Patent Portfolio

This page collects patent-oriented outputs related to General Research, AI/IT/OT integration, dynamic behavior modeling, workflow orchestration, and industrial software architecture.

This page only shares a small set of representative and publication-safe patent examples. It does not list the complete granted, pending, or internally filed patent portfolio, because full portfolio details, unpublished disclosures, filing strategy, and confidential technical records are not public information.

  • Scope: selected public-facing invention summaries, patent-family examples, and application-domain mapping.
  • Focus: executable industrial semantics, workflow generation, digital twin behavior models, and AI-assisted engineering.
  • Boundary: unpublished invention disclosures, non-public filing details, full claim strategy, and confidential technical attachments are intentionally excluded.
  • Status: entries are written as high-level technical descriptions for communication, not as complete legal patent records.
Siemens Inventor of the Year 2025 - Newcomer

I was honored by Siemens as Inventor of the Year 2025 in the Newcomer category for the Workflow Canvas invention family. The nomination materials describe Workflow Canvas as an IT/OT integration solution that connects natural-language intent with machine-executable workflows, enabling semantic interoperability across software, equipment, and digital-factory systems.

ItemProperty
RecognitionSiemens Inventor of the Year 2025 - Newcomer
InventorZhen Hua Zhou
FieldDigital Architectures - IT/OT Integration
Patent basisWorkflow Canvas invention family including WMB, SPIDR, and CREEM related PCT applications
  • Inventive output: 35+ invention disclosures over three years, with Zhen Hua serving as lead inventor for 30+ of them.
  • Patent family: 25 PCT applications based on Workflow Canvas innovations had been filed by the end of 2024.
  • Business signal: the nomination materials cite EUR 3.25 million FY24 order intake from 10+ clients and a further EUR 1.8 million sales pipeline across X/DI/SI/ADV.
  • Ecosystem signal: Workflow Canvas delivered 145 external developer training certifications, attracted 500+ lab visitors, and supported 800+ registered cloud users with 3300+ projects and 300k+ Function Blocks used by the end of 2024.
Workflow Canvas: Public Siemens Story

The public Siemens Inventors page describes Workflow Canvas as a translation bridge for digital factories: a way to convert human language into automated control commands and connect IT systems, OT assets, databases, robots, cloud services, and shop-floor equipment that traditionally speak incompatible languages.

  • Global indexability: every relevant machine, data element, or process object receives a unique identity so it can be found, linked, and reused across systems.
  • Non-intrusiveness: existing machines and software remain unchanged while Workflow Canvas docks onto them from the outside.
  • Multilingualism: the platform translates between IT and OT languages, helping legacy systems and modern AI-supported cloud solutions communicate through open standards.
  • Standardization ambition: the public article describes a path toward national standardization in China and potential international IEC standardization.
Robotic WASM Aggregated Model
ItemProperty
IDR/
FamilyWASM
PCT/

This invention describes a unified pipeline for generating, optimizing, and executing robotic behavior models in web-based and industrial environments. It aggregates URDF, CAD/manufacturer data, inverse-kinematics logic, rendering settings, and WebAssembly compatibility into a modular description that can be used by simulation tools, browser-based 3D editors, and industrial edge runtimes.

  • Technical problem: robotic modeling workflows are fragmented across URDF editing, IK generation, desktop simulation, vendor-specific tools, and inefficient web deployment.
  • Core contribution: a Robotic Behavior Model Aggregate Description that combines behavior mapping, joints, links, render settings, IK serialization, and WASM execution artifacts.
  • Engineering value: enables lightweight browser simulation, virtual commissioning, industrial training, and reusable robot behavior assets across NX MCD, Plant Simulation, Process Simulate, WebGL editors, and edge/cloud deployments.
  • AI/OT value: exposes robot behavior as a structured model that AI agents can inspect, generate, validate, and connect to workflow orchestration.

