Prediction Of Two-Phase Heat Transfer Coefficients Using Artificial Neural Networks

The current Project deals with the prediction of Two-Phase heat transfer coefficients using Artificial Neural Networks (ANN) for natural refrigerant mixtures which are widely used for low-temperature applications such as Cryocoolers, Bio-Freezers, small scale air, and nitrogen liquefaction plants.

About your Instructor

Mr. Venkatesh Dasari is a PhD Scholar at the Refrigeration and Air conditioning laboratory, Mechanical Engineering Department, IIT Madras.

He is currently working on Low Temperature Mixed Refrigerant Joule Thomson(MRJT) Cryocoolers. Familiar with simulation softwares such as Aspen plus and hands on experience in Rating and Design of Helical coil heat exchangers, Capillary tubes, and selection of different Mixed refrigerants using various optimization techniques.

LinkedIn Profile: https://www.linkedin.com/in/venkateshdasari-iitm/

 

About the Project

Heat Exchangers are used in almost all industrial processes that involve either direct or indirect heating or cooling applications. The most familiar applications that we use daily are refrigerators, Air-Conditioners, Pasteurization of Milk, etc. Design and Rating of heat exchangers are crucial for all the above-mentioned processes and process outcomes may deviate significantly if we chose incorrect heat exchangers. There are different types of heat exchangers that are used for specific applications.

The challenging part in the design of heat exchangers is the prediction of the heat transfer coefficient. Heat transfer coefficient is the function of both flow and geometrical parameters. Flow parameters include mass flow rate, pressure, temperature, etc. and geometrical parameters include orientation of heat exchangers, type of heat exchanger, geometrical details such as length, diameter, etc. Prediction of Two-phase heat transfer coefficient is one of the challenging tasks for thermal engineers and most conventional methods are applicable for particular data groups with limited accuracy.

The current Project deals with the prediction of Two-Phase heat transfer coefficients using Artificial Neural Networks (ANN) for natural refrigerant mixtures which are widely used for low-temperature applications such as Cryocoolers, Bio-Freezers, small scale air, and nitrogen liquefaction plants.

 

Outcome of the Project

This project will give you insights into the basic difference between the design and rating of heat exchangers. Basics of two-phase heat transfer flow patterns and calculation of flow boiling heat transfer coefficients. Introduction to Machine learning, ANN, and supervised ANN techniques. At the end of the project, you will gain the following skills.

  1. Basics of heat exchanger design and Rating.
  2. Basics of Two-Phase heat transfer.
  3. Estimation of Two-Phase heat transfer coefficients using different empirical correlations.
  4. Introduction to ML / ANN.
  5. Building ANN models using Keras Python.
  6. Basics of FORTRAN/Python/MATLAB of your choice to write code for the project

Week 1

1
Introduction to heat exchanger rating and design.
2
Introduction to ML/ANN

Week 2

1
Basics of Two-Phase heat transfer coefficients and overview of existing correlations

Week 3

1
Introduction to MATLAB/FORTRAN/PYTHON of your choice and writing simple codes in any of the above language

Week 4

1
Working on literature and gathering experimental data and pre-processing of data

Week 5

1
Building ANN Models using Keras and getting regression models for the Data

Week 6

1
Analysing Results and Comparing with Existing Models and post processing of Results

Week 7

1
Report Creation and Submission
1) Fundamentals of Heat Transfer, Thermodynamics, Basic Engineering Mathematics 2) MATLAB (Basics Will be taught in this course) 3) Weekly 4 to 5 hours of time for reading literature and practice.
1) All Live sessions will happen in a private group for the project in the Workplace 2) A Whatsapp group will also be formed for general discussion and provide remainders.
Course available for 50 days
Get course
30-Day Money-Back Guarantee

Includes

12 LIVE Sessions
Full lifetime access
Access on mobile and TV