A method and system for automatic switching of control parameters of a textile machine drive

Abstract

The present application relates to a kind of textile machinery drive control parameter automatic switching method and system.Use vibration detector to detect the working state of each frequency modulation motor, and the signal collected is analyzed, and features are extracted, so as to realize the collection of frequency modulation motor working state;At the same time, the control parameter is matrixed, and the corresponding adjustment function is calculated, so that the change of each working state matrix under the adjustment of adjustment function can be calculated and predicted;From this, the final working state caused by each control parameter matrix can be obtained;From it, the shortest adjustment time can be selected to achieve the fastest parameter adjustment, ensure the yield rate and reaction speed of production line;Reduce the adjustment times, and only single adjustment can achieve fast adjustment to the best working state.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of textile control, and particularly relates to a textile machine transmission control parameter automatic switching method and system. BACKGROUND

[0002] Textile machinery is a kind of mechanical equipment needed for processing natural fibers or chemical fibers into textiles. Although the machinery for producing chemical fibers includes various chemical machinery, it is now considered as an extension of textile machinery and belongs to the broad sense of textile machinery. A large number of transmission rollers are used in textile machinery, and the working parameters of each transmission roller will affect the yield of the final product. Therefore, it is necessary to monitor the working state of each transmission mechanism in real time to ensure the yield of the product.

[0003] Application No. CN201810048712.9 discloses a embroidery machine for textile, which comprises a textile machine body, a ball screw is arranged at the upper end of the textile machine body, the left end of the ball screw is tightly welded with the textile machine body, a transmission shaft sleeve is arranged at the right side of the textile machine body and is movably connected with the textile machine body, a roller is arranged in the textile machine body and the left end of the roller is fixedly connected with the inner side of the textile machine body. The embroidery machine for textile comprises a textile machine body and a servo motor. The internal instruments of the servo motor are rotated through the work of the servo motor. The servo motor and the textile machine body are connected by wires, which can quickly respond to drive the operation of the entire textile machine. However, this scheme does not disclose how to adjust the working parameters of the transmission mechanism to ensure the yield.

[0004] In the prior art, the control parameters need to be adjusted and converted frequently. If the control parameter matrix is frequently converted, the frequency conversion motor will frequently convert the working state, which will gradually accumulate the deviation caused by accidental factors, so that the adjustment of the production line by the controller gradually deviates from the control, resulting in a decrease in the yield. However, reducing the number of control parameter conversions will easily lead to the failure to adjust the parameters to the best at one time. Therefore, there is a need for a method and system that can adjust the control parameters at one time and ensure that the production line reaches the best in the shortest time. SUMMARY

[0005] In view of the above, to solve the above problems, a textile machine transmission control parameter automatic switching system is provided, which comprises a main controller, a variable frequency motor, a vibration detector and a detection controller.

[0006] The main controller is connected with the variable frequency motor, and the main controller adjusts the voltage U, the current I and the frequency f of the variable frequency motor. The vibration detector is arranged on the variable frequency motor and is used for detecting the vibration data of the variable frequency motor. The vibration detector sends the collected vibration signal to the detection controller. The detection controller performs analog-to-digital conversion on the vibration signal to obtain digital waveform data of the vibration signal.

[0007] The detection controller sends the digital waveform data of the vibration signal to the main controller, the main controller processes the waveform data of all vibration detectors collected to obtain the working state matrix T of the variable frequency motor, and inputs the working state matrix T into the adaptive control model; the adaptive control model outputs the corresponding control parameter matrix M according to the input working state matrix T; the detection controller outputs the control parameter matrix M to the variable frequency motor, so as to complete the automatic switching of the control parameters of the variable frequency motor.

[0008] The control parameter matrix M corresponding to the time when the working state matrix T reaches the standard working state matrix T0 under the adjustment of the control parameter matrix M is calculated in the main controller, that is, the control parameter matrix output by the adaptive control model.

