The Clean Energy Graveyard
summary
The project claims 3,800 cancelled projects in the UK since 2010 and ships an open-source python pipeline & interactive web visualisation that ingests the REPD, enriches council documents with the GeoPandas, Pandas, Gemini API stack plus SQLite, FastAPI, Vite/Svelte and MapLibre, and surfaces locations, citations and probable causes. The demo works and exposes two big findings: Gemini speeds enrichment but hallucinates and changes over time, and NIMBY correlation is weak while a backlog of grid connections appears to explain many delays.
AI-written summary · about
chapters (19)
- 0:00Clean Energy Graveyard project intro
- 1:24Investigating renewable energy developments
- 4:58NIMBYism in renewable projects
- 6:21Government renewable datasets
- 8:05Live demo: Clean Energy Graveyard map
- 9:32Tech stack and AI tools
- 11:16First impressions with Gemini AI
- 13:29AI’s flawed analysis examples
- 14:32AI integration lessons
- 16:17Model Context Protocol workflow
- 17:42Lightning lessons from LLMs
- 19:07Project results overview
- 21:56Renewable project delays analysis
- 23:23Forth site wind farm case study
- 24:50NIMBYism impact on projects
- 26:38Tracker limitations and improvements
- 27:40Open-sourcing data and collaboration
- 28:44Next steps and goals
- 29:25Summary and closing
related talks
22mGovernment dashboards and open data.
23mDataEngBytes 2023 - BRI-T1-11 - Lizzie MacNeill
02h 59mExploring Eco topics with Python
35mVisualizing research data: Challenges of combining different datasources
31m"How we used Python to try and save lives"
15mGeoBI: A data visualisation platform using D3, leaflet, Geoserver, PostGres and php
26mThe Data I Breathe
22mGovernment dashboards and open data.
