An automatic control method and system for discharging of an overhung scraper centrifuge

Abstract

The application relates to the technical field of centrifuges, in particular to an automatic unloading control method and system of an upper-suspension type scraper unloading centrifuge. The method comprises the following steps: acquiring vibration data and rotating speed data of a centrifuge drum; dividing the whole data acquisition time into multiple time windows; acquiring a comprehensive abnormality coefficient of each time window according to the spike pulse characteristics, periodic characteristics, frequency change characteristics and high frequency characteristics of the vibration data of each time window, and acquiring an abnormality influence coefficient of each time window in combination with the discrete degree of the rotating speed data of each time window; acquiring an unloading abnormality coefficient of each time window according to the change characteristics of the abnormality influence coefficients of the adjacent windows of each time window, and then acquiring the optimized prediction range parameters of each time window, so as to optimize the scraper movement control of the next time window. The prediction range parameters of each time window are adaptively adjusted, and the unloading efficiency of the centrifuge is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of centrifuges, in particular to an automatic unloading control method and system for an upper-suspension type scraper unloading centrifuge. BACKGROUND

[0002] The upper-suspension type scraper unloading centrifuge is an industrial equipment widely used in the field of solid-liquid separation. A strong centrifugal force is generated by a high-speed rotating drum to make the solid particles suspended in the liquid settle, thereby realizing solid-liquid separation. Its significant feature is that the drum adopts an upper-suspension type structure, and the scraper device is installed inside the drum, which can automatically scrape and unload the settled solid particles after the end of the centrifugal separation process, realizing continuous or intermittent unloading operation.

[0003] The separated solid particle materials will adhere to the inner wall of the drum. When unloading, the centrifuge will reduce the rotating speed of the drum to an unloading rotating speed state, and then the system program controls the scraper to cut radially in the drum, thereby scraping the materials from the inner wall. The materials hanging down are discharged from the centrifuge through the unloading slot. However, during the centrifugal process, if the feeding speed is too fast or the feeding is uneven, the material distribution on the inner wall of the drum will be uneven, and the particle size of part of the material is large. When the scraper unloads, the material with large particles may cause the scraper to wear out intensively, and even the friction between the scraper and the material increases, which will cause the drum to vibrate greatly, thereby affecting the unloading efficiency and separation effect. The conventional unloading process does not fully consider the influence of uneven material distribution and excessive wear of the scraper, which intensifies the instability of the drum during the unloading process, resulting in low unloading efficiency. SUMMARY

[0004] In order to solve the above technical problems, the purpose of the present application is to provide an automatic unloading control method and system for an upper-suspension type scraper unloading centrifuge, and the technical solution adopted is as follows:

[0005] In the first aspect, the present application provides an automatic unloading control method for an upper-suspension type scraper unloading centrifuge, which comprises the following steps:

[0006] Real-time acquisition of vibration data and rotating speed data of the centrifuge drum;

[0007] Divide the entire data acquisition time into multiple time windows; divide each time window into multiple time periods, and obtain the vibration anomaly significant value of each time window according to the difference between each peak value in the vibration data of each time window and its adjacent data, and the distance of the fitting curve corresponding to the vibration data of each adjacent time period in each time window;

[0008] The vibration data of each period is decomposed into a plurality of modal components, and a frequency component sequence of each period is obtained according to a frequency corresponding to a maximum amplitude in a frequency spectrum diagram of each modal component of each period; a frequency abnormal value of each time window is obtained according to an average level of a distance between frequency component sequences of all arbitrary two periods in each time window, and an average level of a maximum value in the frequency component sequences of all periods in each time window, and a comprehensive abnormal coefficient of each time window is obtained in combination with the vibration abnormal significant value of each time window, and an abnormal influence coefficient of each time window is obtained in combination with a discrete degree of the rotating speed data of each time window;

[0009] A discharge abnormal coefficient of each time window is obtained according to a difference of the abnormal influence coefficients of all adjacent neighbor windows in the neighbor windows of each time window, and an average level of the abnormal influence coefficients of each time window and all neighbor windows thereof, and then an optimized prediction range parameter of each time window is obtained in combination with a preset value range of the prediction range parameter in the model predictive control technology, so as to optimize the scraper movement control of the next time window.

