A method and system for controlling a magnetic levitation fan under high load and large fluctuation conditions

Abstract

The present invention discloses a method and system for controlling a magnetic levitation fan under high-load, high-fluctuation working conditions, which relates to the field of magnetic levitation fan control technology. The method comprises: real-time acquisition of rotor monitoring data and preprocessing thereof; construction of a rotor displacement prediction model based on the preprocessed monitoring data to predict rotor offset balance; dynamic adjustment of the control current based on the monitoring data and the rotor offset balance prediction result, and transmission of the control current to a power amplifier; the power amplifier changes the magnitude of the magnetic bearing current according to the received control current and monitors the working state of the power amplifier in real time; analysis of the received control current and the working state of the power amplifier, and real-time feedback adjustment of the control current based on the analysis results. In high-voltage, high-current application scenarios, the present invention comprehensively considers factors such as withstand voltage, withstand current, switching frequency, and ambient temperature to ensure the reliability and performance of the power amplifier and the reliability of current adjustment.

Description

Technical Field

[0001] The present invention relates to the technical field of magnetic levitation fan control, and more particularly to a method and system for controlling a magnetic levitation fan under high-load and large-fluctuation working conditions. Background Art

[0002] Currently, domestic motors are classified by input voltage into low-voltage motors (less than 1 kV) and high-voltage motors (over 3.3 kV). According to motor energy efficiency guidelines, higher voltage levels reduce total operating costs, lower currents reduce energy losses, and increase motor power density. Converting low-voltage motors to high-voltage motors is a key approach to energy conservation. 1MW high-power, high-voltage motors using three-phase asynchronous motors have disadvantages such as relatively complex structure, low speed, high maintenance costs, and difficulty starting. Their control performance, including control accuracy and torque smoothness, and power density, are inferior to those of permanent magnet synchronous motors. Permanent magnet synchronous motors offer a 5% to 10% improvement in rated efficiency compared to asynchronous motors. Furthermore, high-speed motors often use permanent magnet synchronous motors, with speeds reaching tens of thousands or even hundreds of thousands of revolutions per minute (RPM) and frequencies reaching 1000 Hz or higher. High-power motors using high-speed, high-voltage permanent magnet synchronous motors offer significant advantages, including simple structure, high energy efficiency, and high power density.

[0003] However, high-power, high-speed, and high-voltage permanent magnet synchronous motors involve electromagnetic scheme design, motor structure design, rotor support scheme design, motor drive, and control algorithm design. The power amplifier's withstand voltage and current are key parameters for achieving high load-bearing capacity. When the power amplifier's MOS transistors cannot meet the operating conditions, IGBTs are required for power amplifier design. Especially in high-voltage, high-current applications, the design must comprehensively consider factors such as withstand voltage, current, switching frequency, and ambient temperature to ensure the power amplifier's reliability and performance to meet the requirements of high-speed, high-voltage motors.

[0004] Therefore, how to propose a magnetic levitation fan control method and system under high load and large fluctuation conditions, and comprehensively consider factors such as voltage resistance, current resistance, switching frequency and ambient temperature in high voltage and high current application scenarios to ensure the reliability and performance of the power amplifier, ensure the reliability of current adjustment, and achieve stable control of the magnetic levitation fan is a problem that technical personnel in this field urgently need to solve. Summary of the Invention

[0005] In view of this, the present invention provides a method and system for controlling a magnetic levitation fan under high-load and large-fluctuation working conditions. In high-voltage and high-current application scenarios, factors such as withstand voltage, withstand current, switching frequency, and ambient temperature are comprehensively considered to ensure the reliability and performance of the power amplifier, guarantee the reliability of current adjustment, and realize stable control of the magnetic levitation fan. To achieve the above objectives, the present invention adopts the following technical solutions:

[0006] A method for controlling a magnetic levitation fan under high-load and large-fluctuation working conditions, comprising:

[0007] Collect rotor monitoring data in real time and perform pre-processing;

[0008] Build a rotor displacement prediction model based on pre-processed monitoring data to predict rotor offset balance;

[0009] Combined with monitoring data and rotor offset balance prediction results, the control current is dynamically adjusted and transmitted to the power amplifier;

[0010] The power amplifier changes the magnitude of the magnetic bearing current according to the received control current and monitors the working status of the power amplifier in real time;

[0011] The received control current and the working status of the power amplifier are analyzed, and the control current is adjusted in real time based on the analysis results.

