Local AI Modeling for Industrial Automation Target Variables
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Industrial automation systems face challenges in efficiently monitoring and diagnosing complex processes due to the increasing volume and complexity of process data, which often requires real-time analysis and is hindered by the limitations of cloud-based systems in terms of cost, logistics, and security.
Innovation Solution
A local control system with an AI module that analyzes raw data without context, identifies influential subsets, models target variables, and adjusts operations autonomously, using clustering and directed graphs to optimize data analysis and communication within a secure, encrypted network.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Power
If cloud-based systems are used for real-time data analysis, then processing power and storage capacity are improved, but cost, logistics complexity, and security risks increase
Solution Approach 1:
The system segments the data analysis functionality by deploying distributed edge computing nodes throughout the industrial automation system. Each edge node processes local data independently, dividing the overall processing task across multiple locations rather than relying on a centralized cloud system, thereby reducing logistics complexity while maintaining processing capability
Solution Approach 2:
Edge computing nodes serve as intermediaries between industrial devices and cloud systems. These intermediaries perform preliminary data processing and filtering locally, reducing the amount of data that needs to be transmitted to the cloud and simplifying the logistics of cloud-based processing while maintaining access to advanced analytics when needed
2Power
If cloud-based systems are used for real-time data analysis, then processing power and storage capacity are improved, but security risks increase
Solution Approach 1:
By segmenting data processing across distributed edge nodes, the system reduces the security surface exposed to cloud-based threats. Each edge node processes data locally, creating isolated processing environments that limit the potential impact of security breaches while maintaining collective processing power across the distributed network
Solution Approach 2:
Edge computing nodes perform self-service data processing and analysis locally, reducing the need to transmit sensitive industrial data to external cloud systems. This self-sufficient approach maintains processing capability while minimizing security exposure by keeping data processing within the industrial network boundary
3Measurement precision
If all process data is analyzed in real-time, then monitoring accuracy is improved, but bandwidth consumption and processing overhead increase
Solution Approach 1:
The system extracts and processes only the most critical and relevant data elements at the edge computing nodes. By filtering out redundant or less important data before transmission or detailed analysis, the system maintains monitoring accuracy for key parameters while significantly reducing bandwidth consumption and processing overhead
Solution Approach 2:
The system applies partial action by performing selective real-time analysis on only the most critical data streams at edge nodes, while other data is processed asynchronously or transmitted at lower priorities. This approach maintains adequate monitoring accuracy for safety-critical functions while reducing overall bandwidth and processing requirements
Data Source
AI summary
A method for operating an industrial automation system may include receiving, via a first module of a plurality of modules in a control system, a plurality of datasets via at least a portion of the plurality of modules. The plurality datasets may include raw values without context regarding the plurality datasets. The method may then include identifying a subset of the plurality of datasets that influences a value of a target variable by analyzing the data without regard to the context, modeling a behavior of the target variable over time based on the subset of the plurality of datasets, and adjusting one or more operations of an automation device based on the model.