Rotating speed extraction method of joint speed reducer for industrial robot

Abstract

The invention relates to the technical field of industrial robot monitoring, in particular to a rotating speed extraction method of a joint reducer for an industrial robot, and aims to solve the problem that a rotating speed signal is difficult to directly extract from a servo motor control ring. The invention provides a rotating speed extraction method of a joint speed reducer for an industrial robot. The rotating speed extraction method comprises the steps that S1, current monitoring data of the joint speed reducer are obtained; s2, setting the resolution required by rotating speed analysis as delta F; s3, setting the range of a scale expansion and contraction factor; s4, setting a cycle index N; s5, calculating a value an of a scale expansion and contraction factor; s6, setting a bandwidth B; s7, performing Morlet wavelet transform on the current monitoring data of the joint reducer; s8, extracting the maximum frequency and amplitude corresponding to the power in the current signal; s9, calculating the rotating speed value of the speed reducer; s10, performing linear fitting to obtain a change rate k of the rotating speed along with time and an initial rotating speed q; and S11, comparing the k value with a threshold value tau to obtain the working condition of the speed reducer.

Description

Technical Field

[0001] The present invention relates to the technical field of industrial robot monitoring, and particularly relates to a method for extracting the rotational speed of a joint reducer for an industrial robot. Background Art

[0002] Current signals are widely used in the fault diagnosis of joint reducers for industrial robots. Compared with data types such as torque signals, strain signals, and vibration signals, current signals can be directly collected using Hall sensors, are not limited by the installation space of industrial robots, and avoid interfering with the task execution process of industrial robots, having great technical advantages. Although the current signal contains rich reducer state information, in the current signal, the state information of the reducer is modulated by the rotational speed signal. Therefore, the reducer state information can only be presented in a "modulated" state in the current signal. In order to further extract the state information of the reducer itself and eliminate the rotational speed modulation interference, it is first necessary to demodulate the current signal, and accurate rotational speed information is the key to accurately "demodulating" the reducer state information. In addition, in order to achieve refined diagnosis of the fault state of the joint reducer for an industrial robot, the state of the reducer is usually identified under two working conditions: variable speed and constant speed. At this time, accurate extraction of rotational speed information will be a prerequisite for accurately judging the constant speed / variable speed working conditions of the reducer. Usually, since the commonly used drive motor for industrial robots is a servo motor, its rotational speed signal, as one of the main signal flows for precise feedback control, already exists in the control link of the servo motor. However, for the sake of confidentiality of relevant control technologies, generally, its control module is not open to users, resulting in the fact that it is not realistic to directly extract the rotational speed signal from the servo motor control loop. Summary of the Invention

[0003] In view of this, the purpose of the present invention is to propose a method for extracting the rotational speed of a joint reducer for an industrial robot to solve the problem of difficultly directly extracting the rotational speed signal from the servo motor control loop.

[0004] Based on the above purpose, the present invention provides a method for extracting the rotational speed of a joint reducer for an industrial robot, including

[0005] S1: Obtain the current monitoring data of the joint reducer;

[0006] S2: Set the resolution of the rotational speed analysis requirement as ΔF;

[0007] S3: Set the range of the scale expansion factor;

[0008] S4: Set the number of cycles N;

[0009] S5: Calculate the value a of the scale expansion factor n ;

[0010] S6: Set the bandwidth B, where B = 2;

[0011] S7: Perform Morlet wavelet transform on the current monitoring data of the joint reducer;

[0012] S8: Extract the frequency and amplitude corresponding to the maximum power in the current signal;

[0013] S9: Calculate the rotational speed value of the reducer;

[0014] S10: Conduct linear fitting to obtain the rate of change k of the rotational speed with respect to time and the initial rotational speed q;

[0015] S11: Compare the k value with the threshold τ to obtain the working condition of the reducer.

[0016] As a further improvement of the present application, the current monitoring data includes the rotational speed frequency range and the sampling frequency f s and the central frequency f m .

[0017] As a further improvement of the present application, the range of the scale factor is:

[0018] a n ∈[a min , a max ,

[0019] f min is the minimum value of the input rotational speed frequency of the industrial robot joint reducer, and f max is the maximum value of the input rotational speed frequency of the industrial robot joint reducer.

[0020] As a further improvement of the present application, the number of cycles N is: N = [0, 1, 2,..., N max ,

[0021] N max is the total number of cycles.