[0012] Optionally, the real-time acquisition of rotor monitoring data includes acquiring temperature characteristic data, humidity characteristic data, vibration characteristic data, voltage characteristic data, speed characteristic data, current characteristic data and torque index data, and performing denoising and normalization processing on the data.

[0013] Optionally, the rotor displacement prediction model is constructed based on the preprocessed monitoring data, and the prediction of the rotor offset balance includes: extracting features related to the rotor displacement from the real-time monitoring data of the rotor through LSTM; and constructing a rotor displacement prediction model based on the features related to the rotor displacement prediction.

[0014] Optionally, the rotor displacement prediction model is expressed as:

[0015]

[0016] Among them, α represents the predicted value of rotor displacement, Y is the temperature characteristic, R is the humidity characteristic, V is the vibration characteristic, T is the voltage characteristic, H is the speed characteristic, S is the current characteristic, L is the torque index, τ represents the basic rotor displacement value when all external conditions affecting the rotor displacement are in ideal conditions, λ represents the prediction error of the rotor displacement due to factors that the model fails to capture, β represents the nonlinear effect of temperature on the rotor displacement, γ is the linear coefficient of humidity, δ is the linear coefficient of vibration, ζ represents the nonlinear effect of voltage on the rotor displacement, η represents the nonlinear effect of speed on the rotor displacement, θ is the linear coefficient of current, and E is the denominator coefficient of the torque index.

[0017] Optionally, the method of combining monitoring data and rotor offset balance prediction results, dynamically adjusting the control current, and transmitting it to the power amplifier includes: obtaining a reference signal, converting the rotor displacement prediction value through a demodulation circuit; subtracting the position signal obtained by converting the rotor displacement prediction value into the difference between the position signal and the reference signal, obtaining a control signal through a controller based on the difference signal, and transmitting it to the power amplifier.

[0018] Optionally, the real-time monitoring of the working status of the power amplifier includes: real-time collection of the power amplifier's withstand voltage value, withstand current value, switching frequency and ambient temperature data, and cleaning the collected data using abnormality monitoring, data verification and timestamp synchronization; converting the cleaned data into a consistent format through standardized processing, and fusing it using statistical fusion methods to form a working status.

[0019] Optionally, analyzing the received control current and the operating state of the power amplifier, and adjusting the control current in real time based on the analysis result includes:

[0020] Based on the working status, define the comprehensive solution selection function and calculate the risk cost index;

[0021] Based on the risk cost index, select the option with the lowest risk cost index from all candidate options;

[0022] Based on the selection scheme, control signals are constructed to generate dynamic adjustment instructions;

[0023] Combined with the rotor offset balance prediction results and selection scheme, the magnetic bearing current is adjusted in real time through feedback.

[0024] Optionally, selecting the solution with the lowest risk cost index from all candidate solutions further includes:

[0025] Collect physical information of magnetic levitation fans through sensors and scanning equipment;

[0026] Based on the physical information of the magnetic levitation fan, a three-dimensional model of the magnetic levitation fan is established, and a preliminary current adjustment plan is generated based on the plan with the lowest preset risk cost index;

[0027] Based on the preliminary current adjustment scheme, the optimal current adjustment scheme is calculated using the optimization algorithm, and the stability of the generated current adjustment scheme is analyzed;

[0028] The current adjustment scheme is optimized through an iterative optimization algorithm to obtain the optimal current adjustment scheme.

