Material Testing Load Control With Frequency-Domain Hunting Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Accurately detecting hunting in material testing machines is challenging, especially when time-series data of the target value changes over a time axis, due to difficulties in distinguishing hunting from noise and fluctuations.
Innovation Solution
A material testing machine equipped with a hunting detection unit that compares frequency spectra of measurement and target values, with noise removal from the control and measurement systems, allowing for precise detection of hunting by analyzing frequency domains excluding low or high frequency noise, and using dither signals to enhance detection accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If time-series data of target value changes on a time axis is used for hunting detection, then the material test can proceed with dynamic loading conditions, but hunting detection accuracy deteriorates due to difficulty in distinguishing hunting from target value changes
Solution Approach 1:
The patent transforms the hunting detection problem from the time domain to the frequency domain by applying FFT (Fast Fourier Transform) to both the measurement value and target value time-series data. This dimensional transformation allows hunting oscillations to be identified as distinct frequency components in the spectral domain, even when the target value is changing over time. The frequency domain representation separates hunting signals from target value changes, resolving the detection accuracy issue while maintaining dynamic loading capability.
Solution Approach 2:
The patent introduces frequency spectrum analysis as an intermediary processing step between the raw time-series data and hunting detection. By converting both measurement and target value data to frequency spectra and comparing them, the system can identify hunting oscillations as frequency components present in the measurement spectrum but not in the target spectrum. This intermediary transformation enables accurate hunting detection under dynamic loading conditions.
2Measurement precision
If feedback control is performed to make measurement values follow target values, then control precision is improved, but hunting oscillations occur due to control system instability
Solution Approach 1:
The patent implements a feedback mechanism where the detected hunting oscillations trigger automatic adjustment of feedback control parameters. The control device monitors the frequency spectrum comparison results and, upon detecting hunting, modifies feedback control gains or parameters to suppress the oscillations. This feedback loop stabilizes the control system while maintaining precision, as the system dynamically adjusts to prevent hunting rather than simply reacting to it.
3Measurement precision
If noise removal processing is applied to measurement values, then hunting detection accuracy is improved, but detection time increases due to additional processing steps
Solution Approach 1:
The patent applies noise removal processing as a preliminary step before hunting detection by pre-processing the measurement value time-series data to eliminate known noise components. By removing noise beforehand, the subsequent FFT-based hunting detection operates on cleaner data, improving accuracy without requiring excessive computation during the detection phase. The noise characteristics are identified and filtered in advance, reducing the computational burden during real-time hunting detection.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The solution enables accurate hunting detection even with fluctuating target values, improves detection accuracy by removing noise influences, and prevents accuracy decreases in material tests by suppressing hunting through control parameter adjustments.
Implementation Method 1
a hunting detection unit that detects hunting by comparing a frequency spectrum obtained by converting time-series data of the measurement value with a frequency spectrum obtained by converting time-series data of the target value
Data Source
AI summary
Provided is a material testing machine (1) including: a load mechanism (12) that applies a load to a test object; a load measurement device that measures the load applied to the test object; and a control device (30) that performs a feedback control for the load mechanism (12) based on a deviation between a measurement value of the load and a target value of the load, in which a change in a physical quantity generated in the test object due to the load is measured, and the control device (30) includes a hunting detection unit (66) that detects hunting by comparing a frequency spectrum obtained by converting time-series data of the measurement value with a frequency spectrum obtained by converting the time-series data of the target value.