Method and system for monitoring the state of a device
AssigneeSchaeffler Technologies AG & Co. KG
InventorsHarald Reiners, Oliver Webber, Matti Seppä
Life of this patent
Estimating the term…
Abstract
The disclosure relates to a method of and system for monitoring the state of a device that includes a rotatable component. The system includes a structure-borne noise meter and an evaluation device. The evaluation device ascertains a time range feature for a frequency band, compares the time range feature with at least one threshold to determine a time range feature state, and determines the state of the device based on the time range feature state.
Claims20
A method for monitoring a state of a device having a rotatable component, comprising the steps of:
- measuring a structure-borne noise signal of the device during operation of the device;
- ascertaining a first time range feature for a first frequency band of the structure-borne noise signal;
- ascertaining a second time range feature for a second frequency band of the structure-borne noise signal;
- comparing the first time range feature with an upper first threshold and a lower first threshold to determine a first time range feature state, wherein the first time range feature state is classified as a normal state if the first time range feature is below the lower first threshold and below the upper first threshold, the first time range feature state is classified as an elevated state if the first time range feature is above the lower first threshold and below the upper first threshold, and the first time range feature state is classified as a high state if the first time range feature is above the lower first threshold and above the upper first threshold;
- comparing the second time range feature with an upper second threshold and a lower second threshold to determine a second time range feature state, wherein the second time range feature state is classified as a normal state if the second time range feature is below the lower second threshold and below the upper second threshold, the second time range feature state is classified as an elevated state if the second time range feature is above the lower second threshold and below the upper second threshold, and the second time range feature state is classified as a high state if the second time range feature is above the lower second threshold and above the upper second threshold;
- and
- determining the state of the device based on the first time range feature state and the second time range feature state, wherein the first frequency band is a lower frequency band of the structure-borne noise signal and the second frequency band is an upper frequency band of the structure-borne noise signal, and wherein the first frequency band is below the second frequency band and the first frequency band and the second frequency band are spaced apart from one another.
A system for monitoring a state of a device that has a rotatable component, comprising:
- a structure-borne noise meter that measures a structure-borne noise signal of the device;
- and
- an evaluation device that:
- ascertains, during operation of the device in a training phase, a first time range feature of a first frequency band of the structure-borne noise signal, wherein the first frequency band is a lower frequency band of the structure-borne noise signal and the first time range feature is a mean square deviation of the first frequency band;
- ascertains, during operation of the device in the training phase, a second time range feature of a second frequency band of the structure-borne noise signal, wherein the second frequency band is an upper frequency band of the structure-borne noise signal that is spaced apart from and above the first frequency band, and wherein the second time range feature is a mean square deviation of the second frequency band;
- ascertains, during operation of the device, the first time range feature for the first frequency band of the structure-borne noise signal;
- ascertain, during the operation of the device, the second time range feature for the second frequency band of the structure-borne noise signal;
- compares the first time range feature of the first frequency band ascertained during the operation of the device, with at least one of an upper first threshold and a lower first threshold to determine a first time range feature state, wherein the upper first threshold and the lower first threshold are each determined based on the first time range feature ascertained during operation of the device in the training phase;
- compares the second time range feature of the second frequency band ascertained during the operation of the device, with at least one of an upper second threshold and a lower second threshold to determine a second time range feature state, wherein the upper second threshold and the lower second threshold are each determined based on the second time range feature ascertained during operation of the device in the training phase;
- and
- determines the state of the device based on at least one of the first time range feature state and the second time range feature state.
A method for monitoring a state of a device having a rotatable component, comprising the steps of:
- measuring a structure-borne noise signal during operation of the device in a training phase;
- ascertaining a first time range feature of a first frequency band of the measured structure-borne noise signal;
- ascertaining a second time range feature of a second frequency band of the measured structure-borne noise signal;
- determining at least one first threshold based on the ascertained first time range feature;
- and
- determining at least one second threshold based on the ascertained second time range feature, wherein the first frequency band is a lower frequency band of the structure-borne noise signal and the second frequency band is an upper frequency band of the structure-borne noise signal, and wherein the first frequency band is below the second frequency band and the first frequency band and the second frequency band are spaced apart from one another.
Dependencies are inferred from the claim text.