[0029] Optionally, it also includes: recalculating the minimum risk cost index based on real-time working status:

[0030] By comparing the risk cost index of the updated new plan with the risk cost index of the current optimal plan, the current adjustment plan is adjusted;

[0031] If the risk cost index of the new solution is less than the risk cost index value of the current optimal solution, the new solution will be updated to the new optimal solution;

[0032] If the risk cost index of the new solution is greater than the risk cost index of the current optimal solution, the current optimal solution is maintained and current adjustment is performed.

[0033] Optionally, a magnetic levitation fan control system under high load and large fluctuation conditions includes:

[0034] The first acquisition module: used to collect rotor monitoring data in real time and perform preprocessing;

[0035] Model building module: used to build a rotor displacement prediction model based on pre-processed monitoring data to predict rotor offset balance;

[0036] The first adjustment module is used to dynamically adjust the control current by combining the monitoring data and the rotor offset balance prediction result, and transmit it to the power amplifier;

[0037] The second acquisition module is used for the power amplifier to change the magnitude of the magnetic bearing current according to the received control current and to monitor the working status of the power amplifier in real time;

[0038] Feedback regulation module: used to analyze the received control current and the working status of the power amplifier, and adjust the control current in real time based on the analysis results.

[0039] It can be seen from the above technical solutions that, compared with the prior art, the present invention discloses a method and system for controlling a magnetic levitation fan under high-load and large-fluctuation working conditions, which has the following beneficial effects:

[0040] The present invention proposes a method for controlling a magnetic levitation fan under high-load, high-fluctuation working conditions, comprising: real-time acquisition and preprocessing of rotor monitoring data; constructing a rotor displacement prediction model based on the preprocessed monitoring data to predict rotor offset balance; dynamically adjusting the control current based on the monitoring data and the rotor offset balance prediction result, and transmitting the control current to a power amplifier; the power amplifier changes the magnitude of the magnetic bearing current according to the received control current, and monitors the working state of the power amplifier in real time; analyzing the received control current and the working state of the power amplifier, and adjusting the control current in real time based on the analysis results. The present invention discloses the arrangement of a sensor, a controller, and a power amplifier. When the rotor deviates from the equilibrium position, the sensor detects the rotor displacement and, through a demodulation circuit, subtracts the position signal transmitted back from a reference signal. The subtraction signal is then processed by the controller to obtain a control signal. The control signal changes the magnitude of the magnetic bearing current in real time through the power amplifier, thereby enabling the rotor to levitate in the equilibrium position. By monitoring the working status of the power amplifier in real time, analyzing the received control current and the working status of the power amplifier, and adjusting the control current in real time based on the analysis results, in high-voltage and high-current application scenarios, comprehensively considering factors such as voltage resistance, current resistance, switching frequency and ambient temperature to ensure the reliability and performance of the power amplifier and the reliability of current adjustment. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are merely embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0042] Figure 1 This is a circuit diagram of a magnetic levitation fan control method under high load and large fluctuation conditions provided by the present invention.

[0043] Figure 2 This is a flow chart of a method for controlling a magnetic levitation fan under high-load and large-fluctuation working conditions provided by the present invention.

[0044] Figure 3 This is a structural framework diagram of a magnetic levitation fan control system under high load and large fluctuation conditions provided by the present invention. DETAILED DESCRIPTION

[0045] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0046] The embodiment of the present invention discloses a method for controlling a magnetic levitation fan under high load and large fluctuation conditions. Figure 2 As shown, including:

[0047] Collect rotor monitoring data in real time and perform pre-processing;

[0048] Build a rotor displacement prediction model based on pre-processed monitoring data to predict rotor offset balance;

[0049] Combined with monitoring data and rotor offset balance prediction results, the control current is dynamically adjusted and transmitted to the power amplifier;

[0050] The power amplifier changes the magnitude of the magnetic bearing current according to the received control current and monitors the working status of the power amplifier in real time;

[0051] The received control current and the working status of the power amplifier are analyzed, and the control current is adjusted in real time based on the analysis results.

[0052] Furthermore, the real-time collection of rotor monitoring data includes collecting temperature characteristic data, humidity characteristic data, vibration characteristic data, voltage characteristic data, speed characteristic data, current characteristic data and torque index data, and performing denoising and normalization processing on the data.

