A vibration noise optimization method for a planetary row coaxial reducer

Abstract

The present application relates to the technical field of vibration noise optimization of speed reducer, and discloses a vibration noise optimization method of planetary row coaxial speed reducer, comprising: obtaining initial assembly parameters of the speed reducer, collecting load measurement signals of the planetary gear under order frequency sweep working condition, converting the load measurement signals into equivalent meshing load based on the calibration curve, calculating the load imbalance coefficient according to the equivalent meshing load, and taking the peak value of the load imbalance coefficient as the first evaluation index; synchronously collecting the shell radial acceleration and sound pressure signals under the order frequency sweep working condition, calculating the sideband ratio, and taking the peak value of the sideband ratio as the second evaluation index; diagnosing and shunting based on the first evaluation index and the second evaluation index, determining the assembly parameters to be adjusted preferentially, executing iterative adjustment, and outputting the final assembly parameters after iteration. The method can effectively separate and solve the load imbalance caused by assembly error and the sideband amplification caused by geometric eccentricity, and reduce the vibration and noise of the speed reducer from the root.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of reducer vibration noise optimization, in particular to a planetary row coaxial reducer vibration noise optimization method. BACKGROUND

[0002] The planetary row coaxial reducer is widely used in new energy vehicles, engineering machinery and industrial transmission fields due to its advantages of large transmission ratio, compact structure and high transmission efficiency. However, in actual application, the planetary row reducer generally has vibration noise problems, especially under high-speed operating conditions, the periodic modulation of meshing stiffness can significantly amplify the meshing frequency and its sideband energy, resulting in gear whine and shell resonance. Vibration noise not only affects the comfort and reliability of the whole machine, but also causes the fatigue of the tooth surface to be intensified and the service life of the bearing to be shortened.

[0003] In the prior art, the industry generally adopts tooth surface modification, structure floating design or terminal NVH test screening to reduce noise. These methods have certain effect at the design level or quality inspection link, but have obvious limitations: tooth surface modification needs to determine the parameters in advance and cannot flexibly cope with assembly errors; structure floating design relies on the accuracy of parts, and the floating is often locked and fails in actual assembly; final inspection screening can only find problems, but it is difficult to provide correction basis for the assembly end. In addition, a small number of studies show that bearing pre-tightening and assembly eccentricity are related to noise level, but there is no repeatable process and there is a lack of unified quantitative index as a guide. As a result, the same type of reducer has poor NVH consistency in batch production, the optimization process relies on experience, and it is difficult to achieve standardization. SUMMARY

[0004] In view of the above problems, the present application is proposed.

[0005] To solve the above technical problems, the present application provides the following technical scheme: a planetary row coaxial reducer vibration noise optimization method, comprising: obtaining initial assembly parameters of the reducer, collecting load measurement signals of the planetary gear under the order frequency sweep condition, converting the load measurement signals into equivalent meshing load based on the calibration curve, calculating the load imbalance coefficient according to the equivalent meshing load, and taking the peak value of the load imbalance coefficient as the first evaluation index;

[0006] Synchronously collecting the shell radial acceleration and sound pressure signals under the order frequency sweep condition, calculating the sideband ratio, and taking the peak value of the sideband ratio as the second evaluation index;

[0007] Based on the first evaluation index and the second evaluation index, diagnosis and flow are performed, the assembly parameters to be adjusted preferentially are determined, iterative adjustment is performed, and the final assembly parameters are output after iteration.

[0008] As a preferred scheme of the planet row coaxial reducer vibration noise optimization method, the initial assembly parameters of the reducer include bearing pre-tightening parameters, floating mechanism parameters, positioning phase parameters and meshing baseline parameters.

[0009] As a preferred scheme of the planet row coaxial reducer vibration noise optimization method, the collection of the load measurement signal of the planet wheel of the reducer under the order frequency sweep working condition includes configuring a replaceable load measurement jig for each planet wheel.

[0010] According to the calibration curve corresponding to the configured jig, zero point reset and temperature drift normalization are completed, and the calibration curve is a function curve composed of the corresponding relationship between the load value output by the sensor after the jig is installed and the real load value.

[0011] The order frequency sweep is performed, the input speed is increased from the minimum speed to the maximum speed at a constant speed rate, a synchronous signal is arranged, and the load measurement signal is collected at a fixed sampling rate during the order frequency sweep.

[0012] The load measurement signal is inversely calculated as the equivalent meshing load according to the calibration curve.

[0013] As a preferred scheme of the planet row coaxial reducer vibration noise optimization method, the calculation of the load imbalance coefficient according to the equivalent meshing load includes dividing the speed increasing process of the input speed into multiple speed intervals, performing gear period averaging and planet carrier period averaging in each speed interval, and converting the equivalent meshing load into a nominal load in the speed interval.

[0014] For each speed interval, the maximum value of the nominal load of each planet wheel in the interval is divided by the average value of the nominal load to obtain the load imbalance coefficient.

[0015] According to the initial assembly parameters, a typical working condition speed range of the reducer is selected, and the peak value of the load imbalance coefficient in the typical working condition speed range is taken as the first evaluation index. .

[0016] As a preferred scheme of the planet row coaxial reducer vibration noise optimization method, the calculation of the sideband ratio includes angle resampling of the shell radial acceleration and sound pressure signals to obtain the angular domain data of the shell radial acceleration and sound pressure signals, calculation of the order spectrum of the angular domain data of the shell radial acceleration and sound pressure signals in each speed interval, positioning of the meshing order on the abscissa of the order spectrum and performing peak value tracking of consecutive frames to obtain the main meshing amplitude.

