Mechanical Load Parameter Identification With Reliability Indication
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for identifying physical parameters of mechanical loads in speed- or position-controlled applications face challenges due to reliance on parametric models, which are often unknown, and are affected by noise, limited measurement time, and unmodeled dynamics, leading to inaccurate curve fitting and reliability issues.
Innovation Solution
A method that applies a pseudo-random binary signal to a mechanical device, calculates its power density spectrum, and uses a coherence function to automatically fit a parametric model to the non-parametric frequency response, providing improved robustness and reliability indication by accounting for variance and excitation properties.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If parametric models are used for identifying mechanical parameters, then the identification process can be automated, but the accuracy deteriorates due to noise, limited measurement time, and unmodeled dynamics
Solution Approach 1:
The method performs preliminary actions by applying a control signal to excite the mechanical device before the actual parameter identification measurement. This pre-excitation ensures that the system is adequately stimulated across the relevant frequency range, improving the quality of the measured frequency response and enabling more accurate parameter extraction from the power density spectrum.
Solution Approach 2:
The method uses feedback by continuously monitoring the power density spectrum of the return signal and using it to stipulate the excitation signal. This adaptive feedback loop allows the system to adjust the excitation based on the measured response, ensuring optimal excitation conditions are maintained throughout the measurement process, thereby improving identification accuracy while maintaining automation.
2Measurement precision
If control signals are applied to excite the mechanical device, then the identification accuracy improves, but the reliability deteriorates due to noise and unmodeled dynamics
Solution Approach 1:
The method applies dynamic principles by using a control signal that dynamically excites the mechanical device across a range of frequencies. The excitation signal is designed to cover the relevant operational frequency spectrum, allowing the system to capture the dynamic behavior of the mechanical device under various conditions, thereby improving identification accuracy while maintaining reliability through comprehensive dynamic characterization.
Solution Approach 2:
The method utilizes parameter changes by analyzing the power density spectrum across different frequency parameters. By examining how the system responds to excitations at different frequencies and adjusting the excitation parameters based on the measured spectrum, the method achieves accurate parameter identification while being robust to noise and unmodeled dynamics through multi-parameter analysis.
3Loss of time
If automated parameter identification is implemented, then commissioning time is reduced, but the expertise requirement increases due to the complexity of analyzing power density spectra and stipulating excitation signals
Solution Approach 1:
The method implements self-service by enabling the system to automatically perform parameter identification without requiring external expert intervention. The control circuit autonomously applies the control signal, measures the return signal, analyzes the power density spectrum, and stipulates the excitation signal based on the measured data. This self-contained automated process reduces commissioning time while managing complexity through integrated software algorithms that handle the sophisticated analysis internally.
Data Source
AI summary
A method to identify physical parameters of a mechanical load with integrated reliability indication includes: applying a first control signal to a mechanical device in a control circuit; measuring a first return signal; and using a power density spectrum of the first return signal to stipulate an excitation signal for the mechanical device.