[0053] Furthermore, the rotor displacement prediction model is constructed based on the preprocessed monitoring data, and the prediction of the rotor offset balance includes: extracting features related to the rotor displacement from the real-time monitoring data of the rotor through LSTM; and constructing a rotor displacement prediction model based on the features related to the rotor displacement prediction.

[0054] Specifically, LSTM feature extraction is as follows: Serialization processing: input time series data into LSTM, and use the memory unit of LSTM to capture long-term dependencies. LSTM architecture: Construct an LSTM unit containing input gate, forget gate, cell state, output gate, and possible multi-layer LSTM structure to capture complex time series patterns. Input gate: controls which information will be written to the cell state. Forget gate: determines which information in the cell state will be discarded. Cell state: stores long-term information and is the core of LSTM. Output gate: controls which information will be read and output to the next layer. Output: The output of LSTM is a dense representation of time series features that can capture long-term dependencies and dynamic changes.

[0055] In a specific embodiment, the rotor displacement prediction model is further optimized:

[0056] The loss function is defined to measure the difference between the energy consumption predicted by the energy consumption prediction model and the time energy consumption. The expression is:

[0057]

[0058] Where N is the number of training samples, α i is the true rotor displacement prediction value, is the predicted value of rotor displacement predicted by the model, and B is the difference index between the predicted value and the true value;

[0059] On the basis of adjusting the weights by the back-propagation algorithm, dynamic learning rate decay and adaptive batch size adjustment strategy are introduced to minimize the loss function, which is expressed as:

[0060]

[0061] Among them, w (t+1) is the weight value of the updated weight vector after the t+1th iteration, w (t) is the weight vector at the tth iteration, η0 is the initial learning rate, t is the number of iterations, φ is the constant of learning decay, The instantaneous function L at the current weight w (t) The gradient at .

[0062] Furthermore, the rotor displacement prediction model is expressed as:

[0063]

[0064] Among them, α represents the predicted value of rotor displacement, Y is the temperature characteristic, R is the humidity characteristic, V is the vibration characteristic, T is the voltage characteristic, H is the speed characteristic, S is the current characteristic, L is the torque index, τ represents the basic rotor displacement value when all external conditions affecting the rotor displacement are in ideal conditions, λ represents the prediction error of the rotor displacement due to factors that the model fails to capture, β represents the nonlinear effect of temperature on the rotor displacement, γ is the linear coefficient of humidity, δ is the linear coefficient of vibration, ζ represents the nonlinear effect of voltage on the rotor displacement, η represents the nonlinear effect of speed on the rotor displacement, θ is the linear coefficient of current, and E is the denominator coefficient of the torque index.

[0065] Further, such as Figure 1 As shown, the method of combining monitoring data and rotor offset balance prediction results, dynamically adjusting the control current, and transmitting it to the power amplifier includes: obtaining a reference signal, converting the rotor displacement prediction value through a demodulation circuit; subtracting the position signal obtained by the demodulation circuit from the rotor displacement prediction value and the reference signal, obtaining a control signal through a controller based on the difference signal, and transmitting it to the power amplifier.

[0066] Furthermore, the real-time monitoring of the working status of the power amplifier includes: real-time collection of the power amplifier's withstand voltage value, withstand current value, switching frequency and ambient temperature data, and cleaning the collected data using abnormality monitoring, data verification and timestamp synchronization; converting the cleaned data into a consistent format through standardized processing, and fusing it using statistical fusion methods to form a working status.

[0067] Specifically, the standardization method is:

[0068] Among them, X is the original parameter value, X max is the maximum possible value of the parameter, X min is the minimum possible value for this parameter.

