Process Equipment State Monitoring Using Correlated Reference Models

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

In process industries, it is challenging to determine the health condition of equipment online due to complex and sensitive correlations between numerous variables, leading to difficulties in monitoring equipment states and predicting faults in real-time.

Innovation Solution

A method and apparatus for monitoring the state of a device in a process industry, which involves collecting multi-dimensional state data, performing correlation analysis using algorithms like Pearson or Kendall, establishing a device state reference model using the Dirichlet process algorithm, and determining real-time device states to identify normal or abnormal conditions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If planned routine maintenance is performed based on scheduled time, then maintenance can be performed regularly, but it causes excess maintenance or lack of maintenance due to inability to predict actual fault timing

Engineering Contradiction:
Improveequipment reliabilityVSAvoidmaintenance timing accuracy
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The system performs preliminary analysis of historical operation data to establish equipment state reference models before actual faults occur. By pre-processing data and identifying normal state patterns, the system enables early detection of deviations from normal operation, allowing maintenance to be scheduled based on actual equipment condition rather than fixed time intervals.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system continuously monitors equipment state dimensions in real-time and compares them against the established reference models. When deviations indicate potential faults, the system provides feedback alerts that trigger maintenance actions. This closed-loop feedback mechanism replaces open-loop scheduled maintenance with condition-based maintenance timing.

Inventive Principle:
Principle #23Feedback

2Reliability

If online monitoring of equipment state is implemented, then unplanned stops can be reduced by early fault detection, but it requires accurate definition of boundary between normal and abnormal states which is difficult due to complex variable correlations

Engineering Contradiction:
Improveequipment state monitoring accuracyVSAvoidstate boundary definition complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The system segments the complex equipment state into multiple independent state dimensions (vibration, temperature, pressure, etc.). Each dimension is monitored and analyzed separately against its own reference model, rather than attempting to define a single complex boundary for the entire equipment state. This segmentation simplifies the boundary definition problem while maintaining comprehensive monitoring capability.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system transforms the abstract concept of equipment health into concrete measurable parameters and state dimensions. By changing from a holistic state assessment to specific parameter monitoring with statistically-defined boundaries (reference models), the system makes the boundary between normal and abnormal states objectively definable and measurable.

Inventive Principle:
Principle #35Parameter changes

3Productivity

If equipment runs until scheduled maintenance time, then production continuity is maintained, but potential faults may go undetected between maintenance intervals

Engineering Contradiction:
Improveproduction continuityVSAvoidfault detection capability
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The system enables continuous monitoring of equipment state dimensions throughout operation, rather than periodic inspections. This continuous observation maintains production continuity while simultaneously providing ongoing fault detection capability, eliminating the trade-off between running until maintenance and detecting potential faults.

Inventive Principle:
Principle #20Continuity of useful action

Solution Approach 2:

The system introduces state reference models as intermediaries between raw equipment data and fault detection decisions. These models continuously compare actual state dimensions against normal operation patterns, enabling real-time fault detection without interrupting production. The intermediary model translates continuous data streams into actionable maintenance insights.

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentEP3696693B1Method and apparatus for monitoring state of device in process industry and medium
Publication Date: 2025.04.02 SIEMENS AG
  • EP3696693B1 patent drawingFigure 1
  • EP3696693B1 patent drawingFigure 2
  • EP3696693B1 patent drawingFigure 3~4

AI summary

A method and an apparatus for monitoring the state of a device in process industries and a medium. The method comprises: collecting and saving data of the multi-dimensional state of a device in the process industry; performing correlation analysis on the collected historical data of the multi-dimensional state of the device to obtain state dimensions having correlation; acquiring, according to the obtained state dimensions having correlation, historical data of the dimensional states having correlation, and modeling the acquired historical data of the dimensional states to obtain a device state reference model; acquiring real-time data of the dimensional states having correlation corresponding to the device state reference model in real time, and determining whether the current state of the device is normal according to a preset device state determination condition and the acquired real-time data of the dimensional states having correlation, and in combination with the device state reference model. The invention realizes online monitoring of the state of a device in the process industry.