A method and system for air-cooled motor circulation ventilation and pressure regulation for high altitude conditions

Abstract

The application belongs to the technical field of air-cooled motors, and particularly relates to a circulating ventilation and pressure regulation method and system for air-cooled motors under high-altitude conditions. A sealed inner cavity surrounding a rotor is constructed by a sealed machine shell and an end cover type bearing. Based on the rotor fan driving air circulation, the air is cooled by a top heat exchanger after flowing through a heating component, forming an isolated internal circulation. The inner cavity air pressure and temperature are monitored in real time, and the air supplement threshold is corrected according to the load-temperature model, so that the main machine is instructed to supplement purified air by a variable frequency fan when the air pressure is insufficient, and the internal fan speed is adjusted. Periodic leakage diagnosis is performed, and when leakage or failure occurs, the system can perform load reduction operation to ensure that the air pressure and heat dissipation capacity are always maintained above the safety benchmark. Based on the load, temperature and air pressure linkage control mode, the air supplement is optimized, the air pressure maintenance and heat dissipation efficiency are met, the reliability of the system when the sealing performance deteriorates is improved, and the stable operation of the motor is ensured.

Description

Technical Field

[0001] This invention belongs to the field of air-cooled motor technology, specifically relating to a method and system for circulating ventilation and pressure regulation of air-cooled motors for high-altitude conditions. Background Technology

[0002] The high-altitude special motor industry belongs to the intersection of high-end equipment manufacturing and new energy industry, and mainly serves the fields of plateau wind power generation, pumped storage in cold regions, plateau mining machinery and related regional power supply.

[0003] Currently, in high-altitude environments, when external air directly enters the motor and participates in circulation, the low air density results in a small convective heat transfer coefficient and poor cooling effect. Dust and moisture in the external air penetrate the stator windings and rotor core with the circulating airflow, causing surface contamination and moisture absorption of the insulation, accelerating the deterioration of insulation performance. Related motor technologies employ open or semi-open air duct structures; for example, the sealing structure uses lip seals or labyrinth seals at the shaft penetration points. Over long-term operation, wear leads to leakage, and the reliability of the seal decreases with increasing altitude.

[0004] During operation, the motors in this technology experience frequent load fluctuations and real-time changes in heat generation rates. Meanwhile, at high altitudes, heat dissipation capacity fluctuates with air pressure, and this dynamic coupling complicates temperature control. The current control method uses a defined temperature threshold for triggering protection. At low temperatures and light loads, an excessively high threshold leads to frequent false starts of the air supply system, while at high temperatures and heavy loads, an excessively low threshold causes delayed air supply and uncontrolled temperature rise. Furthermore, it can easily cause a mismatch between the air supply pressure threshold and the actual heat dissipation demand, resulting in either excessive or insufficient air supply, leading to overheating.

[0005] Furthermore, when makeup air is injected into a sealed cavity, the direction, velocity, and temperature of the makeup air differ from the original internal circulating air. If the two flow directions are opposite or their velocities are mismatched, eddies or stagnant zones will form in local areas, reducing the overall convective heat transfer coefficient. If the makeup air directly impacts a local area of ​​the stator winding, it will cause thermal stress concentration in that area. Summary of the Invention

[0006] This invention provides a method for circulating ventilation and pressure regulation of air-cooled motors for high-altitude conditions, which effectively improves the heat dissipation efficiency and insulation strength of motors in high-altitude environments and ensures stable operation.

[0007] The methods include: S1. The sealed housing and the end-cap type sealed bearing are combined and installed to form a closed inner cavity surrounding the rotor assembly, and a heat exchange device is installed on the top of the closed inner cavity. S2. Connect pressure monitoring equipment and temperature monitoring equipment in the sealed cavity, and connect the output of the equipment to the pressure control host respectively to establish a multi-parameter monitoring link for synchronously collecting air pressure data and temperature data of the sealed cavity. S3. Using the rotor assembly's internal fan, the air in the sealed cavity is driven to form an internal circulating airflow. S4. The pressure monitoring device obtains the current internal air pressure value of the sealed cavity in real time, and the temperature monitoring device obtains the current internal temperature value of the sealed cavity in real time, and transmits it to the pressure control host. S5. Pre-set a load-temperature correlation model in the pressure control host; determine the target temperature value based on the current internal temperature value and the load-temperature correlation model; compare the current internal temperature value with the target temperature value; correct the make-up air pressure threshold based on the comparison result; compare the current internal air pressure value with the corrected make-up air pressure threshold; if it is determined that the current air pressure is lower than the corrected threshold, generate a make-up air control command including make-up air volume, make-up air duration, and fan cooperative speed. S6. The pressure control host sends a supplementary air control command to the variable frequency fan. The variable frequency fan draws purified air from the external environment through the supplementary air duct and injects supplementary air into the sealed inner cavity to increase the air density. According to the deviation ratio between the current internal air pressure value and the target air pressure, the speed of the rotor kit internal fan is adjusted to enhance the heat dissipation efficiency of the internal circulating airflow. S7. Periodically control the pressure monitoring equipment to collect pressure decay data of the sealed inner cavity, calculate the pressure decay rate and compare it with the preset leakage threshold. If the pressure decay rate is determined to exceed the threshold or the make-up air system is faulty, a load reduction operation command will be triggered and a maintenance alarm will be issued. Continue performing closed-loop adjustments in steps S4-S6 until the air pressure in the sealed cavity returns to a level not lower than the safe operating baseline for low altitude.

[0008] According to another embodiment of this application, a circulating ventilation and pressure regulation system for an air-cooled motor in high-altitude conditions is provided. The system includes: a sealed housing, an end-cap type sealed bearing, a heat exchange device, a pressure monitoring device, a pressure control host, a variable frequency fan, and a makeup air duct. The sealed housing is equipped with stepped bearing mounting holes and heat exchanger mounting windows; The heat exchanger mounting window is surrounded by a sealing flange face and threaded blind holes; the inner wall of the stepped bearing mounting hole is provided with an annular sealing groove. An O-ring fluororubber seal is provided in the annular sealing groove on the inner wall of the stepped bearing mounting hole. The outer ring of the end cap type sealed bearing is installed in the stepped bearing mounting hole of the sealed housing. The heat exchange device includes: a microchannel heat pipe array and a rectangular metal frame; The rectangular metal frame is connected to the sealing flange face of the heat exchanger mounting window on the top of the sealed housing. End cap type sealed bearings include: bearing outer ring, bearing inner ring, and O-ring fluororubber seal; The outer ring of the bearing is installed in the stepped bearing mounting hole of the sealed housing, and the inner ring of the bearing is connected to the rotor assembly; the O-ring fluororubber seal is embedded in the annular sealing groove of the stepped bearing mounting hole; the microchannel heat pipe array is composed of multiple parallel microchannel heat pipes, filled with acetone working fluid, and the heat pipes are divided into evaporation section and condensation section. The pressure monitoring equipment is connected to the pressure control host, which is connected to the variable frequency fan. The variable frequency fan is connected to the inside of the sealed housing through the air supply pipe.

[0009] As can be seen from the above technical solutions, the present invention has the following advantages: The circulating ventilation and pressure regulation method for air-cooled motors in high-altitude conditions provided by this invention uses a sealed housing assembly and an end-cap type sealed bearing to form a closed internal cavity. This prevents gas exchange between the internal cooling medium and the external high-altitude environment during normal operation, avoiding the reduction in heat dissipation capacity caused by the direct participation of low-density external air in the internal circulation. The top-mounted heat exchange device allows internal air to absorb heat through the heating element and then flow directionally to the external environment. This achieves heat transfer across the wall surface while maintaining the isolation between the internal and external gases, solving the problem of reduced convective heat transfer efficiency caused by the thin air in high-altitude environments.

[0010] The deployment of pressure and temperature monitoring devices and their electrical connection with the pressure control host in this invention establishes multi-parameter monitoring. The rotor-mounted internal fan drives the air in the sealed cavity, creating an internal circulating airflow. This directs heat from the stator windings, rotor core, and other heat-generating components to the heat exchange device, ensuring no leakage of internal air during heat exchange. The preset load-temperature correlation model and the correction of the make-up air pressure threshold based on the current temperature and model output match the make-up air start-up conditions with the actual thermal load of the motor, avoiding ineffective make-up air at low temperatures and light loads or delayed make-up air at high temperatures and heavy loads. The variable frequency fan injects purified air into the sealed cavity through the make-up air duct to increase air density, while simultaneously adjusting the internal fan speed according to the air pressure deviation ratio, allowing the incoming air to quickly integrate into the internal circulation and enhance heat dissipation efficiency. Regular collection of pressure decay data and comparison with leakage thresholds, along with the triggering of load reduction commands and the issuance of maintenance alarms in case of faults, enable the system to have self-diagnostic capabilities for sealing performance and safety redundancy.