[0069] Furthermore, analyzing the received control current and the operating state of the power amplifier and adjusting the control current in real time based on the analysis result includes:

[0070] Based on the working status, define the comprehensive solution selection function and calculate the risk cost index;

[0071] Based on the risk cost index, select the option with the lowest risk cost index from all candidate options;

[0072] Based on the selection scheme, control signals are constructed to generate dynamic adjustment instructions;

[0073] Combined with the rotor offset balance prediction results and selection scheme, the magnetic bearing current is adjusted in real time through feedback.

[0074] In a specific implementation, the method of defining a comprehensive solution selection function based on the working status and calculating the risk cost index specifically includes:

[0075] Based on the results of normalization, a comprehensive scheme selection function is defined to calculate the risk cost index under the conditions of comprehensive starting point m, end point n, torque index L, working state I and predicted rotor displacement prediction value α. The expression is:

[0076]

[0077] Among them, α q is the normalized predicted value of rotor displacement, L q is the normalized torque index, I q is the normalized working status index, D q is the normalized path distance, K is a constant used for scale adjustment, represents the nonlinear influence of rotor displacement on risk cost index P, ε is the path distance influence adjustment coefficient, and υ represents the torque index L q The coefficient of the impact on the risk cost index P, where P represents the risk cost index from the starting point m to the end point n.

[0078] Furthermore, the step of selecting the solution with the lowest risk cost index from all candidate solutions further includes:

[0079] Collect physical information of magnetic levitation fans through sensors and scanning equipment;

[0080] Based on the physical information of the magnetic levitation fan, a three-dimensional model of the magnetic levitation fan is established, and a preliminary current adjustment plan is generated based on the plan with the lowest preset risk cost index;

[0081] Based on the preliminary current adjustment scheme, the optimal current adjustment scheme is calculated using the optimization algorithm, and the stability of the generated current adjustment scheme is analyzed;

[0082] The current adjustment scheme is optimized through an iterative optimization algorithm to obtain the optimal current adjustment scheme.

[0083] In a specific embodiment, the step of calculating the best current adjustment scheme based on the preliminary current adjustment scheme using an optimization algorithm and performing stability analysis on the generated current adjustment scheme includes:

[0084] Based on the current adjustment scheme, the optimal rotor movement path is calculated using a genetic algorithm;

[0085] Stability check: Perform stability analysis on the generated current adjustment scheme, including:

[0086] Center of gravity calculation: Calculate the center of gravity of each path segment using the following formula:

[0087] Among them, G x and G y are the x- and y-coordinates of the center of gravity, w i is the weight of the i-th rotor, x i and y i are the x-coordinate and y-coordinate of the i-th material respectively;

[0088] Overturning analysis: Using the principle of moment balance, calculate the overturning moment of the rotor at each position. The calculation formula is:

[0089] Among them, M t is the total overturning moment, d i is the distance from the rotor at position i to the support surface;

[0090] Sliding analysis: Calculate the friction force of the rotor through friction analysis. The calculation formula is:

[0091] F f =μ×N; where F f is the friction force, μ is the friction coefficient, and N is the positive pressure.

[0092] In a specific embodiment, optimizing the current adjustment scheme by an iterative optimization algorithm to obtain the optimal current adjustment scheme includes:

[0093] Build a simulation model that matches the actual scenario in a virtual environment, including a 3D model of the magnetic levitation wind turbine. Execute the initially generated current adjustment plan in the simulation environment. Use the physics engine to simulate the current adjustment process, monitor the rotor force and stability in real time, verify the feasibility and stability of the plan, and record the simulation data generated during the simulation process.

[0094] Adjust the weight coefficient in the fitness function according to the simulation results to reflect the problems found in the simulation and the optimization goals;

[0095] Generate a new current adjustment plan using the particle swarm optimization algorithm, adjust the time series and adjustment value of the current adjustment, and generate a new plan;

[0096] Repeat the current adjustment simulation, data recording, and iterative optimization algorithm application until the simulation results meet the preset stability and efficiency requirements;

[0097] Based on multiple simulations and optimizations, the final optimized current adjustment scheme is determined.

