Skip to article frontmatterSkip to article content
Site not loading correctly?

This may be due to an incorrect BASE_URL configuration. See the MyST Documentation for reference.

Every page of the notes on one page, in reading order. The Book map explains why the book is organized this way; this page lists what is in each section. Needs says what to read if you jump here directly. Entrances by background and role: Start here.

Maturity follows the labels on each page: Stub (outline and goals only), Draft (readable prose, still growing), In review, Stable. Details: Status and versions. Want to write, review, or extend a chapter? See Join and contribute.

Front matter

PageWhat it covers
Start hereChoose an entrance by background and by role
How to use this bookWhat these notes are, how pages are laid out, how to get help
Advice to readersHow to read a page, work the labs, and avoid common traps
Preview this book locallyBuild and serve the notes on your own machine
Running notebooksLocal (canonical), Binder (primary online), Colab (secondary)
Your first labLab 1 in the browser, with numbered steps and no install
Book mapBig-picture layout and what each application produces
Full outlineThis page
Learning pathsRoutes by background and by role, plus course overlays
Related open resourcesNeighbouring courses, notes, and research codes
Join and contributeHow to report, write, review, or join as a section editor
Status and versionsPublic preview v0.1 (until v1.0), versioning, maturity labels
FeedbackTypos, broken links, unclear pages
LicenseCC BY-NC-SA 4.0 content, MIT code, contribution sign-off

Preparation

PageWhat it coversMaturity
Preparation overviewSelf-check before Foundations; optional maths refreshersDraft
Programming from zeroNotebooks, cells, and a kernel, for readers new to programmingDraft
Power systems for ML peopleGrid, swing equation, faults, and CCT in machine-learning languageDraft
Machine learning for power-systems peopleLoss, training, and generalization in power-system languageDraft
Prep 01: NumPy basicsArray workflowsStub
Prep 02: ODE integrationNumerical integration of small ODEsStub
Prep 03: Tiny neural regressionA minimal neural network fitStub
Prep 04: GradientsGradients and automatic differentiation by handStub

Foundations

ChapterWhat it coversMaturityLabNeeds
Foundations overviewShared SMIB spine used throughout the bookDraftEither bridge primer
SciML method familiesMap of SciML families used for power-system dynamics and stabilityDraftGlossary
Problem framingForward dynamics, stability assessment, and inverse estimation as distinct tasksDraftLab 1Glossary: state, parameter
SMIB dynamicsClassical swing equations in consistent unitsDraftLab 1PS primer if the machine is new
Reference simulationA transparent numerical SMIB reference with an event scheduleDraftLab 1SMIB dynamics; first lab
Data and surrogatesInput/output contracts for supervised trajectory surrogatesDraftLab 2Problem framing; ML primer if supervised learning is new
Automatic differentiationFinite differences versus reverse-mode AD for residual lossesDraftML primer or Prep 04
Forward PINNsData and physics losses for a known ODEDraftLab 2Data and surrogates; automatic differentiation
Classical estimation baselinesSimulation-based least squares for named, bounded parametersDraftLab 3Reference simulation
IdentifiabilityNoise, poor excitation, and structural non-identifiabilityDraftLab 3Classical estimation
Inverse PINNsInverse PINNs with trainable named parametersDraftLab 3Forward PINNs; classical estimation
Neural ODEs and hybrid modelsCollocation PINNs versus solver-based learned dynamicsDraftLab 4Forward PINNs

Dynamics (application 1)

ChapterWhat it coversMaturityLabNeeds
Dynamics pillar overviewHow will the system evolve under inputs, parameters, and disturbances? Planned chaptersStubLab 4SMIB dynamics; Lab 1

Stability (application 2)

ChapterWhat it coversMaturityLabNeeds
Stability pillar overviewWill the system remain stable, and how close is it to the boundary?DraftLab 5Lab 1; SMIB dynamics
Stability labels and CCTRotor-angle security, labels, and critical clearing time by bisectionDraftLab 5Stability overview
Surrogate models for stability tasksBinary labels, margin estimates, and trajectory surrogates as different productsDraftLab 5Data and surrogates; CCT
ANDES as network referenceMoving from SMIB to a network simulator after the SMIB workDraftLab 5Lab 5 SMIB path
Direct methodsEqual-area and energy ideas as an independent CCT checkDraftLab 5CCT; SMIB dynamics

Identification (application 3)

ChapterWhat it coversMaturityLabNeeds
Identification pillar overviewWhich parameters and dynamics explain the measurements?DraftLab 3Classical estimation
Measurements and excitationObservation channels, noise models, and window lengthsDraftLab 3Identifiability
Physical vs equivalent parametersPhysical versus equivalent (effective) fitted parametersDraftLab 3Identifiability
Differentiable simulationGradients with respect to (H,D)(H, D) through a numerical solverDraftLab 3Automatic differentiation; classical estimation

Advanced topics

ChapterWhat it coversMaturityLabNeeds
Advanced topics overviewOptional extensions (operators, uncertainty, converters)StubThree pillar overviews
From assessment to decisionHow the three applications connect in one operating-point loopDraftThree pillar overviews

Laboratories

LabWhat you deliverMaturityNeeds
Laboratories indexHow the five assessed labs fit togetherDraftStart here
Lab 1: Reference simulation and scientific datasetSMIB simulator, event schedule, dataset manifest, convergence checkDraftFirst lab
Lab 2: Supervised surrogate versus physics residual (PINN)Matched comparison with a physics-term ablationStubData and surrogates; Lab 1
Lab 3: Inertia and damping estimationLeast squares versus inverse PINN on the same measurementsDraftLab 1; classical estimation
Lab 4: Generalization of a learned dynamic modelHybrid/neural ODE or DeepONet tested on a held-out domainStubNeural ODEs; Labs 1–3
Lab 5: Stability and CCTReference CCT brackets, then a learned model near the boundaryDraftLab 1; CCT

Reference

PageWhat it covers
Reading listCore textbooks, simulators, and how per-chapter references work
GlossaryPlain meaning plus power-systems and ML readings of the same word
NotationSymbols and unit conventions
Front matter
Book map
Front matter
Learning paths