High-torque multi-stage series motor and structural stability detection process

Abstract

The present application relates to motor manufacturing process technical field, more specifically, it relates to a kind of high torsion multi-stage series motor and structural stability detection process, including the modal response of the axial and radial of the frame of high torsion multi-stage series motor is collected, axial main frequency, axial damping ratio and the phase difference of radial two test points are calculated, and compared with preset threshold value;Continuous sweep frequency sinusoidal excitation is applied to motor frame, the nonlinear index of each frequency point is calculated and compared with preset threshold value, while detecting whether resonance peak is split;Then short-time half-sine pulse impact is applied to motor frame, free decay vibration signal is collected, and attenuation main frequency, logarithmic attenuation rate and cross-correlation peak delay are extracted from the time domain of free decay vibration signal, the pretightening force and connection stiffness index of two connecting points are back calculated, finally ensure that high torsion multi-stage series motor can meet the requirement of structural stability when leaving factory, and can be closed-loop corrected by detection Parameter of assembly process.

Description

Technical Field

[0001] This invention relates to the field of motor manufacturing technology, and in particular to a high-torque multi-stage series motor and a process for testing its structural stability. Background Technology

[0002] A series-wound motor, also known as a series-excitation motor or general-purpose motor, is a type of electric motor in which the excitation winding and armature winding are connected in series. It supports both AC and DC power. Its core characteristics are high starting torque, high speed, and easy speed regulation, making it widely used in power tools and household appliances. Traditional single-phase series-wound motors are generally two-pole motors. When current flows through the stator's excitation winding, it generates a main magnetic field, which consists of a pair of poles consisting of the N pole and the S pole. At the same time, the rotor's armature winding generates an armature magnetic field when current flows through it. The interaction of these two magnetic fields produces electromagnetic torque, thereby driving the rotor to rotate.

[0003] With the development of miniaturized household appliances, traditional two-pole series motors have gradually become unable to meet the demands of providing high torque and high power while maintaining miniaturization. Therefore, four-pole series motors are gradually replacing two-pole series motors in the miniaturized household appliance market. Four-pole series motors have one more pole than two-pole series motors, so under ideal conditions, the torque of a four-pole series motor is twice that of a two-pole series motor. This better meets the demands of miniaturized household appliances for miniaturized, high-torque, and high-power series motors. However, this also means that series motors need a miniaturized frame to cope with the vibrations generated by higher torque and higher power, undoubtedly placing higher demands on the structural stability of the series motor.

[0004] In existing technologies, to address structural stability issues such as those encountered with multi-stage series motors, some manufacturers opt for a four-point cross-shaped connecting bracket to assemble the motor frame. However, compared to a two-point straight connecting bracket, this design suffers from increased cost, weight, and poor heat dissipation. Therefore, a structural stability testing process for high-torque multi-stage series motors based on a two-point connecting bracket is urgently needed to ensure that the motors meet structural stability requirements upon leaving the factory. This process should also allow for closed-loop correction of assembly process parameters, ultimately ensuring that the high-torque multi-stage series motor meets structural stability requirements through optimization of the production process. Summary of the Invention

[0005] To ensure that the high-torque multi-stage series motors leaving the factory meet the structural stability requirements and to enable closed-loop correction of assembly process parameters through testing, this application provides a high-torque multi-stage series motor and a structural stability testing process.

[0006] The high-torque multi-stage series motor provided by this invention adopts the following technical solution: A high-torque multi-stage series-wound motor includes a stator assembly, a rotor assembly, a commutation assembly, and a motor frame. The stator assembly includes a stator core with an outer diameter of 95-99 mm, and the stator core has multiple stator teeth, each of which is wound with a coil. The motor frame includes a first support and a second support, which are connected at two points.

