Book a demo

Planetary Gearbox Fault-Diagnosis AI/ML Patent Landscape

Planetary Gearbox Fault-Diagnosis AI/ML Patent Landscape
Competitive Landscape
Planetary Gearbox Fault-Diagnosis AI/ML Patent Landscape in 2026

The field is in a Growth stage, with Chinese academic institutions collectively holding the dominant position and China accounting for the overwhelming share of filing activity. Activity has expanded significantly on a multi-year basis, though annual volume has eased from its 2021 peak and the most recent period remains understated by publication lag.

72
Patent families in scope
23%
Top-5 share of top-100 filers
+52%
3-yr filing growth (lag-adj.)
China
Leading jurisdiction
↗ Tap any metric to explore the underlying patents and uncover deeper insights in Patsnap Eureka
Published byPatsnap Insights Team··7 min readVerified by Patsnap Eureka data
Overview

Chinese universities lead a concentrated, academically dominated field

Tianjin University of Technology holds the top position with 7 patent records, followed jointly by Xi’an Jiaotong University and Hitachi Construction Machinery Co., Ltd. at 4 patent records each. The top five filers together account for 23% of the hundred largest filers’ combined total, indicating moderate concentration at the head of the ranking.

The leading tier is composed almost entirely of Chinese universities, with Hitachi Construction Machinery as the lone industrial entrant in the top three. This creates a notable tier gap between the academic front-runners and the sparse industrial participation, suggesting that commercialization pathways remain under-exploited.

Leading applicants
#ApplicantPatent recordsShare
1Tianjin University of Technology7
2Xi’an Jiaotong University4
3Hitachi Construction Machinery Co., Ltd.4
4NANJING UNIV OF AERONAUTICS & ASTRONAUTICS3
5BEIJING INFORMATION SCI & TECH UNIV3
6Shanghai University of Electric Power3
7Nanjing Tech University2
8Chongqing University OF POSTS & TELECOMM2
9Huaiyin Institute of Technology2
10Harbin Institute of Technology2
#ApplicantPatent recordsShare
11Fuzhou University2
12Nanjing Agricultural University2
13Dalian University of Technology2
14Jiangsu Guoke Intelligent Electrical Co., Ltd.1
15China Merchants Xinjiang Special Equipment Inspection Technology Research Institute Co., Ltd.1
16HUNAN UNIV OF SCI & TECH1
17Nanjing Gongda CNC Technology Co., Ltd.1
18Hulunbuir University1
19Xiamen University1
20Beijing Oriental Institute of Vibration and Noise Technology1
↗ Hover a row · click a company to ask Eureka

The academic dominance of the leader group implies that proprietary, application-specific implementations — particularly for industrial operators in wind energy, construction machinery, and aerospace — remain a realistic differentiation opportunity for commercial players willing to translate university-grade research into deployable systems.

The most recent 18–24 months of filings are likely under-counted due to standard patent publication lag; apparent softness in 20242026 data should not be read as a genuine slowdown. Longer-window growth, applicant concentration, and technology-route coverage are therefore more reliable signals than the latest-year bar alone.

Source: Patsnap Eureka. Chart shows the top applicants ranked by patent records; the corpus total is measured in patent families. These figures use different units and should not be compared directly. This same dataset is now available on Patsnap Open Platform via MCP.Connect via MCP →
Trends & Structure

Multi-year growth driven by neural-network signal-processing convergence

Annual filing volume and the IPC technology mix together reveal a field that expanded sharply after 2020 and is anchored by the intersection of mechanical testing standards and AI computing methods.

Annual filing trend

Filings were negligible before 2018 and surged to a peak in 2021, after which annual volume eased but remained elevated through 2023–2025. The three-year recent window sits 52% above the prior three-year window on a multi-year basis. Data for 2024–2026 are further suppressed by publication lag and should be treated as lower bounds.

Annual filing trendAnnual values from 2017 to 2026, peaking at 15 in 2025.020172201852019320201320211120221220239202415202522026↗ Hover for values · click a bar to ask Eureka

Technology composition

G01M (testing machine and structural balance) is the dominant branch, reflecting vibration-signal acquisition as the diagnostic entry point. G06N (AI computing models) and G06F (digital data processing) are nearly co-dominant, confirming that the core technical advance is in model architecture rather than sensor hardware. F16H (gearing and transmissions) accounts for a small share, marking the physical drivetrain as an adjacent but under-patented angle.

Technology compositionG01M · Testing machine & structure balance leads with 57; G06N · Computing based on AI models 42.G01M · Testing machine &…57G06N · Computing based o…42G06F · Electric digital …38G06K · Data recognition …15F16H · Gearing & transmi…5G06V · Image/video recog…3E02F · Excavating & eart…2F03D · Wind motors (wind…2↗ Hover for values · click a bar to ask Eureka
Source: Patsnap Eureka. Technology-branch counts are measured in patent records; a single patent family can carry several IPC classes, so class totals can exceed the family total in scope.Explore deeper in Eureka →
Key Patents

Highly 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.

