Signal Selection Agent for Industrial Automation Data Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Non-automation experts face challenges in extracting relevant data from distributed control systems due to the large size of the address space and the need for specialized knowledge, making manual identification time-consuming and often infeasible, especially without the involvement of automation experts.
Innovation Solution
A signal selection agent using a domain knowledge mapping model and intuitive user interface facilitates signal identification by mapping domain concepts to signals, enabling data extraction without requiring engineering expertise, through a guided human-machine interaction process and the construction of a knowledge graph from mined data sources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual identification of relevant signals by browsing the address space is performed, then data extraction can be achieved, but it is time-consuming and requires specialized automation knowledge that non-experts do not possess
Solution Approach 1:
The patent introduces an intermediary system (signal selection agent with domain knowledge mapping model) that mediates between the data scientist and the complex address space. This intermediary automatically maps domain concepts to relevant signals, eliminating the need for data scientists to manually browse and understand the address space structure, thus resolving both the time consumption and expertise requirement issues while maintaining accurate data extraction.
Solution Approach 2:
The system enables self-service by allowing data scientists to configure data extraction using their own domain knowledge without requiring automation experts. The domain knowledge mapping model automatically translates high-level domain concepts into specific signal selections, making the system self-configuring from the perspective of the end user and eliminating dependency on specialized personnel.
2Loss of information
If all accessible data is exported from the DCS, then complete data availability is achieved, but high load is placed on the DCS and most data is irrelevant for analysis
Solution Approach 1:
The patent extracts only the relevant subset of data from the complete address space by mapping domain concepts to specific signals. Instead of exporting all accessible data, the system intelligently selects and extracts only those signals that are relevant to the data scientist's domain concepts, thereby reducing DCS load and eliminating irrelevant data while maintaining data completeness for the intended analysis purpose.
Solution Approach 2:
The system applies partial action by selecting only the necessary portion of data (relevant signals) rather than performing excessive action (exporting all data). The domain knowledge mapping model determines the optimal subset of signals needed for specific analysis tasks, avoiding the waste of resources associated with exporting and processing unnecessary data while ensuring all relevant information is captured.
3Ease of operation
If the address space is made browsable via standardized interface, then data accessibility is improved, but the large size of the address space makes manual identification challenging and infeasible
Solution Approach 1:
The patent adds another dimension to data accessibility by introducing domain knowledge as an additional layer beyond the standardized interface. Instead of navigating the complex hierarchical address space structure, users can access data through domain concepts (e.g., 'pump status', 'temperature sensor') that map to multiple underlying signals, effectively transforming the navigation problem from structural exploration to conceptual querying and simplifying access despite the large address space size.
Data Source
AI summary
A system and method for providing a configuration for data extraction from an automation system includes a signal selection agent configured to: receive a user selection of at least one module of the automation system; generate, for display to the user, a user interface identifying one or more selectable signals associated with the selected module and displaying one or more guidance elements comprising data mined from data sources pertaining to the automation system for guiding the user in the selection of relevant signals; receive a user selection of one or more of the selectable signals; and automatically generate the configuration for data extraction on the basis of the selected signals.