[0007] The structural stability testing process provided by this invention adopts the following technical solution: A structural stability testing process includes the following steps: S1: Collect the axial and radial modal responses of the motor frame, calculate the axial main frequency, axial damping ratio and the phase difference between the two radial test points, and compare them with the preset threshold. If any parameter exceeds the preset threshold, it is judged as unqualified. S2: Apply continuous frequency sweep sinusoidal excitation to the motor frame, calculate the nonlinear index at each frequency point and compare the maximum value with the preset threshold, and at the same time detect whether the resonance peak splits. If the maximum value of the nonlinear index exceeds the preset threshold or the resonance peak splits, it is judged as unqualified. S3: Next, apply a short-duration half-sine pulse impact to the motor frame, collect the free decay vibration signal, extract the decay main frequency, logarithmic decay rate and cross-correlation peak delay from the time domain of the free decay vibration signal, and back-calculate the preload and connection stiffness index of the two connection points. If either preload is lower than the lower limit, the relative difference between the preloads of the two connection points exceeds the threshold, or the connection stiffness is lower than the threshold, it is judged as unqualified and a rework suggestion is given.

[0008] Preferably, in S1, the motor is fixed to the test fixture, and two test points are arranged on the motor frame, including test point A and test point B. Test point A is located on the perpendicular bisector of the line connecting the two connection points, and test point B is located on the connecting cantilever of the motor frame. Acceleration sensors are pre-attached to the two test points respectively. Then, the motor frame is struck twice in different directions, including striking the top of the motor frame axially and striking the side of the motor frame radially. The vibration acceleration signal within 0.5 seconds after each strike is collected, and the axial main frequency, axial damping ratio and the phase difference between the two radial test points are calculated in real time.

[0009] Preferably, in S1, an automatic electromagnetic pulse hammer is used to strike the motor frame twice. The hammer head of the automatic electromagnetic pulse hammer has a built-in force sensor. The force of each strike is monitored by the built-in force sensor of the hammer head. The striking force should be kept within ±10% of the set range; otherwise, the strike is repeated.

[0010] Preferably, in S2, an electromagnetic vibrator is used to apply a continuous sweep frequency sinusoidal excitation to the motor frame, with the excitation direction being radial, which is perpendicular to the shaft and the line connecting the two connection points.

[0011] Preferably, in S2, after applying continuous frequency sweep sinusoidal excitation, the vibration acceleration amplitude of test point A is acquired in real time, the transfer function amplitude at each frequency point is calculated, and the maximum and minimum transfer functions within the excitation force change period are recorded. Then, the nonlinear exponent at that frequency point is calculated, and finally, the maximum value in the entire frequency sweep range is taken as the nonlinear exponent.

[0012] Preferably, in S2, the shape of the transfer function curve is observed during the frequency sweep process. If a double peak or shoulder peak appears, it is determined to be a split resonance peak.

[0013] Preferably, in S3, an electromagnetic vibrator is used to apply a short-duration half-sine pulse impact to the motor frame, with the excitation direction being radial, which is perpendicular to the shaft and the line connecting the two connection points.

[0014] Preferably, in S3, after applying a short-duration half-sine pulse impact, the free decay vibration acceleration signals of test point A and test point B within 0.3 seconds after the pulse ends are simultaneously acquired. The decay main frequency, logarithmic decay rate, and cross-correlation peak delay of the two test points are extracted from the decay signals by time-domain peak detection and counting.

[0015] Preferably, in S3, the preload and connection stiffness index of the two connection points are derived by looking up a pre-calibrated table. The table construction method includes the following steps: taking multiple motors of the same model, setting different preload combinations on the test bench, measuring the attenuation main frequency, logarithmic attenuation rate, and cross-correlation peak delay of the two test points for each combination, recording them to form a database, and discretizing the database into tables; after the construction is completed, a layered table lookup method is used when looking up the table. The first layer determines the stiffness level based on the attenuation main frequency, the second layer corrects it based on the logarithmic attenuation rate, and the third layer fine-tunes it based on the cross-correlation peak delay, finally outputting the preload and connection stiffness index of the two connection points.

