873018 Artificial Intelligence in Geotechnical Engineering


Type
Lecture and exercise
Semester hours
2
Lecturer (assistant)
Soranzo, Enrico
Organisation
Offered in
Wintersemester 2024/25
Languages of instruction
Deutsch

Content

1. Introduction to Artificial Intelligence (AI)
1.1 General description of AI
1.2 Main definitions: Machine Learning, Deep Learning, Reinforcement Learning
1.3 Subsets of the algorithms: Classification/Regression, Supervised/Unsupervised Learning
1.4 Methodology: Train/Test, cross-validation
1.5 Performance: performance metrics, overfitting

2. AI application in Geotechnical Engineering
2.1 Parameter correlation: shear parameters, saturated and partially saturated soils, undrained cohesion, compressibility index, SWCC, Proctor curve, permeability coefficient
2.2 Soil classification based on in-situ tests: SPT, CPT, DLM
2.3 Safety prediction: overall stability of (reinforced/anchored) slopes, tunnel face stability, bearing capacity of shallow and deep foundations, liquefaction, internal forces in tunnel lining
2.4 Displacement prediction: rotations in retaining walls, tunnel settlement
2.5 Machine performance: TBM excavation speed
2.6 Constitutive models

3. Computational implementaion of AI
3.1 Crash Course on AI with Python: installation, first steps
3.2 Data collection for geotechnical engineering: literature data, test data, surrogate models, machines
3.3 Relevant AI-Algorithms for geotechnical engineering: Artificial Neural Networks, Support Vector Machine, etc.
3.4 Joint handling of example applications for geotechnical engineering
3.4.1. Parameter correlation
3.4.2 Soil classification
3.4.3 Overall stability
3.4.4 Displacement prediction
3.4.5 Machine performance

4. Homework project
4.1 Project introduction: Data source and analysis objective
4.2 Content description of the individual project

Previous knowledge expected

Basics of Geomechanics and Geotechnical Engineering, Basics of statistics and scientific programming

Objective (expected results of study and acquired competences)

1. Students will be able to explore and list potential applications of AI in geotechnics.
2. They will comprehensively explain various AI techniques and critically evaluate their pros and cons.
3. Upon completing the course, students will be empowered to independently apply AI methods vigorously for practical applications in civil engineering.
4. Through a project, they will intensively grasp the fundamentals of the Python programming language for Data Science in the context of geotechnics.
5. The skill to construct and meticulously prepare geotechnical databases will be acquired.
6. Based on this data, students will be capable of actively training, extensively testing, and continuously optimizing the performance of AI algorithms.
You can find more details like the schedule or information about exams on the course-page in BOKUonline.