AI-powered photovoltaic GIS prospecting and generation forecasting platform.
Distributed photovoltaic business system with GIS mapping, roof detection, and generation forecasting.
This is an end-to-end open-source system designed for distributed photovoltaic businesses. It combines GIS map workspaces with AI vision and forecasting models to handle everything from initial roof prospecting to post-installation operations.
The platform integrates Leaflet and AMap with YOLO11 and OpenCV to automatically identify rooftops and solar panel potential from satellite imagery. It pairs this spatial data with LightGBM, XGBoost, and Transformer models to project short- and long-term energy generation and financial returns.
It suits distributed solar companies, project developers, and technical teams looking for a self-hosted platform to automate lead generation and site assessment. It does not suit teams lacking the infrastructure capacity to run a multi-service stack with databases and AI inference engines.
Integrates satellite and street map layers with POI recognition and manual point management.
Uses YOLO11 and OpenCV on remote sensing imagery to detect candidate rooftops and solar installations.
Calculates polygon areas on maps to estimate system capacity and component counts for initial quotes.
Applies LightGBM, XGBoost, and Transformer models to predict output based on weather and capacity.
Manages leads, site survey appointments, project construction milestones, and after-sales service tickets.
The source code is available in the repository, but no software license is stated in the repo metadata. The only cost is running and hosting the multi-service infrastructure yourself.
880 open remote AI-training roles, updated November 15, 2026; most pay $20–$175 an hour. Zekai lists them free; the application goes to the employer’s platform.
These reviewed AI tools do the same job instantly — from our scored directory.
It uses YOLO11 alongside OpenCV for computer vision tasks on remote sensing imagery.
It relies on LightGBM and XGBoost for short-term tabular modeling, with an optional Transformer architecture for long-term sequence forecasting.
The system architecture requires PostgreSQL with PostGIS, Redis, Elasticsearch, MinIO, and RabbitMQ.
See Zekai first in Google