# US 12,761,630 B2 · Method and system for using domain knowledge to synchronize signals related to a machine or a machining process > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,761,630 B2 - Document ID: US12761630B2 - Status: Granted - Kind: B2 - Application number: 17/840,644 - Filed: 2022-06-15 - Published: 2022-09-29 - Granted: 2026-10-13 - Earliest claimed: 2019-12-20 - Assignees: TRUMPF Werkzeugmaschinen GmbH + Co. KG - Applicants: TRUMPF Werkzeugmaschinen SE + Co. KG - Inventors: Manuel Kiefer; Thomas Kieweler; Martin Lukas; Martin Schober - Web page: https://patentgrep.com/patent/US12761630B2 ## CPC - G05B 19/4155: characterised by program execution, i.e. part program or machine function execution, e.g. selection of a program - G05B 2219/34397: Synchronize manipulators and machine by using a reference clock for all - G05B: CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS (systems for controlling or regulating non-electric variables G05D; systems for regulating electric or magnetic variables G05F; control devices or systems insofar as characterised by mechanical features only G05G) ## Abstract A method for synchronizing signals which are related to a machine and/or a machining process, including recording data of a first data source to obtain a first signal track, recording data of at least one second data source, which is independent of the first data source, to obtain at least one second signal track, analyzing the first and second signal tracks based on previously known domain knowledge, and temporally connecting the first and second signal tracks. ## Claims ### Claim 1 A method for synchronizing signal tracks recording temporal data from data sources, the signal tracks being related to a machine or a machining process performed using the machine, the method comprising: recording data of a first data source to obtain a first signal track; recording data of a second data source, which is independent of the first data source, to obtain a second signal track; manipulating the machine by physically exciting a mechanical resonance point of the machine in a manner targeted to translate into an expected result in the first signal track and the second signal track; analyzing the first signal track and the second signal track based on previously known domain knowledge to determine a temporal synchronicity between the first signal track and the second signal track, the previously known domain knowledge comprising data for recognizing the expected result of exciting the mechanical resonance point in the manner targeted; and temporally connecting the first signal track and the second signal track based on the determined temporal synchronicity. ### Claim 2 (depends on 1) The method as claimed in claim 1, wherein temporally connecting the first signal track and the second signal track comprises assigning the first signal track and the second signal track to a common time axis. ### Claim 3 (depends on 1) The method as claimed in claim 1, wherein the previously known domain knowledge comprises models, and wherein the analyzing of the first signal track and the second signal track is performed based on the models. ### Claim 4 (depends on 1) The method as claimed in claim 1, wherein the previously known domain knowledge comprises reference patterns, wherein the analyzing of the first signal track and the second signal track comprises performing pattern recognition based on the reference patterns. ### Claim 5 (depends on 1) The method as claimed in claim 1, wherein the first signal track and the second signal track are recorded without temporally synchronizing data recordings of the first data source and the second data sources. ### Claim 6 (depends on 5) The method of claim 5, wherein neither of the first signal track nor the second signal track comprise a time stamp. ### Claim 7 (depends on 1) The method as claimed in claim 1, wherein the first data source is inside the machine and the second data source is outside the machine. ### Claim 8 (depends on 1) The method as claimed in claim 1, wherein a period in which the data of the first data source is recorded overlaps with a period in which the data of the second data source is recorded. ### Claim 9 (depends on 1) The method as claimed in claim 1, wherein each of the first signal track and the second signal track comprises a predetermined number of data points. ### Claim 10 (depends on 1) The method as claimed in claim 1, wherein the recorded data of each of the first data source and the second data source is time-normalized per se and contains Nyquist criterion. ### Claim 11 (depends on 1) The method as claimed in claim 1, the method further comprising carrying out a fault identification, fault diagnosis, state monitoring, or predictive maintenance by analyzing the temporally connected first signal track and second signal track. ### Claim 12 (depends on 1) The method of claim 1, wherein the previously known domain knowledge further comprises: knowledge correlating a relationship between individual components of the machine or used in the machining process, which comprises a relationship of a vibration excitation by machine components, axis dynamics, or an absolute position of a kinematic chain; the machining process; properties of components used in the machining process; reference patterns comprising expected patterns in the first signal track or the second signal track from a known excitation, the known excitation comprising the exciting of the mechanical resonance point or an excitation signal, which comprises an input variable manipulation; previously recorded measurement data; simulation results from a model, historical data of the machine or from other machines; or a temporal relation between the known excitation and the first signal track or the second signal track. ### Claim 13 (depends on 1) The method of claim 1, wherein analyzing the first