[0016] Preferably, in S3, the preload and connection stiffness index of the two connection points are inversely calculated using a pre-trained neural network model. The model input parameters include the motor model, the distance between the two points, the motor frame material, and the stator outer diameter.

[0017] The beneficial effects of this invention are as follows: 1. The first stage tests the connection rigidity and asymmetrical connection of the motor frame to screen out obviously unqualified products and prevent them from entering subsequent tests. The second stage tests the micro-looseness and unstable contact surface of the motor frame to confirm whether the connection points have reached a stable contact state. The third stage quantifies the preload force to guide rework operations. Compared with traditional testing methods that usually only measure no-load current, speed, noise and insulation resistance, this method has better testing effect. Ultimately, it ensures that the high-torque multi-stage series motors leaving the factory can meet the structural stability requirements and can use testing to close the loop and correct the parameters of the assembly process. 2. The entire online testing time can be controlled within 10 seconds, and it can be conducted in parallel or in series with the no-load current test, so the impact on the production line cycle time is controllable; 3. This testing process can run in real time on an embedded controller, eliminating the need for a high-performance industrial computer, thus reducing equipment investment costs. Furthermore, it only requires the addition of corresponding sensors, vibrators, etc. at the testing station and algorithm upgrades, with no consumables and more controllable costs. Attached Figure Description

[0018] Figure 1 This is an exploded view of the high-torque multi-stage series motor in Example 1; Figure 2 This is an overall structural diagram of the high-torque multi-stage series motor in Example 1; Figure 3 This is a flowchart of the online detection process for the structural stability of the high-torque multi-stage series motor in Example 1; Figure 4 This is an overall structural diagram of the high-torque multi-stage series motor in Example 2; Explanation of reference numerals in the attached drawings: 1. First bearing; 2. Screw; 3. First bracket; 4. Rotating shaft; 5. First wire frame; 6. Stator core; 61. Stator teeth; 7. First end plate; 8. Rotor core; 9. Second end plate; 10. Second wire frame; 11. Commutator; 12. Carbon brush bushing; 13. Carbon brush plate; 14. Second bracket; 15. Second bearing; 16. Fan blade. Detailed Implementation

[0019] The following will combine Figures 1-4 The present invention will be further illustrated by the embodiments.

[0020] Example 1 This embodiment discloses a high-torque multi-stage series motor and a structural stability testing process.

[0021] Reference Figures 1 to 2The high-torque multi-stage series motor includes a motor frame, a stator core 6, a rotor core 8, and a rotating shaft 4. The stator core 6 is mounted on the motor frame. The rotor core 8 is fitted inside the stator core 6 and sleeved on the rotating shaft 4. The rotating shaft 4 is fitted with a first end plate 7 and a second end plate 9. The rotor core 8 is located between the first end plate 7 and the second end plate 9. The upper end face of the rotor core 8 is located inside the upper end face of the stator core 6, and the distance between the end faces is 0.5-1mm.

[0022] The stator core 6 has four stator teeth 61 extending from the outside to the inside. The stator teeth 61 are evenly arranged and the stator teeth 61 are offset from the adjacent brushes by an angle of 90 degrees. Each stator tooth 61 is wound with a coil.

[0023] The outer diameter of the stator core 6 is 95-99mm, and the stator core 6 adopts one or a combination of square structure or circular structure.

[0024] The rotating shaft 4 is also provided with a first bearing 1, a commutator 11 and a second bearing 15. The first bearing 1, the first end plate 7, the rotor core 8, the second end plate 9 and the second bearing 15 are arranged in sequence from the lower end to the upper end of the rotating shaft 4.

