Ultrahigh-speed permanent magnet electric drive system heat dissipation cooperation method based on magnetothermal coupling optimization

Abstract

The invention relates to the technical field of intelligent control, and particularly discloses an ultrahigh-speed permanent magnet electric drive system heat dissipation cooperation method based on magnetothermal coupling optimization, and the method comprises the steps: respectively extracting a mechanical thermal deformation feature value and an electromagnetic abnormal noise feature value through collecting vibration spectrum and sound spectrum data in a motor operation process in real time; constructing a comprehensive thermal state feature vector, and inputting the comprehensive thermal state feature vector into an intelligent control model trained based on a random forest algorithm to realize dynamic evaluation and grading prediction of the thermal risk grade of the motor; according to the method, the multi-source information of mechanical structure deformation and electromagnetic performance degradation is fused, the sensitivity and accuracy of a thermal management system are improved, and the reliability of the thermal management system is improved. And high-efficiency energy-saving cooling control on the premise of ensuring safe operation of the motor is realized.

Description

Technical Field

[0001] The present invention relates to the field of intelligent control technology, and in particular to a heat dissipation coordination method for an ultra-high-speed permanent magnet electric drive system based on magnetothermal coupling optimization. Background Art

[0002] In high-performance motor drive systems, especially in the field of ultra-high-speed permanent magnet synchronous motor applications, with the continuous improvement of power density and operating speed, the problem of internal heating of the motor has become increasingly prominent, becoming a key factor restricting its long-term stable operation and performance. Traditional motor thermal management methods mainly rely on temperature sensors to collect the surface temperature of the stator winding or housing, and combine empirical thresholds to perform open-loop or fixed-proportion control of the cooling system operation. However, this method has problems such as response lag, single evaluation dimension, and inability to reflect the thermal deformation of the internal structure of the motor and the degradation of electromagnetic performance. It is difficult to meet the dual requirements of thermal risk prediction accuracy and real-time control under complex working conditions. In addition, existing cooling strategies are mostly passive heat dissipation designs, lacking active control mechanisms based on the actual state of the motor, resulting in high energy consumption and low efficiency, which limits the improvement of the overall performance of the system.

[0003] The existing technology has the following deficiencies:

[0004] The present invention solves the unconventional technical problems of "single perception, lagging evaluation, and extensive regulation" in existing motor thermal management systems. Specifically, traditional methods rely solely on temperature signals to judge the thermal state, ignoring the early abnormal signals released by the mechanical structure deformation and electromagnetic performance changes caused by the temperature rise of the motor, resulting in delayed thermal risk identification or even misjudgment; at the same time, the cooling system control strategy lacks a dynamic response mechanism to the actual thermal risk level of the motor, resulting in irrational cooling capacity allocation, affecting the energy efficiency and stability of the system. The present invention introduces a multi-source information fusion analysis mechanism of vibration spectrum and sound spectrum, and combines it with an intelligent control model to realize a hierarchical cooling strategy driven by thermal risk scoring, constructing a closed-loop system from state perception, risk prediction to active regulation, significantly improving the intelligence level and adaptability of the thermal management system, and filling the current technical gap in the field of motor thermal control in terms of multi-physical field coupling evaluation and adaptive regulation. Summary of the Invention

[0005] The purpose of the present invention is to provide a heat dissipation coordination method for an ultra-high-speed permanent magnet electric drive system based on magnetothermal coupling optimization to solve the problems in the above background.

[0006] The purpose of the present invention can be achieved through the following technical solutions:

[0007] The heat dissipation coordination method of an ultra-high-speed permanent magnet electric drive system based on magnetothermal coupling optimization includes the following steps:

[0008] S1: During the operation of the motor, its vibration spectrum and sound spectrum data are monitored in real time;

[0009] S2: Analyze the vibration spectrum data and calculate the mechanical thermal deformation characteristic value based on the degree of change in the vibration spectrum to evaluate whether there is structural deformation inside the motor caused by motor overheating;

[0010] S3: Analyze the sound spectrum data and calculate the electromagnetic abnormal noise characteristic value based on the abnormality of the sound spectrum to determine whether there is electromagnetic performance degradation caused by the increase in motor temperature;

[0011] S4: The mechanical thermal deformation characteristic value and the electromagnetic abnormal noise characteristic value are constructed into a comprehensive thermal state characteristic vector, which is input into the intelligent control model for fusion analysis to determine the current thermal risk level of the motor.

