Offshore-Wind HVDC Collection-Grid Architecture Comparison
Compare offshore collection-grid architectures through component outages, electrical losses and the energy reaching the export connection.

Software compatibility
The analytical core uses only Python standard-library modules. Docker is optional. No commercial power-system software, wind-farm account or API key is required. Read the supplied documentation without installing software.
Project definition
Problem statement
Comparisons between AC and DC collection can change when turbine outages, shared equipment, export capacity and part-load conversion losses are counted differently.
Redundancy may recover energy while adding auxiliary demand. A larger delivered-energy total does not by itself establish an economical architecture.
Project objectives
- Separate the collection-grid boundary from the shared HVDC export connection.
- Track unavailable, isolated, curtailed, lost and delivered energy without double counting.
- Compare radial AC, radial DC, redundant DC and a series-group DC abstraction.
- Test voltage, auxiliary demand, common-condition dependence and dispatch assumptions.
- Check selected published arithmetic and a pinned open-source cable-count implementation.
Project structure
Project components
Literature review
Examines collection architectures, converter reliability, offshore maintenance and transmission-study boundaries.
State model
Enumerates hypothetical turbine, cable, hub and export availability with explicit radial and group isolation rules.
Energy ledger
Uses exact rational calculations to retain each energy category across loading and export-capacity cases.
Sensitivity studies
Investigates voltage, auxiliary power, component dependence and incremental investment ceilings.
Dispatch study
Bounds feasible uniform dispatch under a sufficient monotonicity condition, without claiming a global optimum.
Methodology
Project workflow
- 01Define the boundary
Read the hypothetical inputs and distinguish turbine collection from offshore transmission.
- 02Review the sources
Follow the annotations, access-depth records and original literature figure.
- 03Reproduce
Verify the retained outputs offline or inspect a selected scenario using the supplied command.
- 04Change an assumption
Use a working copy to test a justified change and trace its effect through the energy ledger.
- 05Present the comparison
Explain the conditional results, model limitations and a defensible extension using the editable documentation.
Demonstration scenario
At the hypothetical 30 MW export limit, the extra DC hub module recovers about 233 MWh per year. An additional constant 50 kW auxiliary load consumes 438 MWh per year, reversing the net gain. Explain the assumptions behind both quantities.
Engineering
Tools and method
- Tools
- The project uses Python 3.12, Matplotlib for subject analysis, simulation, and results.
- Analysis
- Standard-library Python models component states, conversion and cable losses with exact fractions.
- Verification
- 283 checks cover hand calculations, invariants, source provenance, scenario input handling and delivered artifacts.
- Documentation
- Includes introduction, literature review, theory, methodology, results, discussion, conclusions, further work and annotated references.
Testing
Evaluation
Evaluation measures
- Non-overlapping energy conservation across all architecture states
- Redundancy gains at three export-capacity limits
- Auxiliary-power thresholds that reverse a nominal energy gain
- Sensitivity to common operating conditions and voltage assumptions
- Conditional incremental investment ceilings
- Published table arithmetic and pinned implementation checks
- Byte-exact reproduction in an isolated Linux container
Project boundaries
- All local component coefficients and availability values are hypothetical.
- This is not a switching simulation, converter design, field reliability forecast or grid-code certificate.
- Fixed-voltage cable approximations omit charging, thermal limits and detailed controls.
- Uniform dispatch is not a claim of globally optimal power allocation.
- Investment ceilings are conditional calculations, not vendor quotations or financial advice.
- No information collected.
Included
- 0170-page project documentation in PDF and editable Word formats
- 02Nine-page usage guide in PDF and editable Word formats
- 0322-slide editable presentation with source notes, tables and charts
- 0412 annotated references, source matrix and an attributed literature image
- 05Seven numbered figures and eight numbered tables in the thesis
- 06Nine retained numerical outputs, 120 design cases and 24 annual ledgers
- 07Complete Python source code, 283 automated tests and offline reproduction
Project record
No information is collected on this page.
- Permanent project ID
- GP-EE-1UBL76N
- Catalogued
- 21 Aug 2026
- Completed
- 06 Sept 2026
- Verified
- 06 Sept 2026
- Demonstration
- Included in repository
Handover
After purchase
- 01Payment is confirmed
The project is marked unavailable and cannot be purchased again.
- 02Repository access is granted
The buyer's submitted GitHub account receives access to the private repository.
- 03The purchase record is delivered
The certification sheet is prepared from the reviewed buyer details and sent privately by email.