SOFC Prognostics AI/ML Patent Landscape 2026
The SOFC prognostics AI/ML patent field is in a growth phase, with filings accelerating sharply in recent years and Chinese academic institutions commanding the leading positions. The field remains highly concentrated, with a small number of universities and a handful of industrial players holding the bulk of activity.
Chinese universities lead a concentrated but rapidly expanding field
Nanchang University and Huazhong University of Science & Technology share the top rank, each filing 6 patent families — together accounting for a substantial portion of the top-ranked filers’ combined output. Robert Bosch GmbH is the most prominent industrial entrant, ranked third.
The top five filers account for 41% of the combined output of the hundred largest filers, signaling a highly concentrated landscape where two Chinese universities set the pace and industrial challengers remain at much lower filing volumes.
| # | Applicant | Patent families | Share |
|---|---|---|---|
| 1 | Nanchang University | 6 | |
| 2 | HUAZHONG UNIV OF SCI & TECH | 6 | |
| 3 | Robert Bosch GmbH | 2 | |
| 4 | Zhejiang Guoqing Energy Technology Development Co., Ltd. | 2 | |
| 5 | NANJING UNIV OF AERONAUTICS & ASTRONAUTICS | 2 | |
| 6 | Tianfu Yongxing Laboratory | 2 | |
| 7 | Saudi Arabian Oil Company (Saudi Aramco) | 2 | |
| 8 | EZHOU INST OF IND TECH HUAZHONG UNIV OF SCI & TECH | 2 | |
| 9 | Wuhan Huaxia Intelligent Technology Co., Ltd. | 2 | |
| 10 | Kongsberg Maritime AS | 2 |
| # | Applicant | Patent families | Share |
|---|---|---|---|
| 11 | Ningbo Institute of Northwestern Polytechnical University | 2 | |
| 12 | GUANGDONG ENERGY GROUP SCIENCE & TECHNOLOGY RESEAR… | 1 | |
| 13 | Xi’an Jiaotong University | 1 | |
| 14 | Jimei University | 1 | |
| 15 | Yunnan Power Grid Co., Ltd. Electric Power Research Institute | 1 | |
| 16 | Tsinghua University | 1 | |
| 17 | Tianjin University of Technology | 1 | |
| 18 | Northwestern Polytechnical University | 1 | |
| 19 | UNIV OF ELECTRONICS SCI & TECH OF CHINA | 1 | |
| 20 | North China Electric Power University | 1 |
The leaders’ positions imply that diagnostic and prognostic methodology for SOFC stacks is largely being defined in academic settings, leaving industrial players considerable room to build proprietary IP around deployment-ready implementations.
The most recent 18–24 months of data are subject to publication lag and likely undercount actual filing activity; the 2025–2026 figures should be treated as provisional. Longer-window growth, applicant concentration, and technology-route coverage are therefore more reliable signals than the latest-year bar alone.
Filings are accelerating with AI methods dominating the technology mix
Annual filing volume and the IPC composition together reveal both the pace of investment and the technological emphases shaping the field. The trend and branch charts should be read together to understand where activity is concentrated and where gaps remain.
Annual filing trend
Filings were sparse from 2017 to 2021, then stepped up in 2022–2023, and reached a visible peak in 2024. The 2025 and 2026 figures are understated due to publication lag and should not be read as a slowdown.
↗ Hover for values · click a bar to ask EurekaTechnology composition
AI computing methods (G06N) and fuel-cell hardware (H01M) are the two dominant branches, closely followed by digital data processing (G06F), reflecting an applied ML approach to SOFC health monitoring. Measurement (G01R) forms a secondary cluster, while branches such as separation processes (B01D) and data recognition (G06K) are sparsely occupied.
↗ Hover for values · click a bar to ask EurekaHighly cited patent families surfaced by the query
Citation-heavy patent families returned by the query. Use this section as citation context, not as a curated list of the most topic-specific patents.
Intermediate temperature solid oxide fuel cell cat…
An intermediate temperature solid oxide fuel cell (IT-SOFC) includes an anode layer, an electrolyte adjacent to the anode layer, and a cathode layer adjacent to the electrolyte and including a material of formula (I) or (II): Sr<sub>2</sub>OsO<sub>4 </sub>(I) or Ba<sub>2</sub>MO<sub>4 </sub>(II), where M is a transition metal or post-transition metal. (excerpt from the patent abstract)
| # | Patent | Citations |
|---|---|---|
| 1 | 一种固体氧化物燃料电池电堆故障诊断方法和系统 | 33 |
| 2 | 固体氧化物燃料电池电压预测方法、终端设备及存储介质 | 14 |
| 3 | 诊断固体氧化物燃料电池系统故障的方法及设备 | 11 |
| 4 | 基于GA-BP的SOFC电堆性能衰减预测方法及系统 | 7 |
| 5 | 一种固体氧化物燃料电池系统的工况辨识方法 | 7 |
| 6 | 带重整器的燃料电池系统的控制方法、装置及电子设备 | 7 |
| 7 | 基于SOFC系统的多故障诊断方法 | 6 |
| 8 | 一种SOFC系统多故障的诊断方法及系统 | 5 |
Ranked by total forward citations. Citation counts favour older and broadly cited patent families, and broad or adjacent patents may appear when they match the search scope. Treat this section as citation context, not as a curated list of the most topic-specific patents. Some patent titles may be shown in their original, non-English language where an accurate translation could not be guaranteed.