[0012] S5: Dynamically adjusts the coolant flow rate according to the thermal risk level, regulates the cooling system output, achieves closed-loop optimization control of motor temperature rise, and improves system operating efficiency and thermal stability.

[0013] As a further solution of the present invention, the step of evaluating whether there is structural deformation inside the motor caused by motor overheating specifically includes:

[0014] During the operation of the motor, the vibration spectrum data of the motor is collected in real time. According to the degree of change of the vibration spectrum, the mechanical thermal deformation characteristic value is calculated to determine whether the mechanical thermal deformation characteristic value of the motor is greater than or equal to a preset threshold. If so, there is structural deformation inside the motor caused by motor overheating. If not, there is no structural deformation caused by motor overheating.

[0015] As a further solution of the present invention: the process of obtaining the mechanical thermal deformation characteristic value is:

[0016] Discrete sampling is performed on the vibration spectrum data of the collected motor to obtain a discrete time series;

[0017] Perform fast Fourier transform on the discrete time series of vibration spectrum data to obtain frequency domain signal;

[0018] Calculate the energy distribution of each frequency component;

[0019] Set the sensitive frequency range related to the motor structural characteristics and convert it into the corresponding spectrum index interval;

[0020] Calculate the degree of energy change within the sensitive frequency range;

[0021] The mechanical thermal deformation characteristic value is obtained by calculating the ratio of the energy change degree within the sensitive frequency range to the sum of the reference energy levels within the sensitive frequency range.

[0022] As a further solution of the present invention: the determination of whether the electromagnetic performance degradation state exists due to the increase in motor temperature specifically includes:

[0023] During the operation of the motor, the sound spectrum data of the motor is collected in real time, and the sound spectrum data is analyzed. According to the abnormality degree of the sound spectrum, the characteristic value of the electromagnetic abnormal noise is calculated to determine whether the characteristic value of the electromagnetic abnormal noise is greater than or equal to a preset threshold. If so, there is a state of electromagnetic performance degradation due to the increase in motor temperature. If not, there is no state of electromagnetic performance degradation due to the increase in motor temperature.

[0024] As a further solution of the present invention: the process of obtaining the characteristic value of the electromagnetic abnormal noise is:

[0025] During the operation of the motor, the sound spectrum data of the motor is collected in real time according to the time series, and the data is constructed into a sound spectrum time series dataset;

[0026] De-noising and normalization of sound spectrum time series data to ensure data accuracy;

[0027] Apply Haar wavelet transform to the processed sound spectrum time series data and perform J-layer decomposition to obtain the approximate coefficients and detail coefficients of each layer;

[0028] Calculate the energy of detail coefficients and approximation coefficients of each layer;

[0029] The ratio of the energy of all layer detail coefficients to the total energy is calculated to obtain the characteristic value of electromagnetic abnormal noise.

[0030] As a further solution of the present invention, the mechanical thermal deformation characteristic value and the electromagnetic abnormal noise characteristic value are constructed into a comprehensive thermal state characteristic vector, and input into the intelligent control model for fusion analysis, specifically including:

[0031] During the operation of the motor, the mechanical thermal deformation eigenvalues ​​and electromagnetic abnormal noise eigenvalues ​​of the motor are obtained and constructed into a comprehensive thermal state eigenvector as the input of the intelligent control model to minimize the error between the predicted thermal risk score of the motor and the actual thermal risk score of the motor. As the training target of the intelligent control model, the intelligent control model is trained and the motor thermal risk score is output based on the trained intelligent control model. The intelligent control model is a random forest model.

[0032] As a further solution of the present invention: the training process of the intelligent control model is:

[0033] The comprehensive thermal state feature vectors and thermal risk scores of the motor in multiple historical operating cycles are obtained as the training data set. The random forest model is used to train the training data set. Subsets of the training data set and feature subsets are randomly selected to construct multiple decision trees. Each decision tree independently learns the nonlinear mapping relationship between input features and thermal risk scores. During the model training process, the objective function is to minimize the mean square error between the predicted thermal risk score and the actual thermal risk score of the motor. The split node selection and leaf node output value of each decision tree are iteratively optimized until the preset maximum number of iterations is met. Finally, the output of the trained random forest model is the thermal risk score of the motor.

