Worm gear speed reducer comprehensive performance test method based on speed reducer comprehensive test board

Abstract

The invention discloses a worm gear speed reducer comprehensive performance test method based on a speed reducer comprehensive test board, and the method comprises the steps: setting a first test condition parameter of a current test node, and controlling a drive motor to drive a worm gear speed reducer to carry out the positive and negative reciprocating motion under the first test condition parameter corresponding to the current test node, synchronously acquiring position signals of the first encoder and the second encoder; calculating a dynamic reverse gap value and a gap value fluctuation characteristic under the current test node; judging whether the gap value fluctuation characteristic of the current test node meets a preset condition or not, and if not, dynamically adjusting the test condition parameter of the next test node; and repeating the steps until all the test nodes are tested, and outputting the dynamic reverse gap value under each test node for comprehensive performance evaluation of the worm gear speed reducer. According to the invention, the accuracy of gap value detection is improved through low-speed micro-amplitude reciprocating motion test, so that the reliability of the comprehensive performance test result of the speed reducer is improved, and the test efficiency and precision are both considered.

Description

Technical Field

[0001] This invention relates to the field of speed reducer technology, and in particular to a method for comprehensive performance testing of worm gear reducers based on a speed reducer comprehensive test bench. Background Technology

[0002] Worm gear reducers are widely used in precision transmission systems across various industrial automation equipment, machinery, conveying equipment, and packaging machinery due to their compact structure, large transmission ratio, and good self-locking properties. Their transmission accuracy, especially backlash, is a key indicator for evaluating the reducer's performance and control precision. Backlash is defined as the maximum angular displacement that the output shaft can produce when the input shaft is fixed. Excessive backlash can lead to inaccurate system positioning, vibration, and noise, severely affecting the equipment's dynamic response characteristics and final machining or positioning accuracy.

[0003] Currently, the measurement of backlash in worm gear reducers mainly relies on traditional static testing methods. This standard method typically requires: rigidly fixing the reducer housing, connecting and fixing its output shaft to a rigid arm or encoder, or applying a small and constant reverse torque, and then sequentially applying a specified calibrated torque in both the forward and reverse rotational directions of the input shaft. The backlash is calculated by measuring the difference in angular displacement of the input shaft under the action of the two torques.

[0004] However, the above methods typically have the following inherent drawbacks: First, the testing process requires manual or specialized tooling to physically fix or load the output shaft. Clamping, alignment, loading, and unloading are time-consuming and labor-intensive, making rapid online testing difficult, resulting in low testing efficiency and cumbersome operation. Second, the rigidity of the test bench, the clearance of the fixing device, the smoothness of torque application, and the accuracy of the measurement reference can all introduce significant errors, making the testing accuracy highly susceptible to human and tooling factors. Third, traditional static testing is conducted at extremely low speeds or in a stationary state, failing to capture the dynamic backlash characteristics of the reducer during actual operation due to factors such as inertia, vibration, temperature rise, and oil film changes. The static backlash value differs from the dynamic backlash at the moment of motion direction switching in the actual servo control system, leading to unsatisfactory compensation effects in the control system based on static test data. This method typically only provides a single backlash value under a specific torque, failing to obtain a complete curve of backlash variation with load torque, and thus cannot accurately assess the accuracy and stability of the reducer under different load conditions. Summary of the Invention

[0005] To address at least one of the aforementioned technical problems, this invention provides a comprehensive performance testing method for worm gear reducers based on a comprehensive reducer test bench.

[0006] In a first aspect, the present invention provides a method for comprehensive performance testing of a worm gear reducer based on a comprehensive reducer test bench, the method comprising: Step S1: Install the first encoder and the second encoder coaxially on the input shaft and output shaft of the worm gear reducer to be tested, respectively; divide the test process into multiple test nodes, and set the first test condition parameters for the current test node. The first test condition parameters include the input shaft speed, load torque, and data sampling frequency. Step S2: Control the drive motor to drive the input shaft of the worm gear reducer to perform positive and negative reciprocating motion under the first test condition parameters corresponding to the current test node, and simultaneously collect the input shaft position signal of the first encoder and the output shaft position signal of the second encoder. Step S3: Calculate the dynamic angular displacement hysteresis of the output shaft position signal relative to the input shaft position signal, and determine the dynamic backlash value and gap value fluctuation characteristics under the current test node. Step S4: Determine whether the gap value fluctuation characteristics of the current test node meet the preset conditions. If they meet the conditions, use the first test condition parameter as the test condition parameter of the next test node. If they do not meet the conditions, dynamically adjust the test condition parameter of the next test node. Step S5: Repeat steps S2 to S4 until all test nodes are completed, output the dynamic backlash value under each test node, and use the dynamic backlash value under each test node for comprehensive performance evaluation of the worm gear reducer.