[0009] The specific calculation method of the adaptive control model is:

[0010] The control parameter matrix of the main controller is M,

[0011]

[0012] Where n is the number of motors;

[0013] The sampling rate of the vibration signal of the variable frequency motor collected by the vibration detector is 20-30 kHz, and the form of the collected vibration signal is the data of vibration acceleration changing with time;

[0014] The main controller extracts features from the digital waveform data, and constructs the working state matrix T1 using the features extracted from all variable frequency motors;

[0015]

[0016] Where m features are extracted, n is the number of motors, and λ is the value of the extracted features;

[0017] Let the adjustment function be G(M), where M is the adjustment parameter matrix, and G(M) is a function of the matrix M;

[0018] And T1·G(M)=T2;

[0019] Where T1 is the working state matrix at the current time, and T2 is the working state matrix at the next minute;

[0020]

[0021] Where s i , p i , q i (i=1…n) are constant coefficients;

[0022] Let the standard working state matrix be T0, when the adjustment is effective, there is T1·Gk (M) = T0, that is, the working state matrix T1 at the current time will become the standard state matrix T0 after being adjusted by the adjustment function G(M) for k times, wherein k is the number of minutes passed;

[0023] A plurality of control parameter matrices are pre-stored in the main controller, and the adjustment function G(M) of each control parameter matrix can be calculated; the control parameter matrices are traversed in the main controller, and the number of times required for the working state matrix at the current time to reach the standard working state matrix T0 under the adjustment of each adjustment function is calculated;

[0024] When traversing the parameter matrix, the operation speed is relatively fast due to direct use of computers for matrix operation, and the calculation amount is relatively small compared to neural networks, random forests and the like, which is more suitable for lower controllers; at the same time, due to the relatively fast calculation speed, the reaction time of the system is also relatively fast, further improving the adjustment speed.

[0025] The control parameter matrix corresponding to the adjustment function with the least number of required adjustments is the control parameter matrix output by the adaptive control model.

[0026] The time length of the vibration signal collected by the vibration detector is 5s, and the sampling rate is 20-30kHz, that is, the time length of collecting a vibration signal is 5 seconds, and 20000-30000 samples are taken per second;

[0027] The main controller first performs Fourier transform on the waveform data to convert the time domain signal into a frequency domain signal to obtain the spectrum corresponding to the vibration waveform data;

[0028] The main controller extracts features from the converted spectrum, and the number of extracted features is 7, which are as follows:

[0029] λ 1n is the peak value of the first characteristic frequency, λ 2n is the half-height width of the first characteristic frequency, λ 3n is the peak value of the second characteristic frequency, λ 4n is the half-height width of the second characteristic frequency, λ 5n is the extreme value of the spectrum, λ 6n is the variance of the spectrum, λ 7n is the skewness of the spectrum;

[0030] The first characteristic frequency is 24.8Hz-25.7Hz, and the second characteristic frequency is 1.48kHz-1.52kHz.

[0031] A method for automatically switching the control parameters of a textile machinery transmission using the above system, comprising the following steps:

[0032] Step 1, first, the main controller loads the initial control parameter matrix to the frequency conversion motor, and meanwhile, the vibration detector starts to detect the working state of each frequency conversion motor and sends the collected vibration signal to the detection controller;

[0033] Step 2, the detection controller performs analog-to-digital conversion on the vibration signal to obtain a digital waveform signal, and sends the digital waveform signal corresponding to each frequency conversion motor to the main controller;

[0034] Step 3, the main controller performs Fourier transform on the waveform data to convert the time domain signal into a frequency domain signal to obtain a spectrum corresponding to the vibration waveform data; the main controller extracts features from the converted spectrum to obtain a working state matrix T1 at the current time;

[0035] Step 4, the main controller pre-stores a plurality of control parameter matrices and can calculate an adjustment function G(M) of each control parameter matrix; the main controller iterates the control parameter matrix and calculates the number of times required for the working state matrix at the current time to reach a standard working state matrix T0 under the adjustment of each adjustment function; the control parameter matrix corresponding to the adjustment function with the least number of required adjustments is the control parameter matrix output by the adaptive control model;

[0036] Step 5, the main controller completes the automatic switching of the frequency conversion motor control parameter under the control parameter matrix output by the adaptive control model.