[0017] The left side band amplitude and the right side band amplitude are extracted, and the side band ratio is calculated:

[0018] ;

[0019] wherein, represents the side band ratio; represents the left side band amplitude; represents the right side band amplitude; represents the main mesh amplitude;

[0020] The peak value of the side band ratio in the speed range of a typical working condition is taken as the second evaluation index .

[0021] As a preferred scheme of the vibration noise optimization method of the planetary coaxial reducer, wherein: the diagnosis shunting based on the first evaluation index and the second evaluation index comprises setting a low threshold value and a high threshold value of the first evaluation index according to the initial assembly parameters of the reducer, and a low threshold value and a high threshold value of the second evaluation index;

[0022] When , enter branch 1, and load sharing is preferred;

[0023] When , and , enter branch 2, and phase compensation is preferred;

[0024] When , and , enter branch 3, and load sharing is preferred first and then phase compensation;

[0025] When , and , enter branch 4, and light phase compensation is performed;

[0026] When , and , enter branch 5, and skip the iterative adjustment to output the final assembly parameters.

[0027] As a preferred scheme of the vibration noise optimization method of the planetary coaxial reducer, wherein: the execution of the iterative adjustment comprises setting a type of assembly parameter to be adjusted and an assembly parameter step length, the type of assembly parameter comprises an effective floating gap, a bearing pre-tightening target torque and a positioning phase angle, the assembly parameter step length is set to a fixed value based on the initial assembly parameters of the reducer, and each time an assembly parameter step length is increased and decreased during adjustment;

[0028] Iterative adjustment is performed based on the result of the diagnosis shunting:

[0029] For branch 1, only the effective floating gap and bearing pretightening target torque are adjusted; the effective floating gap is first increased, and after each increase, the first evaluation index is retested, if the first evaluation index decreases, the increase operation is retained and the effective floating gap is increased again until the retest is ; if the first evaluation index increases or remains unchanged, the increase operation of the effective floating gap is cancelled and the bearing pretightening target torque is adjusted; the bearing pretightening target torque is specifically adjusted by decreasing the bearing pretightening target torque and retesting the first evaluation index, if the first evaluation index decreases, the increase operation is retained and the bearing pretightening target torque is decreased again until the retest is ; if the first evaluation index increases or remains unchanged, the decrease operation of the bearing pretightening target torque is cancelled;

[0030] When the retest is or the first evaluation index cannot be reduced, branch 1 is exited and branch 2 is entered for rejudgment;

[0031] For branch 2, only the positioning phase angle is adjusted, the left side band amplitude and the right side band amplitude are compared, if the right side band amplitude is greater than the left side band amplitude, the positioning phase angle is decreased, if the right side band amplitude is less than the left side band amplitude, the positioning phase angle is increased, and the maximum adjustment value of the positioning phase angle is preset;

[0032] After each adjustment of the positioning phase angle, the second evaluation index is retested, if the second evaluation index decreases, the adjustment operation is retained, if the second evaluation index increases or remains unchanged, the adjustment operation is cancelled;

[0033] When the retest is or the adjustment amount reaches the maximum adjustment value of the positioning phase angle, branch 2 is exited and branch 3 is entered for rejudgment;

[0034] For branch 3, branch 1 is first executed until the exit condition of branch 1 is met, after the exit of branch 1, the second evaluation index is retested, if branch 2 is entered for adjustment, if branch 3 is exited and branch 4 is entered for rejudgment;

[0035] For branch 4, the adjustment method is the same as that of branch 2, but the adjustment of the positioning phase angle is only performed once and retested, when the retest is the adjustment operation is retained and branch 4 is exited and branch 4 is entered for rejudgment, when the retest is the adjustment operation is cancelled and branch 4 is exited and branch 4 is entered for rejudgment.

[0036] The beneficial effects of the present application: the method of the present application can quickly identify the uneven force of the planetary gear and the meshing modulation strength in the short-range bench test, and according to the determination result, enter different process branches such as load balancing priority, phase compensation or combined path. Each branch only carries out fixed amplitude operation on a single process parameter, such as floating gap, bearing pre-tightening or inner tooth ring phase angle, and immediately retests after operation to ensure controllable and traceable process. It can effectively separate and solve the load imbalance caused by assembly error and the sideband amplification caused by geometric eccentricity, and reduce the vibration and noise of the reducer from the root. BRIEF DESCRIPTION OF DRAWINGS

[0037] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0038] Figure 1 The overall flowchart of a planetary row coaxial reducer vibration noise optimization method provided by an embodiment of the present application. DETAILED DESCRIPTION

[0039] In order to make the above-mentioned purposes, features and advantages of the present application more apparent and easy to understand, the specific embodiments of the present application will be described in detail below with reference to the drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor should be within the scope of protection of the present application.

[0040] Embodiment 1, refer to Figure 1 For an embodiment of the present application, a planetary row coaxial reducer vibration noise optimization method is provided, which comprises:

[0041] S1: obtaining the initial assembly parameters of the reducer, collecting the load measurement signal of the planetary gear under the order frequency scanning condition of the reducer, converting the load measurement signal into equivalent meshing load based on the calibration curve, calculating the load imbalance coefficient according to the equivalent meshing load, and taking the peak value of the load imbalance coefficient as the first evaluation index.

