Embedded Field Diagnostics for Low-Bandwidth Automation Devices
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current automation systems face challenges in interpreting and utilizing raw sensor data due to complex structures and limited communication bandwidth, leading to a high manual engineering effort for diagnostic systems, especially in identifying severe alarms amidst numerous minor issues.
Innovation Solution
A smart embedded control system with a diagnostic application interface, signal evaluation component, complex event processing component, and diagnostic reasoning component that analyzes signal data, identifies events, and predicts impacts, enabling improved diagnostic functionality on a field device without requiring external analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data is transferred to remote systems for analysis, then diagnostic capability is improved, but communication bandwidth requirements increase
Solution Approach 1:
The patent extracts and executes diagnostic reasoning functions directly within the field device controller, removing the need to transfer complete raw data sets to remote systems. Only essential processed diagnostic information is communicated, significantly reducing bandwidth requirements while maintaining diagnostic capability.
Solution Approach 2:
The diagnostic system is segmented into distributed components: local signal evaluation, event identification, and diagnostic reasoning are performed at the field device, while remote systems receive only consolidated diagnostic results. This segmentation enables autonomous local processing that reduces communication demands.
2Measurement precision
If complete raw data is transferred for analysis, then diagnostic accuracy is improved, but manual engineering effort increases
Solution Approach 1:
The field device controller performs self-diagnosis by autonomously evaluating signals, identifying events, and reasoning about fault causes using embedded diagnostic reasoning components. This self-service capability eliminates the need for extensive manual engineering of external diagnostic systems while maintaining high diagnostic accuracy through local contextual understanding.
Solution Approach 2:
The system performs preliminary signal evaluation and event identification at the source before data leaves the field device. This preliminary processing transforms raw data into meaningful diagnostic information locally, reducing the need for complex external analysis infrastructure and manual engineering effort.
3Device complexity
If diagnostic reasoning is performed externally, then system complexity is reduced, but reaction time to faults increases
Solution Approach 1:
Diagnostic reasoning is segmented and distributed to the field device controller, enabling local real-time analysis of faults. This segmentation allows immediate detection and reasoning about events at their source, dramatically reducing reaction time while the modular architecture manages system complexity through distributed intelligence.
Solution Approach 2:
The controller performs preliminary diagnostic reasoning locally as events occur, rather than waiting for external analysis. This preliminary action enables immediate identification of root causes and impacts, reducing fault reaction time while the standardized diagnostic interface manages complexity.
Data Source
Figure 1
Figure 2~3
AI summary
An embedded control system for a field device of an automation system, comprising: a diagnostic application interface to a backend server for signal analytics information, complex event pattern information and diagnostic information; a physical process interface to a signal source for transferring signal data; a signal evaluation component for comparing received signal analytics information with received signal data to identify at a first and a second event; an event processing component for applying received event pattern information to at to the first and second identified events to identify at a first classified event; a diagnostic reasoning component for deriving causal dependencies between the first classified event and a further classified event with regard to diagnostic information to identify a root cause for the first classified event, or predict an impact of the first classified event. The diagnostic reasoning is pushed down to the field level with embedded data analytics functionality.