[0011] The sealing structure of this invention eliminates the direct intrusion of the harsh external environment on the interior, and the combined mechanism of air supply and fan keeps the internal air pressure and air density dynamically stable at low altitude levels, improving the reliability of the motor. Attached Figure Description

[0012] To more clearly illustrate the technical solution of the present invention, the accompanying drawings used in the description will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0013] Figure 1 This is a schematic diagram of a circulating ventilation and pressure regulation system for an air-cooled motor used in high-altitude conditions. Figure 2 This is a flowchart of a circulating ventilation and pressure control method for an air-cooled motor used in high-altitude conditions.

[0014] Explanation of reference numerals in the attached figures: 1 - Sealed housing, 2 - End cap type sealed bearing, 3 - Heat exchange device, 4 - Pressure monitoring equipment, 5 - Pressure control host, 6 - Variable frequency fan, 7 - Makeup air duct. Detailed Implementation

[0015] like Figure 1 As shown, the wind-cooled motor circulating ventilation and pressure regulation system for high-altitude conditions provided by the present invention includes a sealed housing 1, an end-cap type sealed bearing 2, a heat exchange device 3, a pressure monitoring device 4, a pressure control host 5, a variable frequency fan 6, and a makeup air duct 7.

[0016] The sealed housing 1 is provided with stepped bearing mounting holes and heat exchanger mounting windows; the heat exchanger mounting windows are surrounded by sealing flange faces and threaded blind holes; the inner wall of the stepped bearing mounting holes is provided with an annular sealing groove.

[0017] An O-ring fluororubber seal is installed in the annular sealing groove on the inner wall of the stepped bearing mounting hole, and the outer ring of the end cap type sealed bearing 2 is installed in the stepped bearing mounting hole of the sealing housing 1.

[0018] The heat exchange device 3 includes a microchannel heat pipe array and a rectangular metal frame; the rectangular metal frame of the heat exchange device 3 is connected to the sealing flange face of the heat exchanger mounting window on the top of the sealed housing 1 by bolts; the pressure monitoring device 4 is connected to the pressure control host 5, the pressure control host 5 is connected to the variable frequency fan 6, and the variable frequency fan 6 is connected to the inside of the sealed housing 1 through the make-up air pipe 7.

[0019] The end-cap type sealed bearing 2 includes: an outer bearing ring, an inner bearing ring, and an O-ring fluororubber seal. The outer bearing ring is installed in the stepped bearing mounting hole of the sealed housing 1, and the inner bearing ring is connected to the rotor assembly. The O-ring fluororubber seal is embedded in the annular sealing groove of the stepped bearing mounting hole. The microchannel heat pipe array consists of multiple parallel microchannel heat pipes filled with acetone working fluid. The heat pipes are divided into evaporation and condensation sections.

[0020] It should be noted that the evaporation section of the microchannel heat pipe array extends into the sealed housing 1, while the condensation section is exposed to the outside. A hollow metal O-ring is provided between the sealing groove on the bottom surface of the metal frame and the heat exchanger mounting window of the sealed housing 1.

[0021] It should be noted that the sealed housing 1 constitutes the external sealed housing of the motor, isolating the internal cooling airflow from the external environment. The stepped bearing mounting holes provide radial positioning and axial limiting references for the end-cap type sealed bearing 2. The heat exchanger mounting window provides a mounting base for the heat exchange device 3, and the sealing flange face and threaded blind hole ensure the installation strength and sealing performance of the heat exchange device.

[0022] The outer ring of the end cap type sealed bearing 2 rotates within the stepped bearing mounting hole of the sealed housing 1, supporting the rotor assembly, bearing the radial load of the rotor, and ensuring stable rotor operation.

[0023] The inner ring of the end-cap type sealed bearing 2 is interference-fitted with the rotor assembly journal to transmit rotor torque and realize mechanical energy transmission. The O-ring fluororubber seal of the end-cap type sealed bearing 2 fills the gap between the outer ring of the bearing and the stepped bearing mounting hole of the sealing housing 1, preventing internal gas from leaking along the bearing axial direction and enhancing the overall sealing performance of the motor.

[0024] The microchannel heat pipe array in heat exchange device 3 utilizes the phase change of acetone to achieve a cycle of heat absorption in the evaporation section, steam rising, heat release in the condensation section, and working fluid recirculation, efficiently transferring heat from inside the motor while completely isolated from the internal and external air. The microchannel structure improves heat transfer efficiency and solves the problem of poor heat dissipation in high-altitude areas.

[0025] The rectangular metal frame of the heat exchange device 3 provides structural support for the microchannel heat pipe array, ensuring the mechanical strength and positional stability of the heat pipe array.

[0026] Pressure monitoring device 4 collects real-time air pressure signals inside the air-cooled motor, providing data for pressure regulation. Pressure control host 5 receives signals from pressure monitoring device 4; when the internal pressure falls below a threshold, it sends a supplementary air command to variable frequency fan 6 to achieve automatic pressure compensation. Variable frequency fan 6 receives the command from pressure control host 5 and supplements air into the motor through supplementary air duct 7, maintaining internal pressure similar to that of low-altitude areas and addressing the problems of low air pressure, high temperature, and insulation corona at high altitudes. Supplementary air duct 7 serves as the air transport channel between variable frequency fan 6 and the sealed casing 1, regulating internal air density and pressure.

[0027] The following describes in detail the circulating ventilation and pressure regulation method for air-cooled motors used in high-altitude conditions according to this application. Specific details, such as particular system structures and technologies, are presented for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application can also be implemented in other embodiments without these specific details.

[0028] It should be understood that, when used in this specification, the term "comprising" indicates the presence of the described feature, integral, step, operation, element, and / or component, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components, and / or collections thereof. The terms "comprising," "including," "having," and variations thereof all mean "including but not limited to," unless otherwise specifically emphasized.

[0029] The terms "one embodiment" or "some embodiments" used in this application mean that one or more embodiments of this application include the specific features, structures, or characteristics described in that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this application do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized.

[0030] 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.

[0031] Please see Figure 2 The diagram shows a flowchart of a method for circulating ventilation and pressure regulation of an air-cooled motor for high-altitude conditions, as described in a specific embodiment. The method includes: S1. The sealed housing and end-cap type sealed bearing are combined and installed to form a sealed inner cavity surrounding the rotor assembly, achieving isolation between the internal cooling medium of the motor and the external high-altitude environment during normal operation. A heat exchange device is installed at the top of the sealed inner cavity, allowing the internal air to flow directionally to the external environment after heat exchange with the heating components.

[0032] S111: Pre-treat the sealed housing, process the sealing structure of the housing mating surface and perform surface passivation treatment, locate the rotor assembly installation benchmark, hoist the rotor assembly into the preset work position inside the housing and complete the initial fixation.

[0033] In some embodiments, the annular sealing groove provides an installation carrier for the sealing structure, and the passivation treatment can form a dense oxide film on the surface of the housing, improving the flatness and corrosion resistance of the sealing surface and reducing the deformation of the sealing surface caused by the temperature difference between day and night at high altitudes.

[0034] Optionally, crosshair positioning lines and adjustable clamps ensure the rotor assembly is coaxial with the housing, preventing vibration from damaging the sealing structure. Pre-applied sealing primer fills the microscopic gaps in the sealing groove, blocking leakage channels in advance.

[0035] S112: The end cap type sealed bearing is press-fitted to the bearing seats at both ends of the sealing housing using an interference fit method, a flexible sealing bushing is added to fill the assembly gap, and a stepped locating pin is used to fix the bearing end cap to ensure the coaxiality of the bearing and the rotor assembly.

[0036] In some embodiments, the interference fit ensures a tight fit between the bearing and the bearing housing through thermal expansion and contraction, forming the basis of the mechanical seal. The flexible sealing bushing utilizes the elastic deformation of fluororubber to compensate for assembly errors and vibration deformation during operation. Stepped locating pins position the bearing end cap, preventing end cap misalignment that could lead to sealing surface displacement. Locking nuts ensure secure positioning, and coaxiality control prevents accelerated bearing wear during operation, guaranteeing the stability of the sealing structure.

