Local AI Model Retraining for Industrial Process Diagnostics
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 data, leading to inefficiencies in real-time performance monitoring and diagnostics, and there is a need for localized solutions that do not rely on cloud computing resources or human experts.
Innovation Solution
A local control system with an AI module that can analyze data in real-time, identify errors, retrain models, and perform diagnostics without transmitting data outside the local area network, using encryption for security and auto-monitoring capabilities to optimize communication bandwidth and data retention.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Power
If data is transmitted to cloud computing resources for analysis, then processing power and computational resources are improved, but data security and bandwidth usage are worsened
Solution Approach 1:
The system segments the centralized cloud-based analysis architecture into distributed local analysis units deployed at edge devices. Each local AI module independently processes data locally, eliminating the need to transmit sensitive data to centralized cloud resources while maintaining processing capabilities through distributed computation.
Solution Approach 2:
Local AI modules serve as intermediaries between data collection devices and cloud resources. These intermediaries process and analyze data locally, filtering only essential information for cloud transmission, thereby reducing bandwidth usage and minimizing data security risks while preserving access to cloud-based processing power when needed.
2Measurement precision
If more process data is collected and analyzed, then monitoring accuracy and diagnostic capability are improved, but computational complexity and data processing requirements are worsened
Solution Approach 1:
The system implements local quality by deploying specialized AI modules at distributed edge locations, each optimized for specific analysis tasks. This allows complex computational work to be performed locally where data is generated, improving monitoring accuracy without centralizing computational complexity in a single system.
Solution Approach 2:
Local AI modules perform self-service by autonomously processing and analyzing data at the source without requiring centralized computational resources. Each module independently executes monitoring and diagnostic functions, reducing overall system computational complexity while maintaining high measurement precision through localized analysis.
3Loss of time
If real-time analysis is performed, then response time and operational efficiency are improved, but computational resource consumption and energy usage are worsened
Solution Approach 1:
The system performs preliminary action by pre-training AI models offline and deploying them to local edge devices. This allows real-time inference to be performed with minimal computational resource consumption, as the heavy training computations are completed in advance, enabling fast response times without excessive energy usage during operational analysis.
Data Source
AI summary
A method for operating an industrial automation system may involve receiving, via a first module of a plurality of modules in a control system, an indication that an error between a measurement associated with a target variable that corresponds with at least a portion of the industrial automation system and a modeled value for the target variable. The method may then involve determining, via the first module, whether the error is within a first range of values and retraining a model used to generate the modeled value for the target variable based on a portion of a plurality of sets of data points acquired via a plurality of sensors disposed in the industrial automation system in response to the error being within the first range of values.