PLC Fault Detection Using Buffered Process Images and ML

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional manufacturing and automation systems lack effective fault detection and monitoring capabilities, making it difficult for operators to diagnose the root cause of faults in automation devices, leading to increased downtime and costs due to inadequate data collection and analysis.

Innovation Solution

An automation device method and system that includes storing input and output process images in a buffer, using a ring buffer to collect data, and employing a machine learning model to predict faults, allowing for improved data collection and analysis for fault diagnosis.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If conventional error code collection methods are used, then fault detection is simple, but diagnostic capability is insufficient for operators to identify root causes

Engineering Contradiction:
Improvefault detection precisionVSAvoidfault diagnostic information
Core Design Contradiction:
Measurement precisionVSLoss of information

Solution Approach 1:

The system performs preliminary data collection and processing by capturing process images, sensor data, and operational parameters before faults occur. This preparatory action ensures that comprehensive diagnostic information is already available when a fault happens, eliminating the need for post-fault data gathering and enabling immediate root cause analysis.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent introduces an intermediary data processing layer that collects and correlates information from multiple sources including process images, sensor readings, and operational parameters. This intermediary system bridges the gap between raw data and diagnostic insights, providing operators with synthesized fault analysis results rather than just raw error codes.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Loss of information

If detailed data collection is implemented for fault analysis, then diagnostic capability improves, but system complexity and data processing requirements increase

Engineering Contradiction:
Improvefault diagnostic informationVSAvoiddata collection system complexity
Core Design Contradiction:
Loss of informationVSDevice complexity

Solution Approach 1:

The data collection system is segmented into modular components: process image capture modules, sensor data acquisition modules, operational parameter collection modules, and analysis modules. Each segment handles specific types of data independently, which can be configured and maintained separately, reducing overall system complexity while maintaining comprehensive data collection capabilities.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system incorporates automated analysis capabilities that process collected data and generate diagnostic insights without requiring complex external processing. The automated fault detection and analysis features enable the system to self-diagnose issues, reducing the burden on operators and simplifying the overall data processing architecture.

Inventive Principle:
Principle #25Self-service

3Loss of information

If comprehensive monitoring data is collected, then root cause analysis capability improves, but downtime for data processing and maintenance increases

Engineering Contradiction:
Improvefault diagnostic informationVSAvoidmaintenance downtime
Core Design Contradiction:
Loss of informationVSLoss of time

Solution Approach 1:

The system continuously collects and pre-processes monitoring data during normal operation, preparing diagnostic information in advance before faults occur. This preliminary data preparation ensures that when a fault happens, the analysis can be performed immediately using pre-processed information, minimizing downtime while maintaining comprehensive diagnostic capabilities.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent replaces manual data analysis and fault diagnosis procedures with automated electronic data processing and analysis systems. This substitution enables rapid processing of comprehensive monitoring data without requiring extended manual intervention, significantly reducing maintenance downtime while maintaining thorough diagnostic analysis.

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

Data Source

PatentEP4372495A1Systems, methods and devices for detecting errors in programmable logic controllers
Publication Date: 2024.05.22 ATS CORPORATION
  • EP4372495A1 patent drawingFigure 1~2
  • EP4372495A1 patent drawingFigure 3~4
  • EP4372495A1 patent drawingFigure 5

AI summary

Provided are systems, methods and devices for automation devices. The method includes providing an automation device for controlling at least one piece of automation equipment; receiving, at an input of the automation device, input data; storing, at an input process image, the input data; executing, at a processor of the automation device, an automation program to generate output data based on the input data; storing, at an output process image of the automation device, the output data; and storing, in a buffer, the input process image and the output process image.