[0034] As a further solution of the present invention: the determining of the current thermal risk level of the motor specifically includes:

[0035] According to the current thermal risk score of the motor output by the intelligent control model, it is determined whether the thermal risk score is greater than or equal to the preset first threshold. If so, it is recorded as a high thermal risk level. If not, it is determined whether the thermal risk score is less than or equal to the preset second threshold. If so, it is recorded as a low thermal risk level. If not, it is recorded as a medium thermal risk level.

[0036] As a further solution of the present invention: the dynamic adjustment of the coolant flow rate according to the thermal risk level specifically includes:

[0037] When the intelligent control model outputs that the current thermal risk score of the motor is a high thermal risk level, the control system starts the maximum cooling power mode and increases the coolant flow rate to the set upper limit value Q by increasing the speed of the variable frequency pump. max ;

[0038] If the thermal risk score is a medium thermal risk level, a closed-loop feedback control strategy is used to dynamically adjust the coolant flow rate based on the deviation between the score and the baseline value;

[0039] If the thermal risk level is low, maintain the minimum cooling flow Q min To achieve coordinated optimization of energy-saving operation and thermal management.

[0040] Beneficial effects of the present invention:

[0041] (1) The present invention deeply integrates the multi-source heterogeneous data of vibration spectrum and sound spectrum collected in real time during the operation of the motor, and constructs a composite thermal state assessment system based on the linkage analysis of mechanical structure state and electromagnetic performance change. Specifically, a high-sensitivity three-axis acceleration sensor and a digital microphone array are used to synchronously obtain the motor's broadband vibration signal and full-band acoustic signal, and respectively extract the mechanical thermal deformation characteristic value reflecting the thermal-induced structural deformation and the electromagnetic abnormal noise characteristic value representing the degradation of electromagnetic performance, and then fuse them to construct a comprehensive thermal state characteristic vector with clear physical meaning and strong interpretability. This characteristic vector not only effectively captures the microstructure changes and electromagnetic field distortion information caused by temperature rise in the motor under complex operating conditions, but also performs nonlinear mapping learning on the characteristic vector through an intelligent control model, thereby achieving high-precision dynamic prediction and graded identification of the motor's thermal risk state. Compared with the traditional single criterion that only relies on temperature monitoring, this method significantly improves the early abnormality recognition capability and response efficiency of the thermal management system, and provides a solid data support and theoretical basis for the realization of intelligent and adaptive motor thermal control strategies.

[0042] (2) Based on the thermal risk score output by the intelligent control model, the present invention constructs a hierarchical dynamic cooling strategy with adaptive adjustment capability, which achieves a precise match between the cooling system output and the thermal state of the motor. Specifically, the system divides the motor operating state into three risk levels: high, medium, and low according to the thermal risk score, and corresponds to different cooling control modes: in the high thermal risk state, the coolant flow rate is quickly increased to the set upper limit to ensure that the motor is effectively cooled in a short time; in the medium risk state, a proportional feedback control mechanism is introduced to dynamically adjust the cooling intensity according to the score deviation to achieve a smooth transition and refined temperature control; in the low risk state, the minimum cooling flow rate is maintained or the energy-saving standby mode is entered, thereby effectively reducing unnecessary cooling energy consumption. This strategy not only ensures the operating safety of the motor under extreme working conditions, but also significantly improves the overall energy efficiency and thermal stability of the system by providing cooling on demand. It is particularly suitable for ultra-high-speed permanent magnet electric drive systems with high performance, high power, and variable load operation, and has good engineering application prospects and promotion value. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] The present invention will be further described below with reference to the accompanying drawings.