[0010] Preferably, the calculation formula of the vibration abnormal significant value of each time window is: ; in the formula, is the vibration abnormal significant value of the i th time window, is the significant coefficient of the i th time window, is the average of the DTW distances between the fitting curves corresponding to all adjacent periods in the i th time window; wherein, The obtaining process of is as follows: the average of the absolute values of the differences between each peak value of the i th time window and its adjacent two data before and after the peak value is calculated, and the cumulative sum of all averages is taken as the significant coefficient of the i th time window.

[0011] Preferably, the frequency component sequence of each period refers to a sequence composed of frequencies corresponding to maximum amplitudes in the frequency spectrum diagrams of each modal component of each period in ascending order.

[0012] Preferably, the calculation formula of the frequency abnormal value of each time window is: ; in the formula, is the frequency abnormal value of the i th time window, is the difference coefficient of the i th time window, is the average of the ratios of the maximum value to the minimum value in the frequency component sequence of each vibration data in the i th time window; wherein, the difference coefficient of each time window refers to the average of the Euclidean distances between the frequency component sequences of all arbitrary two periods in each time window.

[0013] Preferably, the comprehensive abnormal coefficient of each time window refers to the sum of the vibration abnormal significant value and the frequency abnormal value of each time window.

[0014] Preferably, the calculation formula of the abnormal influence coefficient of each time window is: ; wherein, is the abnormal influence coefficient of the i th time window, is the comprehensive abnormal coefficient of the i th time window, is the standard deviation of all the rotating speed data in the i th time window.

[0015] Preferably, the calculation formula of the discharge abnormal coefficient of each time window is: ; wherein, is the discharge abnormal coefficient of the i th time window, is the average of the abnormal influence coefficients of the i th time window and all the adjacent windows thereof, and are the abnormal influence coefficients of the k th and the k-1 th time windows in the adjacent windows of the i th time window, respectively, and N is the number of the adjacent windows of the i th time window.

[0016] Preferably, the calculation formula of the optimized prediction range parameter of each time window is: ; wherein, is the optimized prediction range parameter of the i th time window, a is the preset minimum value of the prediction range parameter, and b is the preset maximum value of the prediction range parameter, is the discharge abnormal coefficient of the i th time window, is a normalization function.

[0017] Preferably, the specific process of optimizing the scraper movement control of the next time window is: taking the optimized prediction range parameter of the current time window as the prediction range parameter of the model predictive control technology in the next time window, so as to optimize the movement control of the scraper in the next time window.

[0018] In the second aspect, the embodiments of the present application further provide an automatic discharge control system of the top suspension type scraper discharge centrifuge, which comprises a memory, a processor, and a computer program stored in the memory and running on the processor, and the processor implements the steps of the automatic discharge control method of the top suspension type scraper discharge centrifuge according to any one of the above embodiments when executing the computer program.

[0019] The present application has at least the following beneficial effects:

[0020] The application can accurately reflect the abnormal characteristics of the vibration data of the rotating drum in the running process by constructing a comprehensive abnormal coefficient based on the in-depth analysis of the spike pulse and periodic characteristics in the time domain, the frequency component difference and the high frequency characteristics in the frequency domain of the vibration data in the unloading process; then, the abnormal influence coefficient is constructed based on the fluctuation abnormality of the rotating speed data, which can accurately represent the comprehensive abnormal characteristics of the vibration data and the rotating speed data, so that the subsequent evaluation of the abnormal characteristics of the rotating drum can be more accurate; then, the unloading abnormal coefficient is constructed based on the analysis of the abnormal characteristic change rule with the feeding of the scraper, and the real-time optimization and adjustment of the scraper movement control is carried out based on the value, which has the advantages of reducing the influence of uneven material distribution and excessive wear of the scraper, and is helpful to improve the unloading efficiency of the centrifuge. BRIEF DESCRIPTION OF DRAWINGS

[0021] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments or the prior art description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0022] Figure 1 A step flow chart of an automatic unloading control method of an upper suspension type scraper unloading centrifuge provided by an embodiment of the present application is shown in the figure.

