Course syllabus - AI-driven Prognostics for Industrial Systems

Objectives

Equip engineers, scientists, operators, and managers with the skills to apply AI-based methods for condition monitoring and prognostics in industrial systems and high-value assets. Participants will learn to identify common failure causes and predict Remaining Useful Life (RUL), involving tasks such as feature engineering, model development, and uncertainty quantification. Led by industry and academia experts, the course introduces state-of-the-art AI-driven prognostic techniques and advanced signal processing methods.

Learning outcomes

1. Describe the core principles of prognostics for industrial systems, focusing on early detection and estimating remaining useful life (RUL).
2. Utilize AI techniques for signal processing and prognostic tasks and analyze the results.
3. Assess the reliability and effectiveness of prognostic models using standard performance metrics and refine models based on evaluation outcomes.
4. Predict RUL of different system components and high value assets, including degradation tracking and utilization of periodic/dynamic measurements.

Course content

1. Introduction to Industrial prognostics
   * Basics of prognostics
   * Prognostic methods overview
   * Prognostics performance metrics
   * Diagnostics vs prognostics
2. AI-driven prognostics
   * Fundamentals of AI
   * AI methods for prognostics
   * Signal processing and feature selection
   * Illustrative examples of prognostics using AI methods
3. Practical applications of prognostics and timeseries analysis methods.
   * Basics of timeseries analysis-based prognostics and illustrative examples
4. Performance degradation prognostics: Basics and illustrative examples
5. Case studies from a variety of application domains
6. Uncertainty in prognostics
   * Sources of uncertainty
   * Uncertainty quantification and management

Specific requirements

75 credits in energy engineering, mechanical engineering, production technology, product and process development, computer technology and/or computer science or equivalent or 40 credits in technology and at least 2 years of full-time professional experience from a relevant area within industry. In addition, English course 6 or English level 2 are required.

Examination

Assignment (INL1) 2 credits, grade: Pass (G) or Fail (U). Learning outcome: 2, 3, 4.
Home examination (HEM1) 1 credits, grade: Pass (G) or Fail (U). Learning outcome: 1-4.

A student who has a certificate from MDU regarding disability study support, can request adaptions for the examination. It is the examiner who takes decisions on any adaptions, based on the certificate and other conditions.

Grade

Two-grade scale