[0044] Figure 1 It is a flow chart of the heat dissipation coordination method of the ultra-high-speed permanent magnet electric drive system based on magnetothermal coupling optimization of the present invention. DETAILED DESCRIPTION

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

[0046] See also Figure 1 As shown, the present invention is a heat dissipation coordination method for an ultra-high-speed permanent magnet electric drive system based on magnetothermal coupling optimization, comprising the following steps:

[0047] S1: During the operation of the motor, its vibration spectrum and sound spectrum data are monitored in real time;

[0048] S2: Analyze the vibration spectrum data and calculate the mechanical thermal deformation characteristic value based on the degree of change in the vibration spectrum to evaluate whether there is structural deformation inside the motor caused by motor overheating;

[0049] S3: Analyze the sound spectrum data and calculate the electromagnetic abnormal noise characteristic value based on the abnormality of the sound spectrum to determine whether there is electromagnetic performance degradation caused by the increase in motor temperature;

[0050] S4: The mechanical thermal deformation characteristic value and the electromagnetic abnormal noise characteristic value are constructed into a comprehensive thermal state characteristic vector, which is input into the intelligent control model for fusion analysis to determine the current thermal risk level of the motor.

[0051] S5: Dynamically adjusts the coolant flow rate according to the thermal risk level, regulates the cooling system output, achieves closed-loop optimization control of motor temperature rise, and improves system operating efficiency and thermal stability.

[0052] In S1, during the operation of the motor, its vibration spectrum and sound spectrum data are monitored in real time, including:

[0053] During motor operation, a multi-channel sensor system deployed at key locations of the motor is used to collect the motor's vibration and sound signals. Vibration signals are collected using a high-sensitivity three-axis acceleration sensor installed at key structural locations such as the motor housing bearing seat and stator bracket. The sampling frequency is not less than 20kHz to ensure that the wide-band vibration characteristics of the motor under high-speed operation are captured. Sound signals are collected using a digital microphone array, which is arranged around the motor air gap and in the cooling duct outlet area. The sampling frequency is set to 44.1kHz to obtain the full-band acoustic characteristics of the motor's electromagnetic noise and the cooling system's aerodynamic noise.

[0054] The collected vibration and sound signals are synchronously processed by the data acquisition module, and the embedded signal conditioning circuit completes filtering, amplification and analog-to-digital conversion operations to generate raw signal data in digital form; then, the on-site digital signal processor is used to perform real-time spectrum analysis on the signal, extract the time-frequency domain characteristics of the signal, generate corresponding vibration spectrum data and sound spectrum data, and transmit them to the central control unit as the basic input for the subsequent calculation of mechanical thermal deformation characteristic values ​​and electromagnetic abnormal noise characteristic values.

[0055] In S2, the vibration spectrum data is analyzed and the mechanical thermal deformation characteristic value is calculated based on the degree of change in the vibration spectrum. This is used to assess whether there is structural deformation inside the motor caused by motor overheating. Specifically, the following are included:

[0056] During the operation of the motor, the vibration spectrum data of the motor is collected in real time. According to the degree of change of the vibration spectrum, the mechanical thermal deformation characteristic value is calculated to determine whether the mechanical thermal deformation characteristic value of the motor is greater than or equal to a preset threshold. If so, there is structural deformation inside the motor caused by motor overheating. If not, there is no structural deformation caused by motor overheating.

[0057] The process of obtaining the mechanical thermal deformation characteristic value is as follows:

[0058] Discrete sampling is performed on the vibration spectrum data of the collected motor to obtain a discrete time series;

[0059] Perform fast Fourier transform on the discrete time series of vibration spectrum data to obtain frequency domain signal;

[0060] Calculate the energy distribution of each frequency component. The calculation expression is:

[0061] E[k]=|X[k]| 2 ;

[0062] Where k represents the number of frequency components, E[k] represents the energy of the kth frequency component, and X[k] represents the value of the frequency domain signal obtained by fast Fourier transform at the kth frequency component;

[0063] Set the sensitive frequency range [f low , f high ] and converted into the corresponding spectrum index interval [k low , k high ];

[0064] Among them, f low Indicates the lower limit frequency of the sensitive frequency range, f high Indicates the upper limit frequency of the sensitive frequency range, k low Indicates that f low The corresponding spectrum index value, khigh Indicates that f high The corresponding spectrum index value;

[0065] Calculate the degree of energy change within the sensitive frequency range. The calculation expression is:

[0066]

[0067] Where D represents the degree of energy change within the sensitive frequency range, and E0[k] represents the reference energy level of the same frequency component under normal working conditions.