[0042] Further, the initial assembly parameters of the reducer are obtained and parameter archiving is performed to form a traceable initial assembly state and provide a unified reference for subsequent closed-loop optimization. Specifically, the baseline assembly of the reducer assembly is completed on the controlled work station according to the process specification, and four key assembly parameters closely related to load sharing and meshing sideband uniformity are quantitatively archived:

[0043] First, the bearing pre-tightening parameters, the target pre-tightening value is established by torque-angle or equivalent displacement method, and the matching gasket thickness is recorded to ensure that the support stiffness and the initial stress of the rolling body are in the controllable interval;

[0044] Second, the floating mechanism parameters, the effective floating gap is measured and registered, which directly determines the ability of the floating release to manufacture and position deviation, and is the prerequisite for load sharing;

[0045] Third, the positioning phase parameter, the positioning phase reference angle of the inner ring gear relative to the shell is fixed by marking or coding as the angle reference for subsequent micro-eccentricity compensation, to ensure that any position fine adjustment can be performed and reproduced relative to the unified reference;

[0046] Fourth, the meshing baseline parameter, the initial tooth side gap is recorded to ensure that the meshing stiffness and the return characteristic are within the baseline range permitted by the design. The above parameters are written into the assembly card in a unified measurement unit and recording format and are synchronized to the electronic database.

[0047] The application simultaneously fixes the test basic conditions at this stage, such as the short-range sweep frequency procedure and the set value of the initial oil temperature, but does not participate in the adjustment as an assembly parameter. Through the above documentation, the influence amount on the load sharing capability in the assembly link, the angle reference for micro-geometric correction, and the meshing baseline are explicit and standardized, thereby providing a stable starting point for subsequent small-step adjustment, re-measurement comparison, and result convergence, and avoiding criterion drift and optimization invalidation caused by inconsistent initial state.

[0048] In order to convert the real force between planets from an invisible state to a quantifiable and re-measurable control quantity, after the baseline assembly is completed, the load measurement signal of the planetary gear under the order frequency sweep working condition is collected.

[0049] The load measurement signal of the planetary gear under the order frequency sweep working condition of the reducer is collected, and the equivalent meshing load is converted based on the calibration curve, which specifically includes:

[0050] A replaceable load measurement jig is configured for each planetary gear;

[0051] According to the calibration curve corresponding to the configured jig, zero reset and temperature drift normalization are completed, and the calibration curve is a function curve composed of the corresponding relationship between the load value output by the sensor after installing the jig and the real load value;

[0052] The order frequency sweep is performed, the input speed is increased from the minimum speed to the maximum speed at a constant speed rate, and a synchronous signal is arranged, and the load measurement signal is collected at a fixed sampling rate during the order frequency sweep;

[0053] The load measurement signal is inversely calculated into the equivalent meshing load according to the calibration curve.

[0054] Specifically, the reducer is placed on the bench under controlled oil temperature conditions, the input shaft is driven according to a unified short-range order frequency scanning procedure, the speed is smoothly increased from the minimum speed to the maximum speed at a constant rate, and the continuous speed-up process of the input speed is divided into multiple speed intervals. At the same time, replaceable load measuring jigs are arranged at each planetary gear position and an angle synchronization reference is established. The jigs can be customized according to product specifications:

[0055] Planetary pin strain bridge clamps (resistance strain gauges are pre-embedded in thin-walled clamping sleeves to form a Wheatstone full bridge);

[0056] Integrated force measuring bearing (strain area is opened in the outer ring and electrical signal is led out);

[0057] Load indicating film (pressure coloration, read with density-force calibration curve).

[0058] To avoid changing the part body, the jigs are interchangeable with standard parts of the same size; when the measurement channel is limited, the rotation method is allowed to complete the recording of all planetary positions in two times, and through the angle synchronization reference, the data of different batches are mapped to the same phase.

[0059] To ensure the quantification results have engineering comparability, the jigs are first calibrated at two points of no load and small load after assembly and the calibration number is recorded, and the oil temperature curve is collected during the frequency scanning process to compensate for the first-order temperature drift of the signal.

[0060] To avoid the frequency deviation and non-steady-state artifacts introduced by speed-up, the collected load measurement signals are converted to angular domain data of the load measurement signals through angular resampling, and the planetary carrier encoder is used as the main synchronization source, combined with the input shaft primary tooth signal to realize phase alignment, so that the measurement of different planetary positions can be mapped to the same angle reference. Angle resampling refers to reconstructing signals at equal intervals of shaft system rotation angle, so that the order characteristics under non-constant speed conditions can be stably identified.

[0061] It should be noted that the angular domain data refers to the signal sequence obtained by resampling with the shaft system rotation angle as the independent variable. During order frequency scanning, if direct time domain analysis is performed, the frequency position will shift, resulting in spectral broadening. By converting the signal to angular domain data, it can be ensured that each angular sampling point is uniformly distributed, so that the meshing characteristics can still be accurately characterized at different speeds and facilitate order analysis.

[0062] Subsequently, the rotational domain data of the load measurement signal was narrowband smoothed and outliers caused by instantaneous operating condition disturbances were removed to obtain the equivalent meshing load of each planetary gear within the speed range. To highlight the uniform load rather than the slight fluctuations within the elastic period between the teeth, gear period averaging and planetary carrier period averaging were performed for each speed range (planetary carrier period averaging is the average of the load waveforms repeated multiple times at the same planetary position with one revolution of the planetary carrier as the period, in order to suppress random noise and transient impacts), and the equivalent meshing load was converted into the nominal load of the i-th planetary gear at that speed.

[0063] After obtaining the nominal load within the same speed range, calculate the load imbalance coefficient:

[0064] ;

[0065] in, Indicates the load imbalance coefficient; i represents the planetary gear number; This indicates that the planetary gears are numbered from 1 to m. Take the maximum value; m represents the total number of planetary gears; This indicates that the planetary gears are numbered from 1 to m. Summation; This represents the nominal load of the i-th planetary gear.

