Magnetic suspension fan motor winding temperature monitoring device and early warning method

Abstract

The application belongs to the technical field of magnetic suspension motor, and discloses a magnetic suspension fan motor winding temperature monitoring device and a warning method; the method comprises the following steps: laminating a preset temperature-sensitive ferromagnetic material and a magnetic suspension fan motor stator core according to a preset process to obtain a composite core; deploying a high-frequency excitation system on the composite core, adjusting the number of turns of the micro coil of the high-frequency excitation system according to a target magnetic field intensity, collecting real-time excitation current, and analyzing to obtain magnetic induction intensity; analyzing the magnetic induction intensity through multi-physical field coupling iteration to obtain a winding three-dimensional temperature field; performing temperature gradient analysis on the three-dimensional temperature field, performing temperature warning according to the analysis result, and dynamically adjusting motor operating parameters; the application effectively improves the comprehensiveness of magnetic suspension fan motor winding temperature monitoring, the timeliness of warning and the accuracy of cooling control, and guarantees the safe and stable operation of the motor under complex working conditions.

Description

Technical Field

[0001] This invention relates to the field of magnetic levitation motor technology, and more specifically, to a magnetic levitation fan motor winding temperature monitoring device and early warning method. Background Technology

[0002] As a key power equipment in industrial production, wastewater treatment, and energy supply, the stable operation of the core component of magnetic levitation fans directly determines the reliability and energy efficiency of the entire machine. The motor windings, as the core carrier for converting electrical energy into mechanical energy, are prone to generating a large amount of heat due to copper losses, iron losses, and electromagnetic induction under high-speed, high-power-density operating conditions. If the temperature continues to rise beyond the safety threshold, it will not only accelerate the aging of the winding insulation layer and reduce the motor's output power, but in severe cases, it can also cause major failures such as winding burnout and magnetic levitation bearing instability, resulting in equipment downtime and economic losses. However, traditional magnetic levitation fan motor winding temperature monitoring and early warning technologies have significant limitations: On the one hand, temperature measurement methods based on contact sensors such as thermocouples and PT100 are limited by the complex electromagnetic environment and compact structure inside the magnetic levitation motor, resulting in problems such as limited installation location, delayed response, and susceptibility to electromagnetic interference leading to data distortion, making it impossible to accurately capture the real temperature changes of the windings in real time; on the other hand, traditional early warning schemes mostly use fixed temperature threshold triggering mechanisms, which do not consider the nonlinear change law of winding temperature under dynamic operating conditions such as load fluctuations, ambient temperature changes, and start-stop transitions, which easily leads to "false alarms" or "missed alarms," ​​resulting in insufficient early warning reliability and difficulty in effectively preventing motor winding overheating faults, thereby affecting the long-term stable operation and service life of the magnetic levitation fan.

[0003] Chinese patent application CN119519210A discloses a magnetic levitation motor and its cooling method, apparatus, storage medium, and program product: a coolant pipe is provided at the motor winding of the magnetic levitation motor; an electronic expansion valve is provided on the coolant pipe; the method includes: obtaining the temperature of the motor winding and a reference temperature of the motor winding; determining whether to adjust the opening of the electronic expansion valve based on the temperature of the motor winding and the reference temperature of the motor winding; if it is determined that the opening of the electronic expansion valve should be adjusted, then substituting the temperature of the motor winding and the reference temperature of the motor winding into a preset anti-saturation PI algorithm to obtain a calculated value of the opening of the electronic expansion valve; and controlling the opening of the electronic expansion valve to be the calculated value.

[0004] While the above methods can meet most scenarios, research and practical application of the above methods and existing technologies have revealed at least the following shortcomings: The above methods rely on traditional external sensors or single measurement point monitoring, which can only obtain local or one-dimensional temperature information, and are prone to missing local hot spots, resulting in a mismatch between cooling resource allocation and actual heat distribution.

[0005] In view of this, the present invention proposes a magnetic levitation fan motor winding temperature monitoring device and early warning method to solve the above problems. Summary of the Invention

[0006] To overcome the aforementioned deficiencies of the prior art and to achieve the above objectives, the present invention provides the following technical solution: a magnetic levitation fan motor winding temperature monitoring device and early warning method, comprising:

[0007] A composite core is obtained by laminating a pre-set temperature-sensitive ferromagnetic material with the stator core of a magnetic levitation fan motor according to a pre-set process.

[0008] A high-frequency excitation system is deployed on a composite iron core. The number of turns of the micro-coil of the high-frequency excitation system is adjusted according to the target magnetic field strength. Real-time excitation current is collected and analyzed to obtain the magnetic induction intensity.

[0009] The magnetic induction intensity is analyzed through multi-physics coupling iteration to obtain the three-dimensional temperature field of the winding.

[0010] Temperature gradient analysis is performed on the three-dimensional temperature field, temperature warnings are issued based on the analysis results, and motor operating parameters are dynamically adjusted.

[0011] Furthermore, methods for obtaining the three-dimensional temperature field of the winding include:

[0012] Set a reference temperature, and calculate the magnetic permeability of the composite iron core based on the current temperature and the permeability temperature model.

[0013] Based on the magnetic permeability of the composite iron core, the q-axis current, and the motor speed, and combined with Maxwell's equations, the theoretical magnetic flux density is calculated and compared with the actual magnetic flux density. If the deviation is greater than the deviation threshold, the magnetic permeability of the composite iron core is adjusted and the calculation is repeated.

[0014] By substituting the theoretical magnetic flux density into the eddy current loss calculation model and combining the eddy current loss coefficient, magnetic field frequency, magnetic flux density, and silicon steel sheet thickness for analysis, the eddy current loss per unit volume of iron core is obtained.

[0015] The eddy current loss per unit volume of iron core and the flow rate of coolant are analyzed in conjunction with the Fourier heat conduction equation to obtain a three-dimensional temperature field, which is then used as the three-dimensional temperature field of the winding.

[0016] Furthermore, methods for obtaining the permeability of composite iron cores include:

[0017] Determine the effective cross-sectional area and magnetic circuit length of the preset temperature-sensitive ferromagnetic material;

[0018] D temperature points are preset, and each temperature point is kept at a temperature for a period of t0. The test frequency and the first measuring coil are set. The coil inductance is measured and the empty coil inductance is subtracted to obtain the actual inductance of the preset temperature-sensitive ferromagnetic material.

[0019] The relative permeability of the preset temperature-sensitive ferromagnetic material is calculated based on the actual inductance, vacuum permeability, effective cross-sectional area, and magnetic circuit length of the preset temperature-sensitive ferromagnetic material.

[0020] By performing a quadratic polynomial fitting on the relative permeability of the preset temperature-sensitive ferromagnetic material, the relative permeability of the preset temperature-sensitive ferromagnetic material with respect to different temperatures is obtained. The first temperature characteristic fitting model is used.

[0021] Determine the effective cross-sectional area and magnetic circuit length of silicon steel sheets;

[0022] D temperature points are preset, each temperature point is kept at a temperature for a period of t0, the test frequency is set, the coil inductance is measured, and the empty coil inductance is subtracted to obtain the actual inductance of the silicon steel sheet;

[0023] The relative permeability of silicon steel sheets is calculated based on the actual inductance, vacuum permeability, effective cross-sectional area, and magnetic circuit length of the silicon steel sheets.