[0068] The mechanical thermal deformation characteristic value is obtained by calculating the ratio of the energy change degree within the sensitive frequency range to the sum of the reference energy levels within the sensitive frequency range. The calculation expression is:

[0069]

[0070] Where, F mtd Represents the mechanical thermal deformation characteristic value.

[0071] It should be noted that: the present invention analyzes the energy changes of the motor vibration spectrum in the sensitive frequency range, combines fast Fourier transform with energy ratio calculation, and extracts the mechanical thermal deformation characteristic value that can accurately reflect the deformation of the motor structure due to thermal effects; this method fully considers the dynamic evolution law of the motor structure resonance characteristics with the increase of temperature, and uses the frequency domain energy offset trend to identify the occurrence of thermally induced deformation. Compared with the traditional detection method that relies on a single temperature sensor, it has higher sensitivity and early fault identification capability; at the same time, based on the set sensitive frequency range and benchmark energy level, adaptive comparison is performed to effectively eliminate noise interference under normal operating conditions, thereby improving the stability of the characteristic value and the accuracy of judgment, thereby realizing non-invasive, online, quantitative evaluation of the internal thermal deformation state of the motor, and providing a reliable basis for subsequent thermal risk prediction and cooling control.

[0072] In S3, the sound spectrum data is analyzed, and based on the degree of abnormality in the sound spectrum, the characteristic value of electromagnetic abnormal noise is calculated to determine whether there is electromagnetic performance degradation due to the increase in motor temperature. Specifically, the following are performed:

[0073] During the operation of the motor, the sound spectrum data of the motor is collected in real time, and the sound spectrum data is analyzed. According to the abnormality degree of the sound spectrum, the characteristic value of the electromagnetic abnormal noise is calculated to determine whether the characteristic value of the electromagnetic abnormal noise is greater than or equal to a preset threshold. If so, there is a state of electromagnetic performance degradation due to the increase in motor temperature. If not, there is no state of electromagnetic performance degradation due to the increase in motor temperature.

[0074] The process of obtaining the characteristic value of the electromagnetic abnormal noise is as follows:

[0075] During the operation of the motor, the sound spectrum data of the motor is collected in real time according to the time series, and the data is constructed into a sound spectrum time series dataset;

[0076] De-noising and normalization of sound spectrum time series data to ensure data accuracy;

[0077] Apply Haar wavelet transform to the processed sound spectrum time series data and perform J-layer decomposition to obtain the approximate coefficients and detail coefficients of each layer;

[0078] Calculate the energy of each layer detail coefficient, the calculation expression is:

[0079]

[0080] Where, E j Represents the energy of the detail coefficient of the j-th layer decomposition, A j [a] represents the value of the detail coefficient of the j-th layer decomposition, a represents the value of the detail coefficient, M j Represents the number of detail coefficients after the j-th layer decomposition of Haar wavelet transform;

[0081] The energy of all layer detail coefficients is calculated relative to the total energy to obtain the characteristic value of electromagnetic abnormal noise;

[0082] The total energy is the sum of the energy of the detail coefficients of all layers and the energy of the approximate coefficients. The calculation method of the energy of the approximate coefficients is the same as that of the detail coefficients, which will not be repeated here.

[0083] It should be noted that: the present invention collects the sound spectrum data when the motor is running, and combines the Haar wavelet transform to perform multi-scale decomposition of the sound signal, and extracts the detail coefficient energy ratio as the electromagnetic abnormal noise characteristic value, so as to achieve accurate judgment of the electromagnetic performance degradation state caused by temperature rise of the motor; this method utilizes the advantages of wavelet transform in time-frequency localization analysis, and can effectively identify the high-frequency abnormal noise components caused by electromagnetic field distortion. Compared with the traditional evaluation method based on overall sound pressure level, it has stronger feature sensitivity and fault type identification ability; at the same time, by denoising, normalizing and multi-layer decomposition of the sound signal, the robustness and consistency of feature extraction are improved, so that the electromagnetic abnormal noise characteristic value can still maintain good discrimination performance under complex working conditions; this technology realizes non-contact electromagnetic state monitoring based on acoustic characteristics, and provides a new technical path and data support for early warning of motor thermal risks.

