METHOD FOR CHECKING AN ASSEMBLY OF AT LEAST THREE STRAIN GAUGES AND STRAIN WAVE GEARING

Abstract

An assembly includes at least three strain gauges and is attached to an elastic transmission element of a strain wave gearing. The assembly is designed to measure a torque acting on the elastic transmission element. Output signals from each of the strain gauges are measured. The output signal of one of the strain gauges is predicted from the measured output signals of the other strain gauges. An error message is output based on the predicted output signal deviating from the respective measured output signal by more than a predetermined tolerance.

Claims

1. A method for checking an assembly of at least three strain gauges, the assembly including the at least three strain gauges being attached to an elastic transmission element of a strain wave gearing and designed to measure a torque acting on the elastic transmission element, and the method comprising: measuring output signals from each of the strain gauges; predicting the output signal from one of the strain gauges based on the measured output signals of the other strain gauges; and outputting an error message based on the predicted output signals deviating from the respective measured output signal by more than a predetermined tolerance.

2. The method according to claim 1, wherein the assembly includes at least four of the strain gauges.

3. The method according to claim 2, further comprising predicting the output signals from each of the strain gauges based on the measured output signals from the other strain gauges.

4. The method according to claim 1, further comprising predicting the output signals from each of the strain gauges based on the measured output signals from the other strain gauges.

5. The method according to claim 1, further comprising selecting the predetermined tolerance based on a function of a rotational speed of the elastic transmission element.

6. The method according to claim 1, further comprising determining the torque acting on the elastic transmission element from the output signals of the strain gauges based on none of the predicted output signals deviating from the respective measured output signals by more than the predetermined tolerance.

7. The method according to claim 1, further comprising predicting the output signal of the strain gauge from the measured output signals of the other strain gauges based on a mathematical prediction model, wherein the mathematical prediction model is determined by a finite element analysis or by a machine learning method.

8. A strain wave gearing, comprising: an elastic transmission element; an assembly of at least three strain gauges, the assembly being applied to the elastic transmission element and being designed to measure a torque acting on the elastic transmission element; and a microcontroller configured to: measure output signals from each of the strain gauges; predict the output signal from one of the strain gauges based on the measured output signals of the other strain gauges; and output an error message based on the predicted output signal deviating from the respective measured output signal by more than a predetermined tolerance.

9. The strain wave gearing according to claim 8, wherein the strain gauges extend around a circumference on a lateral surface or an axial side surface of the elastic transmission element.

10. The strain wave gearing according to claim 8, wherein the strain gauges are each formed as a coating directly on a metallic surface of the elastic transmission element.

11. The method according to claim 3, further comprising outputting the error message based on at least one of the predicted output signals deviating from the respective measured output signal by more than the predetermined tolerance.

12. The method according to claim 4, further comprising outputting the error message based on at least one of the predicted output signals deviating from the respective measured output signal by more than the predetermined tolerance.

13. The method according to claim 1, wherein the strain gauges extend around a circumference on a lateral surface or an axial side surface of the elastic transmission element.

14. The method according to claim 1, wherein the strain gauges are each formed as a coating directly on a metallic surface of the elastic transmission element.

15. The strain wave gearing according to claim 8, wherein the microcontroller is further configured to output the error message based on at least one of the predicted output signals deviating from the respective measured output signal by more than the predetermined tolerance.

16. The strain wave gearing according to claim 8, wherein the microcontroller is further configured to select the predetermined tolerance based on a function of a rotational speed of the elastic transmission element.

17. The strain wave gearing according to claim 8, wherein the microcontroller is further configured to determining the torque acting on the elastic transmission element from the output signals of the strain gauges based on none of the predicted output signals deviating from the respective measured output signals by more than the predetermined tolerance.

18. The strain wave gearing according to claim 8, wherein the microcontroller is further configured to predict the output signal of the strain gauge from the measured output signals of the other strain gauges based on a mathematical prediction model, wherein the mathematical prediction model is determined by a finite element analysis or by a machine learning method.