[0024] A second-temperature characteristic fitting model of the relative permeability of silicon steel sheet with respect to different temperatures is obtained by performing a quadratic polynomial fitting on the relative permeability of silicon steel sheet;

[0025] The volume of the silicon steel sheet in a single cycle is calculated based on the number of silicon steel sheet layers, the thickness of the silicon steel sheet, and the cross-sectional area of ​​the iron core. The volume of the preset temperature-sensitive ferromagnetic material in a single cycle is calculated based on the number of preset temperature-sensitive ferromagnetic material layers, the thickness of the preset temperature-sensitive ferromagnetic material, and the cross-sectional area of ​​the iron core. The volume of the composite iron core in a single cycle is calculated based on the volume of the silicon steel sheet and the volume of the preset temperature-sensitive ferromagnetic material. The volume ratio of the silicon steel sheet and the volume ratio of the preset temperature-sensitive ferromagnetic material are calculated based on the volume of the silicon steel sheet in a single cycle, the volume of the preset temperature-sensitive ferromagnetic material in a single cycle, and the volume of the composite iron core.

[0026] The magnetic permeability of the composite core is calculated based on the relative permeability of the preset temperature-sensitive ferromagnetic material, the relative permeability of the silicon steel sheet, the current temperature, and the reference temperature, combined with the permeability temperature model. The coefficients of the permeability temperature model are the corresponding coefficients in the first temperature characteristic fitting model and the second temperature characteristic fitting model, combined with the weighted result of the volume ratio of the silicon steel sheet and the volume ratio of the preset temperature-sensitive ferromagnetic material.

[0027] Furthermore, methods for obtaining the theoretical magnetic flux density include:

[0028] Based on Ampere's circuital law, the magnetomotive force of each pole of the q-axis winding is calculated according to the number of turns of each group of the q-axis winding, the q-axis current, the q-axis winding coefficient, and the number of pole pairs of the motor.

[0029] Calculate the magnetic reluctance of each segment in the magnetic circuit separately, and then calculate the total magnetic reluctance:

[0030] The stator tooth reluctance is calculated based on the stator tooth length, effective cross-sectional area of ​​the stator teeth, and the permeability of the composite core; the air gap reluctance is calculated based on the air gap length, effective cross-sectional area of ​​the air gap, and permeability of the composite core; the rotor permanent magnet reluctance is calculated based on the rotor permanent magnet length, effective cross-sectional area of ​​the rotor permanent magnet, relative permeability of the permanent magnet, and relative permeability at the reference temperature; the total reluctance is calculated based on the stator tooth reluctance, air gap reluctance, and rotor permanent magnet reluctance.

[0031] Calculate the ratio of the magnetomotive force per pole of the q-axis winding to the total magnetic reluctance to obtain the total magnetic flux.

[0032] The initial theoretical magnetic flux density is calculated based on the total magnetic flux and the effective cross-sectional area of ​​the stator teeth.

[0033] The frequency is calculated based on the motor speed and the number of pole pairs. If the frequency is greater than the frequency threshold, the eddy current loss increases, leading to a rise in temperature. The corrected permeability is then recalculated based on the increased temperature. The total reluctance is updated based on the corrected permeability, and the theoretical magnetic flux density is recalculated based on the updated total reluctance. Otherwise, the initial theoretical magnetic flux density is used as the theoretical magnetic flux density.

[0034] Furthermore, methods for obtaining a three-dimensional temperature field include:

[0035] The Reynolds number of the cooling channel is calculated based on the coolant flow rate, coolant density, coolant dynamic viscosity, and hydraulic diameter of the cooling channel.

[0036] The Nusselt number is calculated based on the Reynolds number of the cooling channel and the Prandtl number of the coolant.

[0037] The convective heat transfer coefficient is calculated based on the Nusselt number and the thermal conductivity of the coolant.

[0038] The three-dimensional structure of the motor is discretized into X grids with side length Y×Y. Each grid represents a control volume P with a volume of... , The length of the grid in the x-direction; The length of the grid in the y direction; Let be the length of the mesh in the z-direction; obtain the discretized control volume mesh model;

[0039] The control equations of an internal heat source containing eddy current losses in the iron core are constructed. The control volume P is integrated and the terms of the control equations are discretized. The discrete algebraic equations of the control volume are obtained by rearranging them.

[0040] By incorporating boundary conditions into discrete algebraic equations, a complete system of algebraic equations containing boundary conditions is obtained.

[0041] Solve the complete algebraic equations including boundary conditions to obtain the new temperature field; calculate the temperature deviation of all control volumes. If the temperature deviation is lower than the minimum deviation threshold, the iteration terminates and the corresponding updated temperature is obtained as the corresponding three-dimensional temperature field. Otherwise, return to recalculate the magnetic permeability of the composite iron core and update to obtain the new temperature field.

[0042] Furthermore, the method for adjusting the number of turns of the micro-coil in the high-frequency excitation system according to the target magnetic field strength includes:

[0043] The number of turns of the microcoil is calculated based on the excitation current, the length of the composite iron core magnetic circuit, and the target magnetic field strength. The microcoil of the high-frequency excitation system is then deployed based on the number of turns of the microcoil.

[0044] Furthermore, methods for obtaining magnetic field strength include:

[0045] The real-time magnetic field strength is calculated based on the real-time excitation current, the length of the composite iron core magnetic circuit, and the number of turns. Then, the magnetic induction intensity is calculated based on the real-time magnetic field strength, the relative permeability of the composite iron core, and the vacuum permeability.

[0046] Furthermore, methods for obtaining composite cores include:

[0047] The preset temperature-sensitive ferromagnetic material is cut into rectangular micro-sheets with a thickness of c and an area accounting for 5% of the cross-sectional area of ​​the stator tooth root using an ultraviolet laser cutting machine.

[0048] An insulating layer of thickness d was deposited on the surface of a micro-flake using magnetron sputtering to obtain a micro-flake with a uniform coating.

[0049] The basic unit uses an A-layer silicon steel sheet superimposed with a B-layer micro-film. According to the preset stacking cycle, the stacking is periodically and alternately performed along the axial direction of the stator tooth root of the magnetic levitation fan motor.

[0050] The microchips are arranged radially along the root of the stator teeth of the magnetic levitation fan motor, with one microchip corresponding to each stator slot, thus obtaining a composite iron core.

[0051] Furthermore, methods for issuing temperature warnings and dynamically adjusting motor operating parameters based on analysis results include:

[0052] Calculate the partial derivatives of the three-dimensional temperature difference with respect to the x and y directions to obtain the axial and radial temperature change rates, and then calculate the temperature gradient based on the axial and radial temperature change rates.

[0053] The abnormal deviation is calculated based on the temperature gradient and the baseline gradient. The abnormal region is divided by K-means clustering. The number of stator teeth covered by the abnormal region is counted. The gradient change rate of the abnormal region is calculated. If the abnormal deviation is between the first deviation threshold and the second deviation threshold, the number of stator teeth covered by the abnormal region is not higher than the number of teeth threshold, and the change rate of the abnormal region is not higher than the first change rate threshold, then a first-level warning is triggered.

[0054] Real-time updates of abnormal deviation and number of teeth covered in abnormal areas; calculation of gradient change rate in abnormal areas.

[0055] If the abnormal deviation is between the second and third deviation thresholds, the number of stator teeth covered by the abnormal region is higher than the number of teeth threshold, and the rate of change of the abnormal region is between the second and third rate of change thresholds, then a level two warning is triggered.