[0037] S113: Perform segmented inflation and pressure testing on the assembled sealing structure. Inflate the sealing housing with dry nitrogen to the preset pressure, test the leakage of each sealing surface, and repair the leaks with sealant.

[0038] In some embodiments, dry nitrogen is chemically stable and will not corrode the internal components of the casing. Filling it with a pressure higher than the baseline value can simulate the sealing and pressure-bearing state under high-altitude and low-pressure environments, and detect potential leakage hazards in advance.

[0039] Optionally, the soap film leak detection method can accurately locate minute leak points. The leakage calculation formula quantifies the sealing performance by relating pressure drop, internal cavity volume, and time, avoiding missed detections due to subjective judgment. Sealant can fill the microscopic gaps in the sealing surface, forming a dense sealing layer after curing, blocking leakage channels.

[0040] S114: A flange interface is machined on the pre-installed mounting surface at the top of the sealed inner cavity. The heat exchange device is connected to the flange interface by bolts, a high-temperature resistant flexible sealing gasket is installed, and an arc-shaped guide plate is installed at the inlet of the heat exchange device.

[0041] In some embodiments, an arc-shaped guide plate is fixed at the inlet of the heat exchange device by welding. The guide plate is made of stainless steel and has an angle adjustment range of 15°-45°. The initial angle is set to 30°. The arc-shaped guide plate uses the principle of fluid mechanics to guide the internal circulating airflow in a directional manner, avoiding the decrease in heat dissipation efficiency caused by airflow turbulence and ensuring that all airflow flows through the heat exchange device.

[0042] S115: Start the temporary drive device to drive the rotor assembly to rotate at low speed, detect the internal airflow direction and velocity, adjust the angle of the guide plate to direct the airflow to the heat exchange device, check the sealing performance and the installation firmness of the heat exchange device, and complete the curing of the basic ventilation structure.

[0043] In some embodiments, a temporary drive unit rotates the rotor assembly to simulate the airflow state during internal fan operation. A Pitot tube flow meter calculates the airflow velocity by detecting the airflow pressure difference. Adjusting the baffle angle changes the airflow direction, ensuring that the airflow flows directionally through the heating components and heat exchange device, forming a complete circulating air path. Secondary pressurization testing checks for potential seal loosening during assembly, ensuring that the sealing structure and heat exchange device are securely installed.

[0044] S2. Connect pressure monitoring equipment and temperature monitoring equipment inside the sealed cavity, and connect the output of the equipment to the pressure control host respectively to establish a multi-parameter monitoring link for synchronously collecting air pressure data and temperature data of the sealed cavity.

[0045] In some embodiments, the pressure monitoring device uses a piezoelectric pressure sensor, connected to the top of the sealed cavity near the inlet of the heat exchange device, avoiding direct impact from airflow and preventing airflow disturbance from affecting detection accuracy. The temperature monitoring device uses a PT100 resistance temperature detector (RTD) sensor, installed at the stator winding end, rotor core surface, and center of the sealed cavity, to achieve multi-point temperature acquisition and reflect the temperature distribution within the cavity.

[0046] S3. The internal fan of the rotor assembly is used to drive the air in the sealed cavity to form an internal circulating airflow. After the circulating airflow passes through the stator windings, rotor core and other heat-generating components to absorb heat, it flows into the heat exchange device for heat exchange, ensuring that the internal air does not leak directly to the external environment during the heat exchange process.

[0047] In some embodiments, the rotor-mounted internal fan adopts a forward-curved blade structure and is mounted on the middle of the rotor assembly shaft via an interference fit and key connection. When the fan is running, it drives the air in the sealed internal cavity to form a clockwise circulating airflow. The airflow path is preset as follows: fan outlet, stator winding gap, rotor core surface, heat exchange device inlet, heat exchanger interior, and fan inlet, forming an airflow circulation.

[0048] Furthermore, high-temperature resistant insulating tape is wrapped around the surface of the stator windings, and arc-shaped airflow guide plates are added to guide the airflow evenly through the winding gaps. Annular guide grooves are machined on the surface of the rotor core to increase the contact area between the airflow and the core, improving heat absorption efficiency. The guide plate at the inlet of the heat exchange device can adjust its angle according to the airflow velocity, ensuring that the circulating airflow flows directionally into the heat exchange channel. During the heat exchange process, internal air flows through the heat exchanger tube walls, transferring heat to the external heat dissipation medium. The cooled air then flows back into the fan inlet. Throughout this process, the sealing structure between the heat exchanger and the casing ensures that internal air does not leak directly to the external environment, maintaining the integrity of the closed-loop circulation.

[0049] S4. The current internal air pressure value of the sealed cavity is obtained in real time through pressure monitoring equipment, and the current internal temperature value of the sealed cavity is obtained in real time through temperature monitoring equipment; the two types of data are transmitted to the pressure control host as the input basis for control logic.

[0050] In some embodiments, pressure and temperature sensors continuously collect data, and stable preheating ensures stable sensor operation and reduces transient errors. Data binding and timestamp annotation ensure the correlation between air pressure and temperature data, facilitating subsequent analysis of their mutual influence.

[0051] S5. Pre-set a load-temperature correlation model in the pressure control host; determine the target temperature value based on the current internal temperature value and the load-temperature correlation model; compare the current internal temperature value with the target temperature value; correct the make-up air pressure threshold based on the comparison result; compare the current internal air pressure value with the corrected make-up air pressure threshold; if it is determined that the current air pressure is lower than the corrected threshold, generate a make-up air control command including make-up air volume, make-up air duration, and fan cooperative speed.

[0052] S6. The pressure control host sends a supplementary air control command to the variable frequency fan. The variable frequency fan draws purified air from the external environment through the supplementary air duct and injects supplementary air into the sealed inner cavity to increase the air density. According to the deviation ratio between the current internal air pressure value and the target air pressure, the speed of the rotor-mounted internal fan is adjusted to enhance the heat dissipation efficiency of the internal circulating airflow.

[0053] S7. Periodically control the pressure monitoring equipment to collect pressure decay data of the sealed inner cavity, calculate the pressure decay rate and compare it with the preset leakage threshold. If the pressure decay rate is determined to exceed the threshold or the make-up air system is faulty, a load reduction operation command will be triggered and a maintenance alarm will be issued. Continue to perform closed-loop adjustments in steps S4-S6 until the air pressure in the sealed cavity is restored to a level not lower than the safe operating baseline for low altitude, ensuring insulation strength and heat dissipation efficiency.

[0054] In one embodiment of the present invention, in S5, the construction method of the load-temperature correlation model specifically includes the following, and the following will provide a possible embodiment and describe its specific implementation in a non-limiting manner.

[0055] S51. Configure the test bench and define the pressure sensor, platinum resistance temperature sensor, and Hall current sensor. Data acquisition conditions include: applying the load current I from 0 to 120% of its rated value in 5% increments, continuously running at each load point until thermal stability, and recording the real-time ambient air pressure p and ambient air temperature T. amb and relative humidity (RH).

[0056] At each steady-state point, the stator winding hot spot temperature T is collected. stator Rotor core temperature T rotor and casing surface temperature T housing The average value of the preset duration is taken as the actual temperature label of the working condition.

[0057] Furthermore, by recording the temperature response curves from a sudden increase in load from 50% to 100%, the thermal time constant τ is extracted as a dynamic feature. An exponential function is then fitted. .

[0058] All data are tagged with operating conditions according to load level, simulated altitude, and operating phase, forming a structured dataset. , where x i The input feature vector includes I, p, and T. amb , RH, τ, etc., y i Let c be the measured temperature vector. i For operating conditions.

[0059] In some embodiments, the test bench is placed in an adjustable pressure environment chamber, and the chamber pressure is controlled by a vacuum pump and regulating valve to vary in 10 kPa increments within the range of 60 kPa to 101.3 kPa to simulate different altitudes.

[0060] Furthermore, at each air pressure setpoint, the motor load is controlled sequentially at 0%, 25%, 50%, 75%, and 100% of the rated load, and briefly operates under 110% overload.

