Application Examples

Engineering problems, mapped to Python.

Explore the catalog of companion notebooks connecting machine learning methods to engineering data, models, evaluation, and decisions.

56 notebooks

Chapters 6–14 methods and applications

Signed-in access browser, Colab, or local use

Validated in Google Colab

Start here

Explore the complete notebook collection.

All 56 validated notebooks can be read in full after sign-in. Authorized users may also load a personal copy into Google Colab or download the `.ipynb` file after accepting the access and use restrictions.

Browse the catalog Access and use terms

01
Select an AE. Search by engineering problem, method, package, or chapter.
02
Review its scope. Check the method, application, key packages, and validation status.
03
Sign in and use it. Read the executed notebook online, load it into Colab, or download a personal copy.

Companion notebook catalog

Application Examples by chapter

All notebooks have executed successfully and generated results in Google Colab. Viewing and notebook transfer require a verified portal account.

Chapter

56 notebooks

Contribute

Propose an improvement or extension.

Use the catalog to identify an AE, then submit a correction, implementation insight, dataset note, or proposed extension through the contribution workflow.

Contribution routes Notebook access remains personal to each verified account.

Engineering problems, mapped to Python.

Explore the complete notebook collection.

Application Examples by chapter

Predicting Tensile Strength of Steel

Estimating Parameters for Heat Loss in a Building

Evaluating a Regression Model for Battery Lifetime Prediction

Modeling Deflection of a Cantilever Beam

Predicting Automobile MPG with Regularized Regression

Predicting Mechanical Failure Status with Logistic Regression

Classifying Concrete Strength with Decision Trees

SVMs for Rotating Equipment Fault Classification

K-NN for Air Quality Monitoring

Bagging for Steel Plate Quality Control

Random Forest for Equipment Failure Prediction

Gradient Boosting for Fault Detection

Evaluating and Interpreting a Random Forest

Binary Classification of Manufactured Parts

Classifying Manufacturing Quality Using MLP

MLP for Predicting Concrete Strength

Predicting the Onset of Diabetes Based on Diagnostic Measures

CNN for Surface Defect Detection

LSTM for Predictive Maintenance Using Sensor Time Series

LSTM Comparison Using the AI4I 2020 Dataset

Clustering Hydraulic System Operating States with K-Means

Hierarchical Clustering for Anomaly Detection in Oil and Gas Chemical Plants

DBSCAN for Fault Detection in a Power Grid Sensor Network

PCA for Vibration-Based Fault Diagnosis

t-SNE/UMAP for Condition Monitoring of Industrial Bearings

Autoencoder for Structural Health Monitoring with Sensor Arrays

GMM for Industrial Process State Segmentation

EM for Fault Detection with GMMs

KDE for Sensor Drift Detection in Environmental Monitoring

Q-Learning for Adaptive Cruise Control

SARSA for Adaptive Traffic Light Control

REINFORCE for CartPole Control

Actor-Critic RL for Temperature Control in a Chemical Reactor

Deep Q-Learning for Battery Management

DDPG for Robot Arm Control

PPO for Adaptive Cruise Control under Stochastic Traffic

VAE for Sensor Anomaly Detection

GAN for Surface-Defect Augmentation

Diffusion Models for Microstructure Synthesis from Experimental Metallography

An LLM Copilot for Cantilever Beam Sizing

Physics-Constrained Regression for Discharge of a Capacitor

Physics-Constrained Classification for Fault Detection in a Mass-Spring System

PINN for Solving a 1D PDE

Fine-Tuning for Defect Detection in Manufacturing

Uncertainty-Aware Stress-Strain Modeling

Bayesian Linear Regression via MCMC

Uncertainty-Aware Prediction of Airfoil Self-Noise

Collaborative Filtering for Design Preference Learning

Mining Association Rules in Aircraft Engine Maintenance and Failure Logs

GCN for Predicting Stress Distributions on a Mesh

GAT for Critical Node Identification in Transportation

Masked Modeling for Power Grid Prediction

Self-Supervised Contrastive Learning for Bearing Degradation Representation

Symbolic Regression for Discovering Governing Equations of a Nonlinear Mechanical System

Hybrid Symbolic-Neural Modeling for Power System Load Prediction

An LLM-Based Agentic System for Early-Stage Engineering Design Support

Propose an improvement or extension.