Observer-Based Flow Rate Control for Thermal Sensor Hysteresis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional fluid control devices using thermal flow rate sensors suffer from slow response speeds and hysteresis, leading to inaccurate and delayed flow rate control due to changes in the fluid control valve, such as those caused by hysteresis, deterioration over time, or peripheral heat effects.
Innovation Solution
A fluid control device equipped with an observer that includes a valve model to estimate the flow rate of the fluid control valve and a flow rate sensor model to simulate the flow rate sensor, which determines model and flow rate error estimation values to compensate for errors, allowing for accurate feedback control.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If a thermal flow rate sensor is used, then the device complexity is reduced and manufacturing cost is lowered, but the response speed becomes slower and control accuracy deteriorates
Solution Approach 1:
An observer is introduced as an intermediary component that includes a flow rate sensor model to simulate the thermal flow rate sensor's behavior. This observer estimates the actual flow rate by compensating for the sensor's slow response characteristics, allowing the system to use a simple thermal sensor while achieving control performance comparable to expensive fast-response sensors
Solution Approach 2:
The system changes the parameter being controlled from direct sensor output to observer-estimated values. By modifying how the flow rate information is obtained (through mathematical estimation rather than direct measurement), the system overcomes the inherent slow response of thermal sensors without changing the sensor itself
2Ease of manufacture
If a thermal flow rate sensor is used, then manufacturing cost is reduced, but manufacturing precision and control accuracy deteriorate due to hysteresis and deterioration over time
Solution Approach 1:
The observer continuously monitors the deviation between the actual sensor output and the modeled sensor behavior, then feeds back compensation signals to correct for hysteresis and deterioration effects. This feedback mechanism allows the system to maintain high control accuracy despite using a low-cost thermal sensor that exhibits hysteresis and aging
Solution Approach 2:
The flow rate sensor model within the observer serves itself by automatically compensating for its own deficiencies. The model learns and adapts to the thermal sensor's characteristics including hysteresis and deterioration, enabling the sensor to self-correct its measurement errors without external intervention
3Duration of action of stationary object
If the valve aperture changes due to hysteresis or deterioration, then the fluid control valve continues to operate, but flow rate control accuracy deteriorates and overshoot is generated
Solution Approach 1:
The observer proactively estimates and compensates for valve aperture changes before they significantly impact control accuracy. By continuously monitoring valve drive voltage and predicting aperture variations due to hysteresis or deterioration, the system takes preliminary corrective action to maintain accurate flow rate control
Solution Approach 2:
The system replaces direct mechanical feedback from the valve (which would require complex position sensing) with a mathematical model-based estimation approach. The observer uses electrical signals (drive voltage) and flow rate measurements to estimate mechanical aperture changes, substituting complex mechanical measurement with simpler electrical and computational methods
Data Source
AI summary
The present invention makes it possible to perform highly accurate flow rate control while also achieving a fast response. Based on a deviation between a flow rate measurement value output by a flow rate sensor and a flow rate estimation value output by a flow rate sensor model, an observer determines a model error estimation value, which is obtained by estimating a model error in the flow rate sensor model relative to the flow rate sensor, and a flow rate error estimation value, which is obtained by estimating a flow rate error caused by a change in a fluid control valve, and inputs the model error estimation value into the flow rate sensor model, and then outputs an estimation value to a feedback controller based on the flow rate error estimation value and the valve flow rate estimation value output by the valve model.