Featured patent
AU2021105779A4Published 2021-10-21

Planetary Gearbox Fault Diagnosis Method Using Par…

Fuzhou University

The invention relates to a planetary gearbox fault diagnosis method. First of all, the signal is decomposed and reconstructed by using the Salp Swarm Optimization algorithm to optimize the Variational Mode Decomposition algorithm (SSO-VMD). Then, the fault features are extracted from multiple domains, and the improved supervised self-organizing incremental… (excerpt from the patent abstract)

Open this patent in Eureka →
Highly cited patent families surfaced by this query
#PatentCitations
1基于ACGAN的风电机组行星轮齿轮箱故障诊断方法34
2一种基于自适应共振稀疏分解理论的风电齿轮箱故障诊断方法21
3一种基于强化胶囊网络的行星齿轮箱故障诊断方法20
4Process and device for the supervision of the kine…19
5一种基于数据驱动的强噪声干扰下齿轮箱故障诊断方法16
6极小故障样本量下行星齿轮箱故障诊断方法15
7基于多域堆栈去噪自动编码网络的行星齿轮箱故障诊断方法15
8一种基于卷积胶囊网络的行星齿轮箱故障诊断方法10

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.

Source: Patsnap Eureka. Citation-ranked patent families surfaced by this query.Open in Eureka →
Insights

What the competitive structure means for R&D investment decisions

Four structural observations — maturity, concentration, collaboration, and geography — together define where new entrants face headwinds and where gaps remain actionable.

Growth

Growth stage: expanding multi-year base, eased annual peak

The lifecycle evidence places the field in Growth. The recent three-year window is 52% above the prior three-year window, confirming broad expansion. Annual volume peaked in 2021 and has since moderated, suggesting the field is past its initial surge but not yet mature — new differentiated approaches still enter the corpus regularly.

Growth stage
Concentration

Moderate top-tier concentration, thin industrial presence

The top five filers hold 23% of the hundred largest filers’ combined total — a moderate concentration level that leaves meaningful room for challengers. More strategically significant is the near-absence of industrial assignees: only Hitachi Construction Machinery appears in the top three, meaning most commercial operators have no established IP position from which to defend or license.

Moderate concentration
Collaboration

University–industry co-filing is emerging at a small scale

Three co-filing pairs are evidenced: Beijing Information Science & Technology University with Beijing Oriental Institute of Vibration and Noise Technology, Chongqing University of Posts and Telecommunications with Chongqing Industrial Big Data Innovation Center, and Dalian University of Technology with Dalian Boiler Pressure Vessel Inspection and Testing Research Institute. Each pair filed one joint patent record, indicating early-stage university-to-industry technology transfer rather than sustained programmatic collaboration.

Early co-filing
Geography

China-dominant filing with limited international protection

China accounts for 64 patent records — the overwhelming majority of activity — while the United States holds 3, Europe (EPO) holds 2, and Australia, India, Japan, and WIPO each hold 1. This concentration means that most innovations in this corpus carry minimal international protection outside China, creating a geographic gap that global industrial players could exploit by filing defensively or offensively in non-Chinese jurisdictions.

China-dominant
Patsnap Eureka · TRIZ Solution Agent
Facing a specific technical bottleneck in this field?

Go 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.

Solve it in Eureka →
Top collaboration links
ApplicantCollaboratorCo-filings
Beijing Information Science and Technology UniversityBeijing Oriental Institute of Vibration and Noise Technology1
Chongqing University of Posts and TelecommunicationsChongqing Industrial Big Data Innovation Center Co., Ltd.1
Dalian University of TechnologyDalian Boiler Pressure Vessel Inspection and Testing Research Institute Co., Ltd.1

Co-filing pairs, ranked by the number of jointly-filed patent families.

Source: Patsnap Eureka. Lifecycle stage and collaboration data are drawn from PatSnap Eureka analytics on the 72-family corpus.Explore insights →
Leaders

Top players differentiate on sensor-signal testing versus pure AI model routes

Tianjin University of Technology leads on volume; Hitachi Construction Machinery is the only top-ranked industrial filer and the sole player combining gearbox hardware and excavation-application claims.

Leader · Tianjin University of Technology

Tianjin University of Technology

Leads the corpus with 7 patent records. Technology emphasis is concentrated on G01M (mechanical testing and structure balance), G06F (digital data processing), and G06N (AI models), reflecting a broad signal-acquisition-to-classification pipeline. Momentum data classifies it as a new entrant in the recent window, indicating that its position was built rapidly within the last filing cycle rather than over a long accumulated base.