Industrial Logic Emitter for Digital Control
ItemProperty
IDR2023E11953 CN
FamilyIndustrial Logic / Digital Control
PCT

This invention proposes an Industrial Logic Emitter that fuses finite state machines, hierarchical state machines, behavior trees, and rule engines into a unified digital-control description and execution approach. It targets industrial workflows that originate from SOP, BOP, machine skills, controller instructions, and human work steps but are currently implemented through incompatible logical paradigms.

  • Technical problem: FSM/HFSM, behavior tree, and rule-engine systems are hard to combine, understand, modify, and execute in a standardized industrial environment.
  • Core contribution: a logic emitter with behavior descriptions, logic descriptions, a logic dispatcher, mutated-state mapping, and a common orchestration engine.
  • Interoperability value: maps heterogeneous state-machine standards such as PackML, ISA-88 batch states, and IEC 61499 state concepts into a compatible control structure.
  • Runtime value: enables logic of choice so different industrial logic models can be described and executed in one orchestrator while still controlling OT hardware, software, and human actions.

Unified Multi-Vendor AGV Orchestration and Map-Fusion Platform
ItemProperty
IDR
FamilyAGV / AMR Orchestration
PCT

This invention addresses heterogeneous AGV and AMR fleets whose vendor-specific robot control systems use incompatible maps, protocols, and dispatch mechanisms. The proposed middleware sits above native RCS layers, fuses navigation maps into a unified global map, and coordinates high-level missions without replacing each vendor's low-level controller or optimization algorithm.

  • Technical problem: warehouse and factory operators face islands of automation when Hikrobot, Geek+, Quicktron, or other AGV fleets cannot share maps, traffic state, or task plans.
  • Core contribution: a modular middleware architecture with vendor connectors, a global dispatcher, global scheduler/controller, map synchronization layer, and unified shop-floor model.
  • Map-fusion method: aligns sparse key points such as charging docks, intersections, gates, or pallet bays through rigid transform optimization, outlier filtering, and pairwise alignment into a master coordinate frame.
  • Operational value: coordinates trajectories through time-gap constraints, arbitration logic, live-map updates, and conflict resolution while preserving native vendor motion planning.
Federated Data Mesh and Data Fabric

The architecture combines domain ownership from Data Mesh with the shared semantics, governance, and resolution services of Data Fabric. Each industrial domain keeps its data products and ontology vocabulary. Shared services handle cross-domain queries, mapping, lineage, and access for AI tools.

There is no single central ontology. Each domain publishes a governed Domain Data Product and a Semantic Data Product. Versioned Mapping Products define the links between ontologies. Software can discover, query, and compose these mappings at runtime, so they do not become hidden spreadsheet crosswalks.

Layer Technical role
Domain Data Product Publishes source-domain data with ownership, quality boundaries, and ontology-aligned metadata.
Semantic Data Product Exposes ontology terms, entity graphs, contextual semantics, constraints, and lifecycle metadata.
Mapping Product Publishes ontology mappings as versioned products with confidence, provenance, and transformation rules.
Federated Governance Plane Registers schemas, semantic contracts, trust rules, lineage, quality policies, and deprecation rules.
Semantic Resolution Engine Discovers mappings, ranks candidates, composes transformations, normalizes units, and annotates provenance.
Unified Semantic Access Provides SPARQL/GraphQL/API/tool endpoints for applications, ETL pipelines, dashboards, and agentic AI tools.

For process plants, DEXPI can anchor P&ID semantics. The ingestion pipeline converts Proteus XML into an entity graph, resolves RDL class references, normalizes attributes, and binds each element to a CREEM Resource Description. It also maps valves, pumps, instruments, and other active elements to Function Blocks. The fabric can then compare both structure and operating behavior.

  • Reusable mapping products replace hand-maintained pairwise integrations.
  • Industrial agents query asset, behavior, document, and event knowledge through one semantic access layer.
  • Query results can include the mapping version, confidence, source lineage, and semantic-contract constraints.
  • Mappings are resolved on demand, which makes ontology version changes visible to downstream applications.
Work in Progress

Additional patent families and publication-safe records can be organized here after public clearance.