[0056] The criteria for triggering a Level 3 warning include: if the abnormal deviation is greater than the third deviation threshold, or the rate of change of the abnormal region is greater than the third rate of change threshold; or the gradient deviation of the abnormal region is accompanied by q-axis current fluctuation greater than the fluctuation threshold, or the vibration value is greater than C times the normal vibration value under the current operating conditions; or after the cooling flow rate is increased by the first proportional threshold and the load is reduced by the second proportional threshold, the abnormal deviation is still greater than the highest deviation threshold. If any of the criteria is met, a Level 3 warning is triggered.

[0057] Furthermore, when a Level 1 warning is triggered, the cooling orientation adjustment range is calculated based on the current cooling orientation range, abnormal deviation, and flow adjustment coefficient; the temperature field inversion sampling period is shortened to G, and the rate of change of abnormal deviation is calculated every preset time interval. If the rate of change is less than 0, the current adjustment is maintained; if the rate of change is not less than 0, the flow adjustment coefficient is increased to the next level, and the adjustment range is recalculated until the rate of change is less than 0, and the Level 1 warning response is lifted.

[0058] When a Level 2 warning is triggered, the cooling direction adjustment magnitude is calculated based on the current cooling direction magnitude, abnormal deviation, and flow adjustment coefficient; the load adjustment magnitude is calculated based on the current load, abnormal deviation, and load adjustment coefficient.

[0059] If the abnormal deviation is lower than the second deviation threshold and the number of stator teeth covered by the abnormal area is lower than the number of teeth threshold, after maintaining the current adjustment preset time period, the load and cooling flow will be gradually restored according to the preset ratio, and the second-level warning response will be lifted.

[0060] When a Level 3 warning is triggered, the load adjustment range is calculated based on the current load, abnormal deviation, and load adjustment coefficient; the cooling flow is adjusted to the maximum allowable value under the current operating conditions.

[0061] If the abnormal deviation is still greater than the highest deviation threshold after adjustment, the load will be reduced by a preset proportion at predetermined intervals until the load is 0. Then the main power supply will be disconnected and the motor temperature will be cooled down to below the highest temperature threshold to release the Level 3 warning response.

[0062] A magnetic levitation fan motor winding temperature monitoring device, implementing the aforementioned magnetic levitation fan motor winding temperature early warning method, including:

[0063] Structural integration module: The pre-set temperature-sensitive ferromagnetic material is laminated with the stator core of the magnetic levitation fan motor according to the pre-set process to obtain a composite core;

[0064] Magnetic field scanning module: A high-frequency excitation system is deployed on the composite iron core. The number of turns of the micro-coil of the high-frequency excitation system is adjusted according to the target magnetic field strength. Real-time excitation current is collected and analyzed to obtain the magnetic induction intensity.

[0065] Temperature analysis module: Through multi-physics coupling iteration, the magnetic induction intensity is analyzed to obtain the three-dimensional temperature field of the winding;

[0066] Temperature warning module: Performs temperature gradient analysis on the three-dimensional temperature field, issues temperature warnings based on the analysis results, and dynamically adjusts motor operating parameters.

[0067] The technical effects and advantages of the magnetic levitation fan motor winding temperature monitoring device and early warning method proposed in this invention are as follows:

[0068] This invention constructs a three-dimensional temperature field of the windings and associated rotor by stacking a pre-set temperature-sensitive ferromagnetic material with the stator core to form a composite core, deploying a high-frequency excitation system to collect magnetic induction intensity, and combining a permeability temperature model with Maxwell's equations and Fourier heat conduction equations for multi-physics field coupling iteration. Then, based on K-means clustering analysis of temperature gradient, it achieves hierarchical early warning and dynamic adjustment of motor operating parameters. This invention completely solves the problems of existing technologies that rely on point sensors to obtain only limited temperature at a limited point on the shaft, cannot cover the entire temperature field of the rotor, and are prone to missing local hot spots. It realizes real-time perception of the entire motor temperature, high-precision reconstruction of the temperature field under precise multi-physics field coupling, and intelligent hierarchical early warning and dynamic parameter adjustment based on spatial temperature distribution and abnormal development trends. This effectively improves the comprehensiveness of magnetic levitation fan motor winding temperature monitoring, the timeliness of early warning, and the accuracy of cooling control, ensuring the safe and stable operation of the motor under complex working conditions. Attached Figure Description

[0069] Figure 1 This is a schematic diagram of the magnetic levitation fan motor winding temperature early warning method of the present invention;

[0070] Figure 2 This is a schematic diagram of the data flow in this invention;

[0071] Figure 3 This is a schematic diagram of the method for obtaining the three-dimensional temperature field of the winding according to the present invention;

[0072] Figure 4 This is a schematic diagram of the magnetic levitation fan motor winding temperature monitoring device of the present invention. Detailed Implementation

[0073] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0074] Example 1

[0075] Please see Figure 1 , Figure 2 As shown, this embodiment provides a method for early warning of the winding temperature of a magnetic levitation fan motor, including:

[0076] A composite core is obtained by laminating a pre-set temperature-sensitive ferromagnetic material with the stator core of a magnetic levitation fan motor according to a pre-set process.

[0077] Methods for obtaining composite cores include:

[0078] A pre-set temperature-sensitive ferromagnetic material, such as FeGa alloy, is cut into rectangular micro-sheets with a thickness of c and an area accounting for 5% of the cross-sectional area of ​​the stator tooth root using an ultraviolet laser cutting machine.

[0079] An Al2O3 insulating layer with a thickness of d was deposited on the surface of the micro wafer using magnetron sputtering. If a pure aluminum target is selected, Ar / O2 is used as the mixed gas, and 150W of radio frequency power is used to sputter for 30 minutes at a working gas pressure of 0.5Pa to obtain a micro wafer with a uniform coating.

[0080] The basic unit uses an A-layer silicon steel sheet superimposed with a B-layer micro-film. According to the preset stacking cycle, the stacking is periodically and alternately performed along the axial direction of the stator tooth root of the magnetic levitation fan motor.

[0081] The microchips are arranged radially along the root of the stator teeth of the magnetic levitation fan motor, with one microchip corresponding to each stator slot, thus obtaining a composite iron core.

[0082] By cutting a pre-set temperature-sensitive ferromagnetic material into rectangular micro-sheets of a specific thickness, each with an area accounting for 5% of the stator tooth root cross-sectional area, and preparing a uniform insulating layer using magnetron sputtering, a composite iron core is obtained by periodically and alternately stacking A-layer silicon steel sheets superimposed with B-layer micro-sheets along the stator tooth root axis as the basic unit. The micro-sheets are arranged radially along the stator tooth root, with one micro-sheet corresponding to each stator slot. Essentially, this fabrication process constructs a distributed temperature-sensing network within the stator core. Compared to existing technologies that rely on external sensors and can only achieve point measurements along a limited line on the shaft, this embedded, distributed micro-sheet arrangement... The design utilizes radially arranged micro-plates corresponding to each stator slot to cover the key areas of the stator tooth root. Combined with axial periodic stacking, it forms a multi-point temperature sensing system in three-dimensional space, which can capture temperature changes at different radial and axial positions of the stator core in real time. Furthermore, by inferring the overall temperature field distribution through the thermal coupling between the stator and rotor, it effectively avoids the problem of missing local hot spots due to the limited coverage of traditional point measurements. At the same time, the integrated design of the micro-plates and the core structure also eliminates measurement deviations caused by poor thermal contact between external sensors and the measured object, improving the comprehensiveness and accuracy of temperature monitoring.

