Centrifuge rotor stall detection method based on smart sensor

Abstract

The present application relates to the technical field of centrifuge rotor stall detection, and particularly relates to a centrifuge rotor stall detection method based on intelligent sensors. The method acquires the current signal of the centrifuge rotor, the reference rotation frequency of each cycle and the vibration signal of each direction in the unloading stage in real time; the rotation frequency of each direction in the current cycle is acquired according to the vibration signal; the first stall risk degree is acquired according to the difference between the rotation frequency of each direction in the current cycle and the previous adjacent cycle; the second stall risk degree is acquired according to the corresponding amplitude of the frequency of the current signal and the corresponding amplitude in the current signal of the previous adjacent cycle; the overall stall risk degree is acquired according to the first stall risk degree and the second stall risk degree to determine whether the current rotor stalls in the unloading stage. The present application can accurately determine whether the centrifuge rotor stalls in the unloading stage in real time, which is conducive to timely discovering the abnormal condition of the rotor and processing, and effectively reducing the loss caused by stall.

Description

Technical Field

[0001] This invention relates to the field of centrifuge rotor stall detection technology, and more specifically to a centrifuge rotor stall detection method based on intelligent sensors. Background Technology

[0002] In industrial production, unloading centrifuges serve as core equipment for solid-liquid separation and are widely used in chemical, food, and environmental protection fields. Their operation involves acceleration, stable separation, unloading, and cleaning phases. The centrifuge rotor, as the core moving component, must rotate at high speed to achieve the separation function. During long-term operation, the rotor is prone to imbalance due to uneven material distribution and mechanical wear. Failure to monitor this in time can lead to rotor stall, causing severe equipment vibration, spindle deformation, and even shutdowns or safety accidents. Therefore, accurate assessment of rotor stall risk is crucial for ensuring stable equipment operation.

[0003] Current assessments of rotor stall risk in unloading centrifuges primarily revolve around two main technical approaches: mechanical vibration signal monitoring and current signal monitoring. Vibration signals directly reflect the rotor's mechanical state. By first acquiring the rotor's X, Y, and Z axis vibration signals, and then extracting the rotational frequency and corresponding amplitude through Fourier transform, the amplitude corresponding to the rotational frequency will significantly increase when rotor imbalance worsens, serving as a basis for judging stall precursors. Current signals, as a direct reflection of the motor load state, are used to assist in verifying the rotor's mechanical state. The principle is that rotor imbalance causes periodic fluctuations in the motor load, resulting in frequency-related ripple superimposed on the main power line current. By analyzing the changes in ripple amplitude through current signal analysis, abnormal rotor load can be indirectly determined.

[0004] However, in reality, the piston pushing action during the unloading stage will generate periodic strong impact vibrations, the energy of which is much higher than the vibration of the rotor itself. This will further affect the corresponding current signal of the rotor. Therefore, the existing methods for assessing the stall risk of the centrifuge rotor during the unloading stage cannot accurately assess the stall risk of the centrifuge rotor, resulting in the inability to detect abnormalities of the centrifuge rotor in a timely and accurate manner, and thus failing to ensure the safe and stable operation of the unloading centrifuge. Summary of the Invention

[0005] To address the technical problem that existing methods for assessing centrifuge rotor stall risk during the unloading phase cannot accurately evaluate the stall risk of the centrifuge rotor, the present invention aims to provide a centrifuge rotor stall detection method based on intelligent sensors. The specific technical solution adopted is as follows:

[0006] This invention provides a method for detecting centrifuge rotor stall based on intelligent sensors, the method comprising the following steps:

[0007] Real-time acquisition of the centrifuge rotor's current signal, reference rotation frequency for each cycle, and vibration signal in each direction during the unloading phase;

[0008] Based on the vibration signals of the centrifuge rotor in each direction during the current cycle during the unloading phase and the vibration signals of the centrifuge rotor in each direction during each cycle during the historical normal operation of the centrifuge rotor during the unloading phase, the rotational frequency of each direction in the current cycle is obtained; based on the difference in rotational frequency of each direction between the current cycle and the previous adjacent cycle, the first stall risk level of the current cycle is obtained.

[0009] The second stall risk level of the current cycle is obtained based on the difference between the amplitude corresponding to each frequency of the current signal in the current cycle and the amplitude corresponding to the current signal in the previous adjacent cycle, as well as the relationship between each frequency of the current signal in the current cycle and the fixed frequency of the power supply and the reference frequency of the current cycle.

[0010] Based on the first stall risk level and the second stall risk level, the overall stall risk level of the current cycle is obtained, and it is determined whether the centrifuge rotor stalls during the unloading phase.

[0011] Furthermore, the method for obtaining the frequency of rotation in each direction is as follows:

[0012] Based on the vibration signals of the centrifuge rotor in each direction during each cycle of its historical normal operation during the unloading phase, obtain the periodic vibration model in each direction during the current unloading phase.

[0013] The periodic vibration model is separated using blind source separation technology to obtain the unloading impact vibration interference signal in each direction during the current unloading stage.

[0014] For any direction, the difference between the spectrum diagrams corresponding to the vibration signal in that direction and the unloading impact vibration interference signal in that direction during the current period is used as the reference spectrum diagram for that direction during the current period.