[0066] The load imbalance coefficient curve is formed as the frequency sweep progresses. Considering that the low-speed region is susceptible to the effect of tooth backfill and the high-speed region is susceptible to the effects of oil-gas turbulence and structural mode superposition, the typical operating speed range of the reducer is selected based on the initial assembly parameters, and the peak value of the load imbalance coefficient within the typical operating speed range is taken as the first evaluation index. Meanwhile, the full curve is retained for subsequent comparison.

[0067] It should be noted that the load imbalance coefficient can directly quantify the load distribution among the planetary gears within a planetary gearbox transmission system. The ideal operating condition for a planetary gearbox reducer is that all planetary gears evenly distribute the transmitted torque, at which point the meshing forces of each planet are essentially equal, and the load imbalance coefficient is close to 1. However, in actual assembly and operation, factors such as manufacturing errors, improper clearance of the floating mechanism, bearing preload deviation, or slight eccentricity of the internal gear ring often lead to differences in the load borne by different planetary gears, resulting in some planets experiencing excessive force while others experience insufficient force. This imbalance not only causes periodic modulation of the meshing stiffness, amplifying the side-band energy of the meshing frequency and generating significant vibration noise, but also exacerbates wear and fatigue failure of local tooth surfaces, shortening the reducer's lifespan.

[0068] Therefore, the load imbalance coefficient, as a root cause quantity, can directly reflect the real balance of the internal force of the planetary system without relying on acoustic or vibration external manifestations. The greater the value is, the more serious the load concentration is, the more obvious the mesh stiffness fluctuation is, and the higher the NVH risk of the system is; on the contrary, when the load imbalance coefficient is close to 1, it indicates that the load distribution is close to the ideal uniform state, and the stability and quietness of the transmission are better.

[0069] Compared with the criterion that only depends on the terminal noise level or the overall vibration amplitude, the load imbalance coefficient can reveal the internal cause of noise formation earlier and more accurately, and has higher sensitivity and guidance. Based on this characteristic, the load imbalance coefficient is set as the first evaluation index in the assembly optimization closed loop to ensure that the direction of subsequent small step adjustment can be closely related to the system load balancing capability, and fundamentally reduce the noise source.

[0070] S2: synchronously collect the shell radial acceleration and sound pressure signals under the order sweep working condition, calculate the sideband ratio, and take the peak value of the sideband ratio as the second evaluation index.

[0071] Further, after synchronously collecting the shell radial acceleration and sound pressure signals under the order sweep working condition, the order domain analysis is performed on the shell radial acceleration and sound pressure signals to extract the sideband characteristics related to the meshing.

[0072] Specifically, the shell radial acceleration and sound pressure signals collected by the test bench are converted into the rotation angle domain data of the shell radial acceleration and sound pressure signals under the condition of angle resampling, so as to eliminate the frequency shift and spectral broadening caused by the sweep speed.

[0073] In each speed interval, the order spectrum of the rotation angle domain data of the shell radial acceleration and sound pressure signals is calculated. In order to suppress spectral leakage, a window function with a known equivalent noise bandwidth is used for windowing, and then the amplitude is adjusted by bandwidth normalization to ensure the comparability of the amplitudes in different speed intervals. Spectral leakage refers to the energy diffusion phenomenon caused by the finite length data, and if it is not adjusted, it will introduce system deviation of the order amplitude.

[0074] According to the tooth number and the transmission ratio distribution relationship, the meshing order is positioned on the horizontal coordinate (order axis) of the order spectrum and the peak value tracking of the continuous frame is performed, the meshing order reflects the energy concentration position of the periodicity of the tooth meshing stiffness, the false peaks are removed by using the amplitude continuity and phase smoothness constraints to obtain the stable trajectory of the main meshing peak, and the amplitude of the stable trajectory of the main meshing peak is read as the main meshing amplitude. Then the left and right sideband energies are extracted in the neighborhood of the meshing order on both sides at equal distances, the sideband is the amplitude / frequency modulation result generated by the low-frequency modulation (such as carrier rotation or geometric eccentricity) to the high-frequency meshing process. In order to improve the noise resistance, according to the energies of the left and right sidebands, the left sideband amplitude and the right sideband amplitude are obtained by using the narrowband energy integration or the peak-mean fusion method in the sideband neighborhood, and the frequency domain gain correction is performed on the sensor frequency response difference to ensure the consistent caliber of different measuring points and different batches of data.

[0075] The sideband ratio is defined as the ratio of the sum of the left and right sideband amplitudes to the main meshing peak amplitude, and is expressed as:

[0076] ;

[0077] Wherein, represents the sideband ratio; represents the left sideband amplitude; represents the right sideband amplitude; represents the main meshing amplitude; the peak value of the sideband ratio in the typical working condition speed range is taken as the second evaluation index .

[0078] The sideband ratio can directly reflect the amplitude modulation degree of the low-frequency modulation to the periodic energy of the meshing stiffness, and further has a high correlation with the intensity of the hum / whistle related to the hearing.

[0079] It should be noted that the sideband ratio is set as the second evaluation index, which is complementary to the load imbalance coefficient. The appearance of the sideband is essentially a manifestation of the low-frequency modulation. When the planetary wheel load is unevenly distributed, the inner ring gear has a slight eccentricity, or the bearing pre-tightening is improper, periodic fluctuations will be introduced in the gear meshing process, so that the high-frequency meshing signal is modulated by the low-frequency, thereby generating symmetric or asymmetric sidebands on both sides of the main frequency. The size of the sideband ratio directly reflects the modulation strength: when the sideband ratio is large, it indicates that the meshing force is significantly disturbed periodically, the noise energy spreads to the low-frequency side, and the whole machine appears obvious hum or whistle in the hearing; when the sideband ratio is small, it indicates that the meshing energy is concentrated, the system meshing process is stable, and the noise performance is improved.

