Idea Details

Team and Contact Details

SNo Student Name School Degree Year Email
1Hamza ShafiqSCMEMS/M.PhilSecondhshafiq.che5@scme.nust.edu.pk

Inter School Idea ? No
Do you need expertises from another area: Yes
If Yes please provide details of expertises you need: Computer Science, Control System, Control Design

Idea Details

Idea Name: Predictive Analytics
Slogan: Torturing the Data
Supervisor Name: Dr. Arshad Hussain
Supervisor Designation: Principal/Dean/Professor
Supervisor School: School of Chemical and Materials Engineering (SCME)
Supervisor Department: Chemical Engineering Department
Contact number: 051-90855105
Email ID: arshad.hussain@scme.nust.edu.pk
Abstract:
    Model-based predictive control (MPC) is developed for crude oil fouling rate and optimized cleaning schedule. MPC is then incorporated with distributed control system (DCS) for optimum operation.
What is the unmet need in society that your idea will fulfill ?
    For processing 100,000 bbl./day, a drop of 1 degree C in coil inlet temperature due to fouling results in approximately £ 25,000. Heat exchanger fouling costs for highly industrialized countries are about 0.25% of their Gross National Product.
Who needs it ? How many would benefit ?
   Chemical Industries utilizing heat exchangers. Petroleum refineries with crude distillation unit. Engineering design firms - provides better understanding of fouling and help in selecting design layout which diminish fouling rate. Industrial cleaning service provider.
How will the solution works
    Predictive tool is based on mathematical model. Macro-enable sheet incorporated with the control system will gather data to predict fouling rate and generates optimum operating parameters. Data based method is proposed for optimizing heat exchangers cleaning schedule. Artificial Neural Network will be used within the hybrid Taguchi framework and Genetic Algorithm. Optimization is carried out on lowering the furnace energy consumption. The model will cope with the overall process uncertainties.
Who are your competitors ? How is your solution different
    Dynamic modeling softwares like Aspen Plus and UniSim. However they does not take fouling initialization period into account making them less accurate. Such softwares are unable to provide cleaning schedule. All these shortcomings are addressed in my tool.
Status: Approved
Entry Date & Time: 2018-12-22 (1441)