[0037] The vibration detector is an acceleration vibration sensor.

[0038] The sampling rate of the vibration signal of the frequency conversion motor collected by the vibration detector is 25 kHz, and the form of the collected vibration signal is vibration acceleration data varying with time.

[0039] The time length of the vibration signal collected by the vibration detector is 5 s, and the sampling rate is 25 kHz, that is, the time length of collecting a vibration signal is 5 seconds, and 25000 samples are taken per second.

[0040] The first characteristic frequency is 25.2 Hz, and the second characteristic frequency is 1.50 kHz.

[0041] Step 4 further comprises, when the number of adjustments required by the control parameter matrix corresponding to the adjustment function with the least number of required adjustments is greater than a threshold, performing an audible and visual warning.

[0042] The beneficial effects of the present application are:

[0043] The application uses a vibration detector to detect the working state of each frequency conversion motor, and analyzes the collected signals to extract features, so as to realize the collection of the working state of the frequency conversion motor; meanwhile, the control parameter matrix is matrixed, and the corresponding adjustment function is calculated, so that the change of each working state matrix under the adjustment of the adjustment function can be calculated and predicted; thus, the final working state caused by each control parameter matrix can be obtained; the parameter adjustment with the shortest adjustment time can be realized, the yield rate and the reaction speed of the production line are ensured, the adjustment times are reduced, and single adjustment can realize fast adjustment to the optimal working state. BRIEF DESCRIPTION OF DRAWINGS

[0044] Figure 1 It is a schematic diagram of the architecture of the application;

[0045] Figure 2 It is a flow chart of the control method of the application. DETAILED DESCRIPTION

[0046] The advantages, features and methods of achieving the purposes of the application will be clear through the drawings and the following detailed description.

[0047] Example 1:

[0048] The textile machine is a 32-roller long-shaft textile integrated machine, and a total of 32 frequency conversion motors are involved, and each motor is connected to a main controller; the main controller uses an upper industrial computer; a data acquisition card is connected between the industrial computer and the frequency conversion motor;

[0049] A data acquisition card with 32-channel acquisition function is used to directly acquire the working parameters of the frequency conversion motor; the data acquisition card is provided with a parameter control function, and can directly control the working of the frequency conversion motor.

[0050] The system control uses a textile machine transmission control parameter automatic switching system, including a main controller, a frequency conversion motor, a vibration detector and a detection controller.

[0051] The main controller is connected to the frequency conversion motor, the main controller adjusts the voltage U, the current I and the frequency f of the frequency conversion motor, and the working parameters voltage U, current I and frequency f of the frequency conversion motor are directly sent to the control board of the frequency conversion motor through the issuing function of the data acquisition card.

[0052] The vibration detector is arranged on the frequency conversion motor, and is used to detect the vibration data of the frequency conversion motor; the vibration detector sends the collected vibration signals to the detection controller, and the detection controller performs analog-to-digital conversion on the vibration signals to obtain digital waveform data of the vibration signals.

[0053] The vibration detector adopts an acceleration vibration detector, one vibration detector is arranged for each variable frequency motor, or three acceleration detectors are arranged for each variable frequency motor, and the three acceleration detectors detect vibration components of the motor in three dimensions respectively; the three vibration components are superimposed to form waveform data.

[0054] The detection controller sends the digital waveform data of the vibration signal to the main controller, the main controller processes the waveform data of all vibration detectors collected to obtain a working state matrix T of the variable frequency motor, and inputs the working state matrix T into an adaptive control model; the adaptive control model outputs a corresponding control parameter matrix M according to the input working state matrix T; the detection controller outputs the control parameter matrix M to the variable frequency motor, so as to complete automatic switching of the control parameters of the variable frequency motor.

[0055] The control parameter matrix M corresponding to the time when the working state matrix T reaches the standard working state matrix T under the adjustment of the control parameter matrix M in the shortest time is calculated in the main controller, that is, the control parameter matrix output by the adaptive control model.