[0084] In S4, the mechanical thermal deformation characteristic value and the electromagnetic abnormal noise characteristic value are constructed into a comprehensive thermal state characteristic vector, which is input into the intelligent control model for fusion analysis to determine the current thermal risk level of the motor, including:

[0085] During the operation of the motor, the mechanical thermal deformation eigenvalues ​​and electromagnetic abnormal noise eigenvalues ​​of the motor are obtained and constructed into a comprehensive thermal state eigenvector as the input of the intelligent control model to minimize the error between the predicted thermal risk score of the motor and the actual thermal risk score of the motor. As the training target of the intelligent control model, the intelligent control model is trained and the motor thermal risk score is output based on the trained intelligent control model. The intelligent control model is a random forest model.

[0086] The training process of the intelligent control model is:

[0087] The comprehensive thermal state feature vectors and thermal risk scores of the motor in multiple historical operating cycles are obtained as the training data set. The random forest model is used to train the training data set. Subsets of the training data set and feature subsets are randomly selected to construct multiple decision trees. Each decision tree independently learns the nonlinear mapping relationship between input features and thermal risk scores. During the model training process, the objective function is to minimize the mean square error between the predicted thermal risk score and the actual thermal risk score of the motor. The split node selection and leaf node output value of each decision tree are iteratively optimized until the preset maximum number of iterations is met. Finally, the output of the trained random forest model is the thermal risk score of the motor.

[0088] According to the current thermal risk score of the motor output by the intelligent control model, it is determined whether the thermal risk score is greater than or equal to the preset first threshold. If so, it is recorded as a high thermal risk level. If not, it is determined whether the thermal risk score is less than or equal to the preset second threshold. If so, it is recorded as a low thermal risk level. If not, it is recorded as a medium thermal risk level.

[0089] In S5, the coolant flow rate is dynamically adjusted according to the thermal risk level, and the cooling system output is regulated to achieve closed-loop optimization control of motor temperature rise, improving system operating efficiency and thermal stability. Specifically,

[0090] When the intelligent control model outputs that the current thermal risk score of the motor is a high thermal risk level, the control system starts the maximum cooling power mode and increases the coolant flow rate to the set upper limit value Q by increasing the speed of the variable frequency pump. max ;

[0091] If the thermal risk score is medium, a closed-loop feedback control strategy is used to dynamically adjust the coolant flow rate based on the deviation between the score and the baseline value. The calculation expression is:

[0092] Q=Q base +k p ·(T score -T base );

[0093] Among them, Q represents the specific value of the adjusted coolant flow rate, Q baseIndicates the preset reference flow rate, k p Represents the proportional gain coefficient, T score Score the current heat risk, T base A second threshold is preset;

[0094] If the thermal risk level is low, maintain the minimum cooling flow Q min To achieve coordinated optimization of energy-saving operation and thermal management.

[0095] It should be noted that: the present invention realizes refined, closed-loop optimization control of the motor temperature rise process by introducing a multi-level dynamic cooling control strategy based on thermal risk scoring; this method adaptively adjusts the coolant flow according to different thermal risk levels, especially quickly improves the cooling capacity at high thermal risks, adopts proportional feedback control to achieve smooth regulation at medium risks, and reduces cooling energy consumption at low risks, thereby improving the overall energy efficiency of the system while ensuring the safe operation of the motor; innovatively links the intelligent assessment results with the cooling system to construct a complete closed-loop link from state perception, risk prediction to active regulation, breaking through the limitations of traditional fixed threshold or open-loop control methods with delayed response and rough adjustment, and significantly improving the thermal stability and operational reliability of the electric drive system.