What the competitive structure means for R&D investment
Four structural signals — maturity stage, concentration, collaboration patterns, and geographic footprint — each carry distinct implications for where new entrants or incumbents should focus effort.
Growth stage with accelerating annual filings
The lifecycle evidence classifies this field as Growth, with annual filings still rising and the most recent years understated by publication lag. A 286% expansion in the recent window confirms that foundational IP positions are still being established, making early entry relatively more impactful than in a mature field. The window for staking broad claims in core prognostic architectures remains open.
Growth phaseTop five control 41% of leading filers’ output
The top five filers hold 41% of the combined output of the hundred largest filers, with two universities sharing the top rank at 6 patent families each. Industrial players such as Robert Bosch GmbH sit at the tier below with 2 patent families, indicating that corporate IP programs in this space are nascent. A sustained industrial filing program could rapidly close the gap given the low absolute volumes involved.
High concentrationOne active co-filing pair identified: Huazhong UST and its Ezhou institute
The only documented co-applicant relationship in the corpus is between Huazhong University of Science & Technology and the Ezhou Institute of Industrial Technology of Huazhong University of Science & Technology, with 2 jointly filed patent families. This university-to-affiliate structure is typical of Chinese technology transfer pipelines. Broader cross-institutional or industry-academia collaboration networks have not yet emerged at scale, representing a potential coordination advantage for early movers who pursue such partnerships.
Limited co-filingChina-dominant with limited international coverage
China accounts for 33 of the patent records in scope, while the United States shows 4 records, and Germany, Europe (EPO), India, and WIPO (PCT) each account for 1 record. The thin international filing footprint means that IP protection outside China is largely unclaimed, creating both a freedom-to-operate opportunity and a risk for non-Chinese players seeking global exclusivity.
China-centricGo beyond the landscape: Eureka’s TRIZ Solution agent breaks down an R&D problem and returns patented concept solutions, each with a technical approach and cited patent & literature evidence.
| Applicant | Collaborator | Co-filings |
|---|---|---|
| Huazhong University of Science & Technology | Ezhou Institute of Industrial Technology, Huazhong University of Science & Technology | 2 |
Co-filing pairs, ranked by the number of jointly-filed patent families.
Academic institutions lead; industrial entrants are early-stage
The two co-leading applicants are both Chinese universities with complementary technology emphases, while industrial players from. Europe and the Gulf are new entrants filing at lower volumes. All momentum-tracked applicants are classified as new entrants, confirming that the field’s IP community is still forming.
Nanchang University
Nanchang University leads with 6 patent families, classified as a new entrant with 5 recent filings, indicating that its position was built rapidly in the most recent filing window. Its technology emphasis is centered on AI computing methods (G06N 3) and digital data processing (G06F 18 and G06F 30), reflecting a predominantly algorithmic, model-driven approach to SOFC health monitoring rather than hardware-level innovation.
6 patent familiesRobert Bosch GmbH
Robert Bosch GmbH ranks third with 2 patent families and is the most prominent industrial filer in the corpus. Its technology focus spans fuel-cell hardware (H01M 4 and H01M 8) with an adjacent position in metal compounds (C01G 15), suggesting a materials-and-cell-level approach rather than a pure software prognostics strategy. As a new entrant in this sub-field, Bosch’s filing trajectory warrants monitoring for scale-up.
2 patent families| Applicant | Recent (3 yrs) | Trend |
|---|---|---|
| Huazhong University of Science & Technology | 3 | ▲ new entrant |
| Nanchang University | 5 | ▲ new entrant |
| Ningbo Institute of Northwestern Polytechnical University | 2 | ▲ new entrant |
| Saudi Arabian Oil Company (Saudi Aramco) | 2 | ▲ new entrant |
| Wuhan Huaxia Intelligent Technology Co., Ltd. | 2 | ▲ new entrant |
| Kongsberg Maritime AS | 2 | ▲ new entrant |
| Nanjing University of Aeronautics and Astronautics | 1 | ▲ new entrant |
Under-served branches adjacent to the SOFC prognostics core
Several IPC branches appear at low counts relative to the dominant AI and fuel-cell classes. These are observations of relative sparsity; technical value and entry path are assessed individually below.