[0056] The specific calculation method of the adaptive control model is as follows:

[0057] The control parameter matrix of the main controller is M,

[0058]

[0059] Wherein n is the number of motors;

[0060] The sampling rate of the vibration signal of the variable frequency motor collected by the vibration detector is 20 kHz, and the collected vibration signal is in the form of vibration acceleration changing with time data;

[0061] The main controller extracts features from the digital waveform data, and constructs a working state matrix T1 using the features extracted from all variable frequency motors;

[0062]

[0063] Wherein, m features are extracted, n is the number of motors, and λ is the value of the extracted features;

[0064] Let the adjustment function be G(M), wherein M is the adjustment parameter matrix, and G(M) is a function of the matrix M;

[0065] And T1·G(M)=T2;

[0066] Wherein T1 is the working state matrix at the current time, and T2 is the working state matrix at the next minute;

[0067]

[0068] Wherein si , p i , q i (i = 1…n) are constant coefficients;

[0069] Let the standard working state matrix be T0, when the adjustment is effective, there is T1·G k (M) = T0, that is, the working state matrix T1 at the current time will become the standard state matrix T0 after k times of adjustment of the adjustment function G(M), wherein k is the number of minutes elapsed;

[0070] A plurality of control parameter matrices are pre-stored in the main controller, and the adjustment function G(M) of each control parameter matrix can be calculated; the control parameter matrices are traversed in the main controller, and the number of times required for the working state matrix at the current time to reach the standard working state matrix T0 under the adjustment of each adjustment function is calculated;

[0071] The standard working state is pre-stored in the main controller, and the number of times of conversion to the standard working state can be directly obtained by traversing the control parameter matrices. The fewer the number of times, the fewer the number of adjustments required and the shorter the time. The shorter the time, the less time in the non-optimal working state, the best textile effect, the lowest waste rate, and the lowest defect rate.

[0072] The control parameter matrix corresponding to the adjustment function with the least number of adjustments required is the control parameter matrix output by the adaptive control model.

[0073] The time length of the vibration signal collected by the vibration detector is 5s, and the sampling rate is 20-30kHz, that is, the time length of collecting a vibration signal is 5 seconds, and 20000-30000 samples are taken per second;

[0074] The collection time of the vibration signal here can also be adjusted as needed, and it can be within 1-10s; the shorter the time, the shorter the data obtained and the greater the contingency; however, the total amount of signal will also be larger, and the amount of calculation will increase; the longer the time, the smaller the contingency, and the amount of calculation is also reduced, but the disadvantage is that the adjustment period will be longer. In actual operation, the collection time needs to be adjusted according to the actual situation, and a suitable time range is selected. Here, 5s is selected according to experience.

[0075] The main controller first performs Fourier transform on the waveform data, converts the time domain signal into a frequency domain signal, and obtains the spectrum corresponding to the vibration waveform data;

[0076] The main controller extracts features from the converted spectrum, and the number of extracted features is 7, which are as follows:

[0077] λ 1n is the peak value of the first characteristic frequency, λ 2n is the half-height width of the first characteristic frequency, λ3n is a peak value of the second characteristic frequency, λ 4n is a half width of the second characteristic frequency, λ 5n is an extreme value of the wave spectrum, λ 6n is a variance of the wave spectrum, λ 7n is a skewness of the wave spectrum;

[0078] In actual feature extraction, the principal component analysis method can also be combined to perform principal component analysis on the above 7 features in advance, extract the optimal features, and reduce the dimension of the data to improve the operation speed. Of course, the new results obtained by adding, subtracting, multiplying, and dividing the above 7 features can also be brought in as new features. The more features, the better the effect, but the more the calculation amount. It needs to be selected according to the load that the system can bear.

[0079] The first characteristic frequency is 24.8-25.7 Hz, and the second characteristic frequency is 1.48-1.52 kHz.