[0015] The frequency shift in the reference spectrum is taken as the frequency shift in that direction for the current period.

[0016] Furthermore, the method for obtaining the periodic vibration model is as follows:

[0017] The historical normal operating cycle of the unloading phase within the preset historical time period is used as the normal reference cycle; where the end time of the preset historical time period is the current time.

[0018] The normal reference periods are arranged according to the time sequence to obtain the period sequence. The period sequence is then numbered sequentially from left to right, from smallest to largest, to determine the number of each normal reference period.

[0019] The ratio of the label of each normal reference period to the sum of the labels of all normal reference periods is used as the participation weight of each normal reference period.

[0020] For any normal reference period and any direction, the product of the participation weight of the normal reference period and the vibration signal in that direction of the normal reference period is taken as the participation vibration signal in that direction of the normal reference period.

[0021] The sum of all participating vibration signals in this direction during the normal reference period is used as the periodic vibration model in this direction during the current unloading stage.

[0022] Furthermore, the method for obtaining the first stall risk level is as follows:

[0023] For any direction, the difference in frequency of rotation in that direction between the current period and the previous adjacent period is taken as the degree of increase in frequency of rotation in that direction during the current period.

[0024] The difference in the degree of frequency increase between the current main direction and each other direction is obtained and used as the first difference; among them, the main direction is the direction with the strongest vibration signal and the highest correlation with centrifuge rotor failure.

[0025] The difference between any two first differences is taken as the second difference; the mean of the second differences is negatively correlated and normalized, and the result is taken as the first reference weight of the current period.

[0026] The product of the first reference weight and the degree of frequency increase in the main direction of the current cycle is used as the first stall risk level of the current cycle.

[0027] Furthermore, the method for obtaining the second stall risk level is as follows:

[0028] For any frequency of the current signal in the current cycle, based on the relationship between that frequency and the fixed frequency of the power supply and the reference frequency of the current cycle, the degree of correspondence between the frequency of the current cycle and that frequency is obtained.

[0029] The difference between the amplitude of the frequency in the current signal of the current period and the current signal of the previous adjacent period is taken as the first characteristic value of the frequency.

[0030] The normalized result of the product of the corresponding degree and the first feature value is taken as the degree of abnormality of the frequency;

[0031] The sum of the abnormality levels of all frequencies of the current signal in the current cycle and the result of normalization are used as the second stall risk level for the current cycle.

[0032] Furthermore, the method for obtaining the corresponding degree is as follows:

[0033] For any integer in the preset set of integers, the product of that integer and the reference frequency of the current period is used as the first reference value;

[0034] The sum of the first reference value and the fixed frequency of the power supply is used as the second reference value;

[0035] The result of negatively correlating and normalizing the difference between the second reference value and the frequency is used as the reference correspondence degree corresponding to the integer.

[0036] The minimum reference correspondence degree corresponding to the integer in the preset integer set is taken as the correspondence degree between the current cycle frequency and the frequency.

[0037] Furthermore, the method for obtaining the overall stall risk level is as follows:

[0038] The sum of the first stall risk level and the second stall risk level, after normalization, is taken as the overall stall risk level for the current cycle.

[0039] Furthermore, the method for determining whether the centrifuge rotor is stalling during the unloading stage is as follows:

[0040] When the overall stall risk level is greater than the preset stall risk level threshold, it is determined that the centrifuge rotor is stalling during the unloading stage.

[0041] When the overall stall risk level is less than or equal to the preset stall risk level threshold, it is determined that the centrifuge rotor does not stall during the unloading stage.

[0042] Furthermore, the cycle is the duration of one complete action from piston extension (pushing material) to reset (preparing for the next push).

[0043] Furthermore, the directions include three directions: X, Y, and Z.

[0044] The present invention has the following beneficial effects:

[0045] This invention first obtains the rotational frequency of each direction in the current cycle based on the vibration signals of the centrifuge rotor in each direction during the unloading phase and the vibration signals of the centrifuge rotor in each direction during each cycle of its historical normal operation during the unloading phase. This accurately reflects the basic vibration frequency of the centrifuge rotor in each direction during the rotation process of the current cycle, preparing for subsequent identification of whether the centrifuge rotor has stalled. Then, based on the difference in rotational frequency in each direction between the current cycle and the previous adjacent cycle, the first stall risk level of the current cycle is obtained, preliminarily determining the possible stall situation of the centrifuge rotor during the unloading phase. To more accurately analyze the possible stall situation of the centrifuge rotor during the unloading phase, further, based on the amplitude corresponding to each frequency of the current signal in the current cycle and the amplitude corresponding to the current signal in the previous adjacent cycle... By considering the differences in amplitude and the relationship between each frequency of the current signal in the current cycle and the fixed frequency of the power supply and the reference frequency of the current cycle, the second stall risk level of the current cycle is obtained, further determining the possible stall situation of the centrifuge rotor during the unloading stage. Then, based on the first and second stall risk levels, the overall stall risk level of the current cycle is obtained, accurately reflecting the stall situation of the centrifuge rotor during the unloading stage. This accurately determines whether the centrifuge rotor is stalling during the unloading stage, effectively improving the accuracy and timeliness of the centrifuge rotor stall anomaly assessment during the unloading stage. This facilitates timely handling of centrifuge rotor anomalies, effectively reducing losses caused by stalling, and ensuring the stable operation of the unloading centrifuge while effectively improving its working efficiency. Attached Figure Description

[0046] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0047] Figure 1 This is a schematic flowchart illustrating a centrifuge rotor stall detection method based on a smart sensor, provided in one embodiment of the present invention.