[0083] A high-frequency excitation system is deployed on a composite iron core. The number of turns of the micro-coil of the high-frequency excitation system is adjusted according to the target magnetic field strength. Real-time excitation current is collected and analyzed to obtain the magnetic induction intensity.

[0084] Methods for adjusting the number of turns of the micro-coil in a high-frequency excitation system based on the target magnetic field strength include:

[0085] The number of turns of the microcoil is calculated based on the excitation current, the length of the composite iron core magnetic circuit, and the target magnetic field strength. The microcoil of the high-frequency excitation system is then deployed according to this number of turns. (The number of turns of the microcoil is mentioned here, but its relevance to the preceding text is unclear.) ,in, The target magnetic field strength; The length of the magnetic circuit of the composite iron core; This is the excitation current.

[0086] Methods for obtaining magnetic field strength include:

[0087] The real-time magnetic field strength is calculated based on the real-time excitation current, the length of the composite iron core's magnetic circuit, and the number of turns. The magnetic induction intensity is then calculated based on the real-time magnetic field strength, the relative permeability of the composite iron core, and the free permeability. (The real-time magnetic field strength is then used as an example.) , The current is the real-time excitation current; the current temperature is... Magnetic induction intensity ,in, The current temperature is Relative permeability of composite iron core; is the vacuum permeability.

[0088] When deploying a high-frequency excitation system on a composite iron core, the required number of micro-coil turns is first calculated based on the target magnetic field strength, preset excitation current, and the magnetic circuit length of the composite iron core, and then deployed. This ensures that each temperature-sensitive microchip in the composite iron core, radially arranged along the stator tooth root and periodically stacked axially, is in a stable and suitable excitation magnetic field, providing a uniform magnetic signal excitation basis for distributed temperature sensing. Subsequently, the real-time excitation current is collected, and the real-time magnetic field strength is calculated by combining the magnetic circuit length of the composite iron core with the number of deployed micro-coil turns. Then, based on this real-time magnetic field strength and the relative temperature of the composite iron core at the current temperature, the system is further optimized. The magnetic permeability and vacuum permeability are used to calculate the magnetic induction intensity at the corresponding spatial point of each temperature-sensitive microchip. Since each stator slot corresponds to a temperature-sensitive microchip, these distributed magnetic induction intensity data can be used to infer the temperature at each corresponding point, and then integrated to form a three-dimensional temperature field of the stator core. Combined with the thermal coupling characteristics between the stator and rotor, the overall temperature field distribution of the rotor can be inferred. This completely breaks through the limitation of existing technologies that rely on sensor point measurement to obtain the temperature of a limited number of points on a single line on the shaft, effectively covering the temperature monitoring of key areas of the rotor and avoiding the omission of local hot spots due to insufficient monitoring range.

[0089] By using multi-physics coupling iteration, the magnetic induction intensity is analyzed to obtain the three-dimensional temperature field of the winding.

[0090] Reference Figure 3 Methods for obtaining the three-dimensional temperature field of the winding include:

[0091] Set a reference temperature, and calculate the magnetic permeability of the composite iron core based on the current temperature and the permeability temperature model.

[0092] Methods for obtaining the permeability of composite iron cores include:

[0093] Determine the effective cross-sectional area and magnetic circuit length of the preset temperature-sensitive ferromagnetic material;

[0094] D temperature points are preset, each held at a temperature for a period of t0. The test frequency and the first measuring coil are set. The coil inductance is measured, and the inductance of the empty coil is subtracted to obtain the actual inductance of the preset temperature-sensitive ferromagnetic material; such as the actual inductance of FeGa. ,in, The inductance of the first measuring coil; The inductance of the first measuring coil is the empty coil inductance.

[0095] The relative permeability of a preset temperature-sensitive ferromagnetic material is calculated based on its actual inductance, vacuum permeability, effective cross-sectional area, and magnetic circuit length; for example, the relative permeability of FeGa. ,in, The length of the FeGa magnetic circuit; The vacuum permeability; The number of turns of the first measuring coil; This represents the effective cross-sectional area of ​​FeGa.

[0096] A quadratic polynomial fit was performed on the relative permeability of the preset temperature-sensitive ferromagnetic material to obtain the relative permeability of the preset temperature-sensitive ferromagnetic material with respect to different temperatures. The first temperature characteristic fitting model is as follows; such as the first temperature characteristic fitting model. ,in, The relative permeability of the temperature-sensitive ferromagnetic material at the preset reference temperature; The first-order temperature coefficient; These are the first and second order temperature coefficients; The current temperature; For reference temperature;

[0097] Determine the effective cross-sectional area and magnetic circuit length of silicon steel sheets;

[0098] D temperature points are preset, each held at a temperature for a period of t0. The test frequency is set, the coil inductance is measured, and the inductance of the empty coil is subtracted to obtain the actual inductance of the silicon steel sheet. (The actual inductance of the silicon steel sheet is then calculated.) ,in, The inductance of the second measuring coil; The inductance of the second measuring coil is the empty coil inductance.

[0099] The relative permeability of silicon steel sheets is calculated based on the actual inductance, vacuum permeability, effective cross-sectional area, and magnetic circuit length of the silicon steel sheets; for example, the relative permeability of silicon steel sheets. ,in, The length of the magnetic circuit of the silicon steel sheet; The number of turns of the second measuring coil; This refers to the effective cross-sectional area of ​​the silicon steel sheet;

[0100] A second-temperature characteristic fitting model of the relative permeability of silicon steel sheets with respect to different temperatures is obtained by performing a quadratic polynomial fitting on the relative permeability of silicon steel sheets; such as the second-temperature characteristic fitting model. ,in, The relative permeability of the silicon steel sheet at the reference temperature; It is the second-order temperature coefficient; It is the second-order temperature coefficient; For reference temperature;

[0101] The single-cycle volume of the silicon steel sheet is calculated based on the number of silicon steel sheet layers, the thickness of the silicon steel sheet, and the cross-sectional area of ​​the iron core. The single-cycle volume of the preset temperature-sensitive ferromagnetic material is calculated based on the number of preset temperature-sensitive ferromagnetic material layers, the preset temperature-sensitive ferromagnetic material thickness, and the cross-sectional area of ​​the iron core. The single-cycle volume of the composite iron core is calculated based on the volume of the silicon steel sheet and the preset temperature-sensitive ferromagnetic material. The volume ratio of the silicon steel sheet and the preset temperature-sensitive ferromagnetic material is calculated based on the single-cycle volume of the silicon steel sheet, the single-cycle volume of the preset temperature-sensitive ferromagnetic material, and the single-cycle volume of the composite iron core. (The single-cycle volume of the composite iron core is also mentioned.) ,in, This refers to the volume of a single-cycle silicon steel sheet. The preset single-cycle volume of the temperature-sensitive ferromagnetic material; This refers to the number of silicon steel sheet layers per cycle. For the thickness of silicon steel sheets; The cross-sectional area of ​​the iron core; The number of temperature-sensitive ferromagnetic material layers is preset for a single cycle; The thickness of the temperature-sensitive ferromagnetic material is preset; the volume percentage of the temperature-sensitive ferromagnetic material is preset. Silicon steel sheet volume ratio ;

[0102] The magnetic permeability of the composite core is calculated based on the relative permeability of the preset temperature-sensitive ferromagnetic material, the relative permeability of the silicon steel sheet, the current temperature, and the reference temperature, combined with a magnetic permeability temperature model. The coefficients of the magnetic permeability temperature model are corresponding coefficients from the first and second temperature characteristic fitting models, weighted by the volume percentage of the silicon steel sheet and the volume percentage of the preset temperature-sensitive ferromagnetic material. (The text then abruptly shifts to a different topic: the magnetic permeability of the composite core.) ,in, For reference permeability, ; It is a first-order temperature coefficient. , It is a second-order temperature coefficient. .