[0098] Specifically, the particle swarm optimization algorithm includes:

[0099]

[0100] x i (t+1)=x i (t)+v i (t+1);

[0101] Among them, v i (t) and x i (t) represents the velocity and position of the ith particle in the tth generation, ω is the inertia weight, c1 and c2 are acceleration constants, r1 and r2 are random numbers between [0, 1], is the best historical position of the i-th particle, g best is the global optimal position of all particles.

[0102] Furthermore, it also includes: recalculating the minimum risk cost index based on the real-time working status:

[0103] By comparing the risk cost index of the updated new plan with the risk cost index of the current optimal plan, the current adjustment plan is adjusted;

[0104] If the risk cost index of the new solution is less than the risk cost index value of the current optimal solution, the new solution will be updated to the new optimal solution;

[0105] If the risk cost index of the new solution is greater than the risk cost index of the current optimal solution, the current optimal solution is maintained and current adjustment is performed.

[0106] In a specific embodiment, a magnetic suspension fan control system under high load and large fluctuation conditions, such as Figure 3 As shown, including:

[0107] The first acquisition module: used to collect rotor monitoring data in real time and perform preprocessing;

[0108] Model building module: used to build a rotor displacement prediction model based on pre-processed monitoring data to predict rotor offset balance;

[0109] The first adjustment module is used to dynamically adjust the control current by combining the monitoring data and the rotor offset balance prediction result, and transmit it to the power amplifier;

[0110] The second acquisition module is used for the power amplifier to change the magnitude of the magnetic bearing current according to the received control current and to monitor the working status of the power amplifier in real time;

[0111] Feedback regulation module: used to analyze the received control current and the working status of the power amplifier, and adjust the control current in real time based on the analysis results.

[0112] In a specific embodiment, a computer device is also provided, which is suitable for a method for controlling a magnetic levitation fan under high-load and large-fluctuation working conditions, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute computer-executable instructions to implement a method for controlling a magnetic levitation fan under high-load and large-fluctuation working conditions as proposed in the above embodiment.

[0113] The computer device may be a terminal, comprising a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device comprises a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner may be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a button, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse.

[0114] In a specific embodiment, a storage medium is further provided, on which a computer program is stored. When the program is executed by a processor, a method for controlling a magnetic levitation fan under high-load and large-fluctuation working conditions as proposed in the above embodiment is implemented;

[0115] The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination of them, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, disk or optical disk.

[0116] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. Reference can be made to the common and similar parts between the various embodiments. For the devices disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple, and the relevant parts can be referred to the method description.

[0117] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention is not limited to the embodiments shown herein but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for controlling a magnetic levitation fan under high load and large fluctuation conditions, characterized in that: include: Collect rotor monitoring data in real time and perform pre-processing; The real-time acquisition of rotor monitoring data includes acquiring temperature characteristic data, humidity characteristic data, vibration characteristic data, voltage characteristic data, speed characteristic data, current characteristic data and torque index data, and performing denoising and normalization processing on the data; Build a rotor displacement prediction model based on pre-processed monitoring data to predict rotor offset balance; The rotor displacement prediction model is constructed based on the preprocessed monitoring data, and the prediction of the rotor offset balance includes: extracting features related to the rotor displacement from the real-time monitoring data collected by the rotor through LSTM; and constructing the rotor displacement prediction model based on the features related to the rotor displacement prediction; The expression of the rotor displacement prediction model is: Among them, α represents the predicted value of rotor displacement, Y is the temperature characteristic, R is the humidity characteristic, V is the vibration characteristic, T is the voltage characteristic, H is the speed characteristic, S is the current characteristic, L is the torque index, τ represents the basic rotor displacement value when all external conditions affecting the rotor displacement are in ideal conditions, λ represents the prediction error of the rotor displacement caused by factors that the model fails to capture, β represents the nonlinear effect of temperature on the rotor displacement, γ is the linear coefficient of humidity, δ is the linear coefficient of vibration, ζ represents the nonlinear effect of voltage on the rotor displacement, η represents the nonlinear effect of speed on the rotor displacement, θ is the linear coefficient of current, and E is the denominator coefficient of the torque index; Combined with monitoring data and rotor offset balance prediction results, the control current is dynamically adjusted and transmitted to the power amplifier; The power amplifier changes the magnitude of the magnetic bearing current according to the received control current and monitors the working status of the power amplifier in real time; The received control current and the working status of the power amplifier are analyzed, and the control current is adjusted in real time based on the analysis results.

