Magnetic suspension vacuum pump and control method thereof

Abstract

The invention provides a magnetic suspension vacuum pump and a control method thereof. The magnetic suspension vacuum pump comprises a machine shell, a driving shaft arranged in the machine shell in a matched mode, an impeller arranged in the machine shell at one end of the driving shaft in a matched mode, a motor and a combined type magnetic bearing unit arranged outside the driving shaft behind the impeller in a matched mode, and a combined type displacement detection unit matched with the combined type magnetic bearing unit. A controller is matched with the motor, the combined magnetic bearing unit and the combined displacement detection unit; the combined displacement detection unit obtains combined displacement detection data, and the controller adjusts the output of the combined magnetic bearing unit and the motor based on the trained improved prediction model; according to the control method, sufficient historical data are collected through feedback control, a prediction model is trained and improved based on the data, and feedforward control of the magnetic suspension vacuum pump is achieved through the improved prediction model. Feedforward control of the magnetic suspension vacuum pump is achieved, the use efficiency is improved, the service life is prolonged, and different control measures and effect reproduction are achieved; the method can be applied to rapid fault point detection and fault elimination.

Description

Technical Field

[0001] The invention relates to the technical field of non-variable displacement pumps, and in particular to a magnetic suspension vacuum pump and a control method thereof. Background Art

[0002] The magnetic levitation vacuum pump is a vacuum device that uses magnetic force to drive a rotating body. It suspends the impeller (and its drive shaft) in space through magnetic force, forming a relatively stable gap between the impeller and the pump casing, thereby achieving a contactless vacuum working environment. Liquid or gas is then sucked in at the inlet of the pump and accelerated to a high speed under the action of the impeller before being discharged from the outlet.

[0003] The working principle of the magnetic levitation vacuum pump allows it to avoid the friction and wear caused by traditional mechanical bearings during use, thereby significantly improving the efficiency and life of the pump. The durable feature is particularly prominent, greatly reducing the energy loss caused by friction. Compared with traditional vacuum pumps, it can save energy by 30% - 70%, and also greatly reduces the frequency of maintenance; and because the direct contact during the working process is almost zero, the noise generated during operation is relatively low, with less pollution, and it is more environmentally friendly.

[0004] In order to ensure the normal operation and service life of the magnetic levitation vacuum pump, it is necessary to perform real-time adjustments based on the point displacement of the drive shaft during operation, so that the drive shaft and impeller always remain within a reasonable operating range, thereby optimizing the control scheme.

[0005] In the prior art, the axial displacement sensor is generally set at the motor position. The displacement monitoring at a single position is not perfect for monitoring the state of the drive shaft, and the radial displacement monitoring at a single position cannot optimize the control method of the magnetic levitation bearing. Although the axial displacement sensor monitors the movement of the drive shaft in the length direction, in fact the ultimate goal is to monitor the movement of the impeller. The above monitoring methods all have certain deviations.

[0006] A Chinese patent with publication number CN101243263A discloses a magnetic bearing device with an improved vacuum feedthrough, which discloses a first sensor plate, a second sensor plate and a third sensor plate, attempting to achieve accurate monitoring of the displacement of the drive shaft; however, its monitoring is also feedback-type monitoring, and ultimately "feedback control" is performed, which can only be an optimization for the overall control of the drive shaft, and cannot fundamentally solve the problem of position offset of the drive shaft. Summary of the invention

[0007] In order to solve the above technical problems, the present invention provides a magnetic levitation vacuum pump and a control method thereof.

[0008] The technical solution adopted by the present invention is a magnetic levitation vacuum pump, comprising a casing, a driving shaft is arranged in the casing, an impeller is arranged in the casing at one end of the driving shaft, a motor and a combined magnetic bearing unit are arranged outside the driving shaft behind the impeller, a combined displacement detection unit is arranged in the combined magnetic bearing unit; a controller is arranged in conjunction with the motor, the combined magnetic bearing unit and the combined displacement detection unit; The combined displacement detection unit obtains combined displacement detection data, and the controller adjusts the output of the combined magnetic bearing unit and the motor based on the trained improved prediction model.