[0025] The motor frame includes a first bracket 3, a first wire frame 5, a second wire frame 10, and a second bracket 14, which are installed sequentially from bottom to top using two pairs of screws 2 and nuts. Both the first bracket 3 and the second bracket 14 are two-point connection brackets in a "straight" shape. Both the first bracket 3 and the second bracket 14 include a main bearing section in the middle and cantilever mounting sections on both sides. The main bearing section connects the bearing and the rotating shaft 4, and the cantilever mounting sections are used for connection and installation using screws 2 and nuts. The inner end of the second bracket 14 is also provided with a carbon brush plate 13 and a carbon brush copper sleeve 12. The carbon brush copper sleeve 12 passes through the carbon brush plate 13 and is connected to the second bracket 14. The outer end of the second bracket 14 is provided with a fan blade 16, which is installed on the second bracket 14 by a retaining ring.

[0026] This four-pole series motor has one more pole than a two-pole series motor, and its torque under ideal conditions is twice that of a two-pole series motor. Therefore, it can effectively reduce the overall size of the motor, meeting the demand for high torque and high power in household appliances while miniaturizing them. In addition, the two-point connection bracket of this motor has the advantages of lower cost, lighter weight and better heat dissipation compared to the four-point connection bracket.

[0027] Reference Figures 1 to 3 To allow for the safe implementation of stator outer diameter reduction and two-point connection bracket design, a structural stability testing process needs to be designed for this type of motor. This ensures reliability while significantly reducing costs. In this invention, the structural stability testing process for a high-torque multi-stage series motor includes the following steps: The motor is fixed to the test fixture. Two test points, A and B, are arranged on the motor frame. Test point A is located on the first bracket 3 and is moved outward along the perpendicular bisector of the line connecting the two screws 2 to avoid the shaft and bearing. The moving distance is the shaft radius + bearing radius + 3-5mm safety clearance. Test point B is located on the cantilever mounting part of the first bracket 3. The first bracket 3 is simpler than the second bracket 14, which facilitates the arrangement of test points A and B. Test point B is specifically located at the cantilever end far from the connection point, that is, at a position parallel to and far from the plane where the two screws 2 are connected. This position is prone to vibration response.

[0028] Miniature accelerometers, such as piezoelectric accelerometers, are pre-attached to test points A and B. Then, an automatic electromagnetic pulse hammer is used to strike the motor frame twice in different directions: axially striking the top of the motor frame and radially striking the side of the motor frame. The hammerhead of the automatic electromagnetic pulse hammer has a built-in force sensor, which monitors the force of each strike to ensure that the striking force remains within ±10% of a set range; otherwise, the strike is repeated. In some embodiments, a pneumatic hammer can also be used for striking.

[0029] After each impact, vibration acceleration signals at test points A and B are collected within 0.5 seconds, and the axial dominant frequency, axial damping ratio, and radial phase difference between the two test points are extracted. The axial dominant frequency is the frequency point with the largest amplitude in the axial impact response amplitude spectrum. The extraction process is as follows: after axial impact, the amplitude spectrum of the axial impact response is obtained, and the maximum value is searched within the frequency range of 100Hz-2000Hz. The axial damping ratio is a dimensionless measure of the system's vibration energy dissipation rate in the axial impact response, extracted using the half-power bandwidth method. The extraction process is as follows: find the amplitude corresponding to the axial dominant frequency in the amplitude spectrum, calculate the half-power point amplitude, find two frequencies near the half-power point (one less than the axial dominant frequency and the other greater), calculate the half-power bandwidth using these two frequencies, and then calculate the axial damping ratio using the half-power bandwidth and the axial dominant frequency. If the axial dominant frequency is below a threshold or the damping ratio is greater than a threshold, the connection stiffness is deemed insufficient.

[0030] It should be noted that the axial main frequency and axial damping ratio are the overall modal parameters of the motor under axial excitation. For a frame with good rigidity, the vibration frequency and damping ratio at different positions are similar under low-frequency mode. Therefore, test point A does not need to be set at the geometric center of the two connection points.