19. The strain wave gearing according to claim 8, wherein the assembly includes at least four of the strain gauges.

20. The strain wave gearing according to claim 19, wherein the microcontroller is further configured to output the error message based on at least one of the predicted output signals deviating from the respective measured output signal by more than the predetermined tolerance.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Additional advantages, details, and further developments of the present disclosure will become apparent from the below description of an exemplary embodiment, with reference to the drawing. In the figures:

[0035] FIG. 1 shows a flowchart of an exemplary embodiment of a method according to the present disclosure;

[0036] FIG. 2 shows two steps of the method shown in FIG. 1 in detail; and

[0037] FIG. 3 shows a conditional inequality used in the method shown in FIG. 1.

DETAILED DESCRIPTION

[0038] FIG. 1 shows a flowchart of an exemplary embodiment of a method according to the present disclosure, which is used to check an assembly of at least three strain gauges (not shown). The assembly may comprise four of the strain gauges. The assembly, comprising the four strain gauges, is applied to an elastic transmission element (not shown) of a strain wave gearing (not shown). This embodiment of the method also allows a torque Mt acting on the elastic transmission element (not shown) to be measured. This measurement forms the original function of the assembly comprising the four strain gauges.

[0039] In a first step, output signals of all four strain gauges (not shown) are detected. For this purpose, output voltages from the strain gauges (not shown) are converted using an analog to digital (A/D) converter (not shown) and processed using a microcontroller. In the next step, the torque Mt acting on the elastic transmission element (not shown) is determined from all four measured output signals (shown in detail in FIG. 2). In a further step, output signals of the individual strain gauges (not shown) are predicted from the measured output signals of the other three strain gauges (not shown) (shown in detail in FIG. 2), so that the measured output signal and the predicted output signal are available for the strain gauges. In a next step, it is checked whether the four measured output signals are each within a tolerance range around the corresponding predicted output signal (shown in detail in FIG. 3). If this is not the case, there is an error in the strain gauges (not shown) and a status byte is set to an error identifier, which is 0xFF, for example. The error will usually be a failure or defect in one of the four strain gauges (not shown). If the four measured output signals are each in a tolerance range around the corresponding predicted output signal, there is no error in the strain gauges (not shown), such that the previously determined value of the torque Mt is output and used and the status byte is set to an OK identifier, which is 0x00, for example. The method is then repeated.

[0040] FIG. 2 shows two steps of the method shown in FIG. 1 in detail. First, it is shown how the torque Mt acting on the elastic transmission element (not shown) is determined from the four measured output signals K0, K1, K2, K3. For this purpose, an algorithm is used which is based on a mathematical prediction model, which is determined, for example, by a finite element analysis or by a machine learning method. In the case of machine learning, the algorithm is trained using a plurality of training data before it is used. All of the measured output signals are used to determine the torque. It also shows how the predicted values Pk0, Pk1, Pk2, Pk3 of the output signals are determined from the four measured output signals K0, K1, K2, K3. For this purpose, in turn, an algorithm is used in each case, which is based on a mathematical prediction model, which is determined, for example, by a finite element analysis or by a machine learning method. In the case of machine learning, the algorithm(s) are each trained using multiple training data before they are used. The four predicted values Pk0, Pk1, Pk2, Pk3 of the output signals of the four strain gauges (not shown) are determined from the measured output signals K0, K1, K2, K3 of the other three strain gauges (not shown).

[0041] FIG. 3 shows a conditional inequality which is applied in the method shown in FIG. 1 in order to decide whether the four measured output signals K0, K1, K2, K3 are each within a tolerance range around the corresponding predicted output signal Pk0, Pk1, Pk2, Pk3. For each of the four strain gauges (not shown), it is checked whether the measured output value K0, K1, K2, K3 thereof is within the tolerance range around the value Pk0, Pk1, Pk2, Pk3 of the corresponding predicted output signal. A tolerance value T is predefined for this purpose, so that the tolerance range is ±T. Only when the inequality is satisfied for each of the four strain gauges (not shown) is it concluded that the assembly (not shown) comprising the strain gauges is in a fault-free state.