It is moderately concentrated: the five leading assignees hold 46.5% of the 142 records in scope, and the ten leading hold 62.7%. That leaves roughly 37% of records spread across the remainder of a 61-company ranking, with many holding only a handful of filings each. A new entrant should expect to compete against a small cluster at the top rather than a single dominant holder, and there is meaningful room below that cluster.

Filings rose steadily to a peak of 37 in 2022, then eased to 16 by 2024 — a -20% shift over that three-year span. Because publication typically lags actual filing by around 18 months, the 2025 and 2026 figures in any dataset are still incomplete and should not be read as a continued decline. The honest read is that the field passed an early peak around 2022 and the post-2024 trajectory is not yet clear from published data.

H01M (batteries, cells and fuel cells) is the core class, appearing on 42.3% of the 142 records, but it is closely trailed by G01R (electric and magnetic measurement) at 37.3% and B60L (EV propulsion) at 29.6%. H02J (power supply and grid systems) sits close behind at 28.9%. This spread shows that claims in this field are as much about diagnostics, integration and vehicle systems as about the battery cell itself, so a filing strategy limited to cell-level hardware misses most of the contested space.

US9183327B2, assigned to NEC Corporation and granted in 2015, covers a method for scheduling charge and discharge of a second-life battery using Monte Carlo-modelled uncertainty in load, renewable generation and CO2 emissions rate, with an initial state of charge set from a Gaussian distribution. It is narrow to that specific probabilistic scheduling approach rather than to second-life battery use broadly. Anyone building emissions-aware dispatch logic for repurposed batteries should review its claims closely, but deterministic or rule-based scheduling approaches, or scheduling tied to different optimization objectives, sit outside its specific method claims.

The technology composition data shows lighter filing in AI-based diagnostics (G06N at 3.5% of records) and control/regulating systems (G05B at 4.9%) compared with the dominant H01M, G01R and B60L classes. Sub-areas such as automated cell-sorting logic, cross-chemistry reconfiguration, and grid-integration control layered on top of repurposed packs look under-claimed relative to the core hardware and measurement classes. These are reasonable areas to prioritize prior-art search before drafting new claims, though thinner filing density does not by itself guarantee patentability.