[0048] Figure 2 A flowchart illustrating a method for obtaining the frequency of rotation in each direction according to an embodiment of the present invention;

[0049] Figure 3 This is a structural diagram of a centrifuge rotor stall detection system based on a smart sensor, provided in one embodiment of the present invention.

[0050] Figure 4This is a schematic diagram of a computer device provided according to an embodiment of the present invention. Detailed Implementation

[0051] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of the centrifuge rotor stall detection method based on intelligent sensors proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0052] 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 this invention pertains.

[0053] The specific scheme of the centrifuge rotor stall detection method based on intelligent sensors provided by the present invention will be described in detail below with reference to the accompanying drawings.

[0054] Example 1:

[0055] The implementation scenario of this invention is that the unloading centrifuge has been working for at least 3 months and has gone through the unloading stage under normal working conditions.

[0056] This invention proposes a centrifuge rotor stall detection method based on intelligent sensors. Please refer to [link / reference]. Figure 1 The diagram illustrates a schematic flowchart of a centrifuge rotor stall detection method based on a smart sensor, according to an embodiment of the present invention. The method includes the following steps:

[0057] Step S1: Real-time acquisition of the centrifuge rotor current signal, reference rotation frequency for each cycle, and vibration signal in each direction during the unloading stage.

[0058] Specifically, a discharge centrifuge is known to have four operating stages: acceleration, stable separation, discharge, and cleaning. The discharge stage is the core operation of the centrifuge. It should be noted that the acceleration stage involves the centrifuge rotor accelerating from rest to the target speed; the stable separation stage separates materials at a constant rotor speed; the discharge stage involves strong piston-driven material discharge interference; and the cleaning stage washes away residue from the drum. Therefore, only during the discharge stage is the centrifuge rotor signal affected by the discharge impact. Current centrifuge rotor stall detection methods cannot accurately detect rotor stall during the discharge stage because both rotor vibration and current are affected by the discharge impact.

[0059] To accurately detect centrifuge rotor stall during the unloading stage in real time, this embodiment adds a non-intrusive AC current transformer to the main power line of the drive motor. This transformer can be connected without disconnecting the power line and can safely and accurately monitor current changes during motor operation, thereby acquiring the centrifuge rotor current signal in real time. A triaxial integrated industrial accelerometer is installed at the centrifuge rotor bearing position. This accelerometer has three spatially perpendicular sensing elements built in, which can synchronously collect the vibration signals of the centrifuge rotor in each direction in real time, namely the vibration signals in the three orthogonal directions of X, Y and Z, completely capturing the multi-directional vibration characteristics caused by the rotation of the centrifuge rotor and the impact of unloading. Then, the control system of the unloading centrifuge identifies each working stage of the centrifuge rotor in real time, thereby acquiring the current signal and vibration signal in each direction of the centrifuge rotor in real time during the unloading stage.

[0060] It is known that the piston extension (pushing material) to reset (preparing for the next push) during the unloading phase is a complete process. Therefore, in this embodiment, the cycle of the unloading phase is set to the duration of one complete action from piston extension (pushing material) to reset (preparing for the next push). The reference rotational frequency of the centrifuge rotor for each cycle during the unloading phase is acquired in real time by the control system. One cycle corresponds to one reference rotational frequency, which is the centrifuge rotor speed for that cycle divided by 60.

[0061] Step S2: Based on the vibration signals of the centrifuge rotor in each direction during the current cycle during the unloading stage and the vibration signals of the centrifuge rotor in each direction during each cycle during the historical normal operation of the centrifuge rotor during the unloading stage, obtain the rotational frequency of each direction in the current cycle; based on the difference in rotational frequency of each direction between the current cycle and the previous adjacent cycle, obtain the first stall risk level of the current cycle.

[0062] It is known that during the unloading process, the centrifuge rotor will generate periodic vibrations due to the piston pushing material. The frequency pattern of the vibration signal can directly reflect the unloading impact law. However, considering that the vibration signal may be affected by random interference and noise in actual conditions, in order to eliminate the influence of random interference and noise and more accurately obtain the vibration signal corresponding to each direction in a single cycle under normal operating conditions during the unloading stage, this embodiment performs two-step preprocessing on the vibration signal of each direction in each cycle based on historical normal operating condition data during the unloading stage. First, the time domain alignment of the multi-cycle signals is achieved according to the starting point of the unloading action (such as the trigger time of the piston pushing signal) to ensure that the vibration characteristics of each direction in each cycle are synchronized in the time dimension. Second, the vibration signals of each direction in multiple cycles are fused to generate the vibration signal model corresponding to each direction in a single cycle under normal operating conditions during the unloading stage, effectively reducing the interference of random noise in a single cycle.