[0061] Furthermore, each load point was continuously operated until thermal stability was achieved, defined as a stator winding temperature change of less than 0.3°C over 10 consecutive minutes. After stabilization, the following were recorded: ambient air pressure p and ambient air temperature T. amb Relative humidity RH, stator winding current I, input power Pin, rotor speed n, internal cooling airflow velocity v, stator winding hot spot temperature T stator Rotor core temperature T rotor Casing surface temperature T housing .

[0062] Furthermore, a load step test was conducted: the load was suddenly increased from 50% to 100%, and the temperature response curve was recorded at a sampling rate of 100ms. The exponential function was then fitted using the nonlinear least squares method. The thermal time constant τ is extracted as a dynamic feature.

[0063] Furthermore, the above data is tagged with operating conditions according to load level, air pressure value, and operating stage (heating / steady-state / cooling) to form a structured dataset. , where x i Let yi be the input feature vector, and yi be the measured temperature vector [T]. stator , T rotor , T housing ], ci is the operating condition label.

[0064] S52. Perform a transformation on the data according to the air density formula. Calculate the current air density ρ.

[0065] Optionally, ρ0 = 1.225 kg / m³, p0 = 101.3 kPa, and T0 = 293 K.

[0066] Structural parameters are extracted from the motor design drawings; specifically, the hydraulic diameter d of the air duct. h =4A / P, where A is the cross-sectional area of ​​the air duct, P is the wetted perimeter, and A is the heat dissipation surface area. heat It is determined by the sum of the areas of the outer surface of the stator core and the end face of the rotor.

[0067] Define the heat dissipation coefficient correction factor The square of the load current I² represents the copper loss, and the rotational speed n is obtained through the encoder, and the iron loss correction term f(I,n) is calculated. The thermal time constant τ obtained by fitting the temperature response curve is used as the inertial characteristic.

[0068] Z-score normalization is performed on all continuous features: , where μ j σ j represents the mean and standard deviation on the training set.

[0069] The normalized data was divided into a training set Dtrain, a validation set Dval, and a test set Dtest in a ratio of 7:2:1, and stratified sampling was used to ensure that the proportion of each load level in the training set was balanced.

[0070] In some embodiments, the copper loss, iron loss, and heat dissipation coefficient correction factor of the motor are encoded as numerical features based on enhanced features, enabling the model to utilize physical laws and reduce its reliance on purely data-driven approaches.

[0071] Based on parameter dh A heat This reflects the inherent heat dissipation capacity of the motor and serves as a normalization mechanism across different motor models. Normalization eliminates differences in feature dimensions, preventing features with large numerical values ​​from dominating the training process. Stratified sampling ensures the model has sufficient training samples for each load range, preventing overfitting to common load points.

[0072] S53. Construct a hybrid model consisting of a physical information neural network PINN and a gated recurrent unit GRU.

[0073] For the stator, rotor, and casing, establish the heat balance equations. C i For heat capacity, P i For losses, G ij For thermal conductivity between components.

[0074] The equation residuals are used as the physical loss term Lphy. .

[0075] Furthermore, GRU uses the I, p, and T values ​​from the past N time points. amb The input is the predicted temperature increment, which is then combined with the steady-state temperature of the PINN.

[0076] Total loss function Where λ1=0.7 and λ2=0.3. The Adam optimizer is used with an initial learning rate of 10. -4 The batch size is 64, training is performed for 200 epochs, and training stops when the validation loss does not decrease for 20 consecutive epochs. To prevent overfitting, Dropout and L2 regularization are added after the GRU layer. After training, the model weights that minimize the validation loss are saved.

[0077] In some embodiments, a Physical Information Neural Network (PINN) model is constructed. Physical loss terms are designed based on a lumped parameter thermal network of the motor. The motor is simplified to three nodes: stator, rotor, and casing, and a thermal balance differential equation is established.

[0078] Where Cs, Cr, and Ch are the heat capacities of the stator, rotor, and casing, respectively, calculated from the material's specific heat capacity and mass; P cu P fe,s P fe,r For loss terms; G sr G sh G rh G ha The thermal conductivity between nodes is determined by thermal conductivity, contact area, and thickness, with some thermal conductivity related to air density: G ha = h·A heath is the convective heat transfer coefficient. Nu is the Nusselt number, which is related to Re and Pr.

[0079] The residuals of the above equations are treated as physical loss terms.

[0080] Where d Ti / d t It is approximated by the difference in temperature measurements.

[0081] This embodiment also uses a feature sequence of a preset past duration, including I, p, Tamb, ρ, dI / dt, etc., as input to output the predicted temperature increment for a preset future duration. The layer number is defined as 2, the number of hidden units is 128, and a fully connected layer is then used to output the temperature increment.

[0082] The steady-state temperature calculated by PINN is superimposed with the predicted increment to obtain the final predicted temperature: T pred = T ss +ΔT lstm Total loss function

[0083] Where λ1=0.8, λ2=0.2. The Adam optimizer is used with an initial learning rate of 10. -4 The batch size is 64, and the training runs for 300 epochs. Model weights are saved after each epoch. After training, the model with the smallest validation loss is selected as the final model.

[0084] As can be seen, the PINN part constrains the neural network output through physical equations, making it conform to the laws of thermodynamics, and can provide predictions that conform to physical laws even in regions with sparse data. It captures the dynamic temperature response caused by load fluctuations, compensating for local thermal effects ignored by the lumped parameter model. This embodiment achieves a balance between steady-state accuracy and dynamic response. The thermal conductivity Gha in the physical loss term is related to the air density ρ, enabling the model to automatically adapt to heat dissipation changes in high-altitude environments without requiring explicit input of ρ.

[0085] S54. Calculate the root mean square error (RMSE) and coefficient of determination (R²) between the predicted and actual temperatures on the test set. The requirements are: RMSE ≤ 2℃ and R² ≥ 0.95.

[0086] After configuration, newly added running data is collected at preset intervals to form an incremental dataset Dinc. When the Dinc sample size reaches 200, an online update is triggered.

[0087] Add a regularization term to the loss function F i For F isher Information matrix diagonal elements, θi,old These are the parameters for the old model.

[0088] Furthermore, for each predicted output, the uncertainty interval is calculated, and Monte Carlo Dropout is used to obtain the predicted mean and variance. When the half-width of the 95% confidence interval exceeds 1.5°C, the sample is marked as high uncertainty, and active learning is triggered, with the data manually reviewed and verified. Model versions and update logs are stored in the cloud, supporting backtracking and comparative analysis.

[0089] In some embodiments, the load-temperature correlation model is evaluated on a test set before deployment, and the root mean square error (RMSE), mean absolute percentage error (MAPE), and coefficient of determination (R²) are calculated, requiring RMSE ≤ 1.8°C, MAPE ≤ 3%, and R² ≥ 0.97.

[0090] After configuration, an online learning pipeline is established: historical running data for a preset duration is collected, processed, and stored in a temporary database. When the amount of new data reaches 1000 records, model fine-tuning is triggered.

[0091] Add a regularization term to the loss function: L ewc = The expected value is obtained by calculating the squared gradient of the old model on the validation set. The EWC regularization coefficient is set to 0.5. During fine-tuning, the thermal capacity and thermal conductivity of the PINN layer are frozen, the weights of the output layer are updated, and the learning rate is reduced to 10. -5 Train for 10 epochs.

[0092] Furthermore, for each predicted output, the predicted mean and standard deviation are obtained through Monte Carlo Dropout, and the half-width of the 95% confidence interval ΔTconf = 1.96·std is calculated. When ΔTconf exceeds 1.5℃, the sample is marked as high uncertainty, and the system automatically adds this sample to the verification queue. If subsequent sensor data confirms that the sample is correct, it is added to the training set as a high-value sample.

[0093] Furthermore, the model version, update time, and performance metrics after each fine-tuning are recorded to form a model evolution log, supporting version rollback and comparative analysis.

[0094] In one embodiment of the present invention, based on S5, the current internal air pressure value is compared with the corrected make-up air pressure threshold. If it is determined that the current air pressure is lower than the corrected threshold, a make-up air control command is generated, which includes make-up air volume, make-up air duration, and fan cooperative speed. Specifically, the command includes the following: S511. Within the pressure control host, real-time data are collected on the current temperature monitoring Tcur, stator winding current I, and stator input power P of the sealed cavity. in , will (Tcur,I,P) inThis serves as the input vector for the load-temperature correlation model, triggering the model's forward inference process.