[0007] Secondly, the present invention also provides a comprehensive performance testing system for worm gear reducers based on a comprehensive reducer testing bench, the system comprising: The test node division unit is used to perform step S1, which involves coaxially installing the first encoder and the second encoder on the input shaft and the output shaft of the worm gear reducer under test, respectively; dividing the test process into multiple test nodes; and setting the first test condition parameters for the current test node, wherein the first test condition parameters include the input shaft speed, the load torque, and the data sampling frequency. The position signal acquisition unit is used to execute step S2, control the drive motor to drive the input shaft of the worm gear reducer to perform positive and negative reciprocating motion under the first test condition parameters corresponding to the current test node, and synchronously acquire the input shaft position signal of the first encoder and the output shaft position signal of the second encoder. The backlash value calculation unit is used to execute step S3, calculate the dynamic angular displacement hysteresis of the output shaft position signal relative to the input shaft position signal, and determine the dynamic backlash value and gap value fluctuation characteristics under the current test node. The test dynamic adjustment unit is used to execute step S4 and determine whether the gap value fluctuation characteristics of the current test node meet the preset conditions. When the conditions are met, the first test condition parameter is used as the test condition parameter of the next test node. When the conditions are not met, the test condition parameter of the next test node is dynamically adjusted. The comprehensive performance evaluation unit is used to execute step S5, repeat steps S2 to S4 until all test nodes are completed, output the dynamic backlash value under each test node, and use the dynamic backlash value under each test node for the comprehensive performance evaluation of the worm gear reducer.

[0008] Thirdly, the present invention also provides an electronic device including a processor and a memory, the memory being used to store computer program code, the computer program code including computer instructions, wherein when the processor executes the computer instructions, the electronic device performs the method as described in the first aspect above and any possible implementation thereof.

[0009] Fourthly, the present invention also provides a computer-readable storage medium storing a computer program, the computer program including program instructions that, when executed by a processor of an electronic device, cause the processor to perform a method as described in the first aspect above and any possible implementation thereof.

[0010] Compared with the prior art, the beneficial effects of the present invention are as follows: 1) Traditional static testing introduces errors at the moment torque is applied due to mechanism deformation and clearance compression. This method, through low-speed, micro-amplitude reciprocating motion, makes the system approach "quasi-static" testing, greatly reducing measurement errors caused by inertial forces and impacts. Simultaneously, by analyzing the fluctuation characteristics of the clearance value, rather than a single absolute value, it can identify uneven backlash caused by tooth surface damage, assembly defects, etc., which is impossible with single-point measurements. This significantly improves the accuracy and reliability of the test, enabling the detection of potential, intermittent quality problems.

[0011] 2) By dividing the system into multiple test nodes and dynamically adjusting the test condition parameters of the next test node based on fluctuation characteristics, the system's "decision-making" capability is tested. For example, when an abnormal backlash fluctuation is detected at a certain node, a more refined detection mode can be automatically triggered: reducing the rotational speed of the next node to obtain a clearer signal, increasing the sampling frequency to capture more subtle fluctuations, and applying a load to observe the changes in the backlash under stress. This automated process of "identifying the problem - focusing on the problem - in-depth diagnosis" replaces the trial-and-error process relying on human experience, achieving a balance between high efficiency and high precision.

[0012] 3) This solution obtains a multi-dimensional performance dataset by conducting tests at multiple test nodes under different speeds and loads. A "backlash-speed" curve can be plotted to evaluate dynamic stability, and a "backlash-load" curve can be plotted to evaluate rigidity and meshing quality. Abnormal fluctuations in backlash at a specific speed may indicate a dynamic balance problem; abnormal increases in backlash after load increases may indicate a serious meshing defect. This allows the method not only to determine product qualification but also to assess its performance level and even achieve early fault diagnosis, providing data support for quality improvement and predictive maintenance.

[0013] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0014] To more clearly illustrate the technical solutions in the embodiments of the present invention or the background art, the accompanying drawings used in the embodiments of the present invention or the background art will be described below.

[0015] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this disclosure and, together with the specification, serve to illustrate the technical solutions of this disclosure.

