Abstract

Current studies on secondary battery thermal runaway control often overlook the competitive dynamics and infringement risks that arise from portfolio overlaps. This study proposes an applicant-centered framework that integrates patent indices and text-mining-based similarity to map the competitive landscape. We analyzed 1,295 valid patents and utility models (2005–2025) from Korea, China, Japan, and the U.S. using the WIPS ON database. Key patent indices based on citations, patent family, and Technology Strength were utilized to quantify applicant influence. By vectorizing claims and abstracts using Term Frequency–Inverse Document Frequency (TF-IDF) and cosine-similarity networks, we found that while Tesla demonstrated a high citation impact with fewer filings, CATL, LG Energy Solution, Samsung SDI, and BYD formed a dense network core. Universities and Tesla occupied peripheral positions, indicating a niche research focus. Time-series analysis revealed a shift from structural protection (2005–2009) to pack-level management (2010–2019) and data-driven control (2020–2025). This framework supports strategic benchmarking, collaboration targeting, and white-space-based R&D planning.