Digital Twin Deviation Detection in Automated Systems

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Deviations between real automated systems and their digital twins often occur due to local conditions, inaccurate assembly, or system changes, leading to significant manual effort in identifying and correcting these discrepancies, especially when certain areas of the automated system are inaccessible.

Innovation Solution

A procedure and device that attach a sensor unit to the automated system to continuously record measurement variables, compare them with simulated variables from the digital twin, and identify deviations by analyzing the time courses of these variables, allowing for automatic detection and correction of discrepancies.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Extent of automation

If manual inspection and comparison methods are used to identify deviations between the real automated system and the digital twin, then the process can be performed without additional sensors or automated equipment, but the manual effort and time required increase significantly

Engineering Contradiction:
Improveautomation of deviation detectionVSAvoidcomplexity of detection system
Core Design Contradiction:
Extent of automationVSDevice complexity

Solution Approach 1:

The system attaches sensors directly to process items (workpieces, tools, or system components) that are already moving through the automated system. These sensors automatically record measured variables during normal operation without requiring separate inspection equipment or manual intervention. The process item itself serves as the measurement platform, eliminating the need for external detection devices.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The patent replaces manual mechanical inspection methods with automated sensor-based measurement systems. Instead of experts physically examining and comparing system states, electronic sensors continuously record data that is automatically transmitted to and compared against the digital twin simulation results, substituting human mechanical inspection with automated electronic measurement and comparison.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

2Measurement precision

If 3D scanning and laser measurement methods are used to capture the automated system, then detailed geometric data can be obtained, but areas with workspace enclosures or inaccessible locations cannot be measured

Engineering Contradiction:
Improveprecision of system measurementVSAvoidaccessibility of measurement areas
Core Design Contradiction:
Measurement precisionVSEase of operation

Solution Approach 1:

Sensors are attached directly to process items that move through all areas of the automated system including enclosed workspaces and inaccessible locations. The process items themselves carry the measurement tools into areas that would be inaccessible to external scanning equipment, enabling measurement throughout the entire system volume without requiring physical access from outside the enclosures.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The measurement system is nested within the process items (workpieces, tools, or components) that already traverse the automated system. The sensors are integrated into or attached to these moving elements, allowing the measurement apparatus to be carried through the system's workflow and into enclosed areas, rather than requiring external access to measurement zones.

Inventive Principle:
Principle #7Nested doll (Nesting)

3Measurement precision

If simulation experts perform on-site inspections and discussions to clarify consistency between the real system and digital twin, then accurate deviation identification can be achieved, but the time and expert resources required increase considerably

Engineering Contradiction:
Improveaccuracy of deviation identificationVSAvoidtime for deviation detection
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

Sensors attached to process items continuously record measured variables throughout the entire processing cycle as items move through the automated system. This continuous measurement eliminates the need for discrete inspection stops or expert visits, capturing deviation data automatically during normal operation and enabling immediate comparison with digital twin simulations without interrupting the production flow.

Inventive Principle:
Principle #20Continuity of useful action

Solution Approach 2:

The system establishes a feedback loop where sensor measurements from the real system are continuously compared against corresponding simulated measured variables from the digital twin. This automated feedback mechanism immediately identifies deviations and their locations, replacing the iterative process of expert inspection and discussion with automatic real-time comparison and notification.

Inventive Principle:
Principle #23Feedback

4Reliability

If the digital twin is updated manually after expansions or modifications to the real plant, then the digital model can reflect current system state, but the effort and time required for updates increase

Engineering Contradiction:
Improveaccuracy of digital twin representationVSAvoidefficiency of digital twin maintenance
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The continuous comparison between sensor measurements from the real system and simulated variables from the digital twin creates an automatic feedback mechanism. When deviations are detected, the system identifies their locations and can trigger automatic updates to the digital twin model, eliminating the need for manual updates after system modifications. The feedback loop ensures the digital twin remains synchronized with the physical system through automated detection and reporting of changes.

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP3786745B1Identification of deviations between an actual system and a digital twin
Publication Date: 2023.02.22 SIEMENS AG
  • EP3786745B1 patent drawingFigure 1~3

AI summary

The present invention relates to a method and a device for determining deviations between an automated system and a digital twin of the automated system, as well as an automated system comprising the device. A sensor unit is attached to a process item and passes through an automated system together with it. The sensor unit continuously records at least one measured quantity. A time course of the recorded measured quantity is compared with a time course of a corresponding simulated measured quantity. If a deviation exists between the time courses, the location in the automated system where the deviation occurs is identified.