[0016] Figure 1 A flowchart illustrating a comprehensive performance testing method for a worm gear reducer based on a comprehensive reducer test bench, provided as an embodiment of the present invention; Figure 2 This is a schematic diagram of a comprehensive performance testing system for worm gear reducers based on a reducer comprehensive test bench, provided as an embodiment of the present invention. Detailed Implementation

[0017] To enable those skilled in the art to better understand the present invention, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0019] Please see Figure 1 , Figure 1 This is a flowchart illustrating a comprehensive performance testing method for a worm gear reducer based on a comprehensive reducer test bench, as provided in an embodiment of the present invention. Figure 1 As shown, the method includes: Step S1: Install the first encoder and the second encoder coaxially on the input shaft and output shaft of the worm gear reducer under test, respectively; divide the test process into multiple test nodes, and set the first test condition parameters for the current test node. The first test condition parameters include the input shaft speed, load torque, and data sampling frequency.

[0020] The worm gear reducer under test is rigidly fixed on the test bench. High-precision encoders, namely the first encoder and the second encoder, are coaxially mounted on its input and output shafts, respectively. An encoder is a precision sensor that converts mechanical rotational displacement or angle into an electrical signal. Its core function is to accurately measure motion parameters such as the rotational speed, angle, and direction of the shaft, providing quantitative data support for the test. After installation, ensure that the drive motor, encoder, data acquisition card, etc., are correctly connected to the system controller.

[0021] Existing testing methods typically involve static testing under a fixed procedure. This method usually requires: rigidly fixing the reducer housing and connecting and fixing its output shaft to a rigid arm or encoder, or applying a small, constant reverse torque. Then, a specified calibrated torque is applied sequentially in both the forward and reverse rotational directions of the input shaft. Backlash is calculated by measuring the difference in angular displacement of the input shaft under the two torques. Thus, the testing process requires manual or specialized tooling to physically fix or load the output shaft. Clamping, alignment, loading, and unloading are time-consuming and labor-intensive, making rapid online testing difficult, resulting in low testing efficiency, cumbersome operation, and a high risk of measurement errors.

[0022] To address this issue, in step S1, after installing the encoder, the testing process is divided into multiple test nodes, allowing for test state planning based on different test objectives. For example: Node 1 (Quick Screening under No-Load): Set a high speed (e.g., 10 RPM), zero load, and standard sampling rate. The goal is to quickly obtain the basic backlash value for preliminary screening.

[0023] Node 2 (Load Performance Verification): Set a medium speed (e.g., 5 RPM), apply 50% of the rated load, and increase the sampling rate. The goal is to verify the gearbox's backlash retention capability under load.

[0024] Node 3 (Fine Diagnosis): Set extremely low speed (e.g., 0.5 RPM), apply a series of gradient loads (e.g., 25%, 75%, 100% of rated load), and the highest sampling rate to accurately plot the "backlash-load" characteristic curve and perform in-depth diagnosis.

[0025] On the other hand, this embodiment takes into account that if only a single test condition parameter is used, it will lead to many invalid test processes and waste manpower and resources. Therefore, the benefits of multiple test nodes also include that, starting from a certain test node, after setting the parameters of that node and completing the test, the test results under that node can be evaluated. If the requirements are met, the original parameters can be maintained to set the condition parameters for the next test node. However, if the requirements are not met, the test status of the next node can be dynamically adjusted, thereby avoiding the situation of wasting resources due to invalid tests.

[0026] Step S2: Control the drive motor to drive the input shaft of the worm gear reducer to perform positive and negative reciprocating motion under the first test condition parameters corresponding to the current test node, and simultaneously collect the input shaft position signal of the first encoder and the output shaft position signal of the second encoder.

[0027] The system controller instructs the drive motor to rotate the input shaft at a precise and uniform reciprocating speed set at the current node. Preferably, the reciprocating motion is a triangular wave motion or a sine wave motion of forward rotation-reverse rotation-forward rotation, with a speed range of 0.1 RPM to 10 RPM. This speed range setting indicates that the test is conducted under low-speed conditions. The low speed is intended to approximate "quasi-static" test conditions, greatly eliminating the influence of inertial forces and measuring the actual mechanical backlash rather than the dynamic response. The reciprocating motion can repeatedly excite the backlash phenomenon, providing a data basis for subsequent analysis of fluctuation characteristics. The position signals of the input and output shafts are acquired synchronously. This requires not only a synchronous acquisition card in the hardware but also precise timestamp alignment in the software. This is a prerequisite for subsequent accurate phase difference or time delay analysis and calculation of "dynamic angular displacement hysteresis".