[0080] Example 2:

[0081] A method for automatically switching the control parameters of the textile machinery transmission using the above system, comprising the following steps:

[0082] Step 1, first, the main controller loads the initial control parameter matrix to the frequency conversion motor, and at the same time, the vibration detector starts to detect the working state of each frequency conversion motor and sends the collected vibration signal to the detection controller;

[0083] Step 2, the detection controller performs analog-to-digital conversion on the vibration signal to obtain a digital waveform signal, and sends the digital waveform signal corresponding to each frequency conversion motor to the main controller;

[0084] Step 3, the main controller performs Fourier transform on the waveform data to convert the time domain signal into a frequency domain signal, and obtains the wave spectrum corresponding to the vibration waveform data. The main controller extracts the features of the converted wave spectrum to obtain the working state matrix T1 at the current time;

[0085] Step 4, the main controller pre-stores a plurality of control parameter matrices, and can calculate the adjustment function G(M) of each control parameter matrix. The main controller performs traversal on the control parameter matrix, and calculates the number of times required for the working state matrix at the current time to reach the standard working state matrix T0 under the adjustment of each adjustment function. The control parameter matrix corresponding to the adjustment function with the least adjustment times is the control parameter matrix output by the adaptive control model;

[0086] Step 5, the main controller completes the automatic switching of the frequency conversion motor control parameters of the frequency conversion motor under the control parameter matrix output by the adaptive control model.

[0087] In particular, the main controller extracts features from the converted wave spectrum, and the number of extracted features is X, wherein X is determined according to the correlation analysis of the collected waveforms and the yield of the textile processing products in actual control, and the number of X is also determined according to the correlation analysis, that is, the number of extracted features.

[0088] The vibration detector is an acceleration vibration sensor with a sensitivity of 20±5% mV / mm / s, a working direction of 0° to ±100°, and an outer dimension of Ф40*70.

[0089] The sampling rate of the vibration signal of the variable frequency motor collected by the vibration detector is 25 kHz, and the collected vibration signal is in the form of vibration acceleration changing with time data.

[0090] The time length of the vibration signal collected by the vibration detector is 5s, and the sampling rate is 25 kHz, that is, the time length of collecting a vibration signal is 5 seconds, and 25000 samples are taken per second.

[0091] The first characteristic frequency is 25.2 Hz, and the second characteristic frequency is 1.50 kHz.

[0092] Step 4 further includes performing an audible and visual warning when the number of required adjustments of the control parameter matrix corresponding to the adjustment function requiring the least number of adjustments is greater than a threshold value.

[0093] The above only describes the preferred embodiments of the present application, but the protection scope of the present application is not limited thereto, any person skilled in the art can easily think of changes or replacements within the technical scope disclosed by the present application, which should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A system for automatic switching of control parameters of a textile machine drive, characterized in that The system comprises a main controller, a variable frequency motor, a vibration detector and a detection controller. The main controller is connected to the variable frequency motor, and the main controller adjusts the voltage U, the current I and the frequency f of the variable frequency motor. The vibration detector is arranged on the variable frequency motor and is used to detect the vibration data of the variable frequency motor. The vibration detector sends the collected vibration signals to the detection controller. The detection controller performs analog-to-digital conversion on the vibration signals to obtain digital waveform data of the vibration signals. The detection controller sends the digital waveform data of the vibration signals to the main controller. The main controller processes the waveform data collected by all the vibration detectors to obtain a working state matrix T of the variable frequency motor, and inputs the working state matrix T into an adaptive control model. The adaptive control model outputs a corresponding control parameter matrix M according to the input working state matrix T. The detection controller outputs the control parameter matrix M to the variable frequency motor, so as to complete automatic switching of the control parameters of the variable frequency motor. The control parameter matrix corresponding to the working state matrix T that can reach a standard working state matrix in the shortest time under the adjustment of the control parameter matrix M in the main controller is the control parameter matrix output by the adaptive control model. The specific calculation method of the adaptive control model is as follows: The control parameter matrix of the main controller is M, wherein n is the number of motors. The sampling rate of the vibration signals of the variable frequency motor collected by the vibration detector is 20-30 kHz, and the collected vibration signals are in the form of vibration acceleration data varying with time. The main controller extracts features from the digital waveform data, and constructs a working state matrix T1 using the features extracted from all the variable frequency motors. wherein m features are extracted in total, n is the number of motors, and λ is the value of the extracted features. Let the adjustment function be G(M), wherein M is the adjustment parameter matrix, and G(M) is a function of the matrix M. And T1·G(M)=T2. Wherein T1 is the working state matrix at the current time, and T2 is the working state matrix at the next minute. where s i , p i , q i are constant coefficients, where i = 1...n; Let the standard working state matrix be T0, when the adjustment is effective, there is T1·G k (M) = T0, that is, the working state matrix T1 at the current time will become the standard state matrix T0 after k times of adjustment by the adjustment function G(M). A plurality of control parameter matrices are pre-stored in the main controller, and the adjustment function G(M) of each control parameter matrix can be calculated. The control parameter matrices are traversed in the main controller, and the number of times required for the working state matrix at the current time to reach the standard working state matrix T0 under the adjustment of each adjustment function is calculated. The control parameter matrix corresponding to the adjustment function with the least required adjustment times is the control parameter matrix output by the adaptive control model.