[0031] In addition, the phase difference between the two radial test points is the vibration phase difference of test point B relative to test point A at the dominant frequency in the radial impact response. The extraction process is as follows: acquire the signals of test point A and test point B under the radial impact response, perform Fourier transform on the two signals respectively to obtain two complex spectra; then find the radial dominant frequency, and calculate the phase difference at the frequency index of the radial dominant frequency. The phase difference reflects whether there is asymmetrical connection such as warping or loosening on one side of the motor frame. If the phase difference between the two test points exceeds the threshold range during radial impact, it is determined to be an asymmetrical connection. It should be noted that the change in phase difference reflects the degree of connection asymmetry. In this process, test points A and B do not need to be geometrically symmetrical, they only need to have different sensitivities to the stiffness changes of the two support points. Test point A is closer to the center of the support point and is sensitive to stiffness changes on both sides, while test point B is farther from the support point and is more sensitive to stiffness changes on the side closer to it. The combination of the two forms the differential detection of asymmetrical connection. For example, when both connection points of the motor frame are equally secure, the vibrations at both ends of the motor frame are symmetrical. The phase difference between test point A and test point B maintains a fixed value and falls within a preset threshold range. When the two connection points of the motor frame differ significantly, the phase difference deviates from the threshold range, thus indicating an asymmetrical connection. Ultimately, it is possible to determine asymmetrical connections without symmetrically arranging two test points radially (such as test point B1 and test point B2), overcoming the limitation of traditional single-point tapping in distinguishing asymmetrical connections.

[0032] After completing the first step of the above test, a continuous sweeping sinusoidal excitation is applied to the motor frame using an electromagnetic vibrator. The excitation direction is radial, perpendicular to the shaft and the line connecting the two connection points. During the sweeping process, the excitation force amplitude is slowly changed sinusoidally. The vibration acceleration amplitude at test point A is then collected, and the transfer function amplitude at each frequency point (e.g., 100Hz, 101Hz, 102Hz) is calculated. The maximum and minimum transfer functions within the excitation force variation period are recorded. Then, the nonlinearity index at that frequency point is calculated, representing the relative difference between the maximum and minimum values ​​of the transfer function. Finally, the maximum value within the entire sweeping range is taken as the nonlinearity index. In a normal rigid connection, where screw 2 is well-tightened, the transfer function is linear. Therefore, the nonlinearity index reflects the maximum nonlinear response of the motor frame within the sweeping range. A larger value indicates more severe micro-slippage or insufficient preload on the contact surface, suggesting micro-loosening due to insufficient preload of screw 2. This nonlinearity is difficult to detect in conventional torque testing but is more sensitive in this test.

[0033] During frequency sweeping, the shape of the transfer function curve is observed simultaneously. A normal, tight connection will show a single peak near the resonant frequency. If obvious double peaks or shoulder peaks appear, it is determined that the resonant peak is split. This means that a certain connection point has failed to effectively constrain the motor frame, causing that part to generate independent local vibration modes under vibration excitation. For example, in the frequency sweeping transfer function curve, taking the peak value of the main resonant peak as the reference, and taking the horizontal line of the half-power point, if two or more separate local maxima appear above this horizontal line, and the valley value between adjacent peaks is less than 90% of either peak value, it is determined to be an obvious double peak. If a shoulder-like bulge appears on one side of the main peak, and the amplitude of the bulge is greater than 70% of the amplitude of the main peak, it is determined to be a shoulder peak.

