The Soft Sensors component of Capstone’s MACS™ product group is your tool to predict plant quality variables in real time. It can be used to estimate properties that are impractical or impossible to measure online.
By industry, successful applications include:
• RVP
• Distillation
• Cetant
• Brightness
• Freeness
• Tensile strength
• Molecular weight
• Density
MACS Soft Sensors read live data – such as temperatures and pressures – from the plant and use them to calculate estimated quality values from user-entered models. These models can either be based on first principles or empirical models developed through techniques such as PCA, PLS, neural networks or time-series analysis.
The inferred properties can be sent to operators as information, used as inputs for diagnostic systems, or directly used as controlled variables in conventional or predictive control applications.
MACS Soft Sensors can read/write information from virtually any data source including:
- OPCDA, OPCHDA
- Plant historian systems (PI, PHD, IP21...)
- SQL
- Excel
Features of MACS Soft Sensors include dead time compensation, filtering, data validation, model updating from lab analyses and diagnostic health tags.