Industrial Data Stream Mirroring for Predictive Maintenance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing monitoring systems for industrial operations are limited in their ability to predict component failures without interrupting data streams, as they often require modifying PLC programming and hardware, which is costly and disruptive, and cannot perform predictive analysis using historical data due to 'update-in-place' methodologies.
Innovation Solution
A system that copies and stores data from industrial components without interfering with the data stream, converts it into a query-able format, and analyzes it using reference data to predict potential failures, allowing for predictive maintenance without modifying existing PLC programming or hardware.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If PLC programming is modified to enable predictive analysis, then predictive maintenance capability is improved, but system complexity and cost increase
Solution Approach 1:
The patent introduces an intermediary system consisting of a network interface and data capture module that sits between the existing PLC and the predictive analysis functionality. This intermediary captures data packets from the PLC's communication network without modifying the PLC programming, thereby enabling predictive maintenance while avoiding the complexity and cost of PLC modifications.
Solution Approach 2:
The system creates a copy of the data stream from the PLC network and stores it in a database for later analysis. This copying approach allows predictive analysis to be performed on historical data without interfering with the real-time operational data flow, eliminating the need to modify the original PLC system.
2Measurement precision
If data monitoring is implemented using event-based HMI tags, then error detection is improved, but predictive analysis capability deteriorates
Solution Approach 1:
The patent implements continuous data capture of all data packets passing through the PLC network, rather than only capturing event-based data. This continuous capture ensures that all operational data, including pre-failure indicators and historical trends, are preserved in the database, enabling both error detection and predictive analysis.
Solution Approach 2:
The system captures and stores data packets before errors occur, creating a historical record that can be used for predictive analysis. By continuously monitoring and archiving data in advance, the system can identify trends and patterns that precede failures, enabling predictive maintenance before actual errors manifest.
3Reliability
If data packets are captured and stored for analysis, then predictive analysis capability is improved, but data stream interference increases
Solution Approach 1:
The network interface acts as an intermediary that passively monitors and captures data packets from the PLC network without injecting or modifying traffic. This passive capture approach ensures that the original data stream remains uninterrupted, eliminating interference while still enabling comprehensive data collection for predictive analysis.
Solution Approach 2:
The patent replaces active data manipulation methods with passive network packet capture. Instead of using update-in-place methodologies that modify database records and potentially interfere with data flow, the system uses read-only capture of network packets, substituting a non-intrusive mechanism that preserves data stream integrity.
4Quantity of substance
If update-in-place methodology is used for data storage, then storage efficiency is improved, but historical data availability for analysis deteriorates
Solution Approach 1:
The system implements a copy-based storage approach where captured data packets are stored as historical records in a database. Each data packet is preserved in its original form, creating a historical archive that can be queried for analysis without being overwritten. This copying methodology maintains both storage efficiency and historical data availability simultaneously.
Solution Approach 2:
Data packets are captured and stored in advance before they would be overwritten by new incoming data. This preliminary capture and storage creates a historical record that preserves past operational states, enabling retrospective analysis and predictive modeling while maintaining efficient use of storage resources.
Data Source
AI summary
An industrial system and method for monitoring and analyzing industrial operations such that data is obtained from one or more sensors of one or more industrial components and transmitted to a master controller as a data stream through a communication network, the transmitted data is captured, copied, and stored for analysis without interfering with the data stream.