[0063] It is known that the vibration signal in each direction during the unloading phase includes both centrifuge rotor vibration signal and unloading impact vibration signal. To better analyze whether the centrifuge rotor is stalling, it is necessary to separate the vibration signal model corresponding to each direction. This is because the unloading impact vibration signal has high energy and significant interference effects, and its spectral pattern is relatively fixed and easily identifiable. Therefore, the unloading impact vibration interference signal in each direction under normal operating conditions can be separated, preparing for subsequent real-time analysis of centrifuge rotor stall during the unloading phase. Furthermore, by analyzing the difference between the spectral diagrams corresponding to the vibration signal in each direction of the centrifuge rotor in the current cycle and the unloading impact vibration interference signal in the same direction during the unloading phase, the rotational frequency in each direction of the current cycle can be obtained. The method for obtaining the spectral diagram is a well-known technique and will not be elaborated further.

[0064] Considering that the rotational frequency amplitude in each direction should remain stable under normal circumstances, but when the centrifuge rotor becomes unbalanced, i.e., stalls, the rotational frequency amplitude in each direction will show an upward trend, this embodiment obtains the first stall risk level of the current cycle based on the difference in rotational frequency in each direction between the current cycle and the previous adjacent cycle. The greater the first stall risk level, the more likely the centrifuge rotor is to stall during the unloading stage.

[0065] Preferably, in one feasible embodiment, the method for obtaining the frequency shift in each direction is described in reference [reference needed]. Figure 2 The document presents a flowchart of a method for obtaining the frequency of rotation in each direction, as provided in this embodiment. The method includes the following steps:

[0066] Step S201: Based on the vibration signals of the centrifuge rotor in each direction during each cycle of its historical normal operation during the unloading stage, obtain the periodic vibration model in each direction during the current unloading stage.

[0067] In one possible implementation of this embodiment, the method for obtaining the periodic vibration model is as follows: First, the historical normal operating cycles during the unloading phase within a preset historical time period are all taken as normal reference cycles; wherein, the end time of the preset historical time period is the current time; in this embodiment, the duration of the preset historical time period is set to 3 months, and the implementer can set the size of the preset historical time period according to the actual situation, which is not limited here. It should be noted that the normal reference cycle can be directly determined by the control system. Considering that the state of industrial equipment will change dynamically over time, such as rotor wear and performance adjustments after component maintenance, the recently collected normal unloading vibration signals should be given higher weight, and the old vibration signals should be given decreasing weight, so as to track the current operating characteristics of the equipment in real time. If the old vibration signals are over-reliant, the periodic vibration model obtained later will not be able to reflect the latest state changes of the equipment (such as the slight shift in vibration amplitude caused by wear), which will lead to errors in subsequent interference elimination and stall risk assessment, affecting the accuracy of detection.

[0068] Furthermore, in this embodiment, the normal reference cycles are arranged according to time sequence to obtain a cycle sequence. Then, the cycle sequence is numbered sequentially from left to right, from smallest to largest, to determine the number of each normal reference cycle. In this embodiment, the numbering order from smallest to largest is set to 1, 2, 3..., but the implementer can set the numbering according to the actual situation, which is not limited here. The ratio of the number of each normal reference cycle to the sum of the numbers of all normal reference cycles is used as the participation weight of each normal reference cycle. For any normal reference cycle and any direction, the product of the participation weight of the normal reference cycle and the vibration signal in that direction of the normal reference cycle is used as the participation vibration signal in that direction of the normal reference cycle. Finally, the sum of the participation vibration signals in that direction of all normal reference cycles is used as the periodic vibration model in that direction during the current unloading stage.

[0069] Step S202: The periodic vibration model is separated using blind source separation technology to obtain the unloading impact vibration interference signal in each direction during the current unloading stage.

[0070] The vibration signals acquired in step S1 for each direction fall into two main categories: unloading impact vibration signals and other signals such as rotor vibration. The unloading impact vibration signal originates from the periodic pushing action of the piston. It not only has high energy and a significant interference effect on the rotor vibration signal, but its spectral pattern is also relatively fixed (e.g., the characteristic peaks corresponding to the pushing frequency are clear and stable), providing easily identifiable signal characteristics. In contrast, other signals such as rotor vibration (e.g., vibrations caused by imbalance or bearing wear) and environmental noise have complex and variable spectral characteristics (e.g., the rotor imbalance vibration frequency dynamically changes with rotational speed), and their energy is generally lower than that of the unloading impact vibration signal, making them easily masked by the latter. Therefore, this embodiment uses blind source separation technology to separate the periodic vibration model, directly identifying the unloading impact vibration interference signals in each direction during the current unloading stage. This removes interference obstacles for subsequent extraction of the true characteristics of rotor vibration and accurate assessment of stall risk. The blind source separation technology is a well-known technique and will not be elaborated further.

[0071] Step S203: Obtain the rotation frequency in each direction.

