Field Device Diagnostics Using Time-Grouped Fault Pattern Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In industrial automation systems, diagnosing the cause of malfunctions in field devices is challenging due to the large amount of data generated and the irregular occurrence of faults, making it difficult to identify the root cause of communication issues and device state changes.
Innovation Solution
A method that analyzes diagnostic messages from multiple field devices using a heat map or tabular listing, filtered by various criteria, and evaluated with algorithms for image or pattern recognition and cluster detection to suggest corrective actions based on a knowledge database.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If diagnostic messages from a large number of field devices are collected and stored, then the data available for analysis increases, but the complexity of analyzing the data to identify error causes increases
Solution Approach 1:
The patent segments the large volume of diagnostic data by grouping messages according to their temporal distribution patterns. Instead of analyzing all diagnostic messages uniformly, the system divides them into groups based on whether they occur periodically, randomly, or in clusters, thereby making the analysis manageable and systematic.
Solution Approach 2:
The patent introduces an intermediary evaluation process that acts as a mediator between raw diagnostic data and error cause identification. This intermediary layer automatically groups and categorizes diagnostic messages based on temporal patterns, transforming the raw data into structured information that highlights abnormal temporal distributions and facilitates easier analysis of error causes.
2Measurement precision
If manual monitoring and troubleshooting of communication networks is performed using special measuring tools, then detailed diagnostic information can be obtained, but the time required for troubleshooting increases significantly
Solution Approach 1:
The patent performs preliminary action by automatically collecting, storing, and pre-processing diagnostic messages from field devices before actual troubleshooting is needed. The system continuously monitors and groups diagnostic data in the background, so when a fault occurs, the analysis is already partially complete, significantly reducing the time required for actual troubleshooting while maintaining diagnostic precision.
3Difficulty of detecting and measuring
If diagnostic messages are analyzed without grouping by time intervals, then individual error events can be identified, but the ability to detect patterns and recurring issues is reduced
Solution Approach 1:
The patent applies periodic action by dividing the analysis into discrete time intervals and grouping diagnostic messages within each interval. This periodic structuring allows the system to detect recurring patterns and temporal distributions of errors that would be invisible in continuous or unstructured data, while the automated grouping process manages the organizational complexity systematically.
Data Source
Figure 1
Figure 2
AI summary
The invention comprises a method for analysing malfunctions and/or changes of device statuses in a system (A) of automation technology, wherein the system (A) has a plurality of field devices (FA, FB, FC, FD), the field devices (FA, FB, FC, FD) communicating with one another directly via a communication network (D, F), or via at least one communication unit (PLC), more particularly a gateway, a control unit or a remote I/O, and being designed to issue an appropriate diagnostic notice (Diag1, Diag2, Diag3) depending on a malfunction and/or a change of a device status in the system (A), and wherein the diagnostic notices (Diag1, Diag2, Diag3) are transmitted to a data bank (B) and stored in same. The method comprises:- reading the diagnostic notices (Diag1, Diag2, Diag3) from the data bank; filtering the read diagnostic notices (Diag1, Diag2, Diag3) using at least one selection criterion;- linking the filtered diagnostic notices (Diag1, Diag2, Diag3) using time stamps, wherein a time stamp contains a date of the occurrence of a malfunction contained in the corresponding diagnostic notice (Diag1, Diag2, Diag3);- defining time intervals (At), more particularly equal time intervals (At);- grouping the diagnostic notices (Diag1, Diag2, Diag3) linked using the time stamps into the defined time intervals (At) which correspond to their respective time stamps;- evaluating the grouped diagnostic notices (Diag1, Diag2, Diag3) with regard to defined abnormalities (CL1, CL2, CL3, CL4).