|
CE 5250 Special Topics: System Identification |
|
|
Professor: Dr. Andrew Swartz, Ph.D. 201D Dillman Hall (906) 487-2439; raswartz@mtu.edu Office Hours: Monday 11AM-noon, Wednesday and Thursday 9-10 AM, or by arrangement.
|
|
|
[ Meeting Times | Text and Syllabus | Description | Prerequisites | Grading | Policies | Homework | Notes] |
|
|
Lecture Hours: MWF 2:05 - 2:55 PM; Dillman 214 Textbook: Subspace Methods for System Identification: Katayama (2010) Course Syllabus (pdf)
|
|
|
Course Description:
This course it intended to introduce to the student the basic tools required for the process of system identification. The system identification process allows engineers to develop data-driven models mapping system inputs to measured outputs for systems that may be too complicated to model mechanistically or those with mechanistic models that require validation or refinement.
|
|
|
Prerequisites:
Graduate standing. It is expected that all students
understand (and are able to apply) general concepts from integral and
differential calculus, linear algebra, statistics and probability, as well as
partial differential equations as required during the course of an ABET
accredited undergraduate degree in engineering.
|
|
|
Homework Assignments (about 8) ........................................................... Exam (Midterm - Take home) ................................................................... Final Project ................................................................................................ Total ...........................................................................................................
|
40% 25% 35% 100% |
|
Attendance: Attendance is required. Exams: There will be one exam midway through the semester. The exam will be a take home exam with the duration and timing to be determined using input from the class. Computing Policy: It is expected that students will have access to a computer running Matlab for homework assignments, the exam, as well as the final project. Final Project: Students will complete a final project applying the system identification techniques learned in class to a topic selected by the student. In the course of the project, students will form models from data make observations about the models that they have made. Students will present their findings in a report. Topics may be drawn from existing research projects or other areas of interest to the student. Homework: All homework will be collected on the date due. Homework may be turned in in-class or overnight under my office door (201D Dillman Hall). Assignments should be written neatly and legibly. Begin each new problem on a new page. Supporting work and Matlab code is necessary to receive credit. Detailed assignment guidelines specific to each homework set will be distributed via the course website. Scheduling of the Final Exam: There is no final exam for this course. Collaboration Policy: Collaboration on homework sets is encouraged. Multi-student homework collaborations (i.e., a single solution set submitted for multiple students) will be accepted so long as the students represented agree to share the same grade for the assignment. Exams are to be strictly non-collaborative. Course Email List: Dr. Swartz maintains an email list for CE5250 to share information he thinks may benefit the class on an ad-hoc basis. Dr. Swartz considers this information to be a supplement to the classroom experience, not a replacement for lecture attendance. Final Grade Basis: A 93 – 100 AB 87 – 92.9 B 83 – 86.9 BC 77 – 82.9 C 73 – 76.9 CD 67 – 72.9 D 63 – 66.9 F Below 63 ADA Statement: “MTU complies with all federal and state laws and regulations regarding discrimination, including the Americans with Disabilities Act of 1990.” (ADA) If you have a disability and need a reasonable accommodation for equal access to education or services at MTU, please call the Dean of Students at 487-2212. For other concerns about discrimination, you may contact your advisor, department head or the Affirmative Action Office at 487-3310. |
|
|
Homework
Assignments:
|
|
|
Course Notes:
Lesson 0 - Introduction Lesson 1 - Signals and Systems Lesson 2 - Fourier Analysis Lesson 3 - Sampling and Discrete Fourier Transform Lesson 4 - Aliasing and Windowing Lesson 5 - Frequency Domain Representation of LTI Systems and Convolution Lesson 6 - Behavior of LTI Systems Lesson 7 - MDOF LTI Systems Lesson 8 - The Z-Transform Lecture m.files: 1DOF System; 2DOF System Lesson 9 - DT Transfer Function Models Lesson 10 - ARX Models and Least-Squares Projection Lesson 11 - Recursive Least-Squares for ARX Lesson 12 - Model Size Evaluation Lesson 13 - Feedback Systems Lesson 14 - MIMO v. SISO Lesson 15 - Moving Average Models Lesson 16 - Introduction to State-Space Models Lesson 17 - DT State-Space Models Lesson 18 - Features of State-Space Models Lesson 19 - State-Space Realization Preliminaries Lesson 20 - ERA Lesson 21 - Advanced System ID Topics Lesson 22 - Introduction to the Kalman Filter
|
|