Transmission Tower Inspection Drone Hub Deployment Optimization System Based on DeepSeek Large Language Model

Ensuring the safety of high-voltage transmission towers is critical for urban power grid resilience, particularly with increasing urbanization and extreme weather. Traditional manual inspections are inefficient, costly, and hazardous for large-scale networks. This study proposes an intelligent decision support system (DSS) driven by a Large Language Model (LLM) to optimize drone hub deployment for transmission tower inspection. By integrating the DeepSeek LLM with the HiSpot optimization toolkit, our system formulates this task as a p-Hub median problem and automates the end-to-end solution process. The system enables users to describe objectives in natural language, then autonomously identifies the problem, extracts parameters, and invokes solution algorithms. A case study in Zengcheng District, Guangzhou, utilizing OpenStreetMap data for thousands of towers, validates the system’s effectiveness on large-scale p-Hub problems. Results show the system lowers technical barriers for geospatial optimization, offers a novel approach for intelligent infrastructure management, and significantly enhances urban power system resilience.