[0096] Working Principle of the Invention: Based on the magnetothermal coupling mechanism, the present invention constructs a multimodal sensing system that integrates mechanical structural state and electromagnetic performance changes. By collecting vibration and acoustic spectrum data in real time during motor operation, mechanical thermal deformation eigenvalues ​​reflecting heat-induced structural deformation and electromagnetic abnormal noise eigenvalues ​​reflecting electromagnetic performance degradation are extracted respectively. These eigenvalues ​​are then integrated to form a comprehensive thermal state eigenvector, which serves as input to drive an intelligent assessment model trained based on a random forest algorithm, achieving high-precision, dynamic scoring and grading of motor thermal risks. On this basis, the system adaptively regulates coolant flow output based on the thermal risk level, employing a hierarchical control strategy to optimize energy consumption while ensuring safety, forming a closed-loop thermal management system. This method breaks through the traditional thermal control model that relies on a single temperature signal or empirical threshold, achieving a transition from passive heat dissipation to active predictive control. This significantly improves system response speed, control accuracy, and energy efficiency, providing key technical support for the stable and efficient operation of high-performance electric drive systems under complex operating conditions.

[0097] The above formulas are all dimensionless and numerical calculations. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters in the formulas are set by technicians in this field according to actual conditions.

[0098] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer program are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center via wired or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state drive.

[0099] It should be understood that the term "and / or" as used herein simply describes a relationship between associated objects, indicating that three possible relationships exist. For example, "A and / or B" can represent: A alone, A and B together, or B alone. A and B can be singular or plural. Furthermore, the character " / " as used herein generally indicates an "or" relationship between the associated objects, but it may also indicate an "and / or" relationship. For specific understanding, please refer to the context.

[0100] It should be understood that in the various embodiments of the present application, the size of the sequence number of each process does not mean 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 embodiment of the present application. The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equal changes and improvements made in accordance with the scope of the application of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A heat dissipation coordination method for an ultra-high-speed permanent magnet electric drive system based on magnetothermal coupling optimization, characterized in that: The following steps are involved: S1: During the operation of the motor, its vibration spectrum and sound spectrum data are monitored in real time; S2: Analyze the vibration spectrum data and calculate the mechanical thermal deformation characteristic value based on the degree of change in the vibration spectrum to evaluate whether there is structural deformation inside the motor caused by motor overheating; S3: Analyze the sound spectrum data and calculate the electromagnetic abnormal noise characteristic value based on the abnormality of the sound spectrum to determine whether there is electromagnetic performance degradation caused by the increase in motor temperature; S4: The mechanical thermal deformation characteristic value and the electromagnetic abnormal noise characteristic value are constructed into a comprehensive thermal state characteristic vector, which is input into the intelligent control model for fusion analysis to determine the current thermal risk level of the motor. S5: Dynamically adjusts the coolant flow rate according to the thermal risk level, regulates the cooling system output, achieves closed-loop optimization control of motor temperature rise, and improves system operating efficiency and thermal stability.

2. The heat dissipation coordination method of an ultra-high-speed permanent magnet electric drive system based on magnetothermal coupling optimization according to claim 1 is characterized in that: The evaluation of whether there is structural deformation inside the motor due to motor overheating specifically includes: During the operation of the motor, the vibration spectrum data of the motor is collected in real time. According to the degree of change of the vibration spectrum, the mechanical thermal deformation characteristic value is calculated to determine whether the mechanical thermal deformation characteristic value of the motor is greater than or equal to a preset threshold. If so, there is structural deformation inside the motor caused by motor overheating. If not, there is no structural deformation caused by motor overheating.

3. The heat dissipation coordination method of an ultra-high-speed permanent magnet electric drive system based on magnetothermal coupling optimization according to claim 2 is characterized in that: The process of obtaining the mechanical thermal deformation characteristic value is as follows: Discrete sampling is performed on the vibration spectrum data of the collected motor to obtain a discrete time series; Perform fast Fourier transform on the discrete time series of vibration spectrum data to obtain frequency domain signal; Calculate the energy distribution of each frequency component; Set the sensitive frequency range related to the motor structural characteristics and convert it into the corresponding spectrum index interval; Calculate the degree of energy change within the sensitive frequency range; The mechanical thermal deformation characteristic value is obtained by calculating the ratio of the energy change degree within the sensitive frequency range to the sum of the reference energy levels within the sensitive frequency range.