[0009] Preferably, the combined magnetic bearing unit includes radial magnetic bearings respectively arranged outside the driving shaft at both ends of the motor and axial magnetic bearings respectively arranged outside the two radial magnetic bearings, and a fixed plate is provided on the driving shaft to cooperate with the axial magnetic bearings; the radial magnetic bearings and the axial magnetic bearings constitute a balancing system with the impeller.

[0010] Preferably, the combined displacement detection unit comprises a radial displacement sensor arranged in cooperation with any radial magnetic bearing and an axial displacement sensor arranged in cooperation with any fixed plate.

[0011] Preferably, any of the axial displacement sensors comprises one or more measured light points arranged outside the fixed disk, a detection window is provided in conjunction with the measured light points, and the detection window is arranged in conjunction with the controller.

[0012] Preferably, an XOY coordinate system is established in conjunction with the detection window, with the origin of the XOY coordinate system corresponding to the position of the measured light point. When any measured light point is displaced, the controller controls the adjustment and obtains the displacement vectors of the two measured light points and feeds them back to the controller.

[0013] Preferably, the improved prediction model comprises a dual-branch structure; One branch includes several convolutional layers connected in sequence, each convolutional layer is followed by a time series feature aggregation layer, and the last time series feature aggregation layer is followed by a fully connected layer, a softmax layer and an output layer; one or more convolutional blocks are arranged in parallel in any of the convolutional layers; The other branch includes several LSTM units, and the output of any LSTM is associated with the input of the corresponding time series feature aggregation layer.

[0014] Preferably, the trend of continuous data is obtained as a training label; the data used to train the improved prediction model includes detection data of a continuous combined displacement detection unit and corresponding control parameters, and the detection data is used to characterize the spatial position of the drive shaft in the current spatial coordinate system; paired detection data and control parameters with time series characteristics are used as training data pairs, which are input into the improved prediction model and trained in conjunction with the training labels.

[0015] Preferably, the casing comprises a main casing and a volute that are arranged in coordination with each other, the volute is provided with an air inlet axially toward the impeller, and the volute is provided with an air outlet at the blades of the impeller; an elastic member is provided between the assembly points of the main casing and the volute.

[0016] Preferably, a temperature sensor is provided in the main housing, and the temperature sensor is arranged in cooperation with the controller; and a plurality of guide holes are provided on the main housing.

[0017] A control method for a magnetic levitation vacuum pump, the method comprising the following steps: S1 is equipped with a magnetic levitation vacuum pump, the combined displacement detection unit is initialized, and the controller obtains the initial signal; S2 obtains the feedback signal and control parameters of the radial displacement sensor; obtains the signal of the measured light point of the axial displacement sensor captured in the detection window; Obtain the displacement vectors of the two measured light points and their relationship parameters, record the time and corresponding control parameters, and mark them manually; S3 obtains sufficient non-empty detection data and corresponding control parameters of the continuous combined displacement detection unit, obtains manually labeled training labels, and inputs them into the improved prediction model for training; S4 obtains a number of groups of continuous non-empty detection data of the combined displacement detection units and inputs them into the improved prediction model. The improved prediction model outputs classification information to the controller, and the controller adjusts the control parameters to perform control adjustment.

[0018] The present invention provides a magnetic levitation vacuum pump and a control method thereof, comprising a casing, a driving shaft being arranged in the casing, an impeller being arranged in the casing at one end of the driving shaft, a motor and a combined magnetic bearing unit being arranged outside the driving shaft behind the impeller, and a combined displacement detection unit being arranged in the combined magnetic bearing unit; a controller being arranged in the motor, the combined magnetic bearing unit and the combined displacement detection unit; the combined displacement detection unit obtaining combined displacement detection data, and the controller adjusting the output of the combined magnetic bearing unit and the motor based on a trained improved prediction model; the control method collecting sufficient historical data through feedback control, and improving the prediction model based on the training, and finally realizing feedforward control of the magnetic levitation vacuum pump through the improved prediction model.