[0103] Based on the composite core permeability, q-axis current, and motor speed, and combined with Maxwell's equations, the theoretical magnetic flux density is calculated and compared with the actual magnetic flux density. If the deviation is greater than the deviation threshold, the composite core permeability is adjusted and recalculated. The q-axis is the direction perpendicular to the main magnetic field generated by the rotor permanent magnet, used to describe the magnetic field generated by the stator current component perpendicular to the direction of the rotor's main magnetic field.

[0104] Methods for obtaining theoretical magnetic flux density include:

[0105] Based on Ampere's circuital law, the magnetomotive force per pole of the q-axis winding is calculated using the number of turns per group, q-axis current, q-axis winding coefficient, and number of pole pairs of the motor. For example, the magnetomotive force per pole of the q-axis winding... ,in, This is the magnetomotive force correction factor for the distributed winding; This refers to the number of turns per phase of the q-axis winding; This is the q-axis current; This refers to the q-axis winding coefficient; This represents the number of pole pairs of the motor.

[0106] Calculate the magnetic reluctance of each segment in the magnetic circuit separately, and then calculate the total magnetic reluctance: Calculate the stator tooth reluctance based on the stator tooth length, effective cross-sectional area of ​​the stator teeth, and the permeability of the composite core; calculate the air gap reluctance based on the air gap length, effective cross-sectional area of ​​the air gap, and the permeability of the composite core; calculate the rotor permanent magnet reluctance based on the rotor permanent magnet length, effective cross-sectional area of ​​the rotor permanent magnet, the relative permeability of the permanent magnet, and the relative permeability at the reference temperature; calculate the total magnetic reluctance based on the stator tooth reluctance, air gap reluctance, and rotor permanent magnet reluctance; for example, the stator tooth reluctance... air gap magnetoresistance Rotor permanent magnet magnetic reluctance The total magnetic reluctance of a series magnetic circuit is the sum of the reluctances of each segment. Taking a 4-pole motor as an example, each pole's magnetic circuit is independent, so only a single pole needs to be calculated. The air gap has two segments, and the total magnetic reluctance... ,in, This refers to the length of the stator teeth. This represents the effective cross-sectional area of ​​the stator teeth. The current temperature is The permeability of the composite iron core at that time; This is the air gap length; This is the effective cross-sectional area of ​​the air gap; The length of the rotor permanent magnet; The relative permeability of the permanent magnet can be directly extracted from the reference material based on the material of the permanent magnet. For example, the typical value for neodymium iron boron permanent magnets is 1.05. This represents the effective cross-sectional area of ​​the rotor permanent magnet;

[0107] Calculate the ratio of the magnetomotive force per pole of the q-axis winding to the total magnetic reluctance to obtain the total magnetic flux; if the total magnetic flux... ;

[0108] The initial theoretical magnetic flux density is obtained by calculating the total magnetic flux and the effective cross-sectional area of ​​the stator teeth; such as the theoretical magnetic flux density. ;

[0109] The frequency is calculated based on the motor speed and the number of pole pairs. If the frequency exceeds a frequency threshold, increased eddy current losses lead to a temperature rise. In this case, the corrected permeability is recalculated based on the increased temperature. The total reluctance is updated based on the corrected permeability, and the theoretical magnetic flux density is recalculated based on the updated total reluctance. Otherwise, the initial theoretical magnetic flux density is used as the theoretical magnetic flux density. (Frequency...) ,in, This represents the motor speed.

[0110] By substituting the theoretical magnetic flux density into the eddy current loss calculation model and combining the eddy current loss coefficient, magnetic field frequency, magnetic flux density, and silicon steel sheet thickness, the eddy current loss per unit volume of the iron core is obtained. For example, the eddy current loss per unit volume of the iron core... ,in, The eddy current loss coefficient is used. For excitation frequency; For the thickness of silicon steel sheets;

[0111] The eddy current loss per unit volume of iron core and the flow rate of coolant are analyzed in conjunction with the Fourier heat conduction equation to obtain a three-dimensional temperature field, which is then used as the three-dimensional temperature field of the winding.

[0112] Methods for obtaining three-dimensional temperature fields include:

[0113] The Reynolds number of the cooling channel is calculated based on the coolant flow rate, coolant density, coolant dynamic viscosity, and hydraulic diameter of the cooling channel; for example, the Reynolds number of the cooling channel. ,in, The density of the coolant; This refers to the coolant flow rate; The hydraulic diameter of the cooling channel; The dynamic viscosity of the coolant;

[0114] The Nusselt number is calculated based on the Reynolds number of the cooling channel and the Prandtl number of the coolant; for example, the Reynolds number of the cooling channel. At that time, corresponding to laminar flow, Nusselt number Cooling channel Reynolds number At that time, corresponding to turbulence, the Nusselt number ,in, This refers to the Prandtl number of the coolant.

[0115] The convective heat transfer coefficient is calculated based on the Nusselt number and the thermal conductivity of the coolant; for example, the convective heat transfer coefficient... ,in, Thermal conductivity of the coolant;

[0116] The three-dimensional structure of the motor is discretized into X grids with side length Y×Y. Each grid represents a control volume P with a volume of... , The length of the grid in the x-direction; The length of the grid in the y-direction; Let be the length of the mesh in the z-direction; obtain the discretized control volume mesh model;

[0117] The governing equations for an internal heat source containing eddy current losses in the iron core are constructed. The equations are then integrated over the control volume P, and each term is discretized. The resulting discrete algebraic equations of the control volume are then obtained. For example, the governing equations... transient term diffusion term Internal heat source item , ,in, The density of the material; Specific heat capacity of the material; For the diffusion term of the governing equation, For gradient operators, For equivalent thermal conductivity, For temperature, This is the gradient vector of temperature;

[0118] To control the material density of volume P; The specific heat capacity of the material controlling volume P; To control the volume of body P; For the control body P in Temperature at any moment; For the control body P at the current moment Temperature; For time step; To control the heat flux density of volume E; To control the surface area of ​​the body; The equivalent thermal conductivity between control volume P and its adjacent control volume E; For control body E in Temperature at any moment; Let P be the distance between control volumes P and E in the x-direction; For the control body P at the current moment eddy current losses in the iron core; The central coefficient of the temperature of the current control volume P in the discrete equation. ; Let be the center coefficient of the temperature of the adjacent control volume E of P in the discrete equation. ; Let be the center coefficient of the temperature of the adjacent control volume W of P in the discrete equation. The calculation method and same; For the control body W in Temperature at any moment; For constant terms, .