[0072] For any direction, the difference between the frequency spectrum of the vibration signal in that direction and the unloading impact vibration interference signal in that direction during the current period is used as the reference frequency spectrum for that direction during the current period; then, the rotational frequency in the reference frequency spectrum is used as the rotational frequency for that direction during the current period. Here, the rotational frequency represents the fundamental vibration frequency of the centrifuge rotor during normal rotation and is a key indicator for identifying whether the centrifuge rotor has an imbalance problem.

[0073] Preferably, in one feasible embodiment of this invention, the method for obtaining the first stall risk level is as follows: for any direction, the difference between the rotational frequency of the current cycle and that of the previous adjacent cycle in that direction is taken as the degree of increase in rotational frequency in that direction in the current cycle; the greater the degree of increase in rotational frequency, the more likely the centrifuge rotor is to stall during the unloading stage; in order to more accurately analyze the stall situation of the centrifuge rotor during the unloading stage, this embodiment further obtains the absolute value of the difference between the degree of increase in rotational frequency of the main direction of the current cycle and that of each other direction, and takes them as the first difference; when the first difference is more equal, it indicates that the stall analysis of the centrifuge rotor in the current cycle is more accurate. Among them, the main direction is the direction with the strongest vibration signal and the highest correlation with centrifuge rotor failure. It is essentially determined by the mechanical structure of the centrifuge and the rotor motion law. The main direction of a horizontal discharge centrifuge is mostly the horizontal or vertical direction perpendicular to the rotation axis. This is because the rotor of a horizontal centrifuge usually rotates around a horizontal axis (let's assume it's the X-axis, i.e., the rotor rotation axis is the X-axis). At this time, the vibration generated by rotor imbalance and discharge impact will form the strongest signal in the plane perpendicular to the X-axis (YZ plane). If the piston pushes along the Z-axis (horizontal and perpendicular to the rotation axis), the pushing impact force will make the Z-axis vibration amplitude significantly higher than the Y-axis, and the Z-axis vibration amplitude will reflect the fault first. Therefore, the Z-axis is the main direction. If the rotor is more obviously unbalanced in the Y-axis direction (vertical direction) due to material accumulation (such as material accumulating below the rotor), the Y-axis vibration amplitude is the largest and reflects the fault first. Therefore, the Y-axis is the main direction. The dominant direction of a vertical centrifuge is usually an orthogonal direction in the horizontal plane. This is because the rotor of a vertical centrifuge rotates around a vertical axis (let's say the Z-axis), and the vibration is mainly concentrated in the horizontal plane (XY plane) perpendicular to the Z-axis. The dominant direction is then determined from the X and Y axes. If the rotor experiences a continuous imbalance in the X-axis direction (a fixed horizontal direction) due to installation deviation, and the X-axis signal becomes abnormal first during a fault, then the X-axis is the dominant direction. If the mechanical impact during unloading is mainly transmitted along the Y-axis, and the Y-axis vibration signal has the highest signal-to-noise ratio (less interference, stronger effective signal), then the Y-axis is the dominant direction.

[0074] Furthermore, the absolute value of the difference between any two first differences is taken as the second difference; then, the mean of the second differences is negatively correlated and normalized, and the result is used as the first reference weight for the current cycle; in this embodiment, the negative of the mean of the second differences is used as the power of an exponential function with the natural constant as the base, and the output of this exponential function is the result of negatively correlated and normalized mean of the second differences. It should be noted that because this embodiment only has three directions, X, Y, and Z, there is essentially only one second difference, while other embodiments may have more directions. The larger the first reference weight, the more accurate the degree of frequency increase in the main direction of the current cycle. Therefore, the product of the first reference weight and the degree of frequency increase in the main direction of the current cycle is used as the first stall risk level for the current cycle.

[0075] Step S3: Based on the difference between the amplitude corresponding to each frequency of the current signal in the current cycle and the amplitude corresponding to the current signal in the previous adjacent cycle, and the relationship between each frequency of the current signal in the current cycle and the fixed frequency of the power supply and the reference frequency of the current cycle, obtain the second stall risk level of the current cycle.

[0076] In centrifuge rotor stall risk assessment, vibration and current signals are known to be complementary dimensions for verifying the physical phenomenon of rotor imbalance, forming a strong correlation to support the accurate confirmation of stall risk. Rotor imbalance is a key trigger for stall, simultaneously generating two types of measurable physical effects. On the one hand, it causes periodic mechanical vibration of the centrifuge rotor, which is represented by the amplitude corresponding to the rotational frequency in the vibration signal. On the other hand, the resulting vibration exacerbates motor load fluctuations, causing synchronous fluctuations in the motor current, which are represented by current fluctuation amplitudes consistent with the rotational frequency in the current signal. Under normal operating conditions, when the centrifuge rotor imbalance intensifies, the amplitude corresponding to the rotational frequency of the vibration signal increases sharply, and the amplitude corresponding to the frequency of the current signal also increases synchronously. These two signals are strongly coupled through the transmission link of vibration → load fluctuation → current change. It is known that a preliminary risk assessment has been completed through vibration signal analysis in step S2. Further cross-verification using current signals is required to eliminate false judgments. Therefore, in this embodiment, the change of the amplitude corresponding to the frequency of the current signal is analyzed based on the difference between the amplitude corresponding to each frequency of the current signal in the current cycle and the amplitude corresponding to the current signal in the previous adjacent cycle. This analysis further examines the imbalance of the centrifuge rotor in the current cycle.