[0028] Step S3: Calculate the dynamic angular displacement hysteresis of the output shaft position signal relative to the input shaft position signal, and determine the dynamic backlash value and gap value fluctuation characteristics under the current test node.

[0029] In this step, the calculation of dynamic angular displacement hysteresis is the core of this scheme, and it is not simply a matter of reading the difference between a maximum and minimum value. The system performs advanced signal processing on the synchronously acquired encoder signals, including cross-correlation analysis or instantaneous phase demodulation. For example, by calculating the position of the peak of the cross-correlation function, the time hysteresis of the output shaft relative to the input shaft when the direction of motion changes can be accurately obtained. Multiplying this time difference by the instantaneous velocity yields a very high-precision dynamic angular displacement hysteresis, i.e., the dynamic backlash value of this node. To extract the backlash value fluctuation characteristics, the system does not only take a single average hysteresis value, but analyzes all instantaneous backlash values ​​calculated in one or more reciprocating motions. "Backlash value fluctuation characteristics" include, but are not limited to: standard deviation, reflecting the stability of the backlash; the difference between the maximum and minimum values, reflecting extreme cases; and specific frequency components, such as fluctuations appearing at the meshing frequency, which may indicate a defect in a tooth. In this way, a single backlash value can be expanded into a feature vector containing rich quality information for the evaluation of test results in the subsequent step.

[0030] Step S4: Determine whether the gap value fluctuation characteristics of the current test node meet the preset conditions. If they do, use the first test condition parameter as the test condition parameter of the next test node. If they do not meet the conditions, dynamically adjust the test condition parameter of the next test node.

[0031] This step aims to determine whether the test condition parameters for the next test node need adjustment based on the current test node. If the conditions are met, it indicates that the current node test has passed and performance is stable. The system will then maintain the current test condition parameters or make minor adjustments according to a preset template for the next node test, conserving resources. If the conditions are not met, for example, if the fluctuation characteristics show an excessively large standard deviation, indicating backlash instability, the system will dynamically adjust the parameters of the next node for focused diagnostics.

[0032] In one embodiment, dynamically adjusting the test condition parameters of the next test node when the conditions are not met includes: If the standard deviation of the dynamic backlash value calculated by the current test node exceeds the first preset threshold, the data sampling frequency of subsequent test nodes will be increased. If the average value of the dynamic backlash calculated at the current test node exceeds the second preset threshold, then in the subsequent test node, a preset load torque is applied to the output shaft, and the dynamic backlash value under load is retested.

[0033] The system has completed the testing of the current test node and calculated two key results: the average value of the dynamic backlash, reflecting the overall size of the backlash under the test conditions; and the standard deviation of the dynamic backlash, reflecting the degree of fluctuation or dispersion of the backlash values ​​obtained from multiple measurements during the test.

[0034] A high standard deviation indicates unstable backlash values, with significant fluctuations in the measured value within a single reciprocating motion cycle. This typically suggests non-uniform defects, such as localized wear or pitting on a tooth of a worm gear or worm, damage to a ball or raceway in a rolling bearing, or minor foreign object interference. These defects only cause abnormal backlash changes when the machine rotates to a specific position. Therefore, when the standard deviation of the dynamic backlash value calculated at the current test node exceeds a first preset threshold, the adaptive action is to increase the data sampling frequency of the next test node. This is because, at the current test node, the sampling frequency may not be high enough; the system may only "detect" fluctuations but cannot accurately pinpoint the specific angle of the fluctuation or clearly capture the waveform details. Increasing the sampling frequency allows the system to collect a denser number of data points within the same motion cycle. This is equivalent to using a "higher frame rate camera" to capture the fault phenomenon. This enables the system to correlate abnormal backlash fluctuations with the precise angular coordinates of the input / output axes, pinpointing the physical location of the fault. At the same time, it obtains smoother and more realistic fluctuation curves, which facilitates the analysis of fluctuation patterns, determines whether they are spikes, gradual changes or periodic oscillations, and provides a basis for judging defect types such as wear, impact, and eccentricity.

