Batch Distillation Column

Consider a batch distillation column where temperature measurements are available in the pot, middle of the column, top of the column, and condenser. The cooling water flow rate is measured along with inlet and outlet temperatures. To create a new Data Reconciliation simulation, go to the Analysis tab in the CHEMCAD ribbon menu and select Create New Data Reconciliation.

chemcad-dyk-data-reconciliation-1-2025

Integrate Your Data with Data Map

It is possible to use a single set of observed data, or an array of time dependent data. Using a single set of data will generally allow for an easier and faster calculation. Using a larger set of data will generally take more calculation time, but allow for a more accurate fit to the process. The measured values can be stored in an Excel table accessed via the Data Map feature in CHEMCAD. To use Data Map for data reconciliation, under Map Rule, select For Data reconciliation. This setting indicates that the Excel Data Map is used to link a value from Excel to a variable in CHEMCAD.

chemcad-dyk-data-reconciliation-2-2025

Setting Boundaries

To avoid physically unrealistic (but mathematically acceptable) results in the reconciliation process, be sure to set realistic maximum and minimum boundary values for the independent variables. This ensures that the adjustments made are practical and reflective of the real-world process.

chemcad-dyk-data-reconciliation-3-2025