[0119] Incorporating boundary conditions into discrete algebraic equations yields a complete system of algebraic equations containing the boundary conditions; for example, for a control volume P located at the boundary:

[0120] When located at a convective heat transfer boundary, i.e., the outer side of the corresponding boundary is coolant and there is no adjacent control volume, other directional coefficients are calculated normally. If the control volume is E, the equation is corrected as follows: The central coefficient of control volume P ,in, The temperature of control volume P after convective heat transfer boundary correction; W, S, N, U and D are adjacent control volumes of control volume P; , , , and These are the temperatures of the adjacent control bodies W, S, N, U, and D of control body P, respectively. , , and Let be the center coefficient of the temperature of the adjacent control volumes W, S, N, U, and D of control volume P. , , and The calculation method and same; As a contributor to convective heat transfer, The convective heat transfer coefficient is... This represents the boundary area; the corresponding constant term. , This refers to the coolant temperature.

[0121] When located at an adiabatic boundary, meaning the outer side of the corresponding boundary is also an adiabatic boundary with no adjacent control volume (e.g., control volume W), and other directional coefficients are calculated normally, the equation is corrected as follows: The central coefficient of control volume P ,in, The temperature of the control volume P after adiabatic boundary correction; The temperature of the adjacent control volume E of control volume P; the corresponding constant term. ;

[0122] When located at the radiation boundary, i.e., in relation to ambient temperature For radiative heat transfer, without adjacent control volumes, other directional coefficients are calculated normally. If the control volume is S, the equation is corrected as follows: Linearization of radiative heat flux: The radiative heat flux is linearized. Approximately radiative heat flow ,in , For reference temperature, take ,coefficient The central coefficient of control volume P ,in, The temperature of the control volume P after radiative heat transfer correction; It is the Stefan-Boltzmann constant; Surface emissivity is the ratio of the surface emissivity of an object to that of a blackbody. The surface temperature of the object participating in radiative heat transfer; The ambient temperature; This is a contribution term to the radiative heat transfer coefficient. The convective heat transfer coefficient is... The area of ​​the radiating boundary; the corresponding constant term ;

[0123] Solve the complete system of algebraic equations, including boundary conditions, to obtain the new temperature field. For example, using a successive over-relaxation iterative method: for each control volume P, update the temperature sequentially. ,in, The over-relaxation factor can be obtained through natural heuristic optimization algorithms; This represents the number of iterations. For the constant term of the control body P; This represents the sum of the thermal effects of adjacent control entities on the current control entity P. Adjacent control bodies coefficient, Adjacent control bodies The temperature is calculated; the temperature deviation of all control volumes is calculated. If the temperature deviation is lower than the minimum deviation threshold, the iteration terminates and the corresponding updated temperature is obtained as the corresponding three-dimensional temperature field. Otherwise, the process returns to recalculate the magnetic permeability of the composite iron core and update to obtain a new temperature field.

[0124] This method first measures the actual inductance of a preset temperature-sensitive ferromagnetic material and a silicon steel sheet at different temperature points, calculates their relative permeability, and performs quadratic polynomial fitting to obtain their temperature characteristic models. Then, it combines this with a volume-weighted calculation of the composite core permeability within a single cycle, providing a precise basis for the correlation between temperature and magnetic properties. Subsequently, based on the composite core permeability, q-axis current, and motor speed, it calculates the theoretical magnetic flux density using Maxwell's equations and iteratively corrects for deviations to ensure the accuracy of the magnetic field analysis. Finally, it substitutes the theoretical magnetic flux density into the eddy current loss model, combines it with cooling system parameters, and uses the Fourier heat conduction equation to perform a three-dimensional analysis of the motor. The structure is discretized into multiple grid control volumes. Control equations containing internal heat sources are constructed and boundary conditions are incorporated for solution. Iterative optimization is performed until the temperature deviation meets the threshold, ultimately obtaining a three-dimensional temperature field. This process, through the magnetic property feedback of distributed temperature-sensitive microplates in the composite iron core, combined with multi-physics field coupling iteration and three-dimensional grid discretization, overcomes the limitation of existing technologies that rely on sensor point measurements to obtain only the temperature of a limited number of points on the shaft. It can comprehensively cover all regions of the motor stator and the rotor connected by thermal coupling, realizing the reconstruction from local magnetic signals to the global temperature field, accurately capturing the three-dimensional temperature distribution including the rotor, and effectively avoiding the omission of local hot spots due to insufficient monitoring range.

[0125] Temperature gradient analysis is performed on the three-dimensional temperature field, temperature warnings are issued based on the analysis results, and motor operating parameters are dynamically adjusted.

[0126] Methods for issuing temperature warnings and dynamically adjusting motor operating parameters based on analysis results include:

[0127] Calculate the partial derivatives of the three-dimensional temperature difference with respect to the x and y directions to obtain the axial and radial temperature change rates, and then calculate the temperature gradient based on the axial and radial temperature change rates.

[0128] The abnormal deviation is calculated based on the temperature gradient and the baseline gradient. The abnormal region is divided by K-means clustering. The number of stator teeth covered by the abnormal region is counted. The gradient change rate of the abnormal region is calculated. If the abnormal deviation is between the first deviation threshold and the second deviation threshold, the number of stator teeth covered by the abnormal region is not higher than the number of teeth threshold, and the change rate of the abnormal region is not higher than the first change rate threshold, then a first-level warning is triggered.

[0129] The cooling orientation adjustment range is calculated based on the current cooling orientation range, abnormal deviation, and flow rate adjustment coefficient. The temperature field inversion sampling period is shortened to G, and the rate of change of abnormal deviation is calculated every preset time interval. If the rate of change is less than 0, the current adjustment is maintained. If the rate of change is not less than 0, the flow rate adjustment coefficient is increased to the next level, and the adjustment range is recalculated until the rate of change is less than 0.

[0130] Real-time updates of abnormal deviation and number of teeth covered in abnormal areas; calculation of gradient change rate in abnormal areas.

[0131] If the abnormal deviation is between the second and third deviation thresholds, the number of stator teeth covered by the abnormal region is higher than the number of teeth threshold, and the rate of change of the abnormal region is between the second and third rate of change thresholds, then a level two warning is triggered.

[0132] Calculate the cooling orientation adjustment range based on the current cooling orientation range, abnormal deviation, and flow adjustment coefficient; calculate the load adjustment range based on the current load, abnormal deviation, and load adjustment coefficient;

[0133] If the abnormal deviation is lower than the second deviation threshold and the number of stator teeth covered by the abnormal area is lower than the number of teeth threshold, after maintaining the current adjustment preset time period, the load and cooling flow will be gradually restored according to the preset ratio, and the second-level warning response will be lifted.

[0134] The criteria for triggering a Level 3 warning include: if the abnormal deviation is greater than the third deviation threshold, or the rate of change of the abnormal region is greater than the third rate of change threshold; or the gradient deviation of the abnormal region is accompanied by q-axis current fluctuation greater than the fluctuation threshold, or the vibration value is greater than C times the normal vibration value under the current operating conditions; or after the cooling flow rate is increased by the first proportional threshold and the load is reduced by the second proportional threshold, the abnormal deviation is still greater than the highest deviation threshold; if any of the above criteria are met, a Level 3 warning is triggered.

[0135] Calculate the load adjustment range based on the current load, abnormal deviation, and load adjustment coefficient; adjust the cooling flow rate to the maximum allowable value under the current operating conditions.

[0136] If the abnormal deviation is still greater than the highest deviation threshold after adjustment, the load will be reduced by a preset proportion at predetermined intervals until the load is 0, and then the main power supply will be disconnected to cool down the motor temperature to below the highest temperature threshold.

