Lean mono-ethylene glycol (MEG) concentration with presence of inorganic salt virtual meter

A non-transitory, computer-readable medium storing one or more instructions executable by a computer system to perform operations comprising:

• receiving, from sensors, spot samples comprising electrical conductivity and density of mono-ethylene glycol (MEG) concentrations in a transmission pipeline of an industrial plant;
• executing, process simulations to generate intermediate parameters and a correlation model correlating the MEG concentrations and inorganic salt concentrations;
• generating based on the process simulations, a model for estimating a salt concentration build-up rate from the spot samples, the model defining an estimated variation of the MEG concentrations matched over time to plant operating conditions;
• predicting, using hybrid artificial intelligence and physics-guided machine learning, a lean MEG concentration with presence of inorganic salts by correlating the estimated variation of the MEG concentrations at an outlet of a MEG regeneration system, to the salt concentration build-up rate, a reboiler operating pressure, and a temperature of the MEG regeneration system, to determine an amount of the lean MEG concentration at the outlet of the MEG regeneration system;
• and
• providing an instruction to adjust plant operating conditions to resolve an inhibition of the transmission pipeline.

A computer-implemented system, comprising:

• one or more processors;
• and
• a non-transitory computer-readable storage medium coupled to the one or more processors and storing programming instructions for execution by the one or more processors, the programming instructions instructing the one or more processors to perform operations comprising:
• receiving, from sensors, spot samples comprising electrical conductivity and density of mono-ethylene glycol (MEG) concentrations in a transmission pipeline of an industrial plant;
• executing, process simulations to generate intermediate parameters and a correlation model correlating the MEG concentrations and inorganic salt concentrations;
• generating based on the process simulations, a model for estimating a salt concentration build-up rate from the spot samples, the model defining an estimated variation of the MEG concentrations matched over time to plant operating conditions;
• predicting, using hybrid artificial intelligence and physics-guided machine learning, a lean MEG concentration with presence of inorganic salts by correlating the estimated variation of the MEG concentrations at an outlet of a MEG regeneration system, to the salt concentration build-up rate, a reboiler operating pressure, and a temperature of the MEG regeneration system, to determine an amount of the lean MEG concentration at the outlet of the MEG regeneration system;
• and
• providing an instruction to adjust plant operating conditions to resolve an inhibition of the transmission pipeline.