AI Asset Grouping for Semantic Tagging in Industrial Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing automated and industrial control systems, such as BMS and SCADA, face challenges in managing complex environments with dynamic asset changes, requiring labor-intensive manual semantic tagging that hinders efficient energy management and operational optimization.
Innovation Solution
A computer tool and method for automated asset grouping in AIC systems using AI algorithms to identify and group similar assets based on embedded numerical values of textual attributes, simplifying the tagging process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual semantic tagging is performed to accurately identify and group assets in AIC systems, then asset management precision and system understanding are improved, but labor intensity and time consumption increase significantly
Solution Approach 1:
The patent replaces manual semantic tagging (mechanical human labor) with an automated AI-based system that uses machine learning models to analyze asset data, extract features, and perform grouping automatically. This substitution eliminates the need for manual intervention while maintaining or improving tagging accuracy through consistent algorithmic application.
Solution Approach 2:
The system enables assets to be automatically tagged and grouped through self-service mechanisms where the AI model autonomously processes asset data, identifies patterns, and creates groupings without external human assistance. The automated workflow includes data extraction, feature engineering, model inference, and result validation all performed by the system itself.
2Loss of information
If manual asset tagging is performed to understand system changes and monitor performance, then system monitoring capability is improved, but labor costs and operational complexity increase
Solution Approach 1:
The patent replaces manual system monitoring and change tracking with automated AI-based analysis that continuously processes asset data, detects changes, and monitors performance metrics. This automation captures system evolution over time without requiring human intervention, reducing operational complexity while improving information capture.
Solution Approach 2:
The system implements continuous feedback loops where AI models analyze asset data, identify changes, and provide insights about system evolution. This feedback mechanism automatically tracks how systems change over time and provides actionable information for performance optimization without manual analysis.
3Use of energy by moving object
If comprehensive asset tagging is implemented to enable advanced control strategies and energy management, then system efficiency and energy optimization are improved, but implementation cost and resource requirements increase
Solution Approach 1:
The patent applies partial action by implementing asset tagging and grouping selectively based on priority and impact. Rather than requiring complete tagging of all assets, the system focuses on critical assets first, enabling advanced control strategies for high-impact areas while reducing implementation resources. The automated AI approach makes even partial tagging more cost-effective than traditional manual methods.
Solution Approach 2:
The replacement of manual tagging with automated AI processing significantly reduces the resource investment required for comprehensive asset tagging. The automated system processes large volumes of asset data efficiently, making comprehensive tagging feasible with limited resources compared to manual approaches.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A computer method and system for grouping like points in an automated and industrial control system (AIC). A point is identified as a reference for grouping with other similar points in the AIC. Captured from the AIC are textual attributes associated with the identified point. The captured textual attributes are embedded into at least one numeral value. Captured from the computer database are textual attributes associated with other points in the AIC. The captured textual attributes for each other point are then embedded into at least one respective numerical value. The numerical value of the identified point is compared with each numerical value of the other points. Points from the other points are grouped with the identified point that are determined to have a similar embedded numerical value to that of the identified point. Based upon a labelling technique, a semantic tag is assigned to each of the grouped points.