2. The method for controlling a magnetic levitation fan under high load and large fluctuation conditions according to claim 1 is characterized in that: The method of combining monitoring data and rotor offset balance prediction results, dynamically adjusting the control current, and transmitting it to the power amplifier includes: obtaining a reference signal, converting the rotor displacement prediction value through a demodulation circuit; subtracting the position signal obtained by converting the rotor displacement prediction value through the demodulation circuit from the reference signal, obtaining a control signal through a controller based on the difference signal, and transmitting the control signal to the power amplifier.

3. The method for controlling a magnetic levitation fan under high load and large fluctuation conditions according to claim 1 is characterized in that: The real-time monitoring of the working status of the power amplifier includes: real-time collection of the power amplifier's withstand voltage value, withstand current value, switching frequency and ambient temperature data, and cleaning the collected data using abnormality monitoring, data verification and timestamp synchronization; converting the cleaned data into a consistent format through standardization processing, and fusing it using a statistical fusion method to form a working status.

4. The method for controlling a magnetic levitation fan under high load and large fluctuation conditions according to claim 1 is characterized in that: Analyzing the received control current and the working state of the power amplifier and adjusting the control current in real time based on the analysis result includes: Based on the working status, define the comprehensive solution selection function and calculate the risk cost index; Based on the risk cost index, select the option with the lowest risk cost index from all candidate options; Based on the selection scheme, control signals are constructed to generate dynamic adjustment instructions; Combined with the rotor offset balance prediction results and selection scheme, the magnetic bearing current is adjusted in real time through feedback.

5. The method for controlling a magnetic levitation fan under high load and large fluctuation conditions according to claim 4 is characterized in that: The step of selecting the solution with the lowest risk cost index from all candidate solutions further includes: Collect physical information of magnetic levitation fans through sensors and scanning equipment; Based on the physical information of the magnetic levitation fan, a three-dimensional model of the magnetic levitation fan is established, and a preliminary current adjustment plan is generated based on the plan with the lowest preset risk cost index; Based on the preliminary current adjustment scheme, the optimal current adjustment scheme is calculated using the optimization algorithm, and the stability of the generated current adjustment scheme is analyzed; The current adjustment scheme is optimized through an iterative optimization algorithm to obtain the optimal current adjustment scheme.

6. The method for controlling a magnetic levitation fan under high load and large fluctuation conditions according to claim 1 is characterized in that: Also includes: Recalculate the minimum risk cost index based on real-time work status: By comparing the risk cost index of the updated new plan with the risk cost index of the current optimal plan, the current adjustment plan is adjusted; If the risk cost index of the new solution is less than the risk cost index value of the current optimal solution, the new solution will be updated to the new optimal solution; If the risk cost index of the new solution is greater than the risk cost index of the current optimal solution, the current optimal solution is maintained and current adjustment is performed.

7. A magnetic levitation fan control system under high load and large fluctuation conditions, applied to a magnetic levitation fan control method under high load and large fluctuation conditions as claimed in any one of claims 1 to 6, characterized in that: include: The first acquisition module: used to collect rotor monitoring data in real time and perform preprocessing; Model building module: used to build a rotor displacement prediction model based on pre-processed monitoring data to predict rotor offset balance; The first adjustment module is used to dynamically adjust the control current by combining the monitoring data and the rotor offset balance prediction result, and transmit it to the power amplifier; The second acquisition module is used for the power amplifier to change the magnitude of the magnetic bearing current according to the received control current and to monitor the working status of the power amplifier in real time; Feedback regulation module: used to analyze the received control current and the working status of the power amplifier, and adjust the control current in real time based on the analysis results.

Images