Virtual flow meters
Estimate oil, gas, water, or production rates from available operating data.
Soft sensors and virtual measurements
Estimate important values when direct measurement is late, missing, or unreliable.
Soft sensors use industrial models and validated data to estimate process or product values that direct instruments do not provide in time for the decision.
Definition
What is an industrial soft sensor?
An industrial soft sensor is a model that estimates a process or product value from related measurements when the value is not directly measured, is delayed, is expensive to measure, or is unreliable.
Measurement gap
Optimization needs values that plants do not always measure directly.
Many operating decisions depend on values that arrive through lab tests, delayed analyzers, periodic well tests, expensive instruments, inferred product properties, or measurements that fail under real plant conditions.
Virtual measurements close part of that gap. They do not replace engineering judgment or validation. They provide actionable estimates with a known purpose, time basis, and confidence level.
Applications
Soft sensors support decisions across process and production systems.
Virtual analyzers
Estimate product quality or composition before analyzer or lab data catches up.
Product properties
Predict final product characteristics from intermediate measurements and process history.
Subcool estimates
Estimate SAGD subcool from reliable surface measurements when downhole sensors fail.
Sensor validation
Flag abnormal or inconsistent values before they corrupt decision support.
Optimization inputs
Feed estimates into setpoint, recommendation, or constrained-search calculations.
Q&A
Soft sensor and virtual measurement questions
What is an industrial soft sensor?
An industrial soft sensor is a model that estimates a process or product value from related measurements when the value is not directly measured, is delayed, is expensive to measure, or is unreliable.
What is a virtual measurement?
A virtual measurement is an estimated value produced from models and available data. It can support surveillance, diagnostics, control, optimization, or allocation calculations when direct measurement is incomplete.
How are soft sensors used in process optimization?
Soft sensors provide current estimates or predicted values that feed operating decisions, setpoint calculations, quality decisions, or optimization routines.
Are virtual flow meters soft sensors?
Yes. Virtual flow meters are one type of soft sensor. They estimate flow values from related operating data and reference measurements.
What makes a soft sensor reliable?
Reliability depends on data quality, reference measurements, operating coverage, calibration, validation, model monitoring, and clear agreement about how accuracy is assessed.
Measurement problem
Need a reliable estimate where direct measurement falls short?
Bring the value, the available measurements, the reference data, and the decision that depends on the estimate.