[0019] The beneficial effects of the present invention are: (1) Improve the control logic of the magnetic levitation vacuum pump. After collecting sufficient data and implementing feedback control during operation or testing, train an improved prediction model based on the collected data, and implement feedforward control of the magnetic levitation vacuum pump based on the improved prediction model, thereby improving the efficiency of the magnetic levitation vacuum pump and extending its service life. (2) A combined displacement detection unit is used to obtain a more detailed state representation and spatial form of the drive shaft, which is used to fit the detailed characteristics of the drive shaft under different states, thereby achieving the reproduction of different control measures and their effects; (3) It can be used to quickly detect fault points and troubleshoot. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is a cross-sectional structural schematic diagram of the present invention; Figure 2 It is a schematic diagram of the cross-sectional structure of the main casing and the volute assembly surface of the present invention; Figure 3 This is a block diagram of the improved prediction model of the present invention. DETAILED DESCRIPTION

[0021] The present invention is further described in detail below in conjunction with embodiments, but the protection scope of the present invention is not limited thereto.

[0022] Example 1

[0023] A magnetic levitation vacuum pump comprises a casing, a driving shaft 1 is arranged in the casing, an impeller 2 is arranged in the casing at one end of the driving shaft 1, a motor 3 and a combined magnetic bearing unit are arranged outside the driving shaft 1 behind the impeller 2, a combined displacement detection unit is arranged in the combined magnetic bearing unit; a controller is arranged in conjunction with the motor 3, the combined magnetic bearing unit and the combined displacement detection unit; The combined displacement detection unit obtains combined displacement detection data, and the controller adjusts the output of the combined magnetic bearing unit and the motor 3 based on the trained improved prediction model.

[0024] In this embodiment, the impeller 2 is driven to rotate by the rotation of the driving shaft 1. The impeller 2 here is generally a three-dimensional flow impeller 2. After process treatment and optimization, it can accelerate and discharge the gas under high-speed rotation, thereby realizing gas transportation and vacuum generation. It is particularly suitable for medical fields, laboratory equipment, environmental protection fields, industrial fields, aerospace fields, etc.

[0025] In this embodiment, the casing is an assembleable structure. In order to better realize the bearing of the drive shaft 1, the impeller 2, etc., a plurality of supports are arranged in the casing to limit the axial or circumferential position of the motor 3 and the combined magnetic bearing unit, etc., which is easily understood by those skilled in the art; For each independent magnetic bearing of the combined magnetic bearing unit, a plurality of electromagnets are symmetrically arranged at its center, and are used to control the independent electromagnets of each independent magnetic bearing with a controller, thereby realizing that the drive shaft 1 is spatially centered during operation; The motor 3 is implemented as a high-speed permanent magnet synchronous motor 3 , which has the characteristics of high efficiency and energy saving, and can cooperate with the magnetic bearing to make the impeller 2 rotate at a high speed.

[0026] In this embodiment, what is more important is that a combined displacement detection unit is provided for the corresponding combined magnetic bearing unit, and the combined displacement detection data of different magnetic bearing units are fully obtained. After the cold start is switched to normal operation, the collected data is trained and applied to improve the prediction model, and the classification (trend) is obtained through the output of the model, and then the output of the combined magnetic bearing unit and the motor 3 is continuously adjusted in the form of feedforward to obtain a better control result; in actual application, the controller realizes the functions of speed control of the motor 3, stable control of the bearing (electromagnet control), and monitoring and protection of the operating status of the entire vacuum pump, and then automatically adjusts the operating parameters of the vacuum pump according to the actual working conditions to ensure its efficient and stable operation.