[0023] Figure 2 A flow chart of obtaining the unloading abnormal coefficient of each time window provided by an embodiment of the present application is shown in the figure. DETAILED DESCRIPTION

[0024] In order to further illustrate the technical means and effects adopted by the present application to achieve the predetermined invention purpose, the specific embodiments, structure, characteristics and effects of the automatic unloading control method and system of the upper suspension type scraper unloading centrifuge according to the present application are described in detail as follows. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0025] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present application belongs.

[0026] The specific scheme of the automatic unloading control method and system of the upper suspension type scraper unloading centrifuge provided by the present application is specifically described below with reference to the drawings.

[0027] Please refer to Figure 1Fig. 1 shows a flow chart of a method for automatically controlling the unloading of an overhead scraper centrifuge according to an embodiment of the present application, which comprises the following steps:

[0028] Step one: Real-time acquisition of vibration data and rotational speed data of the centrifuge drum.

[0029] During the operation of the overhead scraper centrifuge, low unloading efficiency or incomplete unloading can cause part of the filter cake to remain in the drum, which will mix with the new material during the next feeding, reducing the product purity and affecting the product quality and stability. If the scraper frequently stalls or interferes with the drum during unloading, it can scratch or damage the inner wall of the drum, affecting the balance and service life of the drum, and causing the drum to vibrate and change speed during unloading. In order to monitor the unloading process in real time, the vibration data and rotational speed data of the drum are collected in real time by the parameter monitoring system of the centrifuge. In this embodiment, the collection frequency of vibration data and rotational speed data is 1000HZ.

[0030] Step two: Divide the entire data collection time into multiple time windows; divide each time window into multiple time periods, and obtain the vibration anomaly significant value of each time window according to the difference between each peak value in the vibration data of each time window and its adjacent data, and the distance of the fitting curve corresponding to the vibration data of each adjacent time period in each time window.

[0031] Each time the centrifuge is operated, the motor is first started at low speed, and after reaching the set feeding speed, the feeding valve is automatically opened, and the suspension material enters the high-speed rotating drum through the feeding pipe. Under the action of centrifugal force, the liquid is thrown out of the drum through the screen and the small holes on the drum wall, and the solid is trapped on the surface of the drum screen to form a filter cake. After separation is completed, the motor is reduced to the unloading speed, and the scraper is controlled to move radially and axially by the electrical control system to scrape the filter cake off the inner wall of the drum, and the filter cake is discharged through the discharge port. Due to the simultaneous feeding and separation of the material, the distribution of the material on the inner wall of the drum is not uniform due to the influence of the feeding speed and the material density. Moreover, due to the long-term use of the scraper, the wear of the scraper is aggravated, and the friction between the scraper and the material is increased, which can easily cause abnormal vibration of the drum during unloading. Further analysis of the abnormal vibration characteristics during unloading is as follows.

[0032] The present application takes each preset time length from the beginning to the end of the whole data acquisition time as a time window. Since the vibration changes rapidly and randomly during the discharging process, in order to analyze the abnormal vibration characteristics in the short term in real time, the preset time length is set to 2s in the embodiment, wherein s is the unit of time, second. Due to uneven material distribution and scraper wear, the interference force received by the drum is unstable in time and space, which will cause the phase change of the vibration data to be unstable. Frequent impact forces may be generated during the scraping of the material, which will be manifested as sharp pulse in the time domain waveform of the vibration data. In view of this, taking all the vibration data in the ith time window as an example, in order to reduce the noise interference during data acquisition, the present application first performs high-pass filtering preprocessing on the vibration data in the window, then uses an automatic multi-scale peak finding algorithm to obtain the peak points of the vibration data, and calculates the mean of the absolute values of the difference between each peak and its adjacent two data, respectively. The cumulative sum of all the means is taken as the significant coefficient of the ith time window, denoted as , which is used to represent the significant degree of the sharp pulse feature of the vibration data in the time window. The greater the , the stronger the impact force between the scraper and the material during discharging in the period.

