Master Thesis - Deep Learning for Surrogate CFD modelling

Job description

·  Finspang, Sweden
·  Engineering
·  179984
Gas and Power

Share Job
· 
· 
· 
· 

Job Description

Are you a master student planning to write your Master Thesis during spring 2020? Join us on our journey into the future #Siemens

Be part of an open and dynamic workplace where professional and personal development is high on the agenda. By making sustainable energy solutions more cost effective, developing new technologies for the future's smart industry and electrifying passenger and freight transport, we make reality of our vision of a sustainable world.

We are now looking for a student to take on the assignment “Deep Learning for Surrogate CFD modelling.”

Who we are?

Data Analytics in SIT AB , the digitalization transformation performed by Siemens during the recent years has had many consequences. One of the most important ones has been the establishment of processes to collect and maintain useful data in a database format. On the one hand, the extensive maintenance reports database provides the company with useful information about unexpected events, component repair and operation history. On the other hand, multiple sensors placed along the turbine deliver information about thermodynamic parameters and operating parameters as the amount of produced MWh. The data analytics department has been working extensively using this data to automate decision making process for the power plants operators, as well as to provide useful information to other departments within SIT AB. Some examples of these projects have been: creating visualization tools to investigate the operation profile of the turbines, optimizing the compressor washing time, optimizing the power production of the plants according to market and demand forecasts using machine learning or predicting the remaining life of a component using machine learning.

The assignment:

Numerical simulations on fluid dynamics problems and finite element analysis primarily rely on spatial or/and temporal discretization of the governing equations that dictate the physics of the studied system using polynomials into a finite-dimensional algebraic system. Due to the complex nature of the physics and sensitivity from meshing variable geometries, these numerical simulations are often prohibitive in terms of computational power for real-time applications. Because of this issue developing computationally cost-effective surrogate models are of high interest to allow this near real time calculation.

This master thesis is part of Siemens efforts to develop advanced scalable lifing and stress calculations that can provide near real-time information about the failure probability of different components in the turbine. The final goal of this project is to develop a surrogate model using deep learning that can reproduce the behavior of the CFD calculations under different inputs for a certain geometry or gas turbine component.

The first step of this assignment is to select a simple geometry (for instance a 2D airfoil) and perform the following steps:

· 
Run traditional CFD simulations over the airfoil profile in order to obtain and input/output dataset (this can be velocity profiles, viscosity uncertainty etc..). This will be done changing the inputs of the simulation, the boundary conditions etc…

· 
Rearrange the dataset adding the metadata of the different simulation runs as additional information for the Deep Learning algorithm.

· 
Train and test different Deep Learning algorithms evaluate the performance of them in a quantitative way.

· 
Explain the influence of the different input features to the output of the simulation.

The second step consist of selecting a more complex geometry (simplified blade design for instance) and repeat the experiments of the first step but taking into consideration the following:

· 
Can we include a physic-constrained learning mechanism within our previous deep learning algorithm?

· 
If so, how this affects the performance of the prediction in terms of accuracy, prediction time and training time.

· 
Are we able to explain the predictions in a better way using this new training?

Students will be provided with access to all the needed data. They will be working closely with domain experts with strong backgrounds in statistics, data mining, machine learning and mathematical optimization.

Your Profile:

·  The project is suitable for a student with academic background in energy systems, engineering, computer science, statistics, mathematics or another relevant field .
· 
As a student you have strong analytical skills and solid mathematical background.

· 
Besides, you are interested in data analytics (especially in prescriptive analytics) and hold good programming skills.

· 
We consider meritorious skills the knowledge of machine learning/deep learning-oriented libraries (scikit-learn, caret, mlr, keras, tensorflow, pytorch etc…), data handling libraries (Pandas or tidyverse).

· 
We also consider meritorious CFD knowledge with Opensource tools as OpenFOAM or Fluidsim.

Application:

Do not hesitate - apply today via siemens.se/jobb ref nr 179984 and no later than 2019-11-30. For questions about the role please contact recruiting manager Ronny Nordbergronny.norberg@siemens.com . For questions about the technicalities of the projects please contact:edgar.bahilo_rodriguez@siemens.com , rodriguez@siemens.com ordavood.naderi@siemens.com .

Trade Union representatives:

Christine Lindström, Unionen, 0122-817 28
Simon Bruneflod, Sveriges Ingenjörer, 0122-842 24
Jan Lundgren, Ledarna, 0122-812 33
Kenth Gustavsson, IF Metall, 0122-815 25

--------------------------------------------------------------------------------

In this recruitment we renounce all calls relating to advertising and recruitment support.

Job ID: 179984

Organization: Gas and Power

Company: Siemens Industrial Turbomachinery AB

Experience Level: Student (Not Yet Graduated)

Job Type: Full-time

Apply

Share Job
· 
· 
· 
· 

Can't find what you are looking for?

Let's stay connected

Can't find what you are looking for?

Siemens Careers

US Only:Review Accommodations for Disability

Contact Us

© Siemens AG, 1996 – 2019
Terms Privacy Policy Cookie Policy

Cookies are used on this site to assist in continually improving the candidate experience and all the interaction data we store of our visitors is anonymous. Learn more about your rights on ourPrivacy Policy page.

Okay x