PLC Time-Series Classification for Abnormal Device Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In industrial settings, identifying abnormal devices within a programmable logic controller (PLC) system is time-consuming and requires significant expertise due to the large amount of data generated, necessitating a mechanism to efficiently classify and analyze data for early detection of abnormalities without relying on extensive knowledge or experience.
Innovation Solution
A programmable logic controller system that includes an execution engine, data collection, classification, and identification sections to analyze time-series data, determining detection algorithms based on device types, allowing for automatic identification of abnormal devices by classifying characteristics such as periodicity and continuity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If data from all devices is recorded and analyzed to identify abnormal devices, then identification accuracy is improved, but analysis time increases significantly
Solution Approach 1:
The patent segments the analysis process by classifying devices into different types (input devices, output devices, memory devices, etc.) and analyzing each type separately with type-appropriate methods. This segmentation allows the system to focus analysis on relevant devices rather than processing all device data uniformly, thereby reducing overall analysis time while maintaining identification accuracy.
Solution Approach 2:
The patent applies local quality by using different analysis methods tailored to each device type. For example, input devices are analyzed using specific criteria appropriate to their function, while output devices use different criteria. This localized approach ensures high identification accuracy for each device category while avoiding the time cost of analyzing all devices with a single comprehensive method.
2Reliability
If comprehensive data analysis is performed on all devices, then abnormal device identification reliability is improved, but operational complexity increases
Solution Approach 1:
The patent reduces system complexity by segmenting devices into categories and applying simplified analysis rules to each category. Instead of implementing a single complex analysis system that processes all device data with equal depth, the system uses multiple simple, targeted analysis paths corresponding to different device types, thereby maintaining reliability while reducing operational complexity.
Solution Approach 2:
The patent applies local quality by matching analysis methods to device types, ensuring that each device category receives appropriate analysis depth. This approach maintains high identification reliability for each device type while avoiding the need for a uniformly complex analysis system across all devices.
3Measurement precision
If expert knowledge is used to select specific devices for analysis, then analysis precision is improved, but ease of operation deteriorates
Solution Approach 1:
The patent implements self-service by enabling the system to automatically classify devices into types and select analysis targets based on pre-defined classification rules. The system performs device type identification and analysis target selection autonomously without requiring expert knowledge or manual intervention, thereby maintaining high analysis precision while significantly improving ease of operation.
Solution Approach 2:
The patent applies preliminary action by pre-establishing device classification categories and analysis rules before actual operation. Device types are predetermined (input, output, memory, etc.), and analysis methods are pre-configured for each type. This preliminary setup allows the system to automatically and accurately select analysis targets without requiring expert knowledge during operation.
Data Source
AI summary
An object of the present invention is to efficiently identify an abnormal device by analyzing and classifying characteristics such as periodicity and continuity of data related to each device. A programmable logic controller collects data held by a device as a collection target out of the plurality of devices in accordance with a predetermined collection setting for each scan cycle of the user program, classifies each of the devices into one of a plurality of types on the basis of a characteristic of the collected time-series data of each of the devices, and determines a detection algorithm used in identification of the device as an abnormal device in accordance with the time-series data of the device collected by the collection section and the type of the device classified by the classification section for each of the devices.