[0033] In addition, uneven material distribution may cause material accumulation in some areas, making the hardness of these areas relatively high, so that the vibration caused by the impact between the scraper and the material during the rotation of the material presents a periodic change feature. The amount of material scraped and the resistance will be different each time, and it is difficult to accurately reflect the periodic change feature by directly comparing the relationship between the vibration data. In view of this, the ith time window is divided into q time periods according to a preset number q, which is set to 10 in the embodiment, and the implementer can take the value according to the actual situation; a quadratic polynomial fitting technique is used to obtain the fitting curve of the vibration data in each time period in the ith time window.

[0034] It should be noted that the high-pass filtering algorithm, the automatic multi-scale peak finding algorithm and the quadratic polynomial fitting technique are all known techniques, and the specific process will not be described again.

[0035] As a preferred embodiment, according to the difference between each peak in the vibration data of each time window and its adjacent data, the distance between the fitting curves corresponding to the vibration data of each adjacent time period in each time window, the vibration anomaly significant value of each time window is obtained, which is used to represent the significant degree of the abnormal feature of the vibration data of each time window in the time domain.

[0036] In the embodiment, the vibration anomaly significant value of the ith time window is denoted as , and its specific expression is: ; in the formula, is the vibration anomaly significant value of the ith time window, is the significant coefficient of the ith time window, is the average of the DTW distance between the fitted curves for all adjacent time periods in the ith time window.

[0037] reflects the periodic vibration impact characteristics caused by uneven distribution of the material in the drum; the obtained reflects the sharp pulse and periodic characteristics of the impact vibration of the scraper and the material in the time window. The greater the value of is, the more significant the sharp pulse characteristics and the stronger the periodicity of the vibration data in the ith time window, and the more likely it is that uneven distribution of the material and scraper wear occur in the ith time window.

[0038] Step three: decompose the vibration data of each time period into a plurality of modal components, obtain the frequency component sequence of each time period according to the frequency corresponding to the maximum amplitude in the frequency spectrum of each modal component of each time period, obtain the frequency abnormal value of each time window according to the average level of the distance between the frequency component sequences of any two time periods in each time window and the average level of the maximum value in the frequency component sequence of all time periods in each time window, and obtain the comprehensive abnormal coefficient of each time window in combination with the vibration abnormal significant value of each time window, and obtain the abnormal influence coefficient of each time window in combination with the dispersion degree of the speed data of each time window.

[0039] Further, due to the influence of the hardness change of the material at different positions in the drum, the unbalanced force of the drum will increase, and after the scraper is worn, the interference force received by the drum will be stronger, which can cause the vibration data of the drum to have some new frequency components in addition to the fundamental frequency signal, such as high-frequency components generated by unstable friction between the scraper and the material. Therefore, the greater the frequency variation degree of the vibration data and the higher the partial frequency, the greater the unbalanced force received by the drum.

[0040] Therefore, in order to obtain the frequency information of the vibration data, the empirical mode decomposition algorithm is used to decompose the vibration data of each time period. Taking the vibration data of the jth time period in the ith time window as an example, the number of modal components is set to 5, and the output of the algorithm is 5 modal components. Then the frequency spectrum of each modal component is obtained by using the discrete Fourier transform; the sequence composed of the frequencies corresponding to the maximum amplitudes in the frequency spectrum of each modal component of the jth time period arranged in ascending order is denoted as the frequency component sequence of the jth time period. According to the above method, the frequency component sequence of each time period in the ith time window can be obtained, and the average of the Euclidean distance between the frequency component sequences of any two time periods in the ith time window is taken as the difference coefficient of the ith time window, denoted as The greater the value of is, the greater the frequency variation degree of the vibration data in the window.

[0041] ​​It should be noted that the empirical mode decomposition algorithm and the discrete Fourier transform are both well-known techniques, and thus the specific processes will not be described again.

[0042] In addition, the minimum value in the frequency component sequence corresponds to the fundamental frequency of the vibration data segment, which does not change significantly; and the maximum value in the frequency component sequence corresponds to the high-frequency component of the vibration data segment, which is easily affected by the friction between the scraper and the material and thus becomes large.