[0137] Example 2

[0138] Please see Figure 4 As shown, this embodiment provides a magnetic levitation fan motor winding temperature monitoring device, including:

[0139] Structural integration module: The pre-set temperature-sensitive ferromagnetic material is laminated with the stator core of the magnetic levitation fan motor according to the pre-set process to obtain a composite core;

[0140] Magnetic field scanning module: A high-frequency excitation system is deployed on the composite iron core. The number of turns of the micro-coil of the high-frequency excitation system is adjusted according to the target magnetic field strength. Real-time excitation current is collected and analyzed to obtain the magnetic induction intensity.

[0141] Temperature analysis module: Through multi-physics coupling iteration, the magnetic induction intensity is analyzed to obtain the three-dimensional temperature field of the winding;

[0142] Temperature warning module: Performs temperature gradient analysis on the three-dimensional temperature field, issues temperature warnings based on the analysis results, and dynamically adjusts motor operating parameters.

[0143] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

[0144] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for early warning of winding temperature in a magnetic levitation fan motor, characterized in that, include: A composite core is obtained by laminating a pre-set temperature-sensitive ferromagnetic material with the stator core of a magnetic levitation fan motor according to a pre-set process. Methods for obtaining composite cores include: The preset temperature-sensitive ferromagnetic material is cut into rectangular micro-sheets with a thickness of c and an area accounting for 5% of the cross-sectional area of ​​the stator tooth root using an ultraviolet laser cutting machine. An insulating layer of thickness d was deposited on the surface of a micro-flake using magnetron sputtering to obtain a micro-flake with a uniform coating. The basic unit uses an A-layer silicon steel sheet superimposed with a B-layer micro-film. According to the preset stacking cycle, the stacking is periodically and alternately performed along the axial direction of the stator tooth root of the magnetic levitation fan motor. The micro-chips are arranged radially along the root of the stator teeth of the magnetic levitation fan motor, with one micro-chip corresponding to each stator slot, thus obtaining a composite iron core; A high-frequency excitation system is deployed on a composite iron core. The number of turns of the micro-coil of the high-frequency excitation system is adjusted according to the target magnetic field strength. Real-time excitation current is collected and analyzed to obtain the magnetic induction intensity. The magnetic induction intensity is analyzed through multi-physics coupling iteration to obtain the three-dimensional temperature field of the winding; Methods for obtaining the three-dimensional temperature field of the winding include: Set a reference temperature, and calculate the magnetic permeability of the composite iron core based on the current temperature and the permeability temperature model. Based on the magnetic permeability of the composite iron core, the q-axis current, and the motor speed, and combined with Maxwell's equations, the theoretical magnetic flux density is calculated and compared with the actual magnetic flux density. If the deviation is greater than the deviation threshold, the magnetic permeability of the composite iron core is adjusted and the calculation is repeated. By substituting the theoretical magnetic flux density into the eddy current loss calculation model and combining the eddy current loss coefficient, magnetic field frequency, magnetic flux density, and silicon steel sheet thickness for analysis, the eddy current loss per unit volume of iron core is obtained. The eddy current loss per unit volume of iron core and the flow rate of coolant are analyzed in conjunction with the Fourier heat conduction equation to obtain a three-dimensional temperature field, which is then used as the three-dimensional temperature field of the winding. Temperature gradient analysis is performed on the three-dimensional temperature field, temperature warnings are issued based on the analysis results, and motor operating parameters are dynamically adjusted.

2. The method for early warning of winding temperature of a magnetic levitation fan motor according to claim 1, characterized in that, Methods for obtaining the permeability of composite iron cores include: Determine the effective cross-sectional area and magnetic circuit length of the preset temperature-sensitive ferromagnetic material; D temperature points are preset, and each temperature point is kept at a temperature for a period of t0. The test frequency and the first measuring coil are set. The coil inductance is measured and the empty coil inductance is subtracted to obtain the actual inductance of the preset temperature-sensitive ferromagnetic material. The relative permeability of the preset temperature-sensitive ferromagnetic material is calculated based on the actual inductance, vacuum permeability, effective cross-sectional area, and magnetic circuit length of the preset temperature-sensitive ferromagnetic material. By performing a quadratic polynomial fitting on the relative permeability of the preset temperature-sensitive ferromagnetic material, the relative permeability of the preset temperature-sensitive ferromagnetic material with respect to different temperatures is obtained. The first temperature characteristic fitting model is used. Determine the effective cross-sectional area and magnetic circuit length of silicon steel sheets; D temperature points are preset, each temperature point is kept at a temperature for a period of t0, the test frequency is set, the coil inductance is measured, and the empty coil inductance is subtracted to obtain the actual inductance of the silicon steel sheet; The relative permeability of silicon steel sheets is calculated based on the actual inductance, vacuum permeability, effective cross-sectional area, and magnetic circuit length of the silicon steel sheets. A second-temperature characteristic fitting model of the relative permeability of silicon steel sheet with respect to different temperatures is obtained by performing a quadratic polynomial fitting on the relative permeability of silicon steel sheet; The volume of the silicon steel sheet in a single cycle is calculated based on the number of silicon steel sheet layers, the thickness of the silicon steel sheet, and the cross-sectional area of ​​the iron core. The volume of the preset temperature-sensitive ferromagnetic material in a single cycle is calculated based on the number of preset temperature-sensitive ferromagnetic material layers, the thickness of the preset temperature-sensitive ferromagnetic material, and the cross-sectional area of ​​the iron core. The volume of the composite iron core in a single cycle is calculated based on the volume of the silicon steel sheet and the volume of the preset temperature-sensitive ferromagnetic material. The volume ratio of the silicon steel sheet and the volume ratio of the preset temperature-sensitive ferromagnetic material are calculated based on the volume of the silicon steel sheet in a single cycle, the volume of the preset temperature-sensitive ferromagnetic material in a single cycle, and the volume of the composite iron core. The magnetic permeability of the composite core is calculated based on the relative permeability of the preset temperature-sensitive ferromagnetic material, the relative permeability of the silicon steel sheet, the current temperature, and the reference temperature, combined with the permeability temperature model. The coefficients of the permeability temperature model are the corresponding coefficients in the first temperature characteristic fitting model and the second temperature characteristic fitting model, combined with the weighted result of the volume ratio of the silicon steel sheet and the volume ratio of the preset temperature-sensitive ferromagnetic material.

3. The method for early warning of winding temperature of a magnetic levitation fan motor according to claim 1, characterized in that, Methods for obtaining theoretical magnetic flux density include: Based on Ampere's circuital law, the magnetomotive force of each pole of the q-axis winding is calculated according to the number of turns of each group of the q-axis winding, the q-axis current, the q-axis winding coefficient, and the number of pole pairs of the motor. The magnetic reluctance of each segment in the magnetic circuit is calculated separately, and then the total magnetic reluctance is obtained: the stator tooth magnetic reluctance is calculated based on the stator tooth length, the effective cross-sectional area of ​​the stator teeth, and the permeability of the composite iron core; the air gap magnetic reluctance is calculated based on the air gap length, the effective cross-sectional area of ​​the air gap, and the permeability of the composite iron core; the rotor permanent magnet magnetic reluctance is calculated based on the rotor permanent magnet length, the effective cross-sectional area of ​​the rotor permanent magnet, the relative permeability of the permanent magnet, and the relative permeability at the reference temperature; the total magnetic reluctance is obtained by calculating the stator tooth magnetic reluctance, the air gap magnetic reluctance, and the rotor permanent magnet magnetic reluctance. Calculate the ratio of the magnetomotive force per pole of the q-axis winding to the total magnetic reluctance to obtain the total magnetic flux. The initial theoretical magnetic flux density is calculated based on the total magnetic flux and the effective cross-sectional area of ​​the stator teeth. The frequency is calculated based on the motor speed and the number of pole pairs. If the frequency is greater than the frequency threshold, the eddy current loss increases, leading to a rise in temperature. The corrected permeability is then recalculated based on the increased temperature. The total reluctance is updated based on the corrected permeability, and the theoretical magnetic flux density is recalculated based on the updated total reluctance. Otherwise, the initial theoretical magnetic flux density is used as the theoretical magnetic flux density.