7 patent records
Challenger · Hitachi Construction Machinery Co., Ltd.

Hitachi Construction Machinery Co., Ltd.

Tied second with 4 patent records and is the only top-ranked non-academic entity. Its technology footprint uniquely spans G01M (testing), E02F (excavating and earth-moving), and F16H (gearing and transmissions), indicating application-specific fault-diagnosis development tied to its own construction equipment platforms. Momentum is classified as a new entrant, and its multi-class coverage makes it the most industrially differentiated filer in the ranking.

4 patent records
🔍
See the full applicant breakdown
Unlock rankings for all 100 filers, including momentum scores and technology-route overlaps.
Xi’an Jiaotong UniversityNanjing University of Aeronautics and Astronautics+ more
Unlock full assignee analysis →
Leading-applicant momentum (recent 3 yrs, lag-adjusted)
ApplicantRecent (3 yrs)Trend
Tianjin University of Technology3▲ new entrant
Xi’an Jiaotong University2▲ new entrant
Hitachi Construction Machinery Co., Ltd.4▲ new entrant
Beijing Information Science and Technology University1▲ new entrant
Shanghai University of Electric Power1▲ new entrant
Chongqing University of Posts and Telecommunications1▲ new entrant
Huaiyin Institute of Technology1▲ new entrant
Source: Patsnap Eureka. Patent record counts are from the applicant ranking within the 72-family corpus.Explore players →
Adjacent Branches

Under-served routes adjacent to the dominant AI-signal-processing core

Several IPC branches carry low patent-record counts relative to the corpus total, flagging areas where technical activity exists but dedicated IP coverage is sparse. These are observations of relative sparsity; entry viability depends on each applicant’s specific capability and market context.

F16H · Gearing & transmissions — mechanical design angle

Only 5 patent records are classified under F16H (gearing and transmissions), despite the planetary gearbox being the direct physical subject of the corpus. This sparsity suggests that most filers address the diagnostic algorithm layer while leaving the gearbox hardware, lubrication monitoring, and mechanical-failure-mode modeling comparatively unclaimed. For applicants with drivetrain engineering depth, combining F16H claims with G06N AI models could yield differentiated, harder-to-design-around patents. Hitachi Construction Machinery is the only top-ranked filer with F16H coverage, confirming that this angle is not yet contested.

Search this in Eureka →

F03D · Wind motors — turbine-specific deployment

Wind turbine gearboxes are a primary commercial application for planetary gear fault diagnosis, yet only 2 patent records are classified under F03D (wind motors). The most-cited patent in the corpus explicitly targets wind turbine planetary gearbox diagnosis using ACGAN, showing that the application space is technically active. The gap between citation prominence and F03D filing volume points to an under-claimed application-specific route. Organizations operating or supplying wind turbine drivetrains could build a targeted position by anchoring AI-diagnosis claims to turbine operating conditions, variable-speed loads, and field-deployment constraints.

Search this in Eureka →
🔒
Unlock the full white-space map
Access all five identified adjacent branches with applicant overlap analysis and entry-path scoring.
G06V · Image/video recognition applied to gear surface inspectionE02F · Construction machinery gearbox health monitoring+ more
Unlock full analysis →
Source: Patsnap Eureka. Branch counts are at the patent-record level; low counts indicate sparse but not necessarily zero coverage.Explore emerging →
Route Matrix

How leading applicants differ across technology and application routes

Route coverage across the main technology branches in the current evidence set.

PlayerG01M 13 · Testing machine & structure balanceG06N 3 · Computing based on AI modelsG06F 18 · Electric digital data processingG06K 9 · Data recognition & presentationG06F 17 · Electric digital data processing
Tianjin University of TechnologyStrong · 7Strong · 4Strong · 4Moderate · 2Emerging · 1
Xi’an Jiaotong UniversityStrong · 3Strong · 3Strong · 3AbsentAbsent
Shanghai University of Electric PowerStrong · 2Strong · 2Strong · 2AbsentAbsent
Fuzhou UniversityAbsentStrong · 2AbsentStrong · 2Moderate · 1
Hitachi Construction Machinery Co., Ltd.Strong · 4AbsentAbsentAbsentAbsent
Huaiyin Institute of TechnologyStrong · 2AbsentStrong · 2AbsentAbsent
Chongqing University of Posts and TelecommunicationsAbsentStrong · 2AbsentStrong · 2Absent
Source: Patsnap Eureka. Matrix values are measured in patent records and should not be compared directly with family-level applicant totals.Compare in Eureka →
Frequently asked questions

Frequently asked questions

Still have questions? Patsnap Eureka answers them from patent and research data.Ask Eureka →

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.

Explore in Eureka ↗
Powered by Patsnap Eureka

Help us improve this page

Found incorrect or outdated information? Let us know and we'll get it fixed.