MAE 598/494 Design Optimization | Fall 2019
Faculty: Max Yi Ren
Email: yiren@asu.edu
Office: GWC464
Location & Time: ECGG347, TU/TH 4:30pm-5:45pm
Office hours: TU/TH 2:00pm-4:00pm in GWC464, or by appointment
Prerequisites: Students are expected to have a solid background in calculus and linear algebra (Math review), and be fluent in Matlab and/or Python.
Course Overview
The purpose of this course is to introduce the student to mathematical modeling, optimization theory, and computational methods for analytical and simulation-based optimal engineering design. The student will learn optimization and optimal control theories, formulate mathematical descriptions for real-world design problems, and develop optimization algorithms to solve these problems.
Workload and grading
There are 5 homework assignments, 2 mid-term exams, and a team project.
- Homework (8%*5)
- Exams (15%*2)
- Project review 1 (5%)
- Project review 2 (20%)
- Project final report (5%)
Course schedule
| Date | Topic | Assignments | Note |
|---|---|---|---|
| Aug 22 (TH) | Introduction to optimization | HW 1 in | |
| Aug 27 (TU) | Math review: Linear algebra and matrix calculus | ||
| Aug 29 (TH) | Unconstrained optimization 1: Convexity and optimality conditions | HW 1 due, HW 2 in | |
| Sep 3 (TU) | Unconstrained optimization 2: Gradient descent and convergence | ||
| Sep 5 (TH) | Unconstrained optimization 3: Modern line search techniques | ||
| Sep 10 (TU) | Unconstrained optimization 4: Newton’s method | ||
| Sep 12 (TH) | Unconstrained optimization 5: Practice | HW 2 due | |
| Sep 17 (TU) | Mid-term 1 | ||
| Sep 19 (TH) | Supervised learning 1: Ordinary least square | HW 3 in | |
| Sep 24 (TU) | Supervised learning 2: Design of experiments | ||
| Sep 26 (TH) | Supervised learning 3: Neural networks | ||
| Oct 1 (TU) | Supervised learning 4: Practice | ||
| Oct 3 (TH) | Project review 1 | Mandatory team meetings | |
| Oct 8 (TU) | Guest lecture | Preliminary report due | Max out of town |
| Oct 10 (TH) | Constrained optimization 1: Reduced gradient | ||
| Oct 15 (TU) | Fall break | ||
| Oct 17 (TH) | Constrained optimization 2: Examples | HW 3 due | |
| Oct 22 (TU) | Constrained optimization 3: KKT conditions | HW 4 in | |
| Oct 24 (TH) | Constrained optimization 4: KKT geometry, sensitivity analysis | ||
| Oct 29 (TU) | Constrained optimization 5: Generalized reduced gradient | ||
| Oct 31 (TH) | Constrained optimization 6: Quasi-Newton methods | ||
| Nov 5 (TU) | Constrained optimization 7: Active set strategy | ||
| Nov 7 (TH) | Constrained optimization 8: Sequential Quadratic Programming | HW 4 due, HW 5 in | |
| Nov 12 (TU) | Mid-term 2 | ||
| Nov 14 (TH) | Project review 2 | No class, mandatory team meetings | |
| Nov 19 (TU) | Optimal control 1: Pontryagin’s Maximum Principle | Progress report due | |
| Nov 21 (TH) | Optimal control 2: Markov decision process | ||
| Nov 26 (TU) | Optimal control 3: Bellman Equation | HW 5 due | |
| Nov 28 (TH) | Thanksgiving | ||
| Dec 3 (TU) | Optimal control 4: Practice | ||
| Dec 5 (TH) | |||
| Dec 9 | Final report due |
Project
-
Each team should have no more than 4 students.
-
Students are responsible for defining a design problem and proposing a credible plan to solve it. Past project reports can be found from the course website. Tentative topics can be found here.
-
Preliminary report is due on Oct 8. The report should contain a mathematical description of the optimization problem to be solved, and all validated models needed for solving this problem. Key challenges and mitigation strategies should be identified. See detailed requirements here.
-
Progress report is due on Nov 19. The report should contain a complete analysis on the solutions to the target problem. Parametric studies should be included. See detailed requirements here.
-
Each student will be graded by their own contributions to the project.
Academic Integrity
Each student has an obligation to act with honesty and integrity, and to respect the rights of others in carrying out all academic assignments. MAE 598/494 will follow the process defined by the Office of the Dean of Students, which states that any student who is found to have violated the Student Code of Conduct will, at a minimum, receive an E in the course. The College Policy defines the process to be used if the student wishes to appeal this action.