RedxGrid provides an AI‑driven drone inspection platform for electric utilities that captures georeferenced imagery and high‑frequency partial discharge measurements in a single flight. The data are combined into a unified evidence graph and processed with physics‑based machine‑learning models to generate explainable risk scores and audit‑ready work orders, integrating via standard APIs with existing GIS and asset‑management systems.
Funding
Funding not disclosed
Founders
Product
Problem
Electric utilities often discover partial discharge (PD) issues only after they cause transformer or cable failures, because traditional inspections are manual, fragmented across separate tools, and lack predictive insight. This leads to costly unplanned outages and regulatory compliance challenges.
Solution
RedxGrid delivers an AI‑driven drone inspection platform that captures high‑resolution imagery and PD measurements, then applies physics‑domain machine‑learning models to identify early‑stage discharge signatures. All raw evidence, asset history, and analyst annotations are ingested into a single evidence graph, preserving traceability throughout the workflow. The system generates explainable risk scores with confidence intervals, enabling operators to prioritize interventions confidently. Integrated reporting automatically produces audit‑ready work orders and compliance documentation, streamlining hand‑off to maintenance crews. Live telemetry and mission‑control tools synchronize field crews, reducing inspection turnaround time. The platform is built to interoperate with existing utility GIS and asset‑management systems via standard APIs.
Target Audience
The primary customers are transmission and distribution utilities, grid operators, and asset‑management teams responsible for transformer and cable reliability, as well as compliance groups that require defensible inspection records.
Features
- AI‑powered drone survey integration that captures georeferenced imagery and high‑frequency PD traces in a single flight pass
- Physics‑domain deep‑learning models for anomaly detection and early‑failure risk scoring on transformer and cable assets
- Unified evidence chain linking drone data, PD measurements, asset records, and analyst notes for end‑to‑end traceability
- Explainable alerts with confidence scores and contextual metadata displayed in a web dashboard for operator validation
- Automated generation of audit‑ready work orders and regulatory‑compliant reports, exportable via REST/FHIR APIs
- DJI‑based mission control with live telemetry, real‑time video streaming, and remote flight parameter adjustments
- Centralized Asset Management Center providing versioned asset datasets, quality checkpoints, and role‑based access controls
- Professional services toolkit for pilot setup, data onboarding, and integration with legacy SCADA/GIS platforms