[0095] The load-temperature correlation model is based on a hybrid architecture of graph neural network and LSTM. It processes the historical load-temperature data sequence in the time dimension through LSTM layers and outputs a time-series feature vector ξ. seq ; Based on the PI-GNN layer analysis, the heat conduction topology of the motor stator, rotor, and housing is analyzed, and ξ seq With the equivalent diameter of the air duct deq and the ambient temperature T amb After splicing, the target temperature threshold T under the current operating condition is output through the fully connected layer. target .

[0096] In some embodiments, the LSTM layer captures the hysteresis and periodicity of the temperature response during load abrupt changes, ξ seq It incorporates a coupling mechanism involving load changes, heat generation rate changes, and temperature changes. The PI-GNN layer adjusts the weights of the heat conduction edges, such as the change in thermal resistance between the stator winding and the core with temperature. It explicitly models the physical process of reduced heat dissipation capacity due to decreased air density at high altitudes. Together, these factors enable the model output Ttarget to reflect load fluctuations on the heat generation side and the degradation of the physical field on the heat dissipation side.

[0097] S512, Define temperature deviation ΔT=Tcur T target If ΔT > 0, it indicates that the current temperature exceeds the safe threshold, and the make-up air pressure threshold needs to be adjusted. Establish a heat balance-pressure correlation model: motor heat generation power. ηloss is the motor efficiency, and the heat dissipation power Pcool is calculated as follows:

[0098] h(p) is the heat dissipation coefficient under air pressure p.

[0099] Aheat represents the total heat dissipation area.

[0100] At thermal equilibrium

[0101] ΔPleak is the leakage loss power, which is related to the pressure difference Δp = p. With a positive correlation between pamb and pamb, the pressure correction Δpcorr is derived as follows:

[0102] ρ0 is the air density at low altitude, R is the gas constant, Acavity is the cross-sectional area of ​​the sealed cavity, and Vcavity is the cavity volume. The corrected air supply pressure threshold pcorr is obtained by superimposing Δpcorr with the initial make-up air pressure threshold pbase: pcorr = pbase + Δpcorr.

[0103] In some embodiments, from an energy conservation perspective, the difference between heat generation and heat dissipation is compensated by leakage and make-up air. Using a pressure-dependent model of the heat dissipation coefficient, the reduction in heat dissipation capacity due to pressure is quantified as h(p), and the required pressure compensation is calculated by combining this with the temperature deviation.

[0104] S513. Based on the corrected make-up air pressure threshold pcorr, combined with the closed cavity pressure attenuation model Δp=kleak Δt, where kleak is the leakage coefficient, and the calculation method for the makeup air volume Qv required for pressure recovery is as follows: Based on the flow-pressure characteristic curve of the make-up air system. Where nfan is the fan speed, the target fan speed nfan is calculated. Based on the momentum coupling relationship between the internal circulating airflow and the makeup airflow, vtotal = vinner + kmix VSUPP, vinner, and vsupp represent the airflow velocity driven by the internal fan, respectively; vsupp represents the supplementary airflow velocity; and kmix is ​​the mixing coefficient. Combined with the internal fan speed minus airflow velocity model, vinner = kv ninner allocates the internal fan speed increment Δninner to ensure the total airflow velocity satisfies Qheat = h(pcorr). Aheat The heat dissipation requirements of ΔT.

[0105] In some embodiments, the makeup air volume calculation is anchored to the pressure recovery target and time constraint. The flow-pressure characteristic curve reflects the actual output capacity of the fan, and the momentum coupling relationship ensures efficient mixing of the makeup air and the internal circulating airflow within the sealed cavity, improving heat dissipation efficiency. By solving a multivariate system simultaneously, the parameters of makeup air volume, fan speed, and internal fan speed are coordinated.

[0106] S514. Within Δtval after the air supply control command is executed, continuously collect the pressure pval and temperature Tval data of the sealed cavity, and compare them with the command target value (p... corr T target (Comparison). Calculate the pressure recovery rate ηp: Temperature suppression rate ηT: .

[0107] If ηp < 0.8 or ηT < 0.7, the instruction execution effect is deemed unsatisfactory, triggering a secondary correction process for the instruction parameters: recalculate the leakage coefficient kleak, update the heat dissipation coefficient h(p), and repeat steps S512 - S514 to generate a new instruction; if ηp ≥ 0.8 and ηT ≥ 0.7, the current instruction parameters and operating conditions are stored in the operating condition database for online model learning.

[0108] In some embodiments, the pressure recovery rate and temperature suppression rate quantify the contribution of command execution to the pressure replenishment and cooling targets, and the execution deviation is identified through a real-time feedback loop. The secondary correction process configures the online identified leakage coefficient and thermophysical parameters, enabling the command generation model to adapt to the environment and equipment aging, solving the problem of static mismatch of control commands caused by motor performance degradation and environmental parameter fluctuations in high-altitude environments, and ensuring the long-term effectiveness and robustness of air replenishment regulation.

[0109] In one embodiment of the present invention, based on step S5, determining the target temperature value according to the current internal temperature value and the load-temperature correlation model, comparing the current internal temperature value with the target temperature value, and correcting the make-up air pressure threshold based on the comparison result specifically includes the following steps: S531. The pressure control host reads the instantaneous value I of the current load current output by the motor controller. Using the load current I as input, it outputs the corresponding target temperature value Ttarget. Since the load current and motor losses have a non-linear relationship, piecewise cubic spline interpolation is used: the range of I (0–120% of rated current) is divided into 20 intervals, and four coefficients a are stored in each interval. i , b i , c i , d i The interpolation formula is , where I i It is the left endpoint of the interval.

[0110] The calculation is performed once in each control cycle, and the resulting Ttarget is saved.

[0111] It should be noted that, due to the changes in air pressure at high altitudes altering heat dissipation conditions, the target temperature Ttarget cannot be directly used as the control target for current operation; instead, it serves as a benchmark for subsequent temperature difference calculations. Piecewise cubic spline interpolation ensures the continuity and smoothness of Ttarget's variation with I, and more accurately fits the nonlinear characteristics of motor losses than linear interpolation.

[0112] S532, the pressure control host reads the current internal temperature value Tcur and the current ambient temperature Tamb, and obtains the calculated Ttarget.

[0113] Calculate the temperature deviation ΔT = T cur- T target .

[0114] The rate of change of temperature deviation dΔT / dt is calculated using a five-point numerical differential formula:

[0115] Where Ts is the control period, ΔT(k) is the deviation at the current time, ΔT(k-1) is the deviation at the previous time, and so on, which can effectively suppress sensor noise.

[0116] The pressure correction coefficient is derived based on the physical equations of convective heat transfer.

[0117] The steady-state loss Ploss of the motor is equal to the convective heat transfer: , where h is the convective heat transfer coefficient and A is the heat dissipation surface area.

[0118] h is proportional to the 0.8 power of the air density ρ: .

[0119] And ρ is directly proportional to the air pressure p inside the sealed cavity: , where ρ0, p0, and T0 are the standard sea-level air density, air pressure, and air temperature, respectively.

[0120] Substituting the above relationships, we get... C is a constant.

[0121] Under the same load, Ploss remains approximately constant; therefore, the ratio of the target pressure ptarget to the current pressure pcur should satisfy the following condition. .

[0122] Therefore, the static pressure correction factor is defined. .

[0123] When Tcur > Ttarget, βstatic > 1, indicating that higher air pressure is needed to lower the temperature; otherwise, βstatic < 1.

[0124] Furthermore, define the differential correction term. Where τ is the motor thermal time constant, extracted in step S51, for example τ = 600s, and ΔTnorm is the normalization constant, taken as 10℃. The differential term serves to raise the threshold in advance when the temperature rises rapidly, suppressing overshoot. The final air pressure correction coefficient... The β value was limited to the range of 0.8 to 1.5 to avoid extreme adjustment.

[0125] It can be seen that the static correction coefficient βstatic directly originates from the thermal equilibrium equation and reflects the physical relationship between the temperature difference and the required air pressure. The differential correction term draws on the differential action in PID control, but the parameter τ directly uses the motor thermal time constant, making the correction amplitude match the motor thermal inertia. This enables the threshold correction to respond to dynamic changes and ensures that the threshold always remains within the safe range.

[0126] S533: The pressure control host reads the preset low-altitude reference air supply pressure threshold Pbase from the system parameter area. Multiply the air pressure correction coefficient β calculated in step S532 by Pbase to obtain the dynamic air supply pressure threshold Pth = Pbase ·β.