[0080] Different from the load imbalance coefficient, the sideband ratio is more inclined to reveal the root cause of the internal force distribution, while the sideband ratio directly reflects the external perceptible vibration noise performance. Therefore, the sideband ratio is set as the second evaluation index as a result of the load imbalance coefficient: in assembly optimization, the improvement of the load imbalance coefficient does not necessarily always fully translate into noise reduction, at which time the sideband ratio can be used as a basis for judging whether the acoustic improvement target is reached. At the same time, the sideband ratio has sensitivity to the modulation type and degree, which can help to distinguish different imbalance modes such as load deviation and geometric eccentricity.

[0081] In summary, as the second evaluation index, the sideband ratio can establish an intuitive connection between internal force changes and external noise performance, provide direct acoustic criteria for the convergence of assembly optimization, and prompt the need for further micro-eccentricity compensation or phase adjustment when the load imbalance coefficient is qualified but the noise is still high. The load imbalance coefficient and the sideband ratio respectively bear the roles of root cause and result, forming a double-index system that complements each other in the closed-loop optimization system constructed by the present application.

[0082] S3: Based on the first evaluation index and the second evaluation index, diagnosis is performed, the priority adjustment assembly parameter is determined, the iterative adjustment is performed, and the final assembly parameter is output after the iteration is completed.

[0083] Further, based on the first evaluation index and the second evaluation index, diagnosis is performed, and the diagnosis result is used as an execution instruction for subsequent adjustment. According to the initial assembly parameters of the reducer, the low threshold value and the high threshold value of the first evaluation index are set, and the low threshold value and the high threshold value of the second evaluation index are set. Among them, the first evaluation index reflects the load sharing degree among planets, and the larger the value, the worse the load sharing ability; the second evaluation index reflects the intensity of low-frequency modulation in the meshing process, and the larger the value, the more obvious the sideband amplification and the worse the listening experience.

[0084] When , it is determined that the load sharing is severely imbalanced, and branch 1 is entered, and load sharing is prioritized. Severe imbalance leads to strong modulation of meshing stiffness with rotation angle, which is the root cause of sideband amplification; if phase-based compensation is performed first, the root cause may be hidden and it is difficult to converge. Therefore, in this branch, only small-step adjustment around the adjustable process quantities of floating and pre-tightening is allowed, such as adjusting the effective floating gap to unlock the floating release error, or fine-tuning the bearing pre-tightening to avoid locking the floating, and any phase / micro-eccentricity compensation is prohibited.

[0085] When , and At this point, the load distribution is deemed acceptable, but the side band is severely amplified, leading to branch 2, where phase compensation is prioritized. Since the load is now more evenly distributed, the side band anomaly is more likely caused by slight eccentricity of the internal gear ring or assembly phase misalignment, falling under the category of geometric-phase problems. Correction should be directly applied to the positioning phase. Therefore, only slight eccentricity / phase compensation is performed to minimize disturbance to the established load-sharing state, and a retest is conducted after fine-tuning a single wheel. Verify the inhibitory effect.

[0086] when ,and At that time, it was determined to be a moderate imbalance in load sharing accompanied by sideband amplification, and the circuit entered branch 3, where load sharing was performed first, followed by phase compensation. Moderate imbalance is sufficient to induce a noticeable sideband amplification, but its sensitivity is greater for improving load sharing; if it is not reduced first... Phase compensation often exhibits instability or repetition. Therefore, within this branch, it is first implemented in small steps. Reduce to Next, let's return to the diagnostic triage process, based on the new... Decide whether to switch to branch 2 for light phase correction to ensure monotonic convergence of the optimized path.

[0087] when ,and When the load is deemed acceptable and the side band amplification is moderate, the signal enters branch 4 for light phase compensation. If the side band is too high but not reaching the severe threshold, small-step phase fine-tuning can be used to verify the existence of slight eccentricity / phase shift at minimal cost, avoiding excessive disturbance to the existing assembly state. If a single-round fine-tuning... Down to The following steps will exit the closed loop according to the diagnostic triage and release rules.

[0088] when ,and When the load sharing and side belt are both within the acceptable range, enter branch 5, skip the iteration adjustment, and output the final assembly parameters.

[0089] It should be noted that some interval combinations do not form new branch determination rules separately. This is because these combinations have a very low probability of occurring in the actual assembly and operation of planetary gear reducers, or their performance can be covered by existing branches, so there is no need to set up redundant branches.

[0090] For example, when At any time, regardless The values, regardless of their magnitude, all fall under the category of severe load imbalance without amplification of the side band. However, in reality, severe imbalance inevitably leads to a strong modulation of meshing stiffness with rotation angle, and the side band energy usually increases significantly. It is almost impossible to keep low value, even if it appears, also caused by measurement error or working condition disturbance. Therefore, such cases are merged into the uniform load priority branch, which can be excluded by repeated measurement and verification.

[0091] For example, when , and at a very high value, often indicates that the measurement channel exists distortion or environmental noise interference, rather than the real reflection of the assembly state, such cases should trigger the retest mechanism, and not as a branch alone.

[0092] Therefore, the present application does not set up a branch for these atypical intervals alone, but is handled by the retest checking mechanism and the hysteresis band control. This design avoids the complication of decision logic, while ensuring the stability and engineering operability of the assembly optimization process. In other words, the five branches listed in the diagnostic shunt have covered the main performance interval in the actual assembly working condition, which can deal with high probability problem scenarios, and can handle low probability or measurement abnormal scenarios through abnormal detection means, so that the optimization process is comprehensive and efficient.