signal track and the second signal track based on previously known domain knowledge to determine the temporal synchronicity between the first signal track and the second signal track comprises analyzing the first signal track and the second signal track based on the data for recognizing the expected result of exciting the mechanical resonance point in the manner targeted, a known temporal relation between the first signal track and the excitation of the mechanical resonance point, and a known temporal relation between the second signal track and the excitation of the mechanical resonance point to assign the first signal track and the second signal track to a common time axis. ### Claim 14 (depends on 1) The method of claim 1, wherein after the first signal track and the second signal track are temporally connected to generate temporally connected signal tracks, the method further comprises carrying out a fault identification, a fault diagnosis, state monitoring, a predictive maintenance analysis by analyzing the temporally connected signal tracks in real-time. ### Claim 15 (depends on 1) The method of claim 1, wherein after the first signal track and the second signal track are temporally connected to generate temporally connected signal tracks, the method further comprises analyzing the temporally connected signal tracks based on the domain knowledge. ### Claim 16 (depends on 15) The method of claim 15, wherein analyzing the temporally connected signal tracks based on the domain knowledge comprises analyzing whether the expected result of exciting the mechanical resonance point in the manner targeted is recognized in the first signal track and the second signal track to determine a possible fault source. ### Claim 17 (depends on 1) The method of claim 1, wherein analyzing the first signal track and the second signal track based on previously known domain knowledge to determine the temporal synchronicity between the first signal track and the second signal track comprises analyzing the first signal track and the second signal track based on the data for recognizing the expected result of exciting the mechanical resonance point in the manner targeted, a known temporal relation between the first signal track and the excitation of the mechanical resonance point, and a known temporal relation between the second signal track and the excitation of the mechanical resonance point to assign the first signal track and the second signal track to a common time axis, wherein after the first signal track and the second signal track are temporally connected to generate temporally connected signal tracks, the method further comprises performing a time-frequency transformation on the temporally connected signal tracks to generate transformed signal tracks having harmonized spectra; carrying out a fault identification by analyzing the transformed signal tracks in view of the domain knowledge to determine where a fault is present; and eliminating the fault. ### Claim 18 A system for synchronizing signal tracks recording temporal data from data sources, the system comprising: an analysis device comprising a processor configured to execute a method for synchronizing the signal tracks and memory storing previously obtained domain knowledge, which are related to a machine or a machining process performed using the machine, the method comprising: recording data of a first data source to obtain a first signal track; recording data of a second data source, which is independent of the first data source, to obtain a second signal track; manipulate the first data source or second data source to physically excite a mechanical resonance point of the machine in a manner targeted to translate into an expected result in the first signal track and the second signal track; analyzing, using the processor, the first signal track and the second signal track based on the previously obtained domain knowledge to determine a temporal synchronicity between the first signal track and the second signal track, the previously known domain knowledge comprising data for recognizing the expected result of exciting the mechanical resonance point in the manner targeted; and temporally connecting the first signal track and the second signal track based on the determined temporal synchronicity. ### Claim 19 (depends on 18) The system as claimed in claim 18, wherein the first data source is arranged inside the machine and the second data source is arranged outside the machine, wherein the first data source comprises a sensor, a controller, or a drive of the machine, and wherein the second data source comprises a sensor, a controller, a camera, or a microphone. ### Claim 20 A method for operating a machine, the method comprising: performing a machining process by controlling the machine based on a first data source, the first data source comprising input variables or output variables of a controller; while performing the machine process: recording data of the first data source to obtain a first signal track, recording data of a second data source, which is independent of the first data source, to obtain a second signal track, and physically exciting a mechanical resonance point of the machine in a manner targeted to translate into an expected result in the second signal track by manipulating the first data source; analyzing the first signal track and the second signal track based on previously known domain knowledge to determine a temporal synchronicity between the first signal track and the second signal track, the previously known domain knowledge comprising data for recognizing the expected result of exciting the mechanical resonance point in the manner targeted; and temporally connecting the first signal track and the second signal track based on the determined temporal synchronicity such that the first signal track and the second signal tracks are synchronized.