[0022] As a further improvement of the present application, the value of the scale factor is:

[0023]

[0024] As a further improvement of the present application, when performing Morlet wavelet transform on the current monitoring data of the joint reducer, the Morlet wavelet is:

[0025]

[0026] B = 2δ 2 ,

[0027]

[0028] Among them, n is the number of fluctuations, e is a constant, j is a complex number, and t is time;

[0029] The angular frequency in the current signal of the industrial robot is

[0030] ω = 2πf,

[0031] where f is the frequency in the drive current signal of the industrial robot.

[0032] As a further improvement of the present application, performing a Fourier transform on the Morlet wavelet gives:

[0033]

[0034] where ω m is the central angular frequency.

[0035] As a further improvement of the present application, when performing a continuous wavelet transform using the Morlet wavelet, its mother wavelet function is

[0036]

[0037] where a is the scale factor, b is the translation factor, and t is time;

[0038] Obtaining:

[0039]

[0040] As a further improvement of the present application, let the number of motor pole pairs be P, then the rotational speed n of the input shaft of the reducer

[0041] , then there is Performing linear fitting, the linear regression model is expressed as: n = kt + q, where n is the rotational speed of the reducer after adding noise, and t is time;

[0042] The values of the rate of change k of the rotational speed with respect to time and the initial rotational speed q are required to minimize the sum of the squares of the perpendicular distances from all rotational speed points (t i , n i ) to the fitting line n = kt + q, that is, the optimization objective function is:

[0043]

[0044] where m is the number of rotational speed data points;

[0045] According to the calculation method for finding the extreme value of the function, the specific calculation formulas for k and b are obtained:

[0046]

[0047] The closer the k value is to 0, the less obvious the change of rotational speed with time. Therefore, the k value can accurately reflect the change of rotational speed. The q value directly reflects the specific value of the rotational speed at the corresponding time t. For constant speed, the q value is the rotational speed throughout the entire time period. For variable speed, the q value is the initial rotational speed corresponding to the moment t. Set the threshold for judging the working condition as constant speed as τ. When k is less than τ, it is judged that the joint reducer of the industrial robot is at a constant speed at this time, otherwise it is at a variable speed, and then the working condition of the reducer is obtained.

[0048] Advantages of the present invention: The present invention proposes a method for extracting the rotational speed of a joint reducer for an industrial robot based on wavelet transform, which makes full use of the multi-scale analysis ability and self-adaptability of wavelet transform, and can accurately extract the rotational speed information in the case of complex variable-speed scenarios executed by industrial robots. It solves the problem that the short-time Fourier transform currently used for rotational speed extraction is difficult to balance the frequency resolution and time resolution. Based on the requirement for extracting the rotational speed of the joint reducer for an industrial robot, the range and numerical selection of the wavelet transform scale stretching factor are determined, the rotational speed of the reducer is accurately extracted, and the extracted rotational speed is linearly fitted to obtain the linear fitting slope and intercept. According to the correlation between the fitting slope and intercept and the change of rotational speed, the constant-speed / variable-speed working condition of the reducer is judged. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following will briefly introduce the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings in the following description are only those of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0050] Figure 1 It is a flowchart of a method for extracting the rotational speed of a joint reducer for an industrial robot according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0051] In order to make the purpose, technical solutions and advantages of the present invention clearer, the following further details the present invention in conjunction with specific embodiments.

[0052] It should be noted that unless otherwise defined, the technical terms or scientific terms used in the present invention shall have the ordinary meanings understood by those with ordinary skills in the field to which the present invention pertains. The "first", "second" and similar terms used in the present invention do not denote any order, quantity or importance, but are only used to distinguish different components. The terms such as "comprising" or "including" mean that the elements or objects appearing before this term cover the elements or objects listed after this term and their equivalents, without excluding other elements or objects. The terms such as "connected" or "coupled" are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. The terms such as "upper", "lower", "left" and "right" are only used to represent relative positional relationships, and when the absolute position of the object being described changes, the relative positional relationship may also change accordingly.

[0053] As Figure 1 shown, a method for extracting the rotational speed of a joint reducer for an industrial robot includes

[0054] S1: Obtain the current monitoring data of the joint reducer; the current monitoring data includes the rotational speed frequency range, the sampling frequency f s and the central frequency f m of the wavelet.

[0055] S2: Set the resolution of the rotational speed analysis requirement to ΔF;

[0056] S3: Set the range of the scale scaling factor, a n ∈[a min , a max ,

[0057]

[0058] f min is the minimum value of the input rotational speed frequency of the industrial robot joint reducer, and f max is the maximum value of the input rotational speed frequency of the industrial robot joint reducer.