[0043] As a preferred embodiment, the frequency abnormal value of each time window is obtained according to the average level of the distance between the frequency component sequences of any two time segments in each time window, and the average level of the maximum value in the frequency component sequence of each time segment in each time window, and is used to represent the significant degree of the abnormal feature of the vibration data in the frequency domain of each time window.

[0044] In this embodiment, the frequency abnormal value of the i th time window is denoted as , and the specific expression is: ; in the formula, is the frequency abnormal value of the i th time window, is the difference coefficient of the i th time window, is the average value of the ratio of the maximum value to the minimum value in the frequency component sequence of each vibration data segment in the i th time window.

[0045] The obtained reflects the frequency component difference and the significant high-frequency feature of the drum rotation affected by the unbalanced force. The greater the value is, the greater the frequency component difference between each vibration data segment in the i th time window is, and the more significant the high-frequency component is, and thus it is more likely that the drum abnormal vibration occurs in the time period corresponding to the i th time window.

[0046] Further, the comprehensive abnormal coefficient of each time window is obtained according to the vibration abnormal significant value and the frequency abnormal value of each time window. In this embodiment, the comprehensive abnormal coefficient of the i th time window is denoted as , and the expression is: ; in the formula, is the comprehensive abnormal coefficient of the i th time window, is the vibration abnormal significant value of the i th time window, is the frequency abnormal value of the i th time window. The obtained comprehensively reflects the abnormal features of the drum vibration data in the time domain and in the frequency domain in the i th time window.

[0047] Further, if the centrifuge unloading process is not smooth, under the influence of the structure characteristics of the top suspension type centrifuge, the drum is prone to radial swing, thereby increasing the motor load and reducing the stability of the drum speed. Therefore, the more obvious the abnormality of the drum vibration, and the more unstable the speed, the lower the unloading efficiency of the centrifuge. Therefore, the standard deviation of all speed data in the i-th time window is calculated, denoted as , the greater the fluctuation of the speed. Further, the abnormal influence coefficient of the i-th time window is calculated, and the formula is: ; in the formula, is the abnormal influence coefficient of the i-th time window, is the comprehensive abnormal coefficient of the i-th time window, is the standard deviation of all speed data in the i-th time window. The obtained reflects the synchronous influence characteristics of the abnormal fluctuation of the vibration data and the speed data of the drum during the unloading of the centrifuge, and the greater the value, the greater the abnormality of the vibration data and the speed data of the drum in the i-th time window.

[0048] Step four: According to the difference of the abnormal influence coefficients of all adjacent neighboring windows in the neighboring windows of each time window, and the average level of the abnormal influence coefficients of each time window and all its neighboring windows, the unloading abnormality coefficient of each time window is obtained, and then the preset value range of the prediction range parameter in the model predictive control technology is combined to obtain the optimized prediction range parameter of each time window, so as to optimize the scraper movement control of the next time window.

[0049] In addition, as time goes on, the material is continuously scraped off, and the scraper continuously approaches the inner wall of the drum. However, if the material is unevenly distributed, the material closer to the inner wall of the drum is usually more viscous, increasing the difficulty of completely scraping off the material by the scraper. Therefore, under the influence of uneven material distribution and scraper wear, the abnormal characteristics of drum vibration and speed fluctuation may be more obvious closer to the inner wall of the drum, that is, the abnormality of the later time window is more obvious than that of the former time window. In view of this, the present application sets the first N time windows of the i-th time window as the neighboring windows of the i-th window, wherein N is an integer in the range of [8, 12], and N is 10 in the present embodiment.

[0050] As a preferred embodiment, according to the difference of the abnormal influence coefficients of all adjacent neighboring windows in the neighboring windows of each time window, and the average level of the abnormal influence coefficients of each time window and all its neighboring windows, the unloading abnormality coefficient of each time window is obtained, which is used to represent the significant degree of the abnormal characteristics of the drum affected by uneven material distribution and scraper wear in each time window. The flow chart for obtaining the unloading abnormality coefficient of each time window is as follows: Figure 2is shown.