[0077] Considering that in actual situations, the current signal during the unloading stage may encounter problems such as unstable grid voltage and aging motor windings, and also considering that the current signal is a single-dimensional output and cannot be directly decomposed into independent components such as load fluctuations and grid interference using blind source separation technology like vibration signals, this paper addresses the issue of frequency characteristics of the current signal being clearly related to the centrifuge's operating mechanism. The motor's power supply frequency determines the fundamental frequency of the current, and its integer multiple harmonics are generated due to the motor's electrical characteristics (such as core magnetic saturation). Furthermore, the periodic load fluctuations caused by the centrifuge rotor's rotation will superimpose frequency components related to integer multiples of the reference rotation frequency into the current. Based on these principles, this embodiment extracts the frequencies in the current signal of the current cycle using Fourier transform, then analyzes the relationship between each frequency of the current signal in the current cycle and the fixed power supply frequency and the reference rotation frequency of the current cycle. This allows for a more accurate analysis of the centrifuge rotor's stall condition in the current cycle using the current signal. The Fourier transform is a well-known technique and will not be elaborated upon further.

[0078] Therefore, this embodiment obtains the second stall risk level for the current cycle based on the difference between the amplitude corresponding to each frequency of the current signal in the current cycle and the amplitude corresponding to the current signal in the previous adjacent cycle, as well as the relationship between each frequency of the current signal in the current cycle and the fixed frequency of the power supply and the reference frequency of the current cycle. The greater the second stall risk level, the more likely the centrifuge rotor is to stall during the unloading stage.

[0079] Preferably, in one achievable manner of this embodiment, the method for obtaining the second stall risk level is as follows: for any frequency of the current signal in the current cycle, firstly, based on the relationship between the frequency and the fixed frequency of the power supply and the reference frequency of the current cycle, obtain the degree of correspondence between the frequency of the current cycle and the frequency. The greater the degree of correspondence, the more meaningful the frequency is for reference.

[0080] The method for obtaining the degree of correspondence is as follows: For any integer in a preset set of integers, the product of that integer and the reference frequency of the current cycle is used as the first reference value. In this embodiment, the preset set of integers is set to (-6, -5, -4, -3, -2, -1, 0, 1, 2, 3, 4, 5, 6). Implementers can set the preset set of integers according to actual conditions, and there is no limitation here. Further, the sum of the first reference value and the fixed frequency of the power supply is used as the second reference value. Then, the absolute value of the difference between the second reference value and the frequency is negatively correlated and normalized, and the result is used as the reference degree of correspondence corresponding to that integer. Finally, the smallest reference degree of correspondence corresponding to the integer in the preset set of integers is used as the degree of correspondence between the frequency of the current cycle and the frequency. In this embodiment, the negative of the absolute value of the difference between the second reference value and the frequency is used as the power of an exponential function with the natural constant as the base. The output of the exponential function is the result of negatively correlated and normalized difference between the second reference value and the frequency.

[0081] The difference between the amplitude of the current signal corresponding to this frequency in the current cycle and the current signal in the previous adjacent cycle is taken as the first characteristic value of this frequency. The larger the first characteristic value, the more likely the centrifuge rotor is to stall during the unloading stage. To accurately characterize the centrifuge rotor imbalance corresponding to this frequency, this embodiment normalizes the product of the corresponding degree and the first characteristic value as the degree of anomaly of this frequency. To comprehensively characterize the centrifuge rotor imbalance corresponding to the current signal in the current cycle, this embodiment further normalizes the sum of the degrees of anomaly of all frequencies of the current signal in the current cycle as the second degree of stall risk in the current cycle. This embodiment normalizes the sum of the degrees of anomaly of all frequencies of the current signal in the current cycle using the norm normalization function.

[0082] Step S4: Based on the first stall risk level and the second stall risk level, obtain the overall stall risk level of the current cycle, and determine whether the centrifuge rotor stalls during the unloading stage.

[0083] It is known that the greater the first stall risk level and the greater the second stall risk level, the more likely the centrifuge rotor is to stall during the unloading phase. Therefore, this embodiment obtains the overall stall risk level of the current cycle based on the first and second stall risk levels, and then accurately determines whether the centrifuge rotor is stalling during the unloading phase. This allows for timely and accurate detection of centrifuge rotor stalling, enabling timely handling of the centrifuge rotor's operation, reducing equipment damage caused by centrifuge rotor stalling, and helping to ensure the stable operation of the unloading centrifuge and improve its working efficiency.

[0084] Preferably, in one feasible embodiment of this invention, the overall stall risk level is obtained by adding the first stall risk level and the second stall risk level together and then normalizing the result, which is taken as the overall stall risk level for the current cycle. In this embodiment, the sum of the first and second stall risk levels is normalized using a normalization function.