B01D · Separation processes applied to SOFC diagnostics
With only 3 patent records, separation-process methods (B01D) are sparsely represented despite direct relevance to SOFC fuel reforming and gas-side degradation monitoring. Saudi Arabian Oil Co’s presence in this branch — filing 2 patent families combining B01D 53 with H01M 8 — suggests that gas-stream characterization during prognostics is a viable technical direction. An entrant with fuel-processing expertise could build differentiated IP here by linking AI-based anomaly detection to membrane or filtration performance signals.
Search this in Eureka →G06K · Data recognition and pattern presentation for SOFC health
The data recognition and presentation branch (G06K) holds only 2 patent records, representing a sparsely covered area at the intersection of sensor-data pattern recognition and SOFC condition monitoring. Given that the dominant AI branch (G06N) is well-occupied, applying structured data recognition pipelines — such as signal classification or visual feature extraction — to SOFC electrochemical impedance or voltage profiles could offer a differentiated, defensible position adjacent to the crowded neural-network space.
Search this in Eureka →How leading applicants differ by technology route
Route coverage across the main technology branches in the current evidence set.
| Player | G06N 3 · Computing based on AI models | H01M 8 · Batteries, cells & fuel cells | G06F 18 · Electric digital data processing | G01R 31 · Electric & magnetic measurement | G06F 30 · Electric digital data processing |
|---|---|---|---|---|---|
| Huazhong University of Science & Technology | Strong · 3 | Strong · 5 | Strong · 3 | Moderate · 2 | Absent |
| Nanchang University | Strong · 5 | Moderate · 2 | Moderate · 2 | Absent | Moderate · 2 |
| Tianfu Yongxing Laboratory | Strong · 2 | Absent | Strong · 2 | Strong · 2 | Absent |
| Ningbo Institute of Northwestern Polytechnical University | Absent | Strong · 2 | Strong · 2 | Strong · 2 | Absent |
| Wuhan Huaxia Intelligent Technology Co., Ltd. | Strong · 2 | Absent | Absent | Absent | Moderate · 1 |
| Yunnan Power Grid Co., Ltd. Electric Power Research Institute | Strong · 1 | Absent | Strong · 1 | Absent | Strong · 1 |
| Nanjing University of Aeronautics and Astronautics | Strong · 1 | Strong · 1 | Absent | Absent | Strong · 1 |
Frequently asked questions
The evidence covers 41 patent families in scope globally. This is a small but rapidly growing corpus, with the majority of filings concentrated in the 2022–2024 period.
Nanchang University and Huazhong University of Science & Technology are tied at the top, each with 6 patent families. Robert Bosch GmbH is the leading industrial filer at 2 patent families.
China is the dominant jurisdiction, accounting for 33 of the patent records in scope. The United States is a distant second with 4 records, and Germany, Europe (EPO), India, and WIPO (PCT) each account for 1 record.
AI computing methods (G06N) and fuel-cell hardware (H01M) are the two most represented IPC branches, followed closely by digital data processing (G06F). This reflects a dominant approach of applying neural network and ML models directly to SOFC stack health and fault detection.
The field is classified as Growth. Annual filings have risen substantially, with a 286% expansion in the recent window, and the 2024 filing count represents the highest annual volume observed. Recent-period figures are understated by publication lag.
The only documented co-filing relationship in the corpus is between Huazhong University of Science & Technology and the Ezhou Institute of Industrial Technology of Huazhong University of Science & Technology, with 2 jointly filed patent families. Broader cross-institutional or industry-academia networks have not yet emerged at scale.
Built on Patsnap Open Platform
This report’s underlying patent dataset — filings, assignees, technology clusters — is open for developers via MCP and REST API. Free to start, 10,000 credits, no credit card required.
Disclaimer. This page is generated from Patsnap Eureka data drawn from a limited snapshot of global patent and scientific-literature records, and is provided for general information and reference only.
Patent data carries inherent limitations: recent filings (typically the most recent 18–24 months) are under-counted due to standard publication lag; counts may be reported at either a patent-family or a patent-record basis and are not always directly comparable; classification, applicant-name, and citation data may contain errors, duplicates, or omissions; and the underlying search query defines and constrains the scope shown. As a result, the analysis may be incomplete or inaccurate and may not reflect the full technology landscape.
Nothing on this page constitutes an exhaustive prior-art, novelty, freedom-to-operate, or validity search, nor does it constitute legal, financial, investment, or professional advice, and it should not be relied upon as such. Any patent, commercial, or strategic decision should be verified independently and reviewed with qualified patent, legal, and domain professionals. Patsnap makes no warranties, express or implied, as to the accuracy, completeness, or fitness for any particular purpose of the information presented.