PLC Redundant Device Switching via Feature Quantity Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current factory automation systems using programmable logic controllers (PLCs) face challenges in increasing equipment capacity utilization and reliability while reducing maintenance costs, particularly in preventive maintenance scenarios where continuous monitoring increases operational costs.
Innovation Solution
A control system that includes a controller with a feature quantity generation unit, abnormality detection unit, switch unit, and learning unit to automatically switch redundant devices between working and standby modes based on detected abnormalities, utilizing machine learning to determine abnormality detection parameters and update device configurations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If preventive maintenance monitoring is implemented to detect abnormalities before failure, then equipment reliability is improved, but maintenance cost increases
Solution Approach 1:
The system performs self-diagnosis by automatically generating feature quantities from device data and detecting abnormalities without requiring external monitoring resources. The control device itself generates diagnostic information through its existing data collection functions, eliminating the need for separate monitoring systems and reducing maintenance complexity while improving reliability
Solution Approach 2:
The system changes the parameter of abnormality detection from requiring complex external monitoring to using simple feature quantity calculations based on existing device operating parameters. By transforming device data into feature quantities and comparing them against thresholds, the system achieves reliable abnormality detection using standard control device functions rather than complex monitoring infrastructure
2Productivity
If automatic device switching is implemented upon abnormality detection, then equipment capacity utilization is improved, but device complexity increases
Solution Approach 1:
The control device performs multiple functions using its existing capabilities: it collects device data for normal operation, generates feature quantities for diagnostics, detects abnormalities, and executes device switching. By making the control device universal and multi-functional, the system achieves automatic switching capability without adding dedicated complexity for each function
Solution Approach 2:
The system merges the abnormality detection function and device switching function into the existing control device architecture. Rather than adding separate monitoring and switching systems, the patent combines these functions within the control device's data processing and control execution capabilities, maintaining simplicity while enabling automatic switching for improved capacity utilization
Data Source
AI summary
A control system includes a controller for controlling a control target and redundant devices accessible from the controller. The control system includes a feature quantity generation unit that generates a feature quantity from data associated with the redundant devices, an abnormality detection unit that determines whether an abnormality has occurred in one of the redundant devices based on the feature quantity generated by the feature quantity generation unit and a predetermined abnormality detection parameter, a switch unit that switches the redundant devices between a working mode and a standby mode when the abnormality detection unit determines that an abnormality has occurred, and a learning unit that performs machine learning using the data associated with the redundant devices to determine the abnormality detection parameter.