[0085] Preferably, in one feasible embodiment of this invention, the method for determining whether the centrifuge rotor is stalling during the unloading stage is as follows: It is known that the greater the overall stall risk level, the greater the risk of stalling during the unloading stage. Therefore, this embodiment sets a preset stall risk level threshold of 0.6. The implementer can set the preset stall risk level threshold according to actual conditions, and this is not limited here. When the overall stall risk level is greater than the preset stall risk level threshold, it is determined that the centrifuge rotor is stalling during the unloading stage; when the overall stall risk level is less than or equal to the preset stall risk level threshold, it is determined that the centrifuge rotor is not stalling during the unloading stage.

[0086] In summary, this embodiment acquires the current signal of the centrifuge rotor, the reference rotational frequency for each cycle, and the vibration signal in each direction during the unloading phase in real time. Based on the vibration signal, it acquires the rotational frequency in each direction for the current cycle. Based on the difference in rotational frequency in each direction between the current cycle and the previous adjacent cycle, it acquires the first stall risk level. Based on the amplitude corresponding to the frequency of the current signal in the current cycle and the amplitude corresponding to the current signal in the previous adjacent cycle, it acquires the second stall risk level. Based on the first and second stall risk levels, it acquires the overall stall risk level to determine whether the rotor is stalling during the unloading phase. This invention, by accurately determining whether the centrifuge rotor is stalling during the unloading phase in real time, facilitates timely detection and handling of rotor anomalies, effectively reducing losses caused by stalling.

[0087] Example 2:

[0088] This invention also proposes a centrifuge rotor stall detection system based on intelligent sensors; please refer to [link to relevant documentation]. Figure 3 The diagram illustrates a centrifuge rotor stall detection system based on a smart sensor, according to an embodiment of the present invention. The system includes: a data acquisition module 10, a first stall risk level acquisition module 20, a second stall risk level acquisition module 30, and a stall judgment module 40.

[0089] The data acquisition module 10 is used to acquire in real time the current signal of the centrifuge rotor, the reference rotation frequency for each cycle, and the vibration signal in each direction during the unloading stage.

[0090] The first stall risk level acquisition module 20 is used to acquire the rotational frequency of each direction in the current cycle based on the vibration signal of the centrifuge rotor in each direction during the current cycle and the vibration signal of the centrifuge rotor in each direction during the historical normal operation of the centrifuge rotor in the unloading stage; and to acquire the first stall risk level of the current cycle based on the difference in rotational frequency of each direction between the current cycle and the previous adjacent cycle.

[0091] The second stall risk level acquisition module 30 is used to acquire the second stall risk level of the current cycle based on the difference between the amplitude corresponding to each frequency of the current signal in the current cycle and the amplitude corresponding to the current signal in the previous adjacent cycle, as well as the relationship between each frequency of the current signal in the current cycle and the fixed frequency of the power supply and the reference frequency of the current cycle.

[0092] Stall determination module 40 is used to obtain the overall stall risk level of the current cycle based on the first stall risk level and the second stall risk level, and to determine whether the centrifuge rotor stalls during the unloading stage.

[0093] It should be noted that the system provided in the above embodiments is only an example of the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the computer device can be divided into different functional modules to complete all or part of the functions described above. In addition, the centrifuge rotor stall detection system based on intelligent sensors and the centrifuge rotor stall detection method based on intelligent sensors provided in the above embodiments belong to the same concept. The specific implementation process is detailed in the method embodiments and will not be repeated here.

[0094] Example 3:

[0095] This invention also proposes a centrifuge rotor stall detection device based on intelligent sensors. The device includes a memory and a processor. The memory stores executable program code, and the processor calls and executes this executable program code to perform a centrifuge rotor stall detection method based on intelligent sensors provided in the embodiments of this application. Specifically, the device may be a chip, component, or module. The chip may include a connected processor and memory; the memory stores instructions, and when the processor calls and executes the instructions, the chip can perform the centrifuge rotor stall detection method based on intelligent sensors provided in the above embodiments.

[0096] Furthermore, this application also protects a computer device; please refer to [link to relevant documentation]. Figure 4 The computer device includes a memory 401, a processor 402, and a computer program 403 stored in the memory 401 and running on the processor 402. When the processor 402 executes the computer program 403, the computer device can execute any of the aforementioned centrifuge rotor stall detection methods based on smart sensors.

[0097] Example 4:

[0098] This embodiment also provides a computer-readable storage medium storing computer program code. When the computer program code is run on a computer, the computer executes the above-described related method steps to implement the centrifuge rotor stall detection method based on intelligent sensors provided in the above embodiment.

[0099] Example 5:

[0100] This embodiment also provides a computer program product that, when run on a computer, causes the computer to perform the aforementioned related steps to implement the centrifuge rotor stall detection method based on intelligent sensors provided in the above embodiment.

[0101] In this embodiment, the device, computer-readable storage medium, computer program product, or chip are all used to execute the corresponding methods provided above. Therefore, the beneficial effects they can achieve can be referred to the beneficial effects in the corresponding methods provided above, and will not be repeated here.

