|Instructor(s)||Md Waselul Haque Sadid [Coordinator]|
Office Hours: Wednesday 10:00 - 11:00
|Calendar Description||Introduction to modern methods of linear system identification. Different types of models. Review of classic time- and frequency-based approach to empirical, 'black-box' system modeling. Non-parametric identification: impulse and step weights, spectral analysis. Parametric, discrete transfer function models from I/O data using Least Squares. Data-collection procedures, model structure selection, use of auto- and cross-correlation functions for diagnostics and model validation, overview of different estimation algorithms.|
|Learning Objectives (Indicators)|
At the end of this course, the successful student will be able to:
NOTE:Numbers in parentheses refer to the graduate attributes required by the Canadian Engineering Accreditation Board (CEAB).
3.0 hours of lecture per week for 13 weeks
|Teaching Assistants||Shayan Sepahvand, firstname.lastname@example.org|
Note: In order for a student to pass a course, a minimum overall course mark of 50% must be obtained. In addition, for courses that have both "Theory and Laboratory" components, the student must pass the Laboratory and Theory portions separately by achieving a minimum of 50% in the combined Laboratory components and 50% in the combined Theory components. Please refer to the "Course Evaluation" section above for details on the Theory and Laboratory components (if applicable).
|Examinations||Course evaluation is ongoing and semester-long, and includes both group work (lab/tutorial reports) and individual effort (final project). All reports include simulations. The course professor verifies all individual codes submitted with the final report. If the execution of the code does not support claims in the report, the project will receive a non-negotiable and significant reduction in the grade.|
|Other Evaluation Information||Course activities are part of the ongoing and semester-long evaluation: there are graded activities in every week of classes (on top of scheduled tutorial/lab reports). The graded activities include both individual assessments (D2L quizzes), homework assignments and computer simulations on the application of theory learned, which are then demonstrated to the professor in class.|
|Teaching Methods||Lectures and Tutorials will be conducted in person. Students will have access to course materials on D2L. Students will be required to complete D2L Quizzes and Homework Assignments using upload features of D2L. All tutorials and final project reports have to be uploaded to D2L. Zoom teleconferencing software will be used for individual consultation, office hours, and individual student simulation presentations.|
|Other Information||Students will learn to work MATLAB System Identification Toolbox in the tutorial session with the help of course TA.|
Goals for the course and course logistics. Overview: terminology, objectives, introduction to modern identification procedures (diagnostics, identification, validation), types of models. Data Collection - PRBS signal. Introduction to Matlab System Identification Toolbox. Introduction to Tutorial # 1. Review - frequency response, Bode plots for conventional modeling.
Modeling: Non-parametric models in frequency domain: SPA, ETFE. Review: Transfer function models, conversion between continuous and discrete representations, sampling. Modeling - simple Box-Jenkins model structures: OE Model (deterministic process, white noise). Diagnostic tools in frequency domain - summary. Activity # 1 due.
Introduction to Tutorial # 2: Non-parametric models in time domain. Review - time domain response for conventional modeling (Step and Impulse response plots). Review of basic definitions of stochastic processes. Non-parametric models in time domain: impulse and step weights from de-convolution and from correlation analysis. The effect of noise on non-parametric models in time domain. Simple parametric, non-robust, discrete transfer function models from impulse weights. Hankel Test of system order. Activity # 2 due.
Diagnostic tools in time domain summary. Review of matrix algebra. Introduction to Least Squares methods. Robustness of parametric models, The effect of noise on conventional parametric models (non-robust and robust).
Introduction to time-series modeling. Combined dynamic-stochastic models - Box-Jenkins structures. Overview of different parameter estimation algorithms. Figures of Merit for Model selection: Akaike Index, Loss Function. Refining OE model: ACF, PACF and CCF checks. Activity # 3 (D2L Quiz) due.
Validation for OE Model: Residue whiteness testing - Chi-Square tests, Confidence Intervals. Full examples of OE Model Identification and Validation. Hands-on simulations and group work.
Hands-on simulations and group work - Activity # 4 due. Introduction to Tutorial # 3: Stochastic noise models. Noise models: AR, MA, ARMA, processes. Auto- and Partial Auto-Correlation Functions as diagnostic tools for stochastic noise models.
Hands-on simulations and group work - Activity # 6 due. Summary of all diagnostic tools for all Box-Jenkins models: non-parametric time and frequency domain models, Auto- and Partial Auto-Correlation functions.
Hands-on simulations and group work- Activity # 7 due. Refining BJ model: ACF, PACF and CCF checks. Complete Validation for BJ Model: Residue whiteness testing - Chi-Square tests, Confidence Intervals. Activity # 8 (D2L Quiz) due.
Hands-on simulations and group work - Activity # 9 due. Review - designing data collection experiment, model structure selection, complete diagnostics, structure revisions and final model validation. Examples of a full system identification procedure.
Hands-on simulations and group work - Activity # 10 due. Overview of the Final Project (individual): "Black Box" System Identification of two systems (OE and PEM structures). Questions and answers regarding the project.
Hands-on simulations and group work. Questions and answers regarding the final project, active consultation on final project computer simulations. Activity # 11 (D2L Quiz) due.
Questions and answers regarding the final project, active consultation on final project computer simulations. Final Project due on December 6.
Tutorial 1: Diagnostic Tools in Frequency Domain and Simple Model identification - OE Model (2 sessions)
Tutorial 2: Diagnostic Tools in Time Domain and Simple Model identification - OE Model (2 sessions)
Tutorial 3 - Stochastic Noise Models - Identify structure of four different noise models (2 sessions)
Tutorial 4 - Simple System Identification of a Real-Life System â€“ Servomotor (3 sessions)
Students are reminded that they are required to adhere to all relevant university policies found in their online course shell in D2L and/or on the Senate website
You can submit an Academic Consideration Request when an extenuating circumstance has occurred that has significantly impacted your ability to fulfill an academic requirement. You may always visit the Senate website and select the blue radio button on the top right hand side entitled: Academic Consideration Request (ACR) to submit this request.
For Extenuating Circumstances, Policy 167: Academic Consideration allows for a once per semester ACR request without supporting documentation if the absence is less than 3 days in duration and is not for a final exam/final assessment. Absences more than 3 days in duration and those that involve a final exam/final assessment, require documentation. Students must notify their instructor once a request for academic consideration is submitted. See Senate Policy 167: Academic Consideration.
Academic Accommodation Support (AAS) is the university's disability services office. AAS works directly with incoming and returning students looking for help with their academic accommodations. AAS works with any student who requires academic accommodation regardless of program or course load.
Academic Accommodations (for students with disabilities) and Academic Consideration (for students faced with extenuating circumstances that can include short-term health issues) are governed by two different university policies. Learn more about Academic Accommodations versus Academic Consideration and how to access each.
At Toronto Metropolitan University, we recognize that things can come up throughout the term that may interfere with a student’s ability to succeed in their coursework. These circumstances are outside of one’s control and can have a serious impact on physical and mental well-being. Seeking help can be a challenge, especially in those times of crisis.
If you are experiencing a mental health crisis, please call 911 and go to the nearest hospital emergency room. You can also access these outside resources at anytime:
If non-crisis support is needed, you can access these campus resources:
We encourage all Toronto Metropolitan University community members to access available resources to ensure support is reachable. You can find more resources available through the Toronto Metropolitan University Mental Health and Wellbeing website.