[0027] Example 2

[0028] Based on Example 1, the combined magnetic bearing unit includes radial magnetic bearings 4 respectively arranged outside the driving shaft 1 at both ends of the motor 3 and axial magnetic bearings 5 ​​respectively arranged outside the two radial magnetic bearings 4, and a fixed plate 6 is provided on the driving shaft 1 to cooperate with the axial magnetic bearings 5; the radial magnetic bearings 4 and the axial magnetic bearings 5 ​​form a balancing system with the impeller 2.

[0029] The combined displacement detection unit includes a radial displacement sensor (not shown in the figure) arranged in cooperation with any radial magnetic bearing 4 and an axial displacement sensor 7 arranged in cooperation with any fixed plate 6 .

[0030] Compared with conventional technology, in this embodiment, axial magnetic bearings 5 ​​are respectively arranged on the relative outer sides of the two radial magnetic bearings 4, and its main purpose is to obtain more accurate axial displacement data and the difference between the two; and on the basis of the radial magnetic bearings 4 and the axial magnetic bearings 5 ​​forming a balanced system with the impeller 2, the axial magnetic bearing 5 also plays more balancing functions; the balanced system here means that during operation, the torque of a certain point between the impeller 2 and the drive shaft 1 is equal to or basically equal to the torque of the radial magnetic bearings 4 and the axial magnetic bearings 5 ​​with this point.

[0031] In this embodiment, in order to obtain more accurate axial displacement data, in addition to setting a radial displacement sensor in conjunction with each radial magnetic bearing 4, an axial displacement sensor 7 is also set in conjunction with each axial magnetic bearing 5. Specifically, an axial displacement sensor 7 is set on the fixed plate 6 to obtain more detailed performances of the two ends of the drive shaft 1, including but not limited to vibration, deformation, and jump.

[0032] Example 3

[0033] On the basis of Example 2, any of the axial displacement sensors 7 includes one or more measured light points arranged outside the fixed disk 6, and a detection window 9 is provided in conjunction with the measured light points, and the detection window 9 is arranged in conjunction with the controller.

[0034] Furthermore, an XOY coordinate system is established in conjunction with the detection window 9, with the origin of the XOY coordinate system corresponding to the position of the measured light point. When any measured light point is displaced, the controller controls the adjustment and obtains the displacement vectors of the two measured light points and feeds them back to the controller.

[0035] The radial displacement sensor is applied to the annular radial magnetic bearing 4, and its control is easy to implement. The axial displacement sensor 7 in the prior art often measures a very small displacement change by inductance, but the data obtained in the electromagnetic environment is not necessarily accurate, there may be a certain jitter, and it is not possible to obtain a small longitudinal displacement (vibration, etc.) well. Therefore, in this embodiment, the measured light point is directly set on the outside of the fixed plate 6, and the detection window 9 can be correspondingly set in the packaging shell or the main shell 10; The detection window 9 provides a larger range, within which the horizontal displacement and "jump" of the measured light point at the end of the drive shaft 12 and their differences can be better obtained, thereby obtaining better state characteristics of the drive shaft 1 and / or the impeller 2; during the high-speed movement of the drive shaft 1, in fact, only one measured light point is needed to achieve its theoretical capture at the same position of the detection window 9, and this "capture" can be achieved by adjusting the detection frequency of the detection window 9; Specifically, the detection window 9 is a rectangular window, which is generally calibrated with the corresponding measured light point at its center point, and the origin is used as the starting point of the vector. In principle, the most ideal state is that the measured light point is always at the origin (or initial position), but this is obviously impossible; when the measured light point moves including the X-axis and Y-axis and between quadrants, it indicates that control adjustment is required, and the vector data is recorded after the control adjustment to collect historical data for the subsequent training of the improved prediction model; it is particularly important to note that when there is a difference between two vectors at the same time, the eigenvalue represented by the difference is more important. This difference is reflected in the angle and delay of the displacement vectors at both ends of the drive shaft 1. Of course, the delay here can also be expressed by the angle.