[0059] S4: Set the number of cycles N;

[0060] N = [0, 1, 2,..., N max ,

[0061] N max is the total number of cycles.

[0062] S5: Calculate the value a n of the scale scaling factor,

[0063]

[0064] S6: Set the bandwidth B, where B = 2;

[0065] S7: Perform Morlet wavelet transform on the current monitoring data of the joint reducer. The Morlet wavelet is:

[0066]

[0067] B = 2δ 2 ,

[0068]

[0069] where n is the number of fluctuations, e is a constant, j is a complex number, and t is time;

[0070] The angular frequency in the current signal of the industrial robot is

[0071] ω = 2πf,

[0072] where f is the frequency in the drive current signal of the industrial robot.

[0073] S8: Extract the frequency and amplitude corresponding to the maximum power in the current signal,

[0074] i = 0, (i = [0, 1, 2,..., m]);

[0075]

[0076] i_max = index(A max-i );

[0077] f c-i = F(i_max);

[0078] The formula for converting frequency to speed:

[0079] Judge whether i is less than m. If so, i is incremented by 1 and the calculation is repeated. If not, enter the linear regression model.

[0080] where A is the amplitude, A max-i is the maximum amplitude at time i, i-max is the serial number corresponding to A max-i , F is the frequency corresponding to all times, f c-i is the frequency corresponding to the maximum amplitude at time i, m is the total number of data, and n i is the rotational speed at time i.

[0081] Meaning of the maximum amplitude: The amplitude of the wavelet coefficient ∣W(a,b)∣ is essentially the local energy density of the signal at the time-frequency point (a,b). The larger the amplitude, the more concentrated the energy of the signal at this moment and this frequency, that is, the highest degree of matching between the wavelet and the local characteristics of the signal at this time.

[0082] The process of measuring frequency by wavelet transform is actually to continuously transform the shape of the mother wavelet (the process of wavelet transform - stretching, compressing, and translating) to match it with the measured signal, and then find the shape of the mother wavelet with the maximum amplitude corresponding to each time (the best match). The frequency corresponding to this mother wavelet shape is the measured signal frequency.

[0083] By finding the serial number of the maximum value of the wavelet corresponding amplitude at each time, and then finding the wavelet frequency in another table corresponding to this serial number. In this way, the finally measured frequency of the wavelet is obtained.

[0084] S9: Calculate the rotational speed value of the reducer, including obtaining by performing Fourier transform on the Morlet wavelet:

[0085]

[0086] where ω m is the central angular frequency.

[0087] As a further improvement of the present application, when performing continuous wavelet transform using the Morlet wavelet, its mother wavelet function is

[0088]

[0089] where a is the scale factor, b is the translation factor, and t is time;

[0090] The wavelet basis function is also called the "mother wavelet", and the wavelet shape corresponding to the operation of stretching, compressing, and translating the mother wavelet is also called the "father wavelet". Ψ a，b (t) is the expression of the mother wavelet function, which includes the independent variable t, two wavelet transform parameters a and b, and then is to expand and explain how the wavelet transform parameters are specifically used, This is to associate with the specifically selected wavelet basis function Morlet wavelet mentioned above.

[0091] Obtain:

[0092]

[0093] As a further improvement of the present application, let the number of pole pairs of the motor be P, then the rotational speed n of the input shaft of the reducer

[0094] , then there is

[0095] S10: Perform linear fitting. The linear regression model is expressed as: n = kt + q, where n is the rotational speed of the reducer after adding noise, t is time, and the rate of change k of the rotational speed with respect to time and the initial rotational speed q are obtained; the values of the rate of change k of the rotational speed with respect to time and the initial rotational speed q are required to minimize the sum of the squares of the perpendicular distances from all rotational speed points (t i , n i ) to the fitting line n = kt + q. That is, the optimization objective function is:

[0096]

[0097] where m is the number of rotational speed data points;

[0098] According to the calculation method for the function to take the extreme value, the specific calculation formulas for k and b are obtained:

[0099]

[0100] S11: Compare the value of k with the threshold τ to obtain the operating condition of the reducer.

[0101] The closer the value of k is to 0, the less obvious the change of the rotational speed with respect to time. Therefore, the value of k can accurately reflect the change of the rotational speed. The value of q directly reflects the specific value of the rotational speed at the corresponding time t. For constant speed, the value of q is the rotational speed throughout the time period. For variable speed, the value of q is the initial rotational speed corresponding to the time t. Set the threshold for determining the operating condition as constant speed to be τ. When k is less than τ, it is determined that the joint reducer of the industrial robot is at a constant speed at this time, otherwise it is at a variable speed, and then the operating condition of the reducer is obtained.