2. The system according to claim 1, wherein: The time length of the vibration signals collected by the vibration detector is 5s, and the sampling rate is 20-30 kHz, that is, the time length of one vibration signal collection is 5 seconds. The main controller first performs Fourier transform on the waveform data to convert the time domain signals into frequency domain signals, thereby obtaining the spectrum corresponding to the vibration waveform data. The main controller extracts 7 features from the converted spectrum, and the features are as follows: λ 1n is the peak value of the first characteristic frequency, λ 2n is the half-height width of the first characteristic frequency, λ 3n is the peak value of the second characteristic frequency, λ 4n is the half-height width of the second characteristic frequency, λ 5n is the extreme value of the wave spectrum, λ 6n is the variance of the wave spectrum, λ 7n is the skewness of the wave spectrum; wherein the first feature frequency is 24.8Hz-25.7Hz, and the second feature frequency is 1.48kHz-1.52kHz.

3. A method for automatic switching of control parameters of a textile machine drive using the system of claim 2, characterized in that The system comprises the following steps: Step 1, first the main controller to the frequency conversion motor load initial control parameter matrix, while the vibration detector starts to detect the working state of each variable frequency motor, and sends the collected vibration signal to the detection controller; Step 2, the detection controller to the vibration signal analog to digital conversion, get digital waveform signal, and each variable frequency motor corresponding digital waveform signal sent to the main controller; Step 3, the main controller will waveform data Fourier transform, the time domain signal into frequency domain signal, get the vibration waveform data corresponding to the spectrum; the main controller to the converted spectrum feature extraction to get the current time working state matrix T1; Step 4, the main controller pre stored with a plurality of control parameter matrix, and can calculate each control parameter matrix adjustment function G(M); in the main controller control parameter matrix traversal, and calculate the current time working state matrix in each adjustment function adjustment to reach the standard working state matrix T0 required number of times; The least number of adjustment function corresponding to the control parameter matrix is the control parameter matrix output by the adaptive control model; Step 5, the main controller to the variable frequency motor in the adaptive control model output control parameter matrix complete variable frequency motor control parameter automatic switching.

4. The textile machinery transmission control parameter automatic switching method according to claim 3, wherein: The vibration detector is an acceleration vibration sensor.

5. The method for automatic switching of textile machine drive control parameters according to claim 4, characterized in that: The sampling rate of the vibration signal collected by the vibration detector is 25 kHz, and the form of the collected vibration signal is vibration acceleration data with time.

6. The textile machinery transmission control parameter automatic switching method according to claim 3, wherein: The time length of the vibration signal collected by the vibration detector is 5 s, and the sampling rate is 25 kHz, that is, the time length of collecting a vibration signal is 5 s, and the sampling is 25000 times per second.

7. The textile machinery transmission control parameter automatic switching method according to claim 3, wherein: The first characteristic frequency is 25.2 Hz, and the second characteristic frequency is 1.50 kHz.

8. The textile machinery transmission control parameter automatic switching method according to claim 3, wherein: Step 4 further comprises, when the adjustment function corresponding to the control parameter matrix required by the least number of adjustment times is greater than the threshold value, the sound and light warning is carried out.

Images