[0127] The pressure control host replaces the original threshold with Pth.

[0128] In the next control cycle, the updated Pth will be used for the air pressure ratio comparison; Specifically, compare the current internal air pressure value Pcur with Pth. If Pcur < Pth, a air supply instruction is generated. To ensure the smoothness of the threshold update, a first-order low-pass filter is applied each time an update is made: Pthfiltered(k) = 0.3·Pth + 0.7·Pthfiltered(k - 1), and the filtered value is used for control.

[0129] Furthermore, the pressure control host stores the corrected Pth, the corresponding β, and the timestamp in the cyclic log each time for later analysis of the motor state change trend.

[0130] It can be seen that multiplying the derived correction coefficient directly by the reference threshold realizes the mapping from the temperature deviation to the air supply threshold. Here, based on β being continuously high and the air supply frequency increasing, it indicates that the heat dissipation structure is blocked or the sealing performance has decreased, improving the reliability of the system operation.

[0131] In an embodiment of the present invention, based on step S6, the pressure control host sends an air supply control instruction to the variable-frequency fan. The variable-frequency fan extracts purified air from the external environment through the air supply pipeline and injects supplementary air into the sealed inner cavity to increase the air density. A possible embodiment will be given below to non-restrictively elaborate on its specific implementation scheme.

[0132] S611: The pressure control host analyzes the air supply control instruction, extracts the target air supply flow rate and the target inner cavity pressure value, and outputs a PWM speed control signal with the corresponding amplitude to the variable-frequency fan.

[0133] In some embodiments, the pressure control host reads the target make-up air flow rate value Qset and the target internal cavity pressure value Pset carried in the make-up air control command, and calculates the output signal according to the correspondence between the rated speed of the fan and the PWM duty cycle. The PWM duty cycle is calculated as D = (Qset / Qfa) × 100%, where Qfa is the rated maximum flow rate of the variable frequency fan. The calculated PWM signal is directly transmitted to the motor drive end of the variable frequency fan through the drive interface.

[0134] S612: The variable frequency fan starts running according to the received PWM speed control signal to match the speed, forming a stable negative pressure at the air inlet of the make-up air duct, and drawing the external ambient air into the make-up air pretreatment chamber.

[0135] In some embodiments, the impeller inside the variable frequency fan rotates at a specified speed to perform work, forming a local aerodynamic negative pressure at the air inlet of the make-up air duct. The negative pressure value satisfies ΔP = 0.5 × ρ × v², where ρ is the external ambient air density and v is the air velocity at the air inlet cross section. Under the action of this negative pressure, the external air bypasses the surface of the motor housing and directly enters the make-up air pretreatment cavity without contacting the circulating airflow inside the motor.

[0136] S613: External air entering the pre-treatment chamber is passed sequentially through a metal filter and a dehumidification zone to remove solid particulate matter and free gaseous moisture from the air.

[0137] In some embodiments, a metal filter with an 80μm pore size is used to intercept sand and suspended particles in the air. The dehumidification zone uses a microporous molecular sieve adsorption module in conjunction with a semiconductor refrigeration and condensation structure to lower the air dew point to below -10℃, and the water content of the treated air is no higher than 1.2g / m³.

[0138] S614: Controls the electromagnetic one-way valve assembly in the air supply duct to unidirectionally open when the air supply pressure is greater than the sum of the pressure in the sealed cavity and the valve opening pressure, thus blocking the return flow of air from the cavity to the outside.

[0139] In some embodiments, the valve opening pressure of the electromagnetic check valve is set to 0.4 kPa, and the physical condition for valve conduction is Psu > Pca + 0.4 kPa, where Psu is the air pressure in the air supply duct and Pca is the real-time pressure in the sealed cavity. When the variable frequency fan stops working, the valve automatically closes under the action of its own elastic component, completely blocking the communication channel between the air supply duct and the cavity.

[0140] S615: The instantaneous flow rate of the make-up air medium is detected in real time by the flow acquisition unit built into the make-up air duct, and the flow signal is transmitted back to the pressure control host to make real-time correction of the make-up air output.

[0141] In some embodiments, a turbine-type flow acquisition unit is installed in the middle section of the make-up air duct to output the instantaneous make-up air flow rate Qreal in real time. The flow signal is transmitted to the closed-loop correction module of the pressure control host. The module uses a proportional correction algorithm to adjust the PWM duty cycle, and the correction amount ΔD = Kp × (Qset) Qreal), where Kp is a calibration scaling factor between 0.8 and 1.2. The corrected PWM signal is re-output to the variable frequency fan to ensure that the internal pressure rises steadily to the target value.

[0142] In one embodiment of the present invention, based on step S6, the rotation speed of the rotor-mounted internal fan is adjusted according to the deviation ratio between the current internal air pressure value and the target air pressure to synergistically enhance the heat dissipation efficiency of the internal circulating airflow. The following will provide a possible embodiment and describe its specific implementation in a non-limiting manner.

[0143] S621: The pressure control host extracts the current internal air pressure value, target air pressure value, and current internal temperature value, calculates the air pressure deviation ratio and temperature correction coefficient, and constructs a set of basic parameter information for the speed regulation of the rotor-mounted internal fan.

[0144] In some embodiments, the pressure control unit reads the current internal air pressure value P. cur The target air pressure value P generated by S5 target and the current internal temperature value T cur ; Calculate the pressure deviation ratio δ, the formula is δ = [(P target - P cur ) / P target ] × 100%.

[0145] The value of δ is limited to 0 to 50%. When δ > 50%, it is calculated as 50% to avoid excessive deviation leading to loss of speed regulation. Define a temperature correction factor KT, KT = 1 + 0.01×(T cur - T target ), where T target The target temperature threshold output in step S5, when T cur ≤ T target At that time, KT takes the value of 1.0; Multiplying δ by KT yields the speed regulation coefficient Kreg, i.e., Kreg = δ×KT, thus constructing the basic parameter information set required for speed regulation.

[0146] S622: Based on the air pressure deviation ratio and temperature correction coefficient, the target fan speed is calculated through a segmented speed regulation formula, and the target speed is converted into a PWM speed regulation signal with corresponding amplitude and output to the fan drive module.

[0147] In some embodiments, a segmented speed regulation formula is used to calculate the target speed ntm of the rotor-mounted internal fan, and different regulation methods are set according to different ranges of the speed regulation coefficient Kreg: When 0 < Kreg ≤ 10%, ntm = nbase × (1 + Kreg), where nbase is the fan base speed; When 10% < Kreg ≤ 30%, ntm = nbase × (1.1 + 0.8 × (Kreg - 0.1)); when 30% < Kreg ≤ 50%, ntm = nbase × (1.26 + 0.4 × (Kreg - 0.3)). After calculating ntm, the target speed is converted into a PWM speed control signal with a corresponding duty cycle. The correspondence between the duty cycle D and the target speed is D = (ntm / nmax) × 100%, where nmax is the maximum rated speed of the fan. The PWM signal is transmitted to the fan drive module to drive the fan to run at the target speed.

[0148] S623: The fan speed acquisition unit captures the actual operating speed of the fan in real time, and combines it with the temperature difference data of the inlet and outlet of the heat exchange device to dynamically compensate and correct the target speed, so as to maintain the heat dissipation coordination and matching.

[0149] In some embodiments, a Hall speed acquisition unit is installed at the end of the shaft of the rotor-mounted internal fan, and temperature acquisition points are installed at the inlet and outlet of the heat exchange device, respectively, to collect the inlet and outlet temperature difference ΔTex = Tin -Tout in real time, where Tin is the inlet temperature of the heat exchange device and Tout is the outlet temperature.

[0150] When ΔTex > 8℃, the heat dissipation efficiency is deemed insufficient, and the heat dissipation compensation coefficient Kcomp = 1 + 0.05×(ΔTex - 8) is defined.

[0151] The speed deviation Δn = ntm - nreal is calculated, and the speed correction Δnreg = Δn × Kcomp is obtained by combining Kcomp. The corrected target speed ntm' = ntm + Δnreg is then converted back into a PWM signal and output to the fan drive module to achieve dynamic speed compensation. When ΔTex ≤ 8℃, Kcomp is set to 1.0, and no additional compensation is performed to ensure that the heat dissipation efficiency and air pressure regulation are matched.