[0035] An excessively large average backlash measured under no-load or low-load conditions directly indicates that the overall meshing clearance of the gear pair has exceeded the allowable range. However, this is usually only a preliminary conclusion, and a more accurate judgment requires further consideration of how this clearance changes under actual working loads. Therefore, if the average value of the dynamic backlash calculated at the current test node exceeds the second preset threshold, a preset load torque is applied to the output shaft at the next test node, and the test is repeated. For worm gear reducers, load significantly affects their backlash. After applying a load, the tooth meshing becomes tighter, and some bearing clearance is eliminated, which usually leads to a reduction in backlash. The purpose of this action is to observe the decreasing trend of backlash with increasing load. If the backlash can decrease rapidly and stabilize at an acceptable low value, it indicates that the reducer's rigidity is acceptable, and a large no-load backlash may be due to insufficient design allowance or insufficient bearing preload. If the backlash reduction is not significant, it may mean that the rigidity is severely insufficient or there is severe wear. Determine whether the reducer is "usable". Even if the no-load backlash exceeds the standard, if its working backlash meets the requirements under a certain load, it may still be suitable for some applications that do not require high positioning accuracy. Conversely, if the back gap is still large after loading, it is judged as a defective product.

[0036] In this embodiment, the system moves beyond mere "measurement" to "analysis." Through the logical chain of "high standard deviation → increased sampling frequency," the system automatically transitions from "anomaly detection" to "in-depth diagnosis," proactively locating and characterizing fault features, significantly improving the detection capability and diagnostic accuracy of latent defects. By employing the strategy of "high average value and applied load," test results are transformed from a simple "pass / fail" criterion into a performance evaluation with practical engineering guidance, providing data support for quality assessment, product grading, and downgrading. The entire judgment and adjustment process is automatically completed by preset algorithms and thresholds, completely eliminating the uncertainty and subjectivity caused by human intervention, ensuring high consistency and comparability of test results from different times and batches.

[0037] Step S5: Repeat steps S2 to S4 until all test nodes are completed, output the dynamic backlash value under each test node, and use the dynamic backlash value under each test node for comprehensive performance evaluation of the worm gear reducer.

[0038] Repeat steps S2-S4 until all preset node tests are completed. Output the dynamic backlash value for each node. This step is the ultimate goal. It means that the system output is not a bunch of isolated data, but a diagnostic report. For example, plot the data of each node as a "backlash-load" curve and a "backlash fluctuation-speed" curve. Based on the linearity, smoothness, stability, and other characteristics of the curves, classify the performance of the reducer as excellent, good, medium, or poor. Combined with the fluctuation characteristics, give a preliminary diagnostic conclusion, such as "backlash value is qualified, but fluctuation is large at high speed, dynamic balancing is recommended" or "backlash is abnormally increased under a certain load, suspected poor meshing". Ultimately, the entire process of "measurement-evaluation-diagnosis" is realized, truly achieving comprehensive performance testing.

[0039] Therefore, this embodiment constructs a test framework through "multi-node planning", acquires high-quality data through "low-speed reciprocating motion and synchronous acquisition", deeply mines performance information through "dynamic gap and feature extraction", realizes an intelligent diagnostic process through "feature-based parameter adaptive adjustment", and finally outputs conclusions with high engineering value through "comprehensive performance evaluation".

[0040] In one embodiment, calculating the dynamic angular displacement hysteresis of the output shaft position signal relative to the input shaft position signal includes: Calculate the cross-correlation function between the input axis position signal and the output axis position signal, and determine the time offset corresponding to the peak value of the cross-correlation function; Multiply the time offset corresponding to the peak value by the average angular velocity in the positive and negative reciprocating motion to obtain the dynamic angular displacement hysteresis.

[0041] The core principle of this embodiment is to use a cross-correlation function to accurately measure the time delay between two signals, namely the input shaft and output shaft position signals. Ideally, if the reducer has no backlash, the output shaft position signal should be a perfect, instantaneous replica of the input shaft position signal, only with its amplitude reduced according to the reduction ratio. In this case, the peak of the cross-correlation function of the two signals appears at the position where the time offset is zero. When backlash exists, after the input shaft changes direction, it needs to "traverse" this backlash before driving the output shaft to start moving. This results in a small time lag between the output shaft signal and the input shaft signal each time it turns. This lag time contains the backlash information. Multiplying this lag time by the average angular velocity during the motion process directly yields the dynamic angular displacement lag caused by the backlash, which is precisely the angular displacement equivalent of the backlash.

[0042] Specifically, the calculation process is as follows: 1) Synchronous Signal Acquisition: The system controls the drive motor to perform low-speed reciprocating motion and synchronously acquires the position signals from two encoders mounted on the input and output shafts with high precision. Ensuring that the two signals have strictly synchronized timestamps is the basis for accurate subsequent calculations.

