FSG3130 Uncertainty Analysis 5.0 credits

1. Introduction: Errors, uncertainty, and UQ. Different viewpoints.

2. Basic statistical tools: Errors and uncertainties in a measured variable

3. UQ forward problem: Uncertainty propagation from multiple variables

4. Sensitivity analysis

5. UQ inverse problem: Data analysis and regression

6. Verification and validation of simulations

1. Experimentation, Errors and Uncertainty
2. Errors and Uncertainties in a Measured Variable
3. Uncertainty in a Result Determined from Multiple Variables
4. General Uncertainty Analysis. Planning an Experiment and Application in validation
5. Detailed Uncertainty Analysis: Designing, Debugging, and Executing an Experiment
6. Validation of Simulations
7. Data Analysis, Regression, and Reporting of Results

The student will be able to discuss general issues regarding mainly experimental uncertainties relevant for measurements with special focus to fluid dynamic systems, the difference between systematic and random errors (bias and uncertainty), confidence intervals, calibration errors, error propagation in data reduction equations, regression analysis etc. There will also be a discussion about how to determine the uncertainty propagation by using Monte Carlo analysis. The uncertainty analysis will be exemplified through discussion of various real-life experiments (and to some extent simulations). Although many of the examples are taking from the fluid dynamics field, the discussion of the uncertainty analysis is general and can be applied to many other scientific fields. After completing this course the student should be able to:

• distinguish between random and systematic (uncertainty and bias) error.
• understand basic statistical concepts and the meaning of confidence intervals.
• calculate uncertainty in a measured variable based on the Taylor series method.
• perform a Monte-Carlo based uncertainty analysis.
• evaluate how long time a variable need to be sampled in order to obtain a certain accuracy in the measured/simulated statistics.
• handle outliers in a reliable and systematic way.
• design, debug and execute an experiment.
• understand the difference between validation and verification of simulations, and how validation can be performed.
• do an accurate regression analysis.

A master degree in a mechanics related area is recommended.

A master degree in a mechanics related area.

A. Segalini & H. Alfredsson, Uncertainty Analysis, Lecture notes, 2018

H.W. Coleman & W. Glenn Steele: Experimentation, validation, and Uncertainty Analysis for Engineers, (3rd Edition), Wiley, 2018

Rabinovich: Evaluating Measurement Accuracy, third edition, Springer, 2018

If the course is discontinued, students may request to be examined during the following two academic years.

• DEL1 - Participation, 1.0 credits, grading scale: P, F
• INL1 - Assignment, 4.0 credits, grading scale: P, F

Based on recommendation from KTH’s coordinator for disabilities, the examiner will decide how to adapt an examination for students with documented disability.

The examiner may apply another examination format when re-examining individual students.

DEL1 Participating 1,0 hp (P, F)

INL1 Assignment 4,0 hp (P, F)

The following items have to be approved in order to obtain a pass on the course:

• Compulsory and active attendance during at least 80% of the lecture time
• Successful completion of homework assignment within given time frame

not possible

Renewed examination is possible

Philipp Schlatter

• All members of a group are responsible for the group's work.
• In any assessment, every student shall honestly disclose any help received and sources used.
• In an oral assessment, every student shall be able to present and answer questions about the entire assignment and solution.

The lecture part of the course is given during one week (approximately 20h) in P2. More information will be posted on Canvas:

https://canvas.kth.se/courses/38459