4. The heat dissipation coordination method for an ultra-high-speed permanent magnet electric drive system based on magnetothermal coupling optimization according to claim 1 is characterized in that: The determination of whether the electromagnetic performance degradation state exists due to the increase in motor temperature specifically includes: During the operation of the motor, the sound spectrum data of the motor is collected in real time, and the sound spectrum data is analyzed. According to the abnormality degree of the sound spectrum, the characteristic value of the electromagnetic abnormal noise is calculated to determine whether the characteristic value of the electromagnetic abnormal noise is greater than or equal to a preset threshold. If so, there is a state of electromagnetic performance degradation due to the increase in motor temperature. If not, there is no state of electromagnetic performance degradation due to the increase in motor temperature.

5. The heat dissipation coordination method of an ultra-high-speed permanent magnet electric drive system based on magnetothermal coupling optimization according to claim 4 is characterized in that: The process of obtaining the characteristic value of the electromagnetic abnormal noise is as follows: During the operation of the motor, the sound spectrum data of the motor is collected in real time according to the time series, and the data is constructed into a sound spectrum time series dataset; De-noising and normalization of sound spectrum time series data to ensure data accuracy; Apply Haar wavelet transform to the processed sound spectrum time series data and perform J-layer decomposition to obtain the approximate coefficients and detail coefficients of each layer; Calculate the energy of detail coefficients and approximation coefficients of each layer; The ratio of the energy of all layer detail coefficients to the total energy is calculated to obtain the characteristic value of electromagnetic abnormal noise.

6. The heat dissipation coordination method of an ultra-high-speed permanent magnet electric drive system based on magnetothermal coupling optimization according to claim 1 is characterized in that: The mechanical thermal deformation characteristic value and the electromagnetic abnormal noise characteristic value are constructed into a comprehensive thermal state characteristic vector, which is input into the intelligent control model for fusion analysis, specifically including: During the operation of the motor, the mechanical thermal deformation eigenvalues ​​and electromagnetic abnormal noise eigenvalues ​​of the motor are obtained and constructed into a comprehensive thermal state eigenvector as the input of the intelligent control model to minimize the error between the predicted thermal risk score of the motor and the actual thermal risk score of the motor. As the training target of the intelligent control model, the intelligent control model is trained and the motor thermal risk score is output based on the trained intelligent control model. The intelligent control model is a random forest model.

7. The heat dissipation coordination method for an ultra-high-speed permanent magnet electric drive system based on magnetothermal coupling optimization according to claim 6 is characterized in that: The training process of the intelligent control model is: The comprehensive thermal state feature vectors and thermal risk scores of the motor over multiple historical operating cycles are obtained as a training dataset. The training dataset is then trained using a random forest model. Subsets of the training dataset and feature subsets are randomly selected to construct multiple decision trees. Each decision tree independently learns the nonlinear mapping relationship between input features and thermal risk scores. exist During the model training process, the objective function is to minimize the mean square error between the predicted thermal risk score and the actual thermal risk score of the motor. The split node selection and leaf node output value of each decision tree are iteratively optimized until the preset maximum number of iterations is met. Finally, the trained random forest model outputs the thermal risk score of the motor.

8. The heat dissipation coordination method for an ultra-high-speed permanent magnet electric drive system based on magnetothermal coupling optimization according to claim 1 is characterized in that: Determining the current thermal risk level of the motor specifically includes: According to the current thermal risk score of the motor output by the intelligent control model, it is determined whether the thermal risk score is greater than or equal to the preset first threshold. If so, it is recorded as a high thermal risk level. If not, it is determined whether the thermal risk score is less than or equal to the preset second threshold. If so, it is recorded as a low thermal risk level. If not, it is recorded as a medium thermal risk level.

9. The heat dissipation coordination method for an ultra-high-speed permanent magnet electric drive system based on magnetothermal coupling optimization according to claim 1, characterized in that: The dynamically adjusting the coolant flow rate according to the thermal risk level specifically includes: When the intelligent control model outputs that the current thermal risk score of the motor is a high thermal risk level, the control system starts the maximum cooling power mode and increases the coolant flow rate to the set upper limit value Q by increasing the speed of the variable frequency pump. max ; If the thermal risk score is a medium thermal risk level, a closed-loop feedback control strategy is used to dynamically adjust the coolant flow rate based on the deviation between the score and the baseline value; If the thermal risk level is low, maintain the minimum cooling flow Q min To achieve coordinated optimization of energy-saving operation and thermal management.

Images