[0043] 2) Signal preprocessing: Perform necessary preprocessing on the acquired raw position signals, such as filtering to eliminate high-frequency noise, thereby improving the signal-to-noise ratio and providing clean data for cross-correlation analysis.

[0044] 3) Calculate the cross-correlation function: The system calculates the cross-correlation function for the two preprocessed position signals, namely the input axis signal and the output axis signal. This function describes the degree of similarity between the two signals at different time offsets.

[0045] 4) Finding the peak time offset: The calculated cross-correlation function is a function of time offset. The system will precisely find the global maximum point of this function. The x-coordinate value corresponding to this maximum point is the overall time delay of the output axis signal relative to the input axis signal. The process of finding the peak is usually implemented through interpolation algorithms to achieve a time resolution higher than the sampling interval.

[0046] 5) Calculate the average angular velocity: Based on the time series of the input shaft position signal, calculate the average angular velocity during the current reciprocating motion cycle.

[0047] 6) Solve for dynamic angular displacement hysteresis: Multiply the found peak time offset by the average angular velocity to obtain the high-precision dynamic angular displacement hysteresis, which is the dynamic backlash value calculated under the current test node.

[0048] Compared to traditional methods such as direct subtraction of maximum and minimum values, the cross-correlation algorithm provided in this embodiment has a natural ability to suppress random noise. Even if the encoder signal contains significant noise, as long as the "waveform transition feature" caused by the backlash in the two signals is correlated, the cross-correlation function can clearly extract it from the noisy background and accurately find the delay time. This makes it possible to achieve high-precision measurement in harsh industrial environments. The time delay resolution of this method is no longer limited by the encoder's own single-cycle resolution or sampling interval. Through algorithms such as peak interpolation, sub-pixel level time offset estimation can be obtained, thereby improving the measurement accuracy of angular displacement hysteresis to a level far exceeding the encoder's original resolution. This is crucial for measuring minute backlashes.

[0049] Furthermore, this method does not isolate the maximum and minimum positions of the input and output shafts, but analyzes the relationship between the entire motion waveform. It reflects the average effect of the entire dynamic process from the start of steering to the re-establishment of transmission, resulting in a more realistic picture and a better representation of the reducer's overall dynamic performance. Moreover, this calculation method does not rely on precisely identifying the "instantaneous" change in motion direction, avoiding errors caused by signal jitter or improper threshold settings. It utilizes information from the entire signal segment, making it insensitive to errors or glitches at individual points, resulting in more stable and reliable measurement results. By analyzing the fluctuations in the lag times calculated across different motion cycles, the "backlash fluctuation characteristics" can be extracted more reliably. Because each lag time is already the result of noise suppression, the calculated fluctuation characteristics more accurately reflect the stability of the mechanical state, rather than measurement noise.

[0050] In one embodiment, the dynamic angular displacement hysteresis of the output shaft position signal relative to the input shaft position signal includes: Perform Hilbert transforms on the input axis position signal and the output axis position signal respectively to construct the corresponding input axis analytic signal and output axis analytic signal; Calculate the instantaneous phase of the input axis analytic signal and the output axis analytic signal to determine the instantaneous phase difference; Extract the steady-state step difference value when the instantaneous phase difference reverses the direction of the input shaft motion, and use the steady-state step difference value as the dynamic angular displacement hysteresis.

[0051] The core principle of this embodiment is to use the Hilbert transform to construct the analytical form of the signal, thereby extracting the instantaneous phase of the signal at any given moment. By comparing the instantaneous phase difference between the input and output shaft signals, the phase abrupt change caused by backlash can be directly and continuously observed. The position signal output by the encoder can be considered as a signal whose angle changes with time. The Hilbert transform can convert a real signal into a complex analytic signal, the phase of which is the instantaneous phase of the signal, which increases monotonically with time, accurately reflecting the rotation angle of the shaft. In the ideal case without backlash, the instantaneous phase difference between the input and output shafts should be a constant value, determined by the reduction ratio. When backlash exists and the input shaft changes direction, the output shaft will temporarily stop, causing its instantaneous phase to stop changing, while the phase of the input shaft continues to change. This will instantly widen the phase difference between the two, forming a step. The stable difference of this step is the dynamic angular displacement hysteresis caused by the backlash.

[0052] Specifically, the calculation process is as follows: 1) Synchronous acquisition and preprocessing of signals: Similar to the cross-correlation method, it is first necessary to acquire the position signals of the input shaft and output shaft encoders with high precision and perform preprocessing such as filtering to reduce noise.

