Fault detection methods for power tools
A method performed in a control unit of a power tool driven by an electric motor, for detecting a fault state in the power tool, wherein the electric motor is associated with a motor current drawn over a motor interface, the method comprising monitoring a parameter of the electric motor during operation of the power tool, wherein the parameter is indicative of the motor current, obtaining a fault model, wherein the fault model is configured to classify a state of the power tool into a pre-determined number of states comprising one or more fault states, based on a series of electric motor parameter values, classifying a state of the power tool into the pre-determined number of states based on the fault model and on one or more values of the monitored electric motor parameter, and triggering an action by the power tool in case the state of the power tool is classified as a fault state.
1. A method performed in a control unit of a power tool driven by an electric motor, for detecting a fault condition in the power tool, wherein the electric motor is associated with a motor current drawn over a motor interface, the method comprising
monitoring the motor current of the electric motor drawn over the motor interface during operation of the power tool,
obtaining a fault model, wherein the fault model is configured to classify a state of the power tool into a pre-determined number of power tool states comprising one or more fault states, based on a series of electric motor current values,
classifying a state of the power tool into the pre-determined number of states based on the fault model and on one or more values of the monitored electric motor current, and
triggering an action by the power tool in case the state of the power tool is classified as a fault state.
2. The method according to claim 1 , further comprising initially training the fault model using recorded values of monitored electric motor currents corresponding to the one or more fault states.
3. The method according to claim 1 , wherein the monitored current of the electric motor comprises a D-Q transformed motor current.
4. The method according to claim 3 , further comprising monitoring meta data associated with the D-Q transformed motor current, wherein the meta data comprises any of frequency width, relative magnitude, frequency sub-band power, and frequency sub-band entropy of a Fourier transformed representation of the D-Q transformed motor current.
5. The method according to claim 4 , further comprising adjusting a sample or window size of the Fourier transformed representation in dependence of a motor speed.
6. The method according to claim 1 , further comprising monitoring an output from any of an inertial measurement unit, IMU, a temperature sensor, a dust sensor, and/or a vision-based sensor.
7. The method according to claim 1 , wherein the fault model is based on a random forest ensemble learning method, or a neural network.
8. The method according to claim 1 , comprising configuring the fault model in dependence of a type of power tool.
9. The method according to claim 1 , wherein each fault state is associated with one out of a pre-determined number of severity levels.
10. The method according to claim 9 , wherein the triggered action comprises notifying an operator of the fault state and its associated severity level, or disabling the power tool in case the current state is classified as a fault state associated with a pre-determined severity level.
11. The method according to claim 10 , wherein disabling the power tool comprises executing an emergency brake by the electric motor.
12. The method according to claim 1 , wherein the method further comprises receiving an updated fault model from an external entity and replacing the fault model by the updated fault model.
13. The method according to claim 1 , comprising storing the recorded values of the monitored electric motor current in a memory module, or
configuring an operating parameter of the power tool in dependence of a detected fault state.
14. The method according to claim 13 , comprising configuring a tool speed and/or segment pressure in dependence of a detected fault state associated with tool glazing.
15. The method according to claim 14 , comprising configuring a pulsed drive of the electric motor pulse in dependence of a detected fault state associated with a non-circular cutting disc shape.
16. The method according to claim 1 , wherein the power tool is a hand-held power tool and/or a concrete processing tool comprising a rotatable abrasive member for processing concrete.
17. A control unit for a power tool comprising processing circuitry configured to execute the method according to claim 1 , wherein the processing circuitry comprises a memory module, and wherein the memory module is arranged to store a time history of the monitored current values of the electric motor.
18. A power tool comprising the control unit according to claim 17 , wherein the power tool is a cut-off tool, a core drill, or a wall saw.
19. A control unit for a power tool driven by an electric motor, wherein the electric motor is associated with a motor current drawn over a motor interface,
the control unit comprising a memory module arranged to store a plurality of fault models corresponding to respective rotatable concrete work tools, where each fault model in the plurality of fault models is configured to classify a state of a respective type of rotatable work tool into a pre-determined number of states comprising one or more fault states, based on a series of electric motor current values,
wherein the control unit is arranged to select a fault model from the plurality of fault models in dependence of a current type of rotatable work tool attached to the power tool,
wherein the control unit is arranged to monitor the motor current of the electric motor drawn over the motor interface during operation of the power tool, and
classify a state of the power tool into the pre-determined number of states based on the fault model and on one or more values of the monitored electric motor current.
20. A method performed in a control unit of a power tool driven by an electric motor, for detecting a fault condition in the power tool, wherein the electric motor is associated with a motor current drawn over a motor interface, the method comprising
obtaining a fault model, wherein the fault model is configured to classify a state of the power tool into a pre-determined number of power tool states comprising one or more fault states, based on a comparison of a series of electric motor current values drawn over the motor interface during operation of the power tool to a corresponding series of electric motor current values drawn over a motor interface of a power tool having suffered a known fault condition, and
classifying a state of the power tool into the pre-determined number of states based on the fault model and on one or more values of the monitored electric motor current.