[0036] Considering that the environment inside the casing is closed, there is almost no interference light (background light) at the measured light point, and more accurate detection results can be obtained.

[0037] Example 4

[0038] Based on Example 1 or 3, the improved prediction model includes a double-branch structure; One branch includes several convolutional layers connected in sequence, each convolutional layer is followed by a time series feature aggregation layer, and the last time series feature aggregation layer is followed by a fully connected layer, a softmax layer and an output layer; one or more convolutional blocks are arranged in parallel in any of the convolutional layers; The other branch includes several LSTM units, and the output of any LSTM is associated with the input of the corresponding time series feature aggregation layer.

[0039] In this embodiment, an improved prediction model is proposed. In fact, the prediction model here fully considers the time series characteristics of the recorded characteristic data, and is different from the prior art. In this embodiment, the driving shaft 1 is simulated and placed in a space environment, that is, the theoretical packaging shell is used as the space environment, so that the shape of the driving shaft 1 in space at each moment can be obtained, thereby better obtaining the collection of abnormal characteristics of the driving shaft 1, especially when vibration and jump occur; Based on this, 3 to 5 convolutional layers are connected in sequence in this embodiment. In fact, each convolutional layer also includes 1 to 3 convolutional blocks arranged in parallel, generally 3, which are used to extract the feature expressions of each surface of the driving shaft 1 after it is placed in space. A time series feature aggregation layer is set after each convolutional layer to obtain the associated features of any time series obtained based on LSTM under another branch, so as to better allocate weights. Finally, after the fully connected layer and the softmax layer, the output layer is output. The output here is the probability distribution of the classification, specifically, the probability distribution of different operating problems. Based on this predicted value, the controller can be quickly adjusted in a feedforward manner to avoid the occurrence of serious problems and better realize the operation of the magnetic levitation vacuum pump.

[0040] In this embodiment, when training the data of the improved prediction model, the trend of the continuous data is obtained as a training label; the data for training the improved prediction model includes the detection data of the continuous combined displacement detection unit and the corresponding control parameters, and the detection data is used to characterize the spatial position of the drive shaft 1 in the current spatial coordinate system; the detection data and the control parameters that are paired and have time series characteristics are used as the training data pair, the detection data is input into the first branch, and the control parameters are input into another branch, and the training is performed in conjunction with the training labels; Furthermore, an error function is constructed to adjust the model parameters, including the cross entropy loss L1 of the model under the true label and the regularization loss of the model parameters. The overall loss L satisfies, L=λ1L1+λ2||Θ||2 2 Among them, λ1 and λ2 are adjustable parameters between 0 and 1, and the second term is the L2 regularization loss to prevent the model from overfitting.

[0041] Example 5

[0042] On the basis of Example 1, the casing includes a main casing 10 and a volute 11 that are arranged in conjunction with each other. The volute 11 is provided with an air inlet 12 axially toward the impeller 2, and the volute 11 is provided with an air outlet 13 at the blades of the impeller 2; an elastic member 16 is provided between the assembly points of the main casing 10 and the volute 11.

[0043] A temperature sensor (not shown in the figure) is arranged in the main housing 10 , and the temperature sensor is arranged in cooperation with the controller; a plurality of guide holes 14 are arranged on the main housing 10 .

[0044] In this embodiment, the main casing 10 and the volute 11 can be arranged in an integrated manner, or they can be installed in a split manner to achieve joint operation and facilitate maintenance. In the case of a split manner, close cooperation between the two is achieved through elastic parts. Specifically, alternating and corresponding trapezoidal positive and negative tooth structures 15 are coaxially arranged on the mating surfaces of the main casing 10 and the volute 11 as assembly sites. While ensuring the accuracy of installation, elastic parts 16 are arranged in the tooth structures 15 to retain sufficient vibration margins. After being connected with fasteners 8, the cooperation between the two is ensured while reducing the hard wear between the two. Of course, both are provided with elastic sealing rings 17 on the side facing the main casing 10 to achieve a sealing effect.