[0053] 2) Constructing the analytical signal: Perform a Hilbert transform on the preprocessed input axis position signal to obtain its orthogonal components; construct the analytical signal of the input axis based on the input axis position signal and the orthogonal components, and similarly construct the analytical signal of the output axis. 3) Calculate instantaneous phase and phase difference: Calculate the instantaneous phase of the input axis and the instantaneous phase of the output axis; then calculate the instantaneous phase difference; 4) Extracting the steady-state step difference: The system identifies the zero-crossing point of the differential of the input shaft position signal, i.e., the moment when the motion direction reverses. Near these reversal moments, observe the change in instantaneous phase difference. It will exhibit a clear step response: before reversal, the instantaneous phase difference is basically stable; at the moment of reversal, the instantaneous phase difference begins to increase linearly; when the backlash is eliminated and the output shaft begins to follow the motion, the instantaneous phase difference will stabilize at a new, larger value. Extract the steady-state difference of this step, i.e., the phase difference between the two stable platforms before and after the step; this value is the dynamic angular displacement hysteresis to be measured. It usually needs to be converted to the output shaft side based on the reduction ratio.

[0054] This embodiment provides continuous, real-time phase difference variation curves. This allows us not only to obtain the final size of the backlash but also to fully observe the entire dynamic process of the backlash being "absorbed," including response time and transient characteristics, providing unprecedented detail for studying the dynamic behavior of transmission systems. Since instantaneous phase is an inherent property of the signal, this method is extremely sensitive to minute angular displacement changes. Even with very small backlash, the resulting phase step can be clearly captured and quantified, giving it a significant advantage in measuring high-precision gearboxes. The Hilbert transform and instantaneous phase calculation primarily rely on the signal's phase information and are insensitive to changes in signal amplitude, such as those caused by encoder mounting eccentricity or signal strength fluctuations. This is more robust than some analysis methods that rely on signal amplitude.

[0055] In one embodiment, the method further includes: Perform spectrum analysis on the output shaft position signal of the second encoder; Determine whether any abnormal peaks are found in the spectrum of the current test node at the meshing frequency or at harmonics of the meshing frequency; If not, use the first test condition parameter as the test condition parameter for the next test node; If so, reduce the input shaft speed in the next test node.

[0056] Any specific fault in rotating machinery will generate specific frequency components in the vibration or motion signal. For worm gear reducers, the core is the meshing of the worm and worm wheel. If the worm wheel or worm has defects such as tooth surface wear, pitting, or eccentricity, an impact will occur when the meshing point passes through the defective area, resulting in abnormally high peaks at the meshing frequency and its harmonics in the signal spectrum. At higher speeds, the fault impact signal may be submerged in noise or other vibrations. Reducing the speed is equivalent to "slowing down" the fault characteristics, which has the following benefits: Improved signal-to-noise ratio: Reducing the speed decreases the inherent vibration noise of the system, making the weak fault impact signal more prominent. Separation of frequency components: All speed-related frequencies are proportionally reduced, making them easier to distinguish in the spectrum and avoiding confusion with other high-frequency noise.

[0057] Specifically, in implementing this embodiment, after a certain test node is completed, in addition to calculating the backlash value, the system performs a Fast Fourier Transform on the high-resolution position signal of the second encoder, converting it from the time domain to the frequency domain to obtain the spectrum of the output shaft position signal. Based on the input shaft speed of the current test node, the system automatically calculates the theoretical meshing frequency and its main harmonic positions. The algorithm sets a narrow-band monitoring window near these theoretical frequency points on the spectrum. The system determines whether the peak amplitude within these windows exceeds a threshold set according to the background noise level. If it does, it is determined that "abnormal peaks exist." If there are no abnormal peaks, it indicates that the meshing state of the worm gear pair is good under the current test conditions, and no obvious defects are found. The system then follows the conventional process, transmitting the current test parameters or fine-tuning them according to a preset template to the next node. If abnormal peaks are found: the system triggers diagnostic mode. In the next test node, the input shaft speed is actively reduced, and the aforementioned steps are repeated. After reducing the speed, the system collects data again and generates a spectrum, observing whether the abnormal peaks still exist and are clearer at the new, lower meshing frequency, thereby confirming the fault and assessing its severity.

[0058] This embodiment demonstrates that the method goes beyond simply measuring a backlash value; it proactively diagnoses potential defects in the worm gear replica, enabling predictive maintenance. Through a closed-loop strategy of "detecting suspected anomalies and reducing speed to focus on diagnosis," the signal-to-noise ratio of fault characteristics is effectively improved, avoiding misdiagnosis or missed diagnosis due to signal aliasing or noise interference at high speeds, resulting in more accurate and reliable diagnostic conclusions.