[0093] The diagnostic shunt undertakes the decision layer function, outputs the branch code and the allowed adjustable object set, and specifies the exit threshold; the iterative adjustment undertakes the execution layer function, executes single-factor small-step adjustment according to the branch and threshold given by the diagnostic shunt, retests the new first evaluation index and second evaluation index in the same representative speed window after each adjustment, and returns the results to the diagnostic shunt. The diagnostic shunt judges whether to continue iteration along the current branch, whether to switch to the subsequent branch, or whether to meet the exit condition to enter branch 5.

[0094] Further, based on the results of the diagnostic shunt, iterative adjustment is performed, and the type and step size of the assembly parameters to be adjusted are set. The assembly parameter type includes effective floating gap, bearing pre-tightening target torque, and positioning phase angle. The assembly parameter step size is set to a fixed value based on the initial assembly parameters of the reducer. During adjustment, the assembly parameter step size is increased or decreased by one assembly parameter step size each time.

[0095] It should be noted that the adjustment of the bearing pre-tightening force can be achieved in two ways: one is to adjust the target torque of the bolt tightening by using the torque-angle method; the other is to adjust by replacing the gasket of different thickness by using the gasket method. Both methods are used to control the pre-tightening state and support stiffness of the bearing, so in the description of the method steps, only the bearing pre-tightening target torque is used to represent the adjustment amount of the bearing pre-tightening force. In actual application, both methods will not appear as independent adjustment amounts at the same time, but one of them will be selected according to the process path adopted by the assembly line.

[0096] For branch 1, only the effective floating gap and bearing pretightening target torque are adjusted; the effective floating gap is first increased, and after each increase, the first evaluation index is retested, if the first evaluation index decreases, the increase operation is retained and the effective floating gap is increased again until ; if the first evaluation index increases or remains unchanged, the increase operation of the effective floating gap is cancelled and the bearing pretightening target torque is adjusted; the bearing pretightening target torque is specifically reduced and the first evaluation index is retested, if the first evaluation index decreases, the increase operation is retained and the bearing pretightening target torque is reduced again until ; if the first evaluation index increases or remains unchanged, the reduction operation of the bearing pretightening target torque is cancelled; when or the first evaluation index cannot be reduced, branch 1 is exited and diagnostic shunt rejudgment is entered;

[0097] The reason why the first evaluation index can be improved by increasing the effective floating gap is that the function of the floating mechanism is to release manufacturing and positioning errors, so that the sun gear / planet gear automatically optimizes around the theoretical center; when the effective floating gap is too small or the assembly error is stuck, individual planet gears will bear higher meshing load for a long time, and the meshing stiffness will be strongly modulated with the rotation angle, and the first evaluation index will increase and drive the second evaluation index to rise.

[0098] When the effective floating gap reaches the upper limit of the type or the improvement of the first evaluation index is insufficient, other assembly parameter types are adjusted. The mechanism is that excessive pretightening will increase the support stiffness and inhibit the pose self-adaptability of the floating mechanism, resulting in the so-called locked floating phenomenon; appropriately reducing the pretightening or reducing the gasket thickness can release the small assembly stress without damaging the bearing life and stability, further improve the load sharing among the planets, and fundamentally reduce the first evaluation index. After each parameter change, the first evaluation index is immediately retested, if the first evaluation index does not decrease, the operation is cancelled and the current path is terminated, to avoid continuing to accumulate in the wrong direction.

[0099] For branch 2, only the positioning phase angle is adjusted, the left side band amplitude and the right side band amplitude are compared, if the right side band amplitude is greater than the left side band amplitude, the positioning phase angle is reduced, if the right side band amplitude is less than the left side band amplitude, the positioning phase angle is increased, and the maximum adjustment value of the positioning phase angle is preset; after each adjustment of the positioning phase angle, the second evaluation index is retested, if the second evaluation index decreases, the adjustment operation is retained, if the second evaluation index increases or remains unchanged, the adjustment operation is cancelled; when or the adjustment amount reaches the maximum adjustment value of the positioning phase angle, branch 2 is exited and diagnostic shunt rejudgment is entered;

[0100] The reason why the second evaluation index can be suppressed by rotating the positioning phase angle is that the ±1 order sidebands on both sides of the meshing main peak are essentially derived from the amplitude / frequency modulation of the low-frequency modulation (carrier base order) on the high-frequency meshing. The positioning phase bias caused by the assembly geometric eccentricity or the inner tooth ring will form a fixed phase modulation source. By allowing the inner tooth ring to rotate in a fixed step relative to the shell, the phase relationship between the eccentric error and the meshing force vector can be changed, and without changing the part body and the center distance, the modulation term can be offset or weakened, thereby reducing the energy of the sidebands on both sides and lowering the second evaluation index.

[0101] The step of the positioning phase angle comes from the controllable resolution of the commonly used rotatable positioning bushing / gap positioning pin and the minimum reproducible angle allowed by the hole system arrangement. The maximum adjustment of the positioning phase angle avoids the secondary positioning of the hole system and the poor fit of the sealing engagement surface.

[0102] For branch 3, first execute branch 1 until the exit condition of branch 1 is met, and then retest the second evaluation index after the exit of branch 1. If branch 2 is entered for adjustment, and if branch 3 is exited and the diagnostic shunt is entered for rejudgment.

[0103] The load sharing problem is the upstream root cause of sideband amplification: if the first evaluation index is not first pressed to the qualified area, the adjustment of the positioning phase angle is directly executed, the modulation source is still in place, and the phase only makes a small phase cancellation on the high-energy modulation, which often leads to slow convergence or even repetition. Therefore, the present application first uses the load sharing priority branch to make and continuously monitors the permissible range of the tooth side gap and the bearing temperature rise during this process; after reaching the load sharing target, the positioning phase angle is adjusted according to the phase compensation priority branch to obtain a significant decrease in the second evaluation index with relatively smaller actions. This sequence design ensures the monotonicity and repeatability of the optimization path, which first eliminates the modulation strength and then corrects the modulation phase.