4. The method for early warning of winding temperature of a magnetic levitation fan motor according to claim 1, characterized in that, Methods for obtaining three-dimensional temperature fields include: The Reynolds number of the cooling channel is calculated based on the coolant flow rate, coolant density, coolant dynamic viscosity, and hydraulic diameter of the cooling channel. The Nusselt number is calculated based on the Reynolds number of the cooling channel and the Prandtl number of the coolant. The convective heat transfer coefficient is calculated based on the Nusselt number and the thermal conductivity of the coolant. The three-dimensional structure of the motor is discretized into X grids with side length Y×Y. Each grid represents a control volume P with a volume of... , The length of the grid in the x-direction; The length of the grid in the y direction; Let be the length of the mesh in the z-direction; obtain the discretized control volume mesh model; The control equations of an internal heat source containing eddy current losses in the iron core are constructed. The control volume P is integrated and the terms of the control equations are discretized. The discrete algebraic equations of the control volume are obtained by rearranging them. By incorporating boundary conditions into discrete algebraic equations, a complete system of algebraic equations containing boundary conditions is obtained. Solve the complete algebraic equations including boundary conditions to obtain the new temperature field; calculate the temperature deviation of all control volumes. If the temperature deviation is lower than the minimum deviation threshold, the iteration terminates and the corresponding updated temperature is obtained as the corresponding three-dimensional temperature field. Otherwise, return to recalculate the magnetic permeability of the composite iron core and update to obtain the new temperature field.

5. The method for early warning of winding temperature of a magnetic levitation fan motor according to claim 1, characterized in that, Methods for adjusting the number of turns of the micro-coil in a high-frequency excitation system based on the target magnetic field strength include: The number of turns of the microcoil is calculated based on the excitation current, the length of the composite iron core magnetic circuit, and the target magnetic field strength. The microcoil of the high-frequency excitation system is then deployed based on the number of turns of the microcoil.

6. The method for early warning of winding temperature of a magnetic levitation fan motor according to claim 5, characterized in that, Methods for obtaining magnetic field strength include: The real-time magnetic field strength is calculated based on the real-time excitation current, the length of the composite iron core magnetic circuit, and the number of turns. Then, the magnetic induction intensity is calculated based on the real-time magnetic field strength, the relative permeability of the composite iron core, and the vacuum permeability.

7. The method for early warning of winding temperature of a magnetic levitation fan motor according to claim 1, characterized in that, Methods for issuing temperature warnings based on analysis results include: Calculate the partial derivatives of the three-dimensional temperature difference with respect to the x and y directions to obtain the axial and radial temperature change rates, and then calculate the temperature gradient based on the axial and radial temperature change rates. The abnormal deviation is calculated based on the temperature gradient and the baseline gradient. The abnormal region is divided by K-means clustering. The number of stator teeth covered by the abnormal region is counted. The gradient change rate of the abnormal region is calculated. If the abnormal deviation is between the first deviation threshold and the second deviation threshold, the number of stator teeth covered by the abnormal region is not higher than the number of teeth threshold, and the change rate of the abnormal region is not higher than the first change rate threshold, then a first-level warning is triggered. Real-time updates of abnormal deviation and number of teeth covered in abnormal areas; calculation of gradient change rate in abnormal areas. If the abnormal deviation is between the second and third deviation thresholds, the number of stator teeth covered by the abnormal region is higher than the number of teeth threshold, and the rate of change of the abnormal region is between the second and third rate of change thresholds, then a level two warning is triggered. The criteria for triggering a Level 3 warning include: if the abnormal deviation is greater than the third deviation threshold, or the rate of change of the abnormal region is greater than the third rate of change threshold; or the gradient deviation of the abnormal region is accompanied by q-axis current fluctuation greater than the fluctuation threshold, or the vibration value is greater than C times the normal vibration value under the current operating conditions; or after the cooling flow rate is increased by the first proportional threshold and the load is reduced by the second proportional threshold, the abnormal deviation is still greater than the highest deviation threshold. If any of the criteria is met, a Level 3 warning is triggered.

8. The method for early warning of winding temperature of a magnetic levitation fan motor according to claim 1, characterized in that, Methods for dynamically adjusting motor operating parameters based on temperature warnings include: When a Level 1 warning is triggered, the cooling orientation adjustment range is calculated based on the current cooling orientation range, abnormal deviation, and flow adjustment coefficient. The temperature field inversion sampling period is shortened to G, and the rate of change of abnormal deviation is calculated every preset time interval. If the rate of change is less than 0, the current adjustment is maintained. If the rate of change is not less than 0, the flow adjustment coefficient is increased to the next level, and the adjustment range is recalculated until the rate of change is less than 0, and the Level 1 warning response is lifted. When a Level 2 warning is triggered, the cooling direction adjustment magnitude is calculated based on the current cooling direction magnitude, abnormal deviation, and flow adjustment coefficient; the load adjustment magnitude is calculated based on the current load, abnormal deviation, and load adjustment coefficient. If the abnormal deviation is lower than the second deviation threshold and the number of stator teeth covered by the abnormal area is lower than the number of teeth threshold, after maintaining the current adjustment preset time period, the load and cooling flow will be gradually restored according to the preset ratio, and the second-level warning response will be lifted. When a Level 3 warning is triggered, the load adjustment range is calculated based on the current negative value; the cooling flow is adjusted to the maximum allowable value of the current operating condition, abnormal deviation, and load adjustment system. If the abnormal deviation is still greater than the highest deviation threshold after adjustment, the load will be reduced by a preset proportion at predetermined intervals until the load is 0. Then the main power supply will be disconnected and the motor temperature will be cooled down to below the highest temperature threshold to release the Level 3 warning response.

9. A magnetic levitation fan motor winding temperature monitoring device, implementing the magnetic levitation fan motor winding temperature early warning method according to any one of claims 1-8, characterized in that, include: Structural integration module: The pre-set temperature-sensitive ferromagnetic material is laminated with the stator core of the magnetic levitation fan motor according to the pre-set process to obtain a composite core; Magnetic field scanning module: A high-frequency excitation system is deployed on the composite iron core. The number of turns of the micro-coil of the high-frequency excitation system is adjusted according to the target magnetic field strength. Real-time excitation current is collected and analyzed to obtain the magnetic induction intensity. Temperature analysis module: Through multi-physics coupling iteration, the magnetic induction intensity is analyzed to obtain the three-dimensional temperature field of the winding; Temperature warning module: Performs temperature gradient analysis on the three-dimensional temperature field, issues temperature warnings based on the analysis results, and dynamically adjusts motor operating parameters.

Images