2010
DOI: 10.1002/ceat.200900615
|View full text |Cite
|
Sign up to set email alerts
|

Designing and Testing a Chemical Demulsifier Dosage Controller in a Crude Oil Desalting Plant: An Artificial Intelligence‐Based Network Approach

Abstract: The aim of this paper is to present an artificial neural network (ANN) controller trained on a historical data set that covers a wide operating range of the fundamental parameters that affect the demulsifier dosage in a crude oil desalting process. The designed controller was tested and implemented on-line in a gas-oil separation plant. The results indicate that the current control strategy overinjects chemical demulsifier into the desalting process whereas the proposed ANN controller predicts a lower demulsif… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
4
0

Year Published

2019
2019
2024
2024

Publication Types

Select...
5
1

Relationship

0
6

Authors

Journals

citations
Cited by 6 publications
(4 citation statements)
references
References 26 publications
(20 reference statements)
0
4
0
Order By: Relevance
“…Mostly, the separated oil is placed in a separator if the mixture concentration is much [19]. Widely, the centrifuge has not been used compared to other approaches for emulsion treatment because of its high capital cost and low capacity [8,19,22]. The presence of specifically massive droplets in crude oil emulsions frequently results in diminished flow velocity, enhancing the utilization of gravitational forces to separate small suspended droplets, water and oil; which generally occur within a space of time in the separator or large-volume desalters [22,54].…”
Section: Mechanical Methodsmentioning
confidence: 99%
“…Mostly, the separated oil is placed in a separator if the mixture concentration is much [19]. Widely, the centrifuge has not been used compared to other approaches for emulsion treatment because of its high capital cost and low capacity [8,19,22]. The presence of specifically massive droplets in crude oil emulsions frequently results in diminished flow velocity, enhancing the utilization of gravitational forces to separate small suspended droplets, water and oil; which generally occur within a space of time in the separator or large-volume desalters [22,54].…”
Section: Mechanical Methodsmentioning
confidence: 99%
“…Demulsification by mechanical means entails the disruption of physical barriers or the application of density variation between the water and the oil phases [29]. Mechanical demulsification of crude oil emulsions can be carried out using a series of mechanical equipment, such as a free-water knockout drum, two-and three-phase separators (low-and high-pressure traps), desalters, and settling tanks [31]. The presence of relatively large droplets in crude oil emulsions often results in reduced flow velocity, enabling the use of gravitational forces for the separation of oil, water, and small suspended droplets [127], which usually occurs within a short time in large-volume desalters or separators [127,128].…”
Section: Mechanical Demulsificationmentioning
confidence: 99%
“…Demulsification by mechanical means entails the disruption of physical barriers or the application of density variation between the water and the oil phases for demulsification processes [29,30]. Mechanical demulsification of crude oil emulsion can be carried out using series of mechanical equipment such as free-water knockout drum, two-and three-phase separators (low-and high-pressure traps), desalters, and settling tanks [31,32]. Demulsification by thermal treatment refers to the separation of oil and water phase in crude oil emulsion by increasing the temperature of the emulsion [33].…”
Section: Introductionmentioning
confidence: 99%
“…[ 95 ] In predictive control schemes, ANNs are used to provide a process model, [ 36 ] directly adjust controller parameters, [ 100 ] or indirectly determine control action based on the process model that they identify simultaneously. [ 107 ] It may be noted that all ANN‐based controllers can be rendered adaptive by training the model identifier on‐line. ANNs can even be trained to replace optimization routines of the controllers.…”
Section: Common Ai‐based Process Control Technologiesmentioning
confidence: 99%