[0104] For branch 4, the adjustment method is the same as branch 2, but only one adjustment of the positioning phase angle is performed and retested. When the retest is the adjustment operation is retained and branch 4 is exited to enter the diagnostic shunt, and when the retest is the adjustment operation is cancelled and branch 4 is exited to enter the diagnostic shunt.

[0105] This path only implements one adjustment of the positioning phase angle and then retests it. Only one adjustment is performed because when the sideband is high, it is usually caused by a slight positioning angle deviation or cumulative tolerance of the hole system, which has the characteristics of slight eccentricity and light adjustment. One rotation of the adjustment is sufficient to determine whether this type of deviation exists and whether it is sensitive to noise, and if If the angle is improved, the angle is retained and the process is exited; if not, the operation is cancelled to avoid unnecessary disturbance to the load sharing state and meshing clearance that has met the standard. This lightweight strategy minimizes the assembly rhythm and risk while retaining the ability to quickly address minor phase problems.

[0106] Through the implementation of this step, the load sharing capability and the degree of sideband suppression are improved in a test caliber that is perceptible, controllable and replicable: when the load imbalance coefficient decreases, the corner modulation of meshing stiffness is weakened, and the sideband ratio is lowered; when the load imbalance coefficient is in a reasonable range and the sideband ratio is still high, the sideband energy can be significantly reduced by adjusting the positioning phase angle by a small angle. Finally, the assembly parameters and the corresponding load imbalance coefficient and sideband ratio retest results are solidified as the NVH assembly fingerprint of the unit, which is used to guide and correct the assembly process of the planetary coaxial reducer, and fundamentally reduces the noise.

[0107] In summary, the present application proposes an assembly optimization method for the vibration and noise of the planetary coaxial reducer. By establishing the initial records of key parameters such as effective floating clearance, bearing pre-tightening force or gasket thickness, and inner tooth ring positioning phase in the baseline assembly stage, and extracting two types of evaluation indexes of load imbalance coefficient and sideband ratio under the condition of bench frequency scanning, the load distribution state and meshing modulation intensity of the planets are quantitatively characterized.

[0108] On this basis, the present application uses the combined relationship of the load imbalance coefficient and the sideband ratio for branch determination, respectively enters different optimization branches such as load sharing priority, phase compensation priority, joint path or lightweight path, and iteratively retests and iterates through small step and single variable controllable operation. The whole process guarantees the closed-loop logic of determination-execution-redetermination, forms a causal chain from the root cause (load sharing capability) to the result (sideband performance), and the convergence path is clear and repeatable.

[0109] The method of the present application not only overcomes the shortcomings of relying on experience and single final inspection in the traditional process, but also converts the invisible load sharing state into quantifiable and retestable indexes, and establishes a standardized process that can be executed under the production rhythm. Through this method, the vibration and noise level of the planetary reducer can be significantly reduced without changing the design of the parts, the assembly consistency and reliability are improved, and a traceable data basis and process fingerprint are provided for subsequent process solidification and mass production, which has significant technical effects and application prospects.

[0110] If the above functions are realized in the form of software function units and sold or used as independent products, they can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application essentially or parts of the present application that make contributions to the prior art or parts of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media that can store program codes.

[0111] The logic and / or steps represented in the flowcharts or otherwise described herein, for example, can be considered as a list of executable instructions for implementing logic functions, which can be embodied in any computer-readable medium for use by or in connection with an instruction execution system, apparatus or device, such as a computer-based system, a system including a processor or other system that can fetch instructions from an instruction execution system, apparatus or device and execute the instructions, or in conjunction with such instruction execution system, apparatus or device. For the purpose of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transport a program for use by or in connection with an instruction execution system, apparatus or device or in conjunction with such instruction execution system, apparatus or device.

[0112] More specific examples (a non-exhaustive list) of the computer-readable medium include the following: an electrical connection having one or more wires (electrical devices), a portable computer diskette (magnetic devices), a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CD ROM). In addition, the computer-readable medium can even be paper or other suitable medium on which the program can be printed, as the program can be electronically obtained, for example, by optical scanning of the paper or other medium, followed by editing, interpreting or otherwise processing, if necessary, in other suitable ways to obtain the electronic program, and then storing it in the computer memory.

[0113] It should be understood that portions of the present application can be implemented in hardware, software, firmware, or combinations thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, implementation can be with any or a combination of the following technologies, which are all well known in the art: a discrete logic circuit having logic gates for implementing logic functions upon an application of data signals, an application specific integrated circuit having appropriate combinational logic gates, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0114] It should be noted that the above examples are only used to illustrate the technical solutions of the present application but not limit the present application. Although the present application has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present application can be modified or replaced equivalently without departing from the spirit and scope of the present application, and all of them should be covered in the scope of the claims of the present application.