[0045] Taking into account that a large number of displacement sensors are used in this embodiment, a certain amount of heat dissipation may be generated, so temperature sensors are distributed in the main casing 10. Through the control and feedback of the temperature by the temperature sensors, the controller cooperates to control the speed adjustment of the drive shaft 1, so that the cold air on the outside can be introduced from the guide hole 14 at the right time; in actual application, the guide hole 14 is "obliquely inserted" and consistent with the rectification direction of the blade, so that when the drive shaft 1 moves rapidly, no air is supplied, and when the drive shaft 1 slows down to a certain degree, the channel is opened to realize air supply and cooling treatment; in order to reduce wind noise, the guide hole 14 is generally only arranged near the axial displacement sensor 7.

[0046] Example 6

[0047] On the basis of the above embodiments, the present invention also relates to a control method of the magnetic levitation vacuum pump, and the method is described in detail below: S1 is equipped with a magnetic levitation vacuum pump, the combined displacement detection unit is initialized, and the controller obtains the initial signal; In particular, it is necessary to initialize the cooperation between the detection window 9 and the measured light point and obtain the initial signal. If the measured light point is not captured, the working state of the vacuum pump and the working state of the axial displacement sensor 7 as a whole (including the detection window 9 and the measured light point) are judged to troubleshoot the fault; S2: obtaining the feedback signal and control parameters of the radial displacement sensor; obtaining the signal of the measured light point of the axial displacement sensor 7 captured in the detection window 9; Obtain the displacement vectors of the two measured light points and their relationship parameters, record the time and corresponding control parameters, and mark them manually; Specifically, if the measured light point has a longitudinal displacement, it actually means that the drive shaft 1 has an energy-consuming dynamic state, and the moment needs to be recorded and manually marked, and the corresponding classification is classified into a specific training label; but in actual applications, the relationship parameters of the displacement vectors of the two measured light points, such as the angle, are more important. When the two measured light points have a longitudinal displacement synchronously but the angle is less than the preset value, it is only necessary to synchronously control the current in the radial magnetic bearing 4. However, if the angle is greater than the preset value, the drive shaft 1 may even collide with the bearing, and this situation needs to be avoided in the subsequent feedforward control; If the measured light point only has lateral displacement, this is a normal process adjustment, and you only need to record the time and corresponding control parameters; S3 obtains sufficient non-empty detection data and corresponding control parameters of the continuous combined displacement detection unit, obtains manually labeled training labels, and inputs them into the improved prediction model for training; S4: obtaining a plurality of groups of continuous non-empty detection data of the combined displacement detection units and inputting them into an improved prediction model; the improved prediction model outputs classification information to a controller; and the controller adjusts control parameters to perform control adjustment.

[0048] When the controller of the present invention is working, it also involves a computer-readable storage medium, on which a control program of the magnetic levitation vacuum pump is stored. When the program is executed by a processor, the control method of the magnetic levitation vacuum pump is implemented.

[0049] When the controller of the present invention is working, it may also involve a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the control method of the magnetic levitation vacuum pump is implemented.

[0050] The present invention, in its implementation, also involves training and testing of an improved prediction model, which is easily understood by those skilled in the art, and can be implemented according to actual conditions.

[0051] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0052] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0053] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0054] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0055] Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0056] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A magnetic levitation vacuum pump, characterized in that: It comprises a casing, a driving shaft is arranged in the casing, an impeller is arranged in the casing at one end of the driving shaft, a motor and a combined magnetic bearing unit are arranged outside the driving shaft behind the impeller, a combined displacement detection unit is arranged in the combined magnetic bearing unit; a controller is arranged in conjunction with the motor, the combined magnetic bearing unit and the combined displacement detection unit; The combined displacement detection unit obtains combined displacement detection data, and the controller adjusts the output of the combined magnetic bearing unit and the motor based on the trained improved prediction model.

