Field Device Failure Compensation Using AI Substitute Variables
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Field devices in industrial automation systems often fail unexpectedly, leading to costly downtime and the need for redundant devices, as the failure of a field device can disrupt critical processes, and existing methods do not effectively predict or manage such failures.
Innovation Solution
A method that uses data processing units within the communication network to create a replacement system by analyzing history data and using AI algorithms to determine replacement variables for failing process variables, allowing for proactive compensation of field device failures without shutting down the system, by comparing current process values with target values and transmitting replacement variables to higher-level units.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If redundant field devices are built into the system in advance to replace failed devices, then system reliability is improved, but device complexity and cost increase
Solution Approach 1:
The patent creates a virtual copy (replacement system) of the failed field device using software and data processing rather than a physical redundant device. The data processing unit generates substitute values that mimic the behavior of the failed device, allowing the system to continue operating without the physical redundancy that would increase complexity and cost.
Solution Approach 2:
The patent replaces the mechanical/physical redundancy approach with an information-processing approach. Instead of having backup physical field devices, the system uses data processing units, historical data analysis, and algorithmic generation of substitute values to maintain system functionality, thereby reducing device complexity while preserving reliability.
2Reliability
If field devices are monitored continuously to detect failures, then reliability is improved, but loss of time for failure detection increases
Solution Approach 1:
The patent performs preliminary actions by continuously analyzing historical data and determining time periods when failures are likely to occur before they actually happen. The system pre-calculates replacement variables and prepares substitute values in advance, so when a failure occurs, the replacement is immediate and downtime is minimized.
Solution Approach 2:
The patent implements feedback mechanisms where the data processing unit continuously monitors field device performance, compares actual values with predicted values, and uses this feedback to improve failure prediction accuracy. The system learns from historical data and adjusts its predictions, enabling earlier and more accurate failure detection without increasing response time.
3Measurement precision
If replacement variables are determined using historical data and AI algorithms, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The patent enables the system to self-service by using its own historical data and embedded AI algorithms to generate replacement variables autonomously. The data processing unit within the automation system performs the analysis and generates substitute values without requiring external intervention or complex external systems, balancing measurement precision with manageable device complexity.
4Reliability
If field devices transmit diagnostic data and status information continuously, then reliability monitoring is improved, but loss of energy increases
Solution Approach 1:
The patent implements periodic action by having field devices transmit diagnostic data and status information at scheduled intervals rather than continuously. The data processing unit receives these periodic transmissions and performs batch analysis to update failure predictions and generate replacement variables, significantly reducing communication energy consumption while maintaining reliable failure detection capability.
Data Source
Figure 1
AI summary
The invention relates to a method for compensating for a malfunction of a field device (FG1, FG2, FG3, FG4, FG5) in an automation technology plant (A), comprising: - monitoring the process values transmitted from the field devices (FG1, FG2, FG3, FG4, FG5) to the superordinate unit (PLC); - producing history data on the basis of the transmitted process values for each of the field devices (FG1, FG2, FG3, FG4, FG5), or of the sensor units (SE1, SE2, SE3, SE3', SE4, SE5, SE5'); - setting up a substitution system (ES) on the basis of the history data, wherein the data processing unit (DV1, DV2), in the course of setup of the substitution system (ES), ascertains which process variable of a sensor unit (SE1, SE2, SE3, SE3', SE4, SE5, SE5') can replace a process variable of another sensor unit (SE1, SE2, SE3, SE3', SE4, SE5, SE5') as substitute variable with a predetermined accuracy; - comparing the current process values of each of the field devices (FG1, FG2, FG3, FG4, FG5), or of the sensor units (SE1, SE2, SE3, SE3', SE4, SE5, SE5'), with setpoint values for precalculating periods in which the current process values of a respective field device (FG1, FG2, FG3, FG4, FG5), or of the sensor units (SE1, SE2, SE3, SE3', SE4, SE5, SE5'), differ from the setpoint values by at least one predetermined value; - transmitting the substitute variable for the respective field device (FG1, FG2, FG3, FG4, FG5), or the sensor units (SE1, SE2, SE3, SE3', SE4, SE5, SE5'), to the superordinate unit (PLC) by means of the data processing unit (DV1, DV2) during the precalculated periods.