[0034] After completing the second step of the above test, a short-duration half-sine pulse impact is applied to the motor frame using the same electromagnetic exciter. The excitation direction is also radial, which is perpendicular to the shaft and the line connecting the two connection points. Then, the free decaying vibration acceleration signals of test points A and B within 0.3 seconds after the pulse ends are collected. From the decay signals, the decaying main frequency, logarithmic decay rate, and cross-correlation peak delay of the two test points are extracted through time-domain peak detection and counting. Specifically, the decaying main frequency is obtained by measuring the time interval between two adjacent positive peaks and taking the average of three consecutive intervals. The logarithmic decay rate is calculated by taking the first and fourth peaks of the three periods of the interval. The cross-correlation peak delay of the two test points is obtained by performing a simple cross-correlation between the signals of test points A and B and taking the absolute value of the correlation coefficient. That is, the cross-correlation operation is performed on the signals of test points A and B, and the absolute value of the time delay corresponding to the maximum correlation coefficient is taken as the cross-correlation peak delay. Next, the preload and connection stiffness index of the two connection points are derived by using a pre-calibrated lookup table. A layered lookup method is employed: the first layer determines the stiffness level based on the attenuation frequency; the second layer corrects for it based on the logarithmic attenuation rate; and the third layer fine-tunes it based on the cross-correlation peak delay. The final output includes the preload of the two connection points and a connection stiffness index, which are compared with threshold values. If all conditions are met, the structural stability of the motor frame is deemed acceptable; otherwise, it is deemed unacceptable. Rework suggestions are provided based on the preload difference, such as "the preload of one screw is too low; the tightening angle needs to be increased by approximately 15°," etc. This closed-loop correction of assembly process parameters is achieved through testing. Additionally, if the measured value falls between two cells during the lookup, linear interpolation can be used.

[0035] The table lookup construction method includes the following steps: take multiple motors of the same model, set different preload combinations on the test bench, measure the attenuation main frequency, logarithmic attenuation rate and cross-correlation peak delay of the two test points for each combination, record them to form a database, and discretize the database into tables.

[0036] In other embodiments, step three can also use a pre-trained neural network model to infer the preload and connection stiffness index of the two connection points. The model input parameters include the motor model, the distance between the two points, the motor frame material, and the stator outer diameter.

[0037] It should also be noted that the preset threshold in this invention is determined by multiple batches of calibration tests on the same type of motor.

[0038] The implementation principle of the high-torque multi-stage series motor and structural stability testing process in this embodiment is as follows: The first stage tests the connection stiffness and asymmetrical connection of the motor frame to screen out obviously unqualified products and avoid them from entering subsequent tests. The second stage tests the micro-loosening and unstable contact surface of the motor frame to confirm whether the connection point has reached a stable contact state. The third stage quantifies the pre-tightening force and then guides rework operations. Finally, it ensures that the high-torque multi-stage series motor leaving the factory can meet the structural stability requirements and can use the test to close the loop and correct the parameters of the assembly process.

[0039] Example 2 This embodiment also discloses a high-torque multi-stage series motor, which differs from Embodiment 1 in that: (Refer to...) Figure 4 The stator core has four mounting holes in its circumference. These four mounting holes are grouped in pairs, giving the stator core two sets of mounting holes in its circumference. This allows the motor to be assembled using either set of mounting holes.

[0040] The above are all preferred embodiments of the present invention and are not intended to limit the scope of protection of this application. Therefore, all equivalent changes made in accordance with the structure, shape and principle of this application should be covered within the scope of protection of this application.

Claims

1. A high-torque multi-stage series-wound motor, comprising a stator assembly, a rotor assembly, a commutation assembly, and a motor frame, characterized in that: The stator assembly includes a stator core with an outer diameter of 95-99mm. The stator core is provided with multiple stator teeth, and each of the multiple stator teeth is wound with a winding coil. The motor frame includes a first bracket and a second bracket, which are connected to each other by a two-point connection.