[0152] In one embodiment of the present invention, based on step S7, the pressure monitoring device is periodically controlled to collect pressure decay data of the sealed cavity, calculate the pressure decay rate and compare it with a preset leakage threshold; if it is determined that the pressure decay rate exceeds the threshold or the air supply system is faulty, a load reduction operation command is triggered and a maintenance alarm is issued. The following will give a possible embodiment and describe its specific implementation in a non-limiting manner.

[0153] S711: Set the pressure decay acquisition cycle and acquisition parameters. The pressure control host triggers the pressure monitoring device according to the preset cycle to acquire the air pressure data of the sealed cavity and the internal temperature data at the corresponding time.

[0154] In some embodiments, the pressure control host has a preset pressure attenuation acquisition period, which is adjusted according to the motor load level. For light load (load rate ≤ 30%), the acquisition period is set to 30 minutes; for medium load (30% < load rate ≤ 70%), it is set to 20 minutes; and for heavy load (load rate > 70%), it is set to 10 minutes.

[0155] During data collection, the current air pressure value Pn, the data collection time tn, and the corresponding internal temperature value Tn of the sealed cavity are obtained.

[0156] S712: Preprocess the collected air pressure data, remove outliers, calculate the pressure decay rate, and obtain the actual effective decay rate.

[0157] S713: Preset graded leakage thresholds and make-up air system fault judgment indicators. Compare the calibrated actual attenuation rate with the graded thresholds, detect the variable frequency fan speed and make-up air duct pressure, and determine whether there is excessive leakage or make-up air system fault.

[0158] In some embodiments, three leakage thresholds are preset: a light leakage threshold v1 = 0.002 MPa / h, a moderate leakage threshold v2 = 0.005 MPa / h, and a severe leakage threshold v3 = 0.01 MPa / h. The thresholds are determined based on the volume of the motor's sealed inner cavity and the sealing level. The larger the volume and the higher the sealing level, the lower the threshold is set.

[0159] Optionally, the calibrated actual decay rate v is compared with the three-level threshold. If v≤v1, it is determined that the seal is normal; if v1<v≤v2, it is determined that there is a slight leak; if v2<v≤v3, it is determined that there is a moderate leak; and if v>v3, it is determined that there is a severe leak.

[0160] Optionally, the pressure control host collects the variable frequency fan speed nfan and the air pressure Ppipe in real time, sets the fan speed threshold nmin and the air pressure threshold Pmin. If nfan < nmin and the duration is ≥ the preset speed threshold, or Ppipe < Pmin and the duration is ≥ the preset speed threshold, the air supply system is determined to be faulty.

[0161] Optionally, the fault determination logic uses excessive leakage or air supply failure, and records the attenuation rate, fan speed, and duct air pressure data at the time of determination.

[0162] S714: If leakage exceeds the standard or the air supply system fails, output graded load reduction control signals according to the fault level to limit the motor output power, start the audible and visual alarm module and send the fault signal to the operation and maintenance terminal, and record the fault parameters and timestamp.

[0163] In some embodiments, graded load reduction adjusts the motor output power according to the severity of the fault. The more severe the fault, the greater the load reduction, which effectively reduces the motor's heat generation and internal air pressure requirements, and slows down the rate of pressure decay.

[0164] Furthermore, power regulation rate control can avoid the mechanical impact on the motor rotor and stator caused by sudden power drops, protect the motor structure, and ensure that faults are detected in a timely manner, thereby improving the safety and reliability of motor operation.

[0165] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0166] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those 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 invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for circulating ventilation and pressure regulation of an air-cooled motor for high-altitude conditions, characterized in that the method... include: S1. The sealed housing and the end-cap type sealed bearing are combined and installed to form a closed inner cavity surrounding the rotor assembly, and a heat exchange device is installed on the top of the closed inner cavity. S2. Connect pressure monitoring equipment and temperature monitoring equipment in the sealed cavity, and connect the output of the equipment to the pressure control host respectively to establish a multi-parameter monitoring link for synchronously collecting air pressure data and temperature data of the sealed cavity. S3. Using the rotor assembly's internal fan, the air in the sealed cavity is driven to form an internal circulating airflow. S4. The pressure monitoring device obtains the current internal air pressure value of the sealed cavity in real time, and the temperature monitoring device obtains the current internal temperature value of the sealed cavity in real time, and transmits it to the pressure control host. S5. Pre-set a load-temperature correlation model in the pressure control host; determine the target temperature value based on the current internal temperature value and the load-temperature correlation model; compare the current internal temperature value with the target temperature value; correct the make-up air pressure threshold based on the comparison result; compare the current internal air pressure value with the corrected make-up air pressure threshold; if it is determined that the current air pressure is lower than the corrected threshold, generate a make-up air control command including make-up air volume, make-up air duration, and fan cooperative speed. S6. The pressure control host sends a supplementary air control command to the variable frequency fan. The variable frequency fan draws purified air from the external environment through the supplementary air duct and injects supplementary air into the sealed inner cavity to increase the air density. According to the deviation ratio between the current internal air pressure value and the target air pressure, the speed of the rotor kit internal fan is adjusted to enhance the heat dissipation efficiency of the internal circulating airflow. S7. Periodically control the pressure monitoring equipment to collect pressure decay data of the sealed inner cavity, calculate the pressure decay rate and compare it with the preset leakage threshold. If the pressure decay rate is determined to exceed the threshold or the make-up air system is faulty, a load reduction operation command will be triggered and a maintenance alarm will be issued. Continue performing closed-loop adjustments in steps S4-S6 until the air pressure in the sealed cavity returns to a level not lower than the safe operating baseline for low altitude.

2. The method for circulating ventilation and pressure regulation of an air-cooled motor for high-altitude conditions according to claim 1, characterized in that, In S5, the construction method of the load-temperature correlation model specifically includes the following: S51: The test bench is placed in an adjustable pressure environment chamber. The chamber pressure is controlled by a vacuum pump and regulating valve. The motor temperature and environmental parameters under different altitudes and load conditions are collected through load step test, and the thermal time constant is extracted to form a structured dataset. S52: The collected data is processed by feature transformation and normalization using the air density formula and motor structure parameters, and divided into training set, validation set and test set according to the proportion. S53: Construct a hybrid model consisting of a physical information neural network and a gated loop unit, and predict motor temperature through residual constraints of thermal balance equations and data-driven training. S54: After evaluating the model accuracy on the test set, the model is periodically fine-tuned with new running data using the elastic weight consolidation method, and the prediction uncertainty is estimated by Monte Carlo Dropout.

3. The method for circulating ventilation and pressure regulation of an air-cooled motor for high-altitude conditions according to claim 2, characterized in that, Based on S5, the current internal air pressure value is compared with the corrected make-up air pressure threshold. If it is determined that the current air pressure is lower than the corrected threshold, a make-up air control command is generated, which includes make-up air volume, make-up air duration, and fan cooperative speed. Specifically, the command includes the following: S511. In the pressure control host, the current sealed cavity temperature, stator winding current and input power collected in real time are input into the load-temperature correlation model. The historical data sequence is processed by the long short-term memory network layer and the time sequence features are extracted. Then, the heat conduction topology relationship is analyzed by the physical information graph neural network layer, and the target temperature threshold under the current working condition is output. S512. Based on the deviation between the current temperature and the target temperature threshold, configure the physical relationship between the heat dissipation coefficient and the air pressure through the heat balance-pressure correlation model, calculate the pressure correction amount required to compensate for the temperature deviation, and superimpose the correction amount with the initial make-up air pressure threshold to obtain the corrected make-up air pressure threshold. S513. Calculate the required make-up air volume based on the corrected make-up air pressure threshold and the pressure attenuation model of the sealed cavity, calculate the target speed of the fan by combining the flow-pressure characteristic curve of the make-up air system, and allocate the internal fan speed increment according to the coupling relationship between the make-up air flow and the internal circulating air flow, and generate make-up air control commands that include make-up air volume, fan speed and internal fan speed. S514. During the monitoring period after the air supply control command is executed, continuously collect the pressure and temperature data of the sealed cavity, calculate the pressure recovery rate and temperature suppression rate respectively. If both reach the set threshold, store the current command parameters in the working condition database for model learning. Otherwise, trigger the recalculation of the leakage coefficient and heat dissipation coefficient and repeat the correction process to generate a new command.

