US 2026/0257310 A1 · AUDIO-BASED MODAL ESTIMATION

Abstract

A method for machine tool natural frequency detection which requires no accelerometers or other data acquisition equipment to be attached to the machine tool. A simple hammer or other impact device is used to strike the machine tool, and sound data from the resulting vibration is collected and analyzed. FFT analysis of the sound data reveals natural frequencies of vibration. Data from multiple impacts may be aggregated, and other data processing techniques such as spectral subtraction are employed to reduce the prominence of false peaks in the FFT results and improve the natural frequency prediction accuracy. Workpiece natural frequencies may be predicted in the same manner. The natural frequency information is used to tune machining conditions-in particular, to select a spindle speed resulting in a cutting frequency which has a specified relationship to the identified natural frequencies.

Claims20

Dependent claims

An audio-based method for natural frequency identification of a machine tool, said method comprising:

collecting audio data from at least one impact event on the machine tool;
performing a fast Fourier transform (FFT) on the audio data, using a computing device having a processor and memory, to produce frequency response data for each of the impact events;
aggregating the frequency response data for all of the impact events into an aggregated frequency response;
identifying at least one natural frequency of vibration of the machine tool from the aggregated frequency response;
and
selecting a machine tool spindle speed, for a machining operation on a workpiece, such that a corresponding cutting frequency is related to the at least one natural frequency of vibration of the machine tool according to predefined criteria.

3 more elements

An audio-based method for natural frequency identification of a machining operation, said method comprising:

collecting audio data from a plurality of impact events on a machine tool;
performing a fast Fourier transform (FFT) on the audio data, using a computing device having a processor and memory, to produce frequency response data for each of the impact events;
processing the frequency response data for each of the impact events, including performing a spectral subtraction of background noise and calculating a moving average across a frequency spectrum;
aggregating the frequency response data for all of the impact events into an aggregated frequency response, including creating the aggregated frequency response by selecting, at each frequency across the frequency spectrum, a mean, a median or a minimum of frequency response values for all of the impact events;
identifying at least one natural frequency of vibration of the machine tool from the aggregated frequency response;
identifying at least one natural frequency of vibration of a workpiece, using the audio-based method for natural frequency identification;
and
selecting a machine tool spindle speed for the machining operation such that a corresponding cutting frequency and select cutting frequency harmonics are related to the at least one natural frequency of vibration of the machine tool and the at least one natural frequency of vibration of the workpiece according to predefined criteria.

5 more elements

A machine tool natural frequency identification system, said system comprising:

a machine tool configured for performing an operation on a workpiece;
a microphone placed proximal the machine tool;
and
a computing device in communication with the microphone, said computing device being configured to identify natural frequencies by performing steps including;
collecting audio data from at least one impact event on the machine tool;
performing a fast Fourier transform (FFT) on the audio data, to produce frequency response data for each of the impact events;
aggregating the frequency response data for all of the impact events into an aggregated frequency response;
and
identifying at least one natural frequency of vibration of the machine tool from the aggregated frequency response.

6 more elements

Dependencies are inferred from the claim text.

Audio-based modal estimation

Life of this patent

Audio-based modal estimation

Life of this patent

Abstract

Claims20