DSpace Repository

A Bayesian network model for controlling hydrogen peroxide production process

Show simple item record

dc.contributor.advisor Dr. Peter Haddawy (Member) en_US
dc.contributor.advisor Dr. Qi Yulu (Member) en_US
dc.contributor.advisor Prof. Ramakoti Sadananda (Chairperson) en_US
dc.contributor.author Kyaw Hlaing Aye en_US
dc.date December 2000 en_US
dc.date.accessioned 2015-01-12T10:39:18Z
dc.date.available 2015-01-12T10:39:18Z
dc.identifier.other AIT Thesis no.CS-00-6 en_US
dc.identifier.uri http://www.cs.ait.ac.th/xmlui/handle/123456789/193
dc.description 38 leaves en_US
dc.description.abstract Bayesian belief networks have proven to be an effective technique for representing and reasoning with uncertain knowledge. A Bayesian network is a graphical representation of a probability distribution. Therefore statistical techniques can be used to construct Bayesian networks from data. This thesis presents a Bayesian network model of the hydrogen peroxide production process. The purpose of the model is to aid in controlling the process in order to maximize production. The quantity of hydrogen peroxide produced is a function of the relative concentration of the components of the working solution. But measuring these components takes an inordinate amount of time. Therefore, the a Bayesian network was build that can estimate the concentrations of the solution components based on quantities that can be easily measured during production. The network was built using four and a half years of data from one particular production plant. A user interface was constructed that permits the user to enter observations and in response suggests adjustments to the concentrations of the working solution components. The interface is domain specific and does not require the user to have any knowledge of Bayesian networks. The model is empirically evaluated by comparing the predictions to actual values for 60 days of readings. en_US
dc.description.sponsorship Asian Institute of Technology en_US
dc.relation.ispartof Thesis no. CS-00-6 en_US
dc.relation.ispartof Asian Institute of Technology. Thesis no. CS-00-6 en_US
dc.source SAT/CS en_US
dc.subject Bayesian statistical decision theory en_US
dc.title A Bayesian network model for controlling hydrogen peroxide production process en_US
dc.type Thesis en_US

Files in this item

This item appears in the following Collection(s)

Show simple item record

Search DSpace

Advanced Search


My Account