"From Rooftop Solar Discovery to Design in Just Minutes"

Ham Ilhan, CEO of H Energy
Operating Mohat, SolarOnCare, SolarShare, and SolarShareBaro
Using the AI Agent 'Helios' for Rooftop Discovery
Automatically Recommending Optimal Layouts for Maximum Power Generation

"By utilizing the AI agent Helios, we can address the shortage of renewable energy supply in Korea."

Ham Ilhan, CEO of H Energy, is introducing the company at the "PR Day" held on the 16th at the office in Gangnam-gu, Seoul. H Energy

Ham Ilhan, CEO of H Energy, made this statement at the "PR Day" hosted by the Korea Venture Business Association on the 16th at the company’s office in Gangnam-gu, Seoul. H Energy is a platform company that covers the entire process of investment, production, consumption, and trading of renewable energy. Through its core platforms—Mohat, SolarOnCare, SolarShare, and SolarShareBaro—the company connects everything from sourcing rooftop solar assets nationwide to their operation.

According to H Energy, the available rooftop solar potential in Korea currently stands at around 52 GW. The company sees this as a volume that can be tapped into between 2030 and 2040. With the recent Middle East conflict highlighting the importance of energy security, H Energy is focusing its capabilities on identifying idle rooftops nationwide, estimating their generation capacity, and advancing the entire process from operation to management.

Against this backdrop, the recently developed AI agent "Helios" has made these tasks significantly easier. By entering a building’s address into Helios’s AI engine, Pathfinder, the AI analyzes satellite images to assess the building’s orientation, shading, and roof structure, then recommends the optimal array layout for maximum power generation.

CEO Ham explained, "Previously, electrical equipment companies had to visit sites in person to design rooftop solar installations, which required considerable labor and expense. By using Helios, design work that used to take two to three hours can now be completed in just a few minutes."

Panoramic view of the Mohat Power Plant by H Energy in Jincheon-gun, Chungbuk. H Energy

The permitting process after the design phase has also been greatly streamlined. The document generation engine "Synapse," which is equipped with a large language model (LLM), automatically creates customized documents that match the varying forms required by each local government. Previously, these tasks were done manually by professional consultants, but according to the company, they can now be processed on the same day.

Currently, H Energy operates 5,500 solar power plants nationwide, with a combined capacity of 700 MW, using its AI-based asset management platform, SolarOnCare. The platform analyzes real-time inverter voltage and current data to classify five types of fault patterns—normal, voltage drop, current drop, module aging, and partial shading—with an accuracy of 90.9%. Even for power plants without design drawings, it can diagnose abnormalities by reverse-estimating panel orientation and tilt based solely on actual power generation data.

H Energy forecasts sales of 162.7 billion won and operating profit of 14.6 billion won for this year. The company aims to achieve sales of 400 billion won by 2028. To reach this goal, it plans to increase the share of business-to-business (B2B) transactions to 70% and to raise the annual number of new SolarOnCare installations to around 8,000 sites per year.

CEO Ham stated, "If AI technology is applied to the design, diagnostics, and asset evaluation of solar power plants, we can resolve the longstanding structural issues in the renewable energy market. Through a new platform economy that enables anyone to own, trade, and share profits from energy assets, we will revolutionize the ownership and distribution structure of energy capital."

This content was produced with the assistance of AI translation services.