Process Dynamics Monitoring via Low-Dimensional Parameter Mapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Industrial process control systems face challenges in detecting changes in dynamic behavior due to noisy data and multiple parameters, making it difficult to determine when process behavior has shifted, which can lead to poor quality production if not addressed promptly.
Innovation Solution
The system maps process model parameters to a lower dimensional space by combining time delay and time constant parameters into a response time parameter or examining the frequency response at a key frequency, reducing the parameter space to two more stable parameters that are easier to monitor for changes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multiple process model parameters (time delay, time constant, gain) are monitored separately, then comprehensive process behavior coverage is achieved, but noise and variability increase making change detection difficult
Solution Approach 1:
The patent combines multiple process model parameters (time delay and time constant) into a single response time parameter by addition. This merging reduces the parameter space from three dimensions (gain, time delay, time constant) to two dimensions (gain, response time), thereby reducing noise and variability while maintaining comprehensive process behavior monitoring capability.
Solution Approach 2:
The patent transforms the parameter space by changing its dimensional structure. Instead of monitoring three separate parameters (gain, time delay, time constant), the system projects them into a lower dimensional space with two parameters (gain, response time). This dimensional reduction simplifies change detection while preserving the essential dynamic behavior information.
2Measurement precision
If process model parameters are monitored with high precision, then accurate detection of process changes is achieved, but the complexity of analysis and interpretation increases
Solution Approach 1:
By combining time delay and time constant into response time, the patent reduces the number of parameters that need to be analyzed independently. This merging maintains measurement precision for detecting process changes while simplifying the analysis and interpretation burden, as fewer parameters need to be tracked and compared.
Solution Approach 2:
The patent extracts the essential dynamic behavior characteristic from multiple parameters by creating response time as a composite parameter. This extraction isolates the key dynamic information from the noise of individual parameter variations, making change detection more straightforward while preserving accuracy.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method includes acquiring (601) process data collected in an industrial process control and automation system (100). The method also includes reducing (603) a dimension space of the process data by combining two or more parameters (405, 410, 415) of the process data or examining a frequency response of the process 5 data. The method further includes determining (605) a change in a process based on a change in the process data in the reduced dimension space. The method also includes outputting (607) a result based on the determined change in the process.