Sensor Signature Analysis for Unexpected Automation Operations
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current control systems for industrial automation processes face inefficiencies in training and retraining process models, particularly in terms of energy and processing power usage, which hinders real-time optimization and monitoring.
Innovation Solution
The system receives processed sensor data from multiple sources, identifies signatures indicating unexpected operations, and performs root cause analysis to determine correlations, allowing for real-time adjustments and model updates to improve operational efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a model predictive control (MPC) system is used to optimize process performance, then the process can transition from current operating state to desired operating state, but training the process model involves inefficient use of computing resources (energy, processing power, storage)
Solution Approach 1:
The patent segments the model training process by identifying and separating unexpected operations from normal operations. The system divides sensor data into expected and unexpected categories, training the process model only on unexpected operations. This segmentation reduces the volume of training data required, thereby decreasing computing resource consumption while maintaining the model's ability to optimize process transitions.
Solution Approach 2:
The system applies partial action by selectively processing only the necessary portion of sensor data - specifically, only unexpected operations are used for model training. Instead of processing all sensor data equally, the system identifies and focuses computational resources on the anomalous portions that actually contribute to model improvement, reducing overall energy and processing power consumption.
2Speed
If processed sensor data from multiple sources is analyzed in real-time to identify unexpected operations, then faster determination of control actions is achieved, but the complexity of data processing and analysis increases
Solution Approach 1:
The system extracts and isolates unexpected operations from the bulk of sensor data by comparing against expected operation patterns. By taking out only the anomalous data portions for detailed analysis, the system reduces the complexity of real-time processing while maintaining fast response capabilities. The extraction of unexpected operations from normal operational data simplifies the analysis burden.
Solution Approach 2:
The patent introduces an intermediary layer that processes sensor data before final analysis - a system that identifies and flags unexpected operations as intermediates between raw sensor data and control decisions. This intermediary processing stage simplifies the overall system complexity by pre-categorizing data, allowing faster downstream processing of only the relevant unexpected operations.
Data Source
AI summary
A non-transitory computer-readable medium comprising computer-executable instructions that, when executed, cause a processor to perform operations including receiving a first dataset from a first automation component, the first dataset corresponds to raw data acquired by a first sensor; receiving a second dataset from a second automation component, the second dataset corresponds to raw data acquired by a second sensor; receiving data indicating an expected operation related to operations of an industrial automation system including the first and second automation components; determining a signature based on the first and second datasets and the data indicating the expected operation, wherein the signature indicates an unexpected operation as compared to the expected operation; performing a root cause analysis using the signature to determine a relationship indicating a first set of changes of the first dataset corresponding to a second set of changes in the data indicating the expected operation.