Modelling

Process data and models

We prepare historical data and build models for a defined question and operating range.

When it applies

Data are available, but changes in material, manual actions and different regimes are mixed together.

Ask about this scope

Starting information

  • Historical data with units and timestamps
  • Records of material and operating changes
  • Model objective and validation periods

Proposed work

We review data, separate regimes and reserve periods for validation. We document exclusions and test the model with data different from those used for fitting.

Deliverable

Prepared data, an evaluated model and a report of assumptions, error and range of application.

Plant trials, equipment installation and control changes require a specific scope and authorisation. This description does not replace operating procedures.

Knowledge

Related reading

Modelling and control 3 min

Why having data does not mean having control

Data describe measurements. Control requires interpreting the process state, distinguishing disturbances, choosing an action and observing its effect. A dataset may contain many rows and still fail to reconstruct…

Published

Instrumentation and data 2 min

The process historian is not the plant

A historian stores values and timestamps. It does not necessarily store the operating regime, instrument condition, disturbances, manual actions or constraints that explain those values.

Published

Modelling and control 2 min

How to validate a model before using it for a control decision

It remains to be evaluated whether the model retains usefulness on data that did not participate in fitting and whether it represents the intended decision. A good retrospective fit may depend on future information,…

Published