[0051] In this embodiment, the discharge abnormality coefficient of the i-th time window is denoted as , and its specific expression is: ; in the formula, is the discharge abnormality coefficient of the i-th time window, is the average of the abnormality influence coefficients of the i-th time window and all its neighboring windows, and are the abnormality influence coefficients of the k-th and k-1-th time windows in the neighboring windows of the i-th time window, and N is the number of neighboring windows of the i-th time window.

[0052] The obtained reflects the characteristics that the abnormal features of the vibration data and the speed data in the time window increase with the increase of the scraper feed and the abnormal influence state characteristics of the discharge, The greater the value of the i-th time window, the more obvious the abnormal features of the drum caused by uneven material distribution and scraper wear.

[0053] Further, in order to improve the discharge efficiency, the application optimizes the movement control of the scraper based on the discharge abnormality coefficient. Specifically, the model predictive control technology is used to control the movement of the scraper. If the calculated discharge abnormality coefficient is greater, it means that the drum is more affected by uneven material distribution and scraper wear during the discharge process, and then a larger prediction range parameter should be set in the model predictive control technology to improve the response efficiency of the feed adjustment of the scraper. If the calculated discharge abnormality coefficient is smaller, it means that the drum is less affected by uneven material distribution and scraper wear during the discharge process, and then a smaller prediction range parameter should be set to improve the control accuracy of the scraper. The preset value range of the prediction range parameter is [a, b], wherein a is the preset minimum value of the prediction range parameter, and b is the preset maximum value of the prediction range parameter. The commonly used value of the prediction range parameter is within 30, so in this embodiment, a is 10 and b is 30.

[0054] Further, according to the discharge abnormality coefficient of the i-th time window, the preset minimum value and the preset maximum value, the optimized prediction range parameter of the i-th time window is calculated, and its expression is: ; in the formula, is the optimized prediction range parameter of the i-th time window, a is the preset minimum value of the prediction range parameter, and b is the preset maximum value of the prediction range parameter, is the discharge abnormality coefficient of the i-th time window, is a normalization function, and in this embodiment, a sigmoid function is used for normalization.

[0055] The optimized prediction range parameter of each time window can be calculated according to the above method, and the optimized prediction range parameter of the current time window is taken as the prediction range parameter of the model prediction control technology in the next time window, so as to optimize the movement control of the doctor blade in the next time window. The prediction range parameter of each time window is adjusted adaptively, which helps to avoid strong impact between the doctor blade and the material, and improves the automatic discharge efficiency of the centrifuge.

[0056] Based on the same inventive concept as the above method, the embodiments of the present application also provide an automatic discharge control system of a top suspension type doctor blade discharge centrifuge, which comprises a memory, a processor and a computer program stored in the memory and running on the processor, and the processor implements the steps of any one of the above automatic discharge control methods of the top suspension type doctor blade discharge centrifuge when executing the computer program.

[0057] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description, and does not represent the advantages and disadvantages of the embodiments. Moreover, the above description is made for specific embodiments of the present application. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multi-task processing and parallel processing are possible or may be advantageous.

[0058] Each of the embodiments in the present specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other. Each embodiment mainly describes the difference from other embodiments.

[0059] The above is only the preferred embodiment of the present application, and is not used to limit the present application. Any modification, equivalent replacement, improvement, etc. made within the principle of the present application shall be included in the protection scope of the present application.