4. The method for circulating ventilation and pressure regulation of an air-cooled motor for high-altitude conditions according to claim 2, characterized in that, In S5, the target temperature value is determined based on the current internal temperature value and the load-temperature correlation model. The current internal temperature value is compared with the target temperature value, and the make-up air pressure threshold is adjusted based on the comparison result. This specifically includes the following: S531, The pressure control host determines the target temperature value under the current operating condition based on the real-time collected load current through a piecewise cubic spline interpolation model. S532. The pressure control host inputs the difference between the current internal temperature and the target temperature value and the rate of change of the difference into the air pressure correction model based on the heat balance equation to calculate the make-up air pressure correction coefficient. S533: The pressure control host multiplies the preset low-altitude reference air supply pressure threshold with the air supply pressure correction coefficient and performs low-pass filtering on the product to obtain the corrected air supply pressure threshold.

5. The method for circulating ventilation and pressure regulation of an air-cooled motor for high-altitude conditions according to claim 1, characterized in that, In S6, the pressure control host sends a makeup air control command to the variable frequency fan. The variable frequency fan draws purified air from the external environment through the makeup air duct and injects makeup air into the sealed cavity to increase the air density. Specifically, this includes the following: S611: The pressure control host parses the make-up air control command, extracts the target make-up air flow rate and the target internal cavity pressure value, and outputs a PWM speed control signal with the corresponding amplitude to the variable frequency fan; S612: The variable frequency fan starts running according to the received PWM speed control signal to match the speed, forming a stable negative pressure at the air inlet of the make-up air duct, and drawing the external ambient air into the make-up air pretreatment chamber; S613: External air entering the pre-treatment chamber is passed sequentially through a metal filter and a dehumidification zone to remove solid particulate matter and free gaseous moisture from the air. S614: Controls the electromagnetic one-way valve assembly in the air supply duct to unilaterally open when the air supply pressure is greater than the sum of the pressure in the sealed cavity and the valve opening pressure, thus blocking the return flow of air from the cavity to the outside. S615: The instantaneous flow rate of the make-up air medium is detected in real time by the flow acquisition unit built into the make-up air duct, and the flow signal is transmitted back to the pressure control host to make real-time correction of the make-up air output.

6. The method for circulating ventilation and pressure regulation of an air-cooled motor for high-altitude conditions according to claim 5, characterized in that, In S6, the rotational speed of the rotor-mounted internal fan is adjusted according to the deviation ratio between the current internal air pressure and the target air pressure to enhance the heat dissipation efficiency of the internal circulating airflow. This includes the following: S621: The pressure control host extracts the current internal air pressure value, target air pressure value and current internal temperature value, calculates the air pressure deviation ratio and temperature correction coefficient, and constructs a set of basic parameter information for the speed regulation of the rotor-mounted internal fan. S622: Based on the air pressure deviation ratio and temperature correction coefficient, the target fan speed is calculated through a segmented speed regulation formula, and the target speed is converted into a PWM speed regulation signal with corresponding amplitude and output to the fan drive module; S623: The fan speed acquisition unit captures the actual operating speed of the fan in real time, and combines it with the temperature difference data of the inlet and outlet of the heat exchange device to dynamically compensate and correct the target speed, so as to maintain the heat dissipation coordination and matching.

7. The method for circulating ventilation and pressure regulation of an air-cooled motor for high-altitude conditions according to claim 6, characterized in that, The method for calculating the pressure deviation ratio and temperature correction factor is as follows: The pressure deviation ratio δ is calculated using the formula δ = [(Ptarget - Pcur) / Ptarget] × 100%; The method also employs a segmented speed regulation formula to calculate the target speed ntm of the rotor-mounted internal fan, and sets different regulation methods according to different ranges of the speed regulation coefficient Kreg: When 0 < Kreg ≤ 10%, ntm = nbase × (1 + Kreg), where nbase is the fan base speed; When 10% < Kreg ≤ 30%, ntm = nbase × (1.1 + 0.8 × (Kreg - 0.1)); When 30% < Kreg ≤ 50%, ntm = nbase × (1.26 + 0.4 × (Kreg - 0.3)); After calculating ntm, the target speed is converted into a PWM speed control signal with a corresponding duty cycle. The correspondence between the duty cycle D and the target speed is D = (ntm / nmax) × 100%, where nmax is the maximum rated speed of the fan. The PWM signal is transmitted to the fan drive module to drive the fan to run at the target speed.

8. The method for circulating ventilation and pressure regulation of an air-cooled motor for high-altitude conditions according to claim 1, characterized in that, S1 specifically includes the following: S111: Pre-process the sealing housing, process the sealing structure of the housing mating surface and perform surface passivation treatment, locate the installation benchmark of the rotor assembly, hoist the rotor assembly into the preset position in the housing and complete the initial fixation. S112: The end cap type sealed bearing is press-fitted to the bearing seats at both ends of the sealing housing using an interference fit method, a flexible sealing bushing is added to fill the assembly gap, and a stepped locating pin is used to fix the bearing end cap to ensure the coaxiality of the bearing and the rotor assembly. S113: Perform segmented inflation and pressure testing on the assembled sealing structure. Inflate the sealing housing with dry nitrogen to the preset pressure, test the leakage of each sealing surface, and repair the leaks with sealant. S114: Machining a flange interface on the pre-set mounting surface at the top of the sealed inner cavity, connecting the heat exchange device to the flange interface with bolts, installing a high-temperature resistant flexible sealing gasket, and installing an arc-shaped guide plate at the inlet of the heat exchange device; S115: Starting the temporary drive device to drive the rotor assembly to rotate at low speed, detecting the internal airflow direction and velocity, adjusting the angle of the guide plate to direct the airflow to the heat exchange device, verifying the sealing performance and the installation firmness of the heat exchange device, and completing the solidification of the basic ventilation structure.

9. The method for circulating ventilation and pressure regulation of an air-cooled motor for high-altitude conditions according to claim 1, characterized in that, S7 specifically includes the following: S711: Set the pressure decay acquisition cycle and acquisition parameters. The pressure control host triggers the pressure monitoring device according to the preset cycle to acquire the air pressure data of the sealed cavity and the internal temperature data at the corresponding time. S712: Preprocesses the collected air pressure data to obtain the actual effective attenuation rate; S713: Preset graded leakage threshold and make-up air system fault judgment index. Compare the calibrated actual attenuation rate with the graded threshold, detect the variable frequency fan speed and make-up air duct pressure, and determine whether there is excessive leakage or make-up air system fault. S714: If leakage exceeds the standard or the air supply system fails, output a graded load reduction control signal according to the fault level to limit the motor output power, activate the audible and visual alarm module and send a fault signal to the operation and maintenance terminal.

10. A circulating ventilation and pressure regulation system for an air-cooled motor used in high-altitude conditions, characterized in that, The system is used to implement the steps of the method for circulating ventilation and pressure regulation of an air-cooled motor for high-altitude conditions as described in any one of claims 1 to 9; The system includes: a sealed housing, end-cap sealed bearings, a heat exchange device, pressure monitoring equipment, a pressure control unit, a variable frequency fan, and a makeup air duct; The sealed housing is equipped with stepped bearing mounting holes and heat exchanger mounting windows; The heat exchanger mounting window is surrounded by a sealing flange face and threaded blind holes; the inner wall of the stepped bearing mounting hole is provided with an annular sealing groove. An O-ring fluororubber seal is provided in the annular sealing groove on the inner wall of the stepped bearing mounting hole. The outer ring of the end cap type sealed bearing is installed in the stepped bearing mounting hole of the sealed housing. The heat exchange device includes: a microchannel heat pipe array and a rectangular metal frame; The rectangular metal frame is connected to the sealing flange face of the heat exchanger mounting window on the top of the sealed housing. End cap type sealed bearings include: bearing outer ring, bearing inner ring, and O-ring fluororubber seal; The outer ring of the bearing is installed in the stepped bearing mounting hole of the sealed housing, and the inner ring of the bearing is connected to the rotor assembly; the O-ring fluororubber seal is embedded in the annular sealing groove of the stepped bearing mounting hole; the microchannel heat pipe array is composed of multiple parallel microchannel heat pipes, filled with acetone working fluid, and the heat pipes are divided into evaporation section and condensation section. The pressure monitoring equipment is connected to the pressure control host, which is connected to the variable frequency fan. The variable frequency fan is connected to the inside of the sealed housing through the air supply pipe.

Images