2. A magnetic levitation vacuum pump according to claim 1, characterized in that: The combined magnetic bearing unit includes radial magnetic bearings respectively arranged outside the driving shaft at both ends of the motor and axial magnetic bearings respectively arranged outside the two radial magnetic bearings. A fixed plate is provided on the driving shaft to cooperate with the axial magnetic bearings; the radial magnetic bearings and the axial magnetic bearings form a balancing system with the impeller.

3. A magnetic levitation vacuum pump according to claim 2, characterized in that: The combined displacement detection unit comprises a radial displacement sensor arranged in cooperation with any radial magnetic suspension bearing and an axial displacement sensor arranged in cooperation with any fixed disk.

4. A magnetic levitation vacuum pump according to claim 3, characterized in that: Any of the axial displacement sensors comprises one or more measured light points arranged outside the fixed disk, a detection window is arranged in conjunction with the measured light points, and the detection window is arranged in conjunction with the controller.

5. A magnetic levitation vacuum pump according to claim 4, characterized in that: An XOY coordinate system is established in conjunction with the detection window, with the origin of the XOY coordinate system corresponding to the position of the measured light point. When any measured light point is displaced, the controller controls the adjustment and obtains the displacement vectors of the two measured light points and feeds them back to the controller.

6. A magnetic levitation vacuum pump according to claim 1, characterized in that: The improved prediction model includes a double-branch structure; One branch includes several convolutional layers connected in sequence, each convolutional layer is followed by a time series feature aggregation layer, and the last time series feature aggregation layer is followed by a fully connected layer, a softmax layer and an output layer; one or more convolutional blocks are arranged in parallel in any of the convolutional layers; The other branch includes several LSTM units, and the output of any LSTM is associated with the input of the corresponding time series feature aggregation layer.

7. A magnetic levitation vacuum pump according to claim 6, characterized in that: Obtaining the trend of continuous data as a training label; the data used to train the improved prediction model includes detection data of the continuous combined displacement detection unit and corresponding control parameters, the detection data being used to characterize the spatial position of the drive shaft in the current spatial coordinate system; Paired detection data and control parameters with time series characteristics are used as training data pairs, which are input into the improved prediction model and trained in conjunction with the training labels.

8. A magnetic levitation vacuum pump according to claim 1, characterized in that: The casing comprises a main casing and a volute which are arranged in coordination with each other. The volute is provided with an air inlet in the axial direction of the impeller, and the volute is provided with an air outlet at the blades of the impeller. An elastic member is provided between the assembly points of the main casing and the volute.

9. A magnetic levitation vacuum pump according to claim 8, characterized in that: A temperature sensor is arranged in the main housing, and the temperature sensor is arranged in cooperation with the controller; a plurality of guide holes are arranged on the main housing.

10. A control method for a magnetic levitation vacuum pump according to any one of claims 1 to 9, characterized in that: The method comprises the following steps: S1 is equipped with a magnetic levitation vacuum pump, the combined displacement detection unit is initialized, and the controller obtains the initial signal; S2 obtains the feedback signal and control parameters of the radial displacement sensor; obtains the signal of the measured light point of the axial displacement sensor captured in the detection window; Obtain the displacement vectors of the two measured light points and their relationship parameters, record the time and corresponding control parameters, and mark them manually; S3 obtains sufficient non-empty detection data and corresponding control parameters of the continuous combined displacement detection unit, obtains manually labeled training labels, and inputs them into the improved prediction model for training; S4: obtaining a plurality of groups of continuous non-empty detection data of the combined displacement detection units and inputting them into an improved prediction model; the improved prediction model outputs classification information to a controller; and the controller adjusts control parameters to perform control adjustment.

Images