Claims

1. An automatic discharge control method for a top suspension type scraper discharge centrifuge, characterized by, The method comprises the following steps: Real-time acquisition of vibration data and rotation speed data of the centrifuge drum; The entire data acquisition time is divided into multiple time windows, and each time window is divided into multiple time periods. According to the difference between each peak value in the vibration data of each time window and its adjacent data, and the distance of the fitting curve corresponding to the vibration data of each adjacent time period in each time window, the vibration anomaly significant value of each time window is obtained. The vibration data of each time period is decomposed into multiple modal components, and the frequency component sequence of each time period is obtained according to the frequency corresponding to the maximum amplitude in the frequency spectrum of each modal component of each time period. According to the average level of the distance between the frequency component sequences of any two time periods in each time window, and the average level of the maximum value in the frequency component sequence of all time periods in each time window, the frequency anomaly value of each time window is obtained, and the comprehensive anomaly coefficient of each time window is obtained in combination with the vibration anomaly significant value of each time window. The abnormal influence coefficient of each time window is obtained in combination with the dispersion degree of the rotation speed data of each time window. According to the difference between the abnormal influence coefficients of all adjacent near-neighbor windows in the near-neighbor window of each time window, and the average level of the abnormal influence coefficients of each time window and all its near-neighbor windows, the unloading anomaly coefficient of each time window is obtained, and then the optimized prediction range parameter of each time window is obtained in combination with the preset value range of the prediction range parameter in the model predictive control technology, so as to optimize the scraper movement control of the next time window.

2. The automatic discharge control method of an overhead scraper discharge centrifuge according to claim 1, wherein The formula for calculating the vibration anomaly significant value of each time window is: ; wherein, is the vibration anomaly significant value of the i th time window, is the significant coefficient of the i th time window, is the mean value of the DTW distance between the fitting curves corresponding to all adjacent time periods in the i th time window; wherein, The acquisition process of is as follows: calculating the mean value of the absolute value of the difference between each peak value of the i th time window and the adjacent two data before and after the peak value, and taking the cumulative sum of all mean values as the significant coefficient of the i th time window.

3. The automatic discharge control method of a top suspension type scraper discharge centrifuge according to claim 1, wherein The frequency component sequence of each time period refers to a sequence composed of frequencies corresponding to the maximum amplitude in the frequency spectrum of each modal component of each time period arranged in ascending order.

4. The automatic discharge control method of an overhead scraper discharge centrifuge according to claim 1, wherein The formula for calculating the frequency outliers of each time window is: ; wherein, is the frequency outlier of the i th time window, is the difference coefficient of the i th time window, is the average of the ratio of the maximum value to the minimum value in the frequency component sequence of each vibration data in the i th time window; wherein the difference coefficient of each time window refers to the average of the Euclidean distances between the frequency component sequences of any two time periods in each time window.

5. The automatic discharge control method of an overhead scraper discharge centrifuge according to claim 1, wherein The comprehensive anomaly coefficient of each time window refers to the sum of the vibration anomaly significant value and the frequency anomaly value of each time window.

6. The automatic discharge control method of an overhead scraper discharge centrifuge according to claim 1, wherein The calculation formula of the abnormal influence coefficient of each time window is: ; wherein, is the abnormal influence coefficient of the i th time window, is the comprehensive abnormal coefficient of the i th time window, is the standard deviation of all the rotating speed data in the i th time window.

7. The automatic discharge control method of an overhead scraper discharge centrifuge according to claim 1, wherein The calculation formula of the unloading abnormality coefficient of each time window is: ; wherein, is the unloading abnormality coefficient of the i th time window, is the average of the abnormality influence coefficients of the i th time window and all the adjacent windows, and are the abnormality influence coefficients of the k th and k-1 th time windows in the adjacent windows of the i th time window, respectively, and N is the number of the adjacent windows of the i th time window.

8. The automatic discharge control method of an overhead scraper discharge centrifuge according to claim 1, wherein The calculation formula of the optimized prediction range parameter of each time window is: ; wherein, is the optimized prediction range parameter of the i th time window, a is a preset minimum value of the prediction range parameter, b is a preset maximum value of the prediction range parameter, is the i th time window, and is the unloading abnormality coefficient of the i th time window, is a normalization function.

9. The automatic discharge control method of an overhead scraper discharge centrifuge according to claim 1, wherein The specific process of optimizing the scraper movement control of the next time window is: taking the optimized prediction range parameter of the current time window as the prediction range parameter of the model predictive control technology in the next time window, to realize the optimization of the movement control of the scraper in the next time window.

10. An automatic discharge control system for an overhung scraper discharge centrifuge, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, The processor executes the computer program to realize the steps of the unloading automatic control method of the overhung scraper unloading centrifuge according to any one of claims 1-9.

Images