Claims

1. A method for optimizing vibration noise of a planetary row coaxial reducer, characterized in that, The method comprises the following steps: Obtaining initial assembly parameters of the reducer, collecting load measurement signals of the planetary gear under the order frequency sweep working condition, converting the load measurement signals into equivalent meshing load based on the calibration curve, calculating the load imbalance coefficient according to the equivalent meshing load, and taking the peak value of the load imbalance coefficient as the first evaluation index; Synchronously collecting the shell radial acceleration and sound pressure signals under the order frequency sweep working condition, calculating the sideband ratio, and taking the peak value of the sideband ratio as the second evaluation index; Based on the first evaluation index and the second evaluation index, the diagnostic shunt is carried out, the priority adjustment assembly parameter is determined, the iterative adjustment is carried out, and the final assembly parameter is output after the iteration is completed; The method comprises the following steps: Dividing the speed-up process of the input rotation speed into multiple speed intervals, performing gear cycle averaging and planetary carrier cycle averaging in each speed interval, and converting the equivalent meshing load into the nominal load in the speed interval; According to the initial assembly parameter, a typical working condition speed range of the speed reducer is selected, and a peak value of a load imbalance coefficient in the typical working condition speed range is taken as a first evaluation index ; For each speed interval, the maximum value of the nominal load of each planetary gear in the interval is divided by the average value of the nominal load to obtain the load imbalance coefficient; The method comprises the following steps: ； wherein, represents the sideband ratio; represents the left sideband amplitude; represents the right sideband amplitude; represents the primary web amplitude; Taking the peak value of the sideband ratio in the typical operating speed range as the second evaluation index .

2. The vibration noise optimization method for a planetary row coaxial reducer according to claim 1, characterized in that: Angle resampling is performed on the shell radial acceleration and sound pressure signals to obtain the rotation angle domain data of the shell radial acceleration and sound pressure signals, the order spectrum of the rotation angle domain data of the shell radial acceleration and sound pressure signals in each speed interval is calculated, the meshing order is positioned on the abscissa of the order spectrum, and the peak value tracking of the continuous frames is performed to obtain the main meshing amplitude; 3. The vibration noise optimization method for a planetary row coaxial reducer according to claim 2, characterized in that: The left sideband amplitude and the right sideband amplitude are extracted, and the sideband ratio is calculated: The initial assembly parameters of the reducer include bearing pre-tightening parameters, floating mechanism parameters, positioning phase parameters and meshing baseline parameters. The method comprises the following steps: According to the calibration curve corresponding to the configured fixture, zero reset and temperature drift normalization are completed, and the calibration curve is a function curve composed of the corresponding relationship between the load value output by the sensor after the fixture is installed and the true load value; 4. The vibration noise optimization method for a planetary row coaxial reducer according to claim 3, characterized in that: The diagnosis and shunting based on the first evaluation index and the second evaluation index comprises setting a low threshold value of the first evaluation index according to the initial assembly parameters of the speed reducer and a high threshold value , a low threshold value and a high threshold value of the second evaluation index; When branch 1, load sharing is preferred. When , and , the branch 2 is entered, and phase compensation is preferred. When , and , the branch 3 is entered, and the first is equalized and then the phase is compensated. When , and , the branch 4 is entered, and light weight phase compensation is performed; When , and , enter branch 5, skip iteration, adjust output final assembly parameters.

5. The vibration noise optimization method for a planetary row coaxial reducer according to claim 4, characterized in that: The order frequency sweep is performed, the input rotation speed is increased from the minimum rotation speed to the maximum rotation speed at a constant speed-up rate, and a synchronous signal is arranged, the load measurement signal is collected at a fixed sampling rate during the order frequency sweep; The load measurement signal is inversely calculated into the equivalent meshing load according to the calibration curve. For branch 1, only the effective floating gap and bearing pretightening target torque are adjusted; the effective floating gap is first increased, and after each increase, the first evaluation index is retested, if the first evaluation index decreases, the increase operation is retained and the effective floating gap is increased again until the retest is ; if the first evaluation index increases or remains unchanged, the increase operation of the effective floating gap is cancelled and the bearing pretightening target torque is adjusted; the adjustment of the bearing pretightening target torque is specifically that the bearing pretightening target torque is reduced and the first evaluation index is retested, if the first evaluation index decreases, the increase operation is retained and the bearing pretightening target torque is reduced again until the retest is ; if the first evaluation index increases or remains unchanged, the decrease operation of the bearing pretightening target torque is cancelled; When the retest is or the first evaluation index cannot be reduced, branch 1 is exited, and the diagnosis shunt re-judgment is entered. The method comprises the following steps: Setting the type of assembly parameters that need to be adjusted and the assembly parameter step length, the type of assembly parameters includes effective floating gap, bearing pre-tightening target torque and positioning phase angle, the assembly parameter step length is set to a fixed value based on the initial assembly parameters of the reducer, and each assembly parameter step length is increased or decreased during adjustment; Based on the results of the diagnostic shunt, the iterative adjustment is carried out: For branch 2, only the positioning phase angle is adjusted, the left sideband amplitude and the right sideband amplitude are compared, if the right sideband amplitude is greater than the left sideband amplitude, the positioning phase angle is reduced, if the right sideband amplitude is less than the left sideband amplitude, the positioning phase angle is increased, and the maximum value of the positioning phase angle is preset. After adjusting the positioning phase angle each time, the second evaluation index is retested. If the second evaluation index decreases, the adjustment operation is retained. If the second evaluation index increases or remains unchanged, the adjustment operation is cancelled; When the retest is completed Or the adjustment reaches the maximum adjustment of the positioning phase angle, exit branch 2, and enter the diagnostic shunt re-determination. For branch 3, first execute branch 1 until the exit condition of branch 1 is met, retest the second evaluation index after branch 1 exits, if then turn to branch 2 for adjustment, if then exit branch 3 and turn to the diagnosis shunt rejudgment; For branch 4, the adjustment mode is the same as branch 2, but only one adjustment of the positioning phase angle is performed and a retest is made, when the retest is , the adjustment operation is retained and branch 4 is exited to enter the diagnostic shunt, when the retest is , the adjustment operation is cancelled and branch 4 is exited to enter the diagnostic shunt.

Images