Controller Program Error Prediction Using Digital Twin and AI
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for detecting and debugging programmatical errors in engineering programs for technical installations, such as industrial plants, are inefficient and prone to missing errors, leading to unplanned downtime and labor-intensive resolution processes.
Innovation Solution
A method and system that utilize a processing unit to capture and analyze input-output signals from a controller device, generating knowledge graphs to predict input and output signals, and simulating future executions to detect potential errors. This system applies an artificial intelligence model to eradicate identified programmatical errors from the engineering program.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual validation and debugging is performed by code developers, then programmatic errors can be detected, but the process is time-consuming and tedious
Solution Approach 1:
The patent replaces manual mechanical debugging processes with an automated AI-based system. The processing unit executes the engineering program, captures I/O signals, generates knowledge graphs, and automatically detects programmatic errors without human intervention, thereby substituting the mechanical manual debugging process with an automated computational system.
Solution Approach 2:
The system enables self-service error detection by automatically executing the engineering program, capturing I/O signals, generating knowledge graphs, and identifying programmatic errors without requiring code developer intervention. The system serves itself by performing validation and debugging tasks autonomously.
2Measurement precision
If code developers manually review gigantic blocks of code, then errors can be identified, but it becomes impossible to complete code development in time
Solution Approach 1:
The patent replaces manual code review processes with automated AI-based analysis. The processing unit automatically executes the program, captures I/O signals, generates knowledge graphs, and detects errors with high precision without requiring developers to manually review gigantic code blocks, thereby maintaining error detection accuracy while dramatically improving development speed.
Solution Approach 2:
The system creates a digital copy of the engineering program execution by generating knowledge graphs that represent the program's behavior and I/O signals. This copy allows automated error detection without requiring developers to examine the actual gigantic code blocks, enabling precise error identification while preserving code development productivity.
3Productivity
If automated assistance is provided to eradicate programmatic errors, then labor and time are reduced, but system complexity increases
Solution Approach 1:
The processing unit is designed with multi-functionality, serving as both the executor of the engineering program and the analyzer of its behavior. It captures I/O signals, generates knowledge graphs, detects errors, and validates resolutions all through a single unified system, reducing overall system complexity despite the advanced capabilities provided.
Data Source
AI summary
A method and system for eradicating programmatical errors in engineering programs for a controller device is provided. The method includes capturing, by a processing unit, a plurality of input-output signals associated with a controller device. Further, the method includes simulating, by the processing unit, a plurality of input signals which are predicted to be received by the controller device during a future scan cycle of execution of the engineering program. The method further includes predicting an error state in the controller device in the future scan cycle, by execution of the engineering program in a digital twin of the controller device. The method further includes generating corrected engineering program by application of an Artificial intelligence model on the engineering program.