[0102] Those of ordinary skill in the art should understand that: The discussion of any of the above embodiments is only exemplary and is not intended to imply that the scope of the present invention (including the claims) is limited to these examples; Under the concept of the present invention, the technical features in the above embodiments or different embodiments can also be combined, and the steps can be implemented in any order, and there are many other variations in different aspects of the present invention as described above, which are not provided in detail for the sake of brevity.

[0103] The present invention aims to cover all such substitutions, modifications, and variations that fall within the broad scope of the appended claims. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for extracting the rotational speed of a joint reducer for an industrial robot, characterized in that, including S1: Obtain the current monitoring data of the joint reducer; S2: Set the resolution of the rotational speed analysis requirement as ΔF; S3: Set the range of the scale factor; S4: Set the number of cycles N; S5: Calculate the value a of the scale factor n ; S6: Set the bandwidth B, B = 2; S7: Perform Morlet wavelet transform on the current monitoring data of the joint reducer; S8: Extract the frequency and amplitude corresponding to the maximum power in the current signal; S9: Calculate the rotational speed value of the reducer; S10: Perform linear fitting to obtain the change rate k of the rotational speed with time and the initial rotational speed q; S11: Compare the k value with the threshold τ to obtain the working condition of the reducer.

2. The rotational speed extraction method of the joint speed reducer for industrial robots according to claim 1, characterized in that The current monitoring data includes a rotational speed frequency range, a sampling frequency fs, and the central frequency f of the wavelet m .

3. The rotational speed extraction method of the joint speed reducer for industrial robots according to claim 2, characterized in that The range of the scale factor is: f min is the minimum value of the input rotational speed frequency of the industrial robot joint reducer, f max is the maximum value of the input rotational speed frequency of the industrial robot joint reducer.

4. The rotational speed extraction method of the joint speed reducer for industrial robots according to claim 3, characterized in that The number of cycles N is: N = [0, 1, 2,..., N max , N max is the total number of cycles.

5. The rotational speed extraction method of the joint speed reducer for industrial robots according to claim 4, characterized in that The value of the scale factor is:

6. The rotational speed extraction method of the joint speed reducer for industrial robots according to claim 2, characterized in that, For performing Morlet wavelet transform on the current monitoring data of the joint reducer, the Morlet wavelet is: B = 2δ 2 , where n is the number of fluctuations, e is a constant, j is a complex number, and t is time; The angular frequency in the industrial robot current signal is ω = 2πf, where f is the frequency in the industrial robot drive current signal.

7. The rotational speed extraction method of the joint speed reducer for industrial robots according to claim 6, characterized in that Performing Fourier transform on the Morlet wavelet gives: Among them, ω m is the central angular frequency.

8. The rotational speed extraction method of the joint speed reducer for industrial robots according to claim 7, characterized in that When using Morlet wavelet for continuous wavelet transform, its mother wavelet function is where a is the scale factor, b is the translation factor, and t is time; Obtain:

9. The rotational speed extraction method of the joint speed reducer for industrial robots according to claim 7, characterized in that Let the number of pole pairs of the motor be P, then the rotational speed n of the input shaft of the reducer, Then there is Perform linear fitting, and the linear regression model is expressed as: n = kt + q, where n is the rotational speed of the reducer after adding noise, and t is time; The value requirements of the rate of change of rotational speed k with respect to time and the initial rotational speed q are such that the sum of the squares of the perpendicular distances from all rotational speed points (t i , n i ) to the fitting line n = kt + q is minimized. That is, the optimization objective function is: where m is the number of rotational speed data points; According to the calculation method of taking the extreme value of the function, obtain the specific calculation formulas for k and b: The closer the k value is to 0, the less obvious the change of the rotational speed with time. Therefore, the k value can accurately reflect the change of the rotational speed. The q value directly reflects the specific value of the rotational speed at the corresponding time t. For constant speed, the q value is the rotational speed in the whole time period. For variable speed, the q value is the initial rotational speed corresponding to the time t. Set the threshold for judging the working condition as constant speed as τ. When k is less than τ, it is judged that the industrial robot joint reducer is in constant speed at this time, otherwise it is in variable speed, and then obtain the working condition of the reducer.

Images