Plant Infrastructure Semantic Model Update Across Engineering Domains
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current infrastructure models for industrial plants are siloed by engineering discipline, leading to duplicated data, high engineering costs, and ineffective maintenance due to lack of consistent models across domains, with maintenance relying heavily on human expertise and obsolete documentation.
Innovation Solution
A method for updating a semantic model of a plant infrastructure that represents physical assets from different engineering domains by determining logical attributes and values, applying pattern-matching rules to create semantic assets, and linking them to identify common attributes and relationships, thereby creating a unified asset-centric view across domains.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If dedicated models and tools are used for each engineering domain, then domain-specific engineering precision is improved, but data duplication and engineering cost increase
Solution Approach 1:
The patent merges multiple domain-specific logical assets (first logical asset from updated application, second logical asset from outdated application) into a unified physical asset representation in the semantic model. This consolidation eliminates data duplication while preserving domain-specific details through the semantic asset abstraction layer.
Solution Approach 2:
The semantic model creates a universal physical asset representation that serves multiple engineering domains simultaneously. The pattern-matching rules enable the same physical asset to be represented consistently across different domains (process, electrical, mechanical) without requiring separate dedicated models for each domain.
2Manufacturing precision
If dedicated models and tools are used for each engineering domain, then domain-specific engineering precision is improved, but device complexity increases
Solution Approach 1:
The patent segments the complex multi-domain asset representation problem into manageable components: domain-specific logical assets, semantic assets with pattern-matching rules, and unified physical assets. This segmentation allows each domain to maintain its precision requirements while the overall system complexity is managed through modular processing steps.
Solution Approach 2:
The semantic asset acts as an intermediary layer between domain-specific logical assets and the unified physical asset representation. The pattern-matching rules serve as mediation mechanisms that translate between different domain representations, reducing overall system complexity by providing a standardized intermediate format.
3Adaptability or versatility
If human expertise and exchanges are used for cross-domain interactions, then adaptability is maintained, but loss of time increases
Solution Approach 1:
The semantic model with pattern-matching rules enables self-service automation of cross-domain asset matching and integration. The system automatically determines physical assets by matching semantic attributes across different domains without requiring manual human intervention, thereby reducing time loss while maintaining adaptability through the flexible pattern-matching mechanism.
4Adaptability or versatility
If obsolete documentation and human experience are used for maintenance, then adaptability is maintained, but productivity decreases
Solution Approach 1:
The system performs preliminary action by pre-establishing the semantic model with pattern-matching rules that encode domain knowledge and asset relationships. This preliminary structuring of maintenance information enables automated asset identification and cross-domain queries, significantly improving maintenance productivity while retaining adaptability through the flexible semantic framework.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The invention relates to an application server (AS) for updating a semantic model of a plant infrastructure comprising physical assets that belong to different engineering domains, wherein applications are dedicated to the engineering domains, wherein physical assets are represented respectively by first logical assets in updated applications and respectively by second logical assets in outdated applications, configured to: determine, for at least one updated application, first logical attributes and corresponding first logical values of a first logical asset from first metadata associated with the first logical asset, determine, for at least one outdated application, second logical attributes and corresponding second logical values of a second logical asset from second metadata associated with the second logical asset, determine a first semantic asset with first semantic attributes and corresponding first semantic values using pattern-matching rules related to the semantic model on the first logical attributes and corresponding first logical values, and determining a second semantic asset with second semantic attributes and corresponding second semantic values using pattern-matching rules related to the semantic model on the second logical attributes and corresponding second logical values, determine common attributes between the first semantic attributes and the second semantic attributes, determine a physical asset associated with the first semantic asset and the second semantic asset, if said common attributes have similar values based on semantic and/or similarity comparison, create an identifier for the determined physical asset associated with the common attributes and values for the common attributes in the semantic model, update the semantic model with the physical asset and with relationship between the physical asset, the first semantic asset and the second semantic asset.