Statistical Modeling of Separator Processes - An Application of Gaussian Processes with Bayesian Optimization
(2020) In Master's Theses in Mathematical Sciences FMSM01 20201Mathematical Statistics
- Abstract
- The separator is a machine with many applications, commonly used to separate liquids or solids into components with different density. Each application demands its own unique set of process parameters to achieve optimal results. Often the procedure of finding the best process parameters is conducted empirically, which can be very time consuming. This thesis aims to address this problem by providing a statistical model of separator processes, which can be used to find the optimal process parameters more efficiently.
Different sensors are mounted on two separators used in regular production. The sensors measure values of the inputs, outputs and process parameters. A Gaussian process is used to model the regression relationships between... (More) - The separator is a machine with many applications, commonly used to separate liquids or solids into components with different density. Each application demands its own unique set of process parameters to achieve optimal results. Often the procedure of finding the best process parameters is conducted empirically, which can be very time consuming. This thesis aims to address this problem by providing a statistical model of separator processes, which can be used to find the optimal process parameters more efficiently.
Different sensors are mounted on two separators used in regular production. The sensors measure values of the inputs, outputs and process parameters. A Gaussian process is used to model the regression relationships between the process parameters and outputs for two separators. Bayesian optimization is then used to find optimal process parameters, which are shown to be accurate in simulations. In four models, one for each output of the two separators, the optimal process parameters are seen to improve the outputs. In the first separator only small improvements can be seen, as the optimal process parameter is near the middle of the data used to build the model. In the second separator large improvements can be seen. Here, the optimal process parameter is at the upper endpoint of the interval, implicating that a higher value of the process parameter could further improve the outputs. Thus, further experiments with a higher value of the process parameter are needed in order to draw conclusions on the optimal process parameter for the second separator.
These optimal process parameters will be used in the real separators to possibly improve the separator performance. The data used in this thesis is supplied by the manufacturer of the separators, Alfa Laval. (Less) - Popular Abstract
- By using real data from regular product production of two separators, it is shown that the statistical model manages to create a good model of the separation process. Further, the results show that the process parameters can be altered in order to improve multiple objectives of the separation process simultaneously.
The separator is a machine with many applications, commonly used to separate liquids or solids into components with different density. In order to improve its performance, testing is often conducted empirically, which can be costly and time-consuming. Each application demands its own unique set of process parameters to achieve optimal results. To find these process parameters deep knowledge of the specific application is... (More) - By using real data from regular product production of two separators, it is shown that the statistical model manages to create a good model of the separation process. Further, the results show that the process parameters can be altered in order to improve multiple objectives of the separation process simultaneously.
The separator is a machine with many applications, commonly used to separate liquids or solids into components with different density. In order to improve its performance, testing is often conducted empirically, which can be costly and time-consuming. Each application demands its own unique set of process parameters to achieve optimal results. To find these process parameters deep knowledge of the specific application is needed. By using a statistical approach, the optimal process parameters can be found more efficiently with less specific knowledge required.
In order to build a statistical model, data is needed. This data is collected by mounting many different sensors on a separator. The sensors measure values of the inputs, outputs and process parameters of the separator. It is easy to understand that the outputs depend on both what is put in the separator and the settings of the separator, called process parameters. Therefore, it is of interest for the operator to find the best process parameters, in order to achieve the best possible outputs.
A statistical model which can be used for most data, from financial applications to industrial separation, is the Gaussian process. It can be used to simulate the real separation process. This allows for quick testing of different settings and scenarios. These tests and simulations can be used to shed light on possible improvements of the process parameters.
In this work, two outputs of each separator were regulated with only one process parameter. Four different Gaussian process models, one for each output of the two separators, were built. These models were then used to find four different optimal process parameters. It was shown that both outputs had almost exactly the same optimal process parameter, which means there is a single best setting for both outputs of each separator.
This statistical approach is aimed to increase the understanding of the separator process, but more importantly lessen the need of real testing. It is easy use for any separator application and would be ideal to help guide an inexperienced user. (Less)
Please use this url to cite or link to this publication:
http://lup.lub.lu.se/student-papers/record/9021987
- author
- Svensson, Tim LU
- supervisor
- organization
- course
- FMSM01 20201
- year
- 2020
- type
- H2 - Master's Degree (Two Years)
- subject
- keywords
- Gaussian process, Bayesian optimization, Separator
- publication/series
- Master's Theses in Mathematical Sciences
- report number
- LUTFMS-3394-2020
- ISSN
- 1404-6342
- other publication id
- 2020:E54
- language
- English
- id
- 9021987
- date added to LUP
- 2020-07-03 16:28:38
- date last changed
- 2021-06-04 18:13:09
@misc{9021987,
abstract = {{The separator is a machine with many applications, commonly used to separate liquids or solids into components with different density. Each application demands its own unique set of process parameters to achieve optimal results. Often the procedure of finding the best process parameters is conducted empirically, which can be very time consuming. This thesis aims to address this problem by providing a statistical model of separator processes, which can be used to find the optimal process parameters more efficiently.
Different sensors are mounted on two separators used in regular production. The sensors measure values of the inputs, outputs and process parameters. A Gaussian process is used to model the regression relationships between the process parameters and outputs for two separators. Bayesian optimization is then used to find optimal process parameters, which are shown to be accurate in simulations. In four models, one for each output of the two separators, the optimal process parameters are seen to improve the outputs. In the first separator only small improvements can be seen, as the optimal process parameter is near the middle of the data used to build the model. In the second separator large improvements can be seen. Here, the optimal process parameter is at the upper endpoint of the interval, implicating that a higher value of the process parameter could further improve the outputs. Thus, further experiments with a higher value of the process parameter are needed in order to draw conclusions on the optimal process parameter for the second separator.
These optimal process parameters will be used in the real separators to possibly improve the separator performance. The data used in this thesis is supplied by the manufacturer of the separators, Alfa Laval.}},
author = {{Svensson, Tim}},
issn = {{1404-6342}},
language = {{eng}},
note = {{Student Paper}},
series = {{Master's Theses in Mathematical Sciences}},
title = {{Statistical Modeling of Separator Processes - An Application of Gaussian Processes with Bayesian Optimization}},
year = {{2020}},
}