Industrial Plant Data Semantics via Ontology Embedding Evaluation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Industrial plant data evaluation during engineering and operation is largely manual due to disparate and unconnected information streams across various modalities, lacking machine-understandable formats.
Innovation Solution
A computer-implemented method transforms data records of different modalities into a unified embedding space using a trained encoder, comparing them to semantic information models to evaluate their semantic meaning and relationships, enabling automatic data evaluation and improving modality integration.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If data from multiple modalities are kept separate in independent formats, then data integrity and modality-specific processing are maintained, but automatic evaluation and integration of plant data are hindered
Solution Approach 1:
The patent merges multiple independent data modalities (P&ID diagrams, I/O lists, control narratives, images, videos, sensor data) into a unified machine-understandable representation. This is achieved by transforming diverse data types into a common format that can be automatically processed and integrated, enabling comprehensive plant evaluation without manual intervention while maintaining the distinct characteristics of each data source.
Solution Approach 2:
The patent introduces an intermediary processing layer that acts as a mediator between raw multi-modal data and the evaluation system. This intermediary transforms and standardizes data from different modalities into a unified representation, facilitating automatic integration and evaluation while preserving the original data integrity and modality-specific properties.
2Productivity
If manual methods are used to evaluate plant data, then flexibility and adaptability to different data formats are maintained, but time consumption and labor requirements increase
Solution Approach 1:
The patent replaces manual evaluation processes with an automated computer-implemented system. The system uses algorithms and computational methods to process, integrate, and evaluate plant data across multiple modalities, eliminating the need for manual data collection, processing, and analysis while significantly improving evaluation efficiency and reducing time consumption.
Solution Approach 2:
The patent enables the evaluation system to automatically perform data processing, integration, and analysis without requiring manual intervention. The system self-manages the transformation of multi-modal data into unified representations and conducts comprehensive plant evaluations autonomously, freeing operators from time-consuming manual tasks while maintaining high productivity.
3Loss of information
If diverse data formats are used for different plant information, then specific data requirements and modalities are satisfied, but machine understandability and data integration are reduced
Solution Approach 1:
The patent transforms data from various modalities by changing their representation parameters into a unified machine-understandable format. This transformation process preserves the essential information and semantic meaning of each data type while converting them into a common structure that can be automatically processed and integrated, enabling both full information accessibility and machine understandability.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A computer-implemented method (100) for evaluating at least one record (2a, 2b) of data, and in particular measurement and/or plant specification data, that characterize at least one component of an industrial plant (1), and/or an operational state thereof, the method comprising the steps of: • transforming (110), by a trained encoder (3a, 3b), the at least one record (2a, 2b) of data into a representation (4a, 4b) in an embedding space (5); • comparing (120) this representation (4a, 4b) to representations (7) of semantic entities (6) from a semantic information model (8) in the same embedding space (5), wherein each semantic entity (6) carries a semantic meaning (6*) with respect to the structure, the construction, and/or the functioning, of the industrial plant (1), and the semantic information model (8) comprises relationships between the semantic entities (6); and • evaluating (130), from the result (120a) of the comparison, the semantic meaning (2a*, 2b*) of the record (2a, 2b) of data with respect to the structure, the construction, and/or the functioning, of the industrial plant (1).