Asynchronous Sensor Synchronization via Measurement Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional control systems for paper machines struggle with synchronizing measurements from asynchronous sensors, which are often unsynchronized and have different sampling times, leading to inefficiencies in controlling multiple sheet properties simultaneously.
Innovation Solution
A method and apparatus that predict and update measurements using a process model, allowing for the synchronization of asynchronous sensor data with the controller's sampling rate, even when the sensors have different sampling times and time delays, enabling effective control of paper machine properties like dry weight, moisture, and caliper.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional CD control systems are used with asynchronous sensors, then each controller can control one sheet property independently, but multivariable coordinated control of multiple sheet properties cannot be implemented
Solution Approach 1:
The patent introduces a measurement predictor as an intermediary component that receives asynchronous measurements from multiple sensors and generates predicted synchronized measurements. This mediator transforms the asynchronous sensor data into a format suitable for multivariable controllers, enabling coordinated control without requiring direct synchronization of the sensors themselves.
Solution Approach 2:
The measurement predictor performs preliminary action by predicting what the sensor measurements would be at the controller's sampling instants. This allows the system to prepare synchronized measurement data in advance, before the actual control decision is made, enabling the multivariable controller to operate with synchronized data even though the sensors operate asynchronously.
2Adaptability or versatility
If sensors with different sampling times are used, then measurement flexibility is improved, but synchronization with the controller's sampling rate becomes problematic
Solution Approach 1:
The measurement predictor changes the time parameter of the measurements by predicting what the sensor values would be at the controller's sampling instants. This transformation allows measurements taken at different times to be converted into synchronized data points, eliminating timing mismatches while preserving the flexibility of using sensors with different sampling rates.
3Device complexity
If asynchronous sensor measurements are used directly in control, then system simplicity is maintained, but control precision and reliability deteriorate
Solution Approach 1:
The measurement predictor serves as an intermediary that bridges the simple asynchronous sensor system and the precision requirements of the controller. It adds the necessary synchronization function without requiring changes to the simple sensor architecture, thereby maintaining system simplicity while achieving control precision.
Data Source
AI summary
A method includes predicting measurements or states to be used by a controller to control a process. The predicted measurements or states are generated using a model of the process. The method also includes providing the predicted measurements or states to the controller such that the controller uses the predicted measurements or states at a sampling rate of the controller. In addition, the method includes updating at least some of the predicted measurements or states using measurements associated with a characteristic of an item from a sensor. The model may represent a discrete time model, and the method may also include generating the discrete time model using a continuous time model of the process. The measurements could be received from a plurality of sensors, where at least two of the sensors have different sampling times.