2. A structural stability testing process for detecting the structural stability of the high-torque multi-stage series-wound motor as described in claim 1, characterized in that, Includes the following steps: S1: Collect the axial and radial modal responses of the motor frame, calculate the axial main frequency, axial damping ratio and the phase difference between the two radial test points, and compare them with the preset threshold. If any parameter exceeds the preset threshold, it is judged as unqualified. S2: Apply continuous frequency sweep sinusoidal excitation to the motor frame, calculate the nonlinear index at each frequency point and compare the maximum value with the preset threshold, and at the same time detect whether the resonance peak splits. If the maximum value of the nonlinear index exceeds the preset threshold or the resonance peak splits, it is judged as unqualified. S3: Next, apply a short-duration half-sine pulse impact to the motor frame, collect the free decay vibration signal, extract the decay main frequency, logarithmic decay rate and cross-correlation peak delay from the time domain of the free decay vibration signal, and back-calculate the preload and connection stiffness index of the two connection points. If either preload is lower than the lower limit, the relative difference between the preloads of the two connection points exceeds the threshold, or the connection stiffness is lower than the threshold, it is judged as unqualified and a rework suggestion is given.

3. The structural stability testing process according to claim 2, characterized in that: In S1, the motor is fixed to the test fixture. Two test points are arranged on the motor frame, including test point A and test point B. Test point A is located on the perpendicular bisector of the line connecting the two connection points, and test point B is located on the connecting cantilever of the motor frame. Accelerometers are pre-attached to the two test points. Then, the motor frame is struck twice in different directions, including striking the top of the motor frame axially and striking the side of the motor frame radially. Vibration acceleration signals are collected within 0.5 seconds after each strike, and the axial main frequency, axial damping ratio and the phase difference between the two radial test points are calculated in real time.

4. The structural stability testing process according to claim 3, characterized in that: In S1, an automatic electromagnetic pulse hammer is used to strike the motor frame twice. The hammer head of the automatic electromagnetic pulse hammer has a built-in force sensor. The force of each strike is monitored by the built-in force sensor of the hammer head. The striking force should be kept within ±10% of the set range; otherwise, the strike is repeated.

5. The structural stability testing process according to claim 3, characterized in that: In S2, an electromagnetic vibrator is used to apply a continuous sweep frequency sinusoidal excitation to the motor frame. The excitation direction is radial, which is perpendicular to the shaft and the line connecting the two connection points.

6. The structural stability testing process according to claim 5, characterized in that: In S2, after applying continuous frequency sweep sinusoidal excitation, the vibration acceleration amplitude of test point A is collected in real time, the transfer function amplitude at each frequency point is calculated, and the maximum and minimum transfer functions within the excitation force change period are recorded. Then, the nonlinear exponent at that frequency point is calculated, and finally, the maximum value in the entire frequency sweep range is taken as the nonlinear exponent.

7. The structural stability testing process according to claim 6, characterized in that: In S2, observe the shape of the transfer function curve during the frequency sweep process. If double peaks or shoulder peaks appear, it is determined to be a split resonance peak.

8. The structural stability testing process according to claim 3, characterized in that: In S3, an electromagnetic vibrator is used to apply a short-duration half-sine pulse impact to the motor frame. The excitation direction is radial, which is perpendicular to the shaft and the line connecting the two connection points.

9. The structural stability testing process according to claim 8, characterized in that: In S3, after a short-duration half-sine pulse impact, the free decay vibration acceleration signals of test point A and test point B within 0.3 seconds after the pulse ends are simultaneously acquired. The decay main frequency, logarithmic decay rate and cross-correlation peak delay of the two test points are extracted from the decay signals by time-domain peak detection and counting.

10. The structural stability testing process according to claim 9, characterized in that: In S3, the preload and connection stiffness index of the two connection points are derived by looking up a pre-calibrated table. The table construction method includes the following steps: take multiple motors of the same model, set different preload combinations on the test bench, measure the attenuation main frequency, logarithmic attenuation rate and cross-correlation peak delay of the two test points for each combination, record them to form a database, and discretize the database into tables; after the construction is completed, a hierarchical table lookup method is used when looking up the table. The first layer determines the stiffness level based on the attenuation main frequency, the second layer corrects it based on the logarithmic attenuation rate, and the third layer fine-tunes it based on the cross-correlation peak delay. Finally, the preload and connection stiffness index of the two connection points are output.

Images