[0102] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0103] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A method for detecting centrifuge rotor stall based on intelligent sensors, characterized in that, The method includes the following steps: Real-time acquisition of the centrifuge rotor's current signal, reference rotation frequency for each cycle, and vibration signal in each direction during the unloading phase; Based on the vibration signals of the centrifuge rotor in each direction during the current cycle during the unloading phase and the vibration signals of the centrifuge rotor in each direction during each cycle during the historical normal operation of the centrifuge rotor during the unloading phase, the rotational frequency of each direction in the current cycle is obtained; based on the difference in rotational frequency of each direction between the current cycle and the previous adjacent cycle, the first stall risk level of the current cycle is obtained. The second stall risk level of the current cycle is obtained based on the difference between the amplitude corresponding to each frequency of the current signal in the current cycle and the amplitude corresponding to the current signal in the previous adjacent cycle, as well as the relationship between each frequency of the current signal in the current cycle and the fixed frequency of the power supply and the reference frequency of the current cycle. Based on the first stall risk level and the second stall risk level, the overall stall risk level of the current cycle is obtained, and it is determined whether the centrifuge rotor stalls during the unloading phase. The method for obtaining the frequency of rotation in each direction is as follows: Based on the vibration signals of the centrifuge rotor in each direction during each cycle of its historical normal operation during the unloading phase, obtain the periodic vibration model in each direction during the current unloading phase. The periodic vibration model is separated using blind source separation technology to obtain the unloading impact vibration interference signal in each direction during the current unloading stage. For any direction, the difference between the spectrum diagrams corresponding to the vibration signal in that direction and the unloading impact vibration interference signal in that direction during the current period is used as the reference spectrum diagram for that direction during the current period. The frequency shift in the reference spectrum is taken as the frequency shift in that direction for the current period; The method for obtaining the first stall risk level is as follows: For any direction, the difference in frequency of rotation in that direction between the current period and the previous adjacent period is taken as the degree of increase in frequency of rotation in that direction during the current period. The difference in the degree of frequency increase between the current main direction and each other direction is obtained and used as the first difference; among them, the main direction is the direction with the strongest vibration signal and the highest correlation with centrifuge rotor failure. The difference between any two first differences is taken as the second difference; the mean of the second differences is negatively correlated and normalized, and the result is taken as the first reference weight of the current period. The product of the first reference weight and the degree of frequency increase in the main direction of the current cycle is used as the first stall risk level of the current cycle. The method for obtaining the second stall risk level is as follows: For any frequency of the current signal in the current cycle, based on the relationship between that frequency and the fixed frequency of the power supply and the reference frequency of the current cycle, the degree of correspondence between the frequency of the current cycle and that frequency is obtained. The difference between the amplitude of the frequency in the current signal of the current period and the current signal of the previous adjacent period is taken as the first characteristic value of the frequency. The normalized result of the product of the corresponding degree and the first feature value is taken as the degree of abnormality of the frequency; The sum of the abnormality levels of all frequencies of the current signal in the current cycle and the result of normalization are used as the second stall risk level for the current cycle.

2. The centrifuge rotor stall detection method based on intelligent sensors as described in claim 1, characterized in that, The method for obtaining the periodic vibration model is as follows: The historical normal operating cycle of the unloading phase within the preset historical time period is used as the normal reference cycle; where the end time of the preset historical time period is the current time. The normal reference periods are arranged according to the time sequence to obtain the period sequence. The period sequence is then numbered sequentially from left to right, from smallest to largest, to determine the number of each normal reference period. The ratio of the label of each normal reference period to the sum of the labels of all normal reference periods is used as the participation weight of each normal reference period. For any normal reference period and any direction, the product of the participation weight of the normal reference period and the vibration signal in that direction of the normal reference period is taken as the participation vibration signal in that direction of the normal reference period. The sum of all participating vibration signals in this direction during the normal reference period is used as the periodic vibration model in this direction during the current unloading stage.

3. The centrifuge rotor stall detection method based on intelligent sensors as described in claim 1, characterized in that, The method for obtaining the corresponding degree is as follows: For any integer in the preset set of integers, the product of that integer and the reference frequency of the current period is used as the first reference value; The sum of the first reference value and the fixed frequency of the power supply is used as the second reference value; The result of negatively correlating and normalizing the difference between the second reference value and the frequency is used as the reference correspondence degree corresponding to the integer. The minimum reference correspondence degree corresponding to the integer in the preset integer set is taken as the correspondence degree between the current cycle frequency and the frequency.

4. The centrifuge rotor stall detection method based on intelligent sensors as described in claim 1, characterized in that, The method for obtaining the overall stall risk level is as follows: The sum of the first stall risk level and the second stall risk level, after normalization, is taken as the overall stall risk level for the current cycle.

5. The centrifuge rotor stall detection method based on intelligent sensors as described in claim 1, characterized in that, The method for determining whether the centrifuge rotor is stalling during the unloading phase is as follows: When the overall stall risk level is greater than the preset stall risk level threshold, it is determined that the centrifuge rotor is stalling during the unloading stage. When the overall stall risk level is less than or equal to the preset stall risk level threshold, it is determined that the centrifuge rotor does not stall during the unloading stage.

6. The centrifuge rotor stall detection method based on intelligent sensors as described in claim 1, characterized in that, The cycle is the time corresponding to one complete action from piston extension (pushing material) to reset (preparing for the next pushing of material).

7. The centrifuge rotor stall detection method based on intelligent sensors as described in claim 1, characterized in that, The directions include the three directions: X, Y, and Z.

Images