[0059] In one embodiment, the method further includes: Plot the backlash-load torque characteristic curve of the worm gear reducer with the load torque as the abscissa and the corresponding dynamic backlash value as the ordinate. Feature extraction is performed on the back clearance-load torque characteristic curve to obtain at least one feature parameter, which includes at least one of the following: initial back clearance value, back clearance change rate, back clearance stability value, curve linearity, and hysteresis area. The characteristic parameters are compared with the preset performance level standards, and the comparison results are used for the comprehensive performance evaluation of the worm gear reducer.

[0060] The performance of a worm gear reducer is not determined by a single static backlash value, but rather by its dynamic characteristics of backlash variation with load. A high-quality reducer should exhibit a well-defined pattern in its backlash variation with load. Plotting a "backlash-load" curve visualizes the mapping relationship between load torque and dynamic backlash. The shape of this curve directly reflects the reducer's core performance indicators such as rigidity, meshing quality, and preload effect. Numerical characteristic parameters extracted from the curve transform the complex curve shape into comparable objective indicators. Each parameter points to a specific performance dimension. Comparing the extracted parameters with a database based on extensive experiments and industry standards allows for the classification of reducer performance into different levels, such as excellent, good, medium, and poor, or a precise judgment on whether it meets the requirements of a specific application scenario.

[0061] The system measures the dynamic backlash value at multiple test points, such as no-load, 25% rated load, 50%, 75%, and 100% load. Then, points are plotted on a coordinate system with the load torque as the x-axis and the corresponding dynamic backlash value as the y-axis. The system uses curve fitting algorithms, such as the least squares method, to fit these discrete points into a smooth backlash-load torque characteristic curve. This curve visually demonstrates the overall trend of backlash variation with increasing load.

[0062] Then, extract one or more key feature parameters from the plotted characteristic curves, including: Initial backlash value: The backlash value of the curve when the load is zero or close to zero. It reflects the basic meshing clearance of the reducer under no-load conditions.

[0063] Backlash variation rate: The slope of the curve in the main load range (e.g., 0-50% rated load). It reflects the rigidity of the reducer. The smaller the variation rate (the flatter the curve), the better the rigidity and the less the load affects the accuracy.

[0064] Backlash stability value: The value at which the backlash no longer decreases significantly after the load increases to a certain extent (e.g., exceeding 80% of the rated load). It reflects the final accuracy level of the reducer under heavy load.

[0065] Curve linearity: measures the degree of deviation of the actual curve from an ideal fitted straight line. The better the linearity, the more predictable the accuracy of the reducer across the entire load range.

[0066] Hysteresis area: If curves are plotted separately during loading and unloading, the two curves will form a hysteresis loop. The hysteresis area reflects the nonlinear characteristics of friction, elastic deformation, and clearance recovery in the transmission system; the smaller the area, the better the performance.

[0067] The system has a pre-set performance level standard database. This database contains the acceptable range or grading thresholds (excellent, good, average, poor) for the aforementioned characteristic parameters for different models and precision levels of speed reducers. The system automatically compares the extracted characteristic parameters with the standards in the database. Finally, it generates a comprehensive performance evaluation report. The report not only provides a "pass / fail" conclusion but also the performance level.

[0068] Thus, through the above embodiments, by extracting multiple characteristic parameters, a comprehensive quantitative evaluation of the reducer is conducted from multiple dimensions, including static accuracy (initial value), rigidity (rate of change), heavy-load accuracy (stable value), accuracy consistency (linearity), and transmission efficiency / hysteresis (hysteresis area), which is far superior to judging based on a single backlash value. Based on the evaluation results of the quantitative parameters, accurate quality grading can be performed on the product. Performance evaluation is entirely based on data and algorithms, completely eliminating the influence of subjective human factors and ensuring highly consistent, fair, and comparable results from evaluations conducted by different batches, at different times, and by different operators.

[0069] See Figure 2 In one embodiment, the present invention also provides a comprehensive performance testing system for worm gear reducers based on a reducer comprehensive test bench, the system comprising: The test node division unit 100 is used to perform step S1, which involves coaxially installing a first encoder and a second encoder on the input shaft and output shaft of the worm gear reducer under test, respectively; dividing the test process into multiple test nodes; and setting the first test condition parameters for the current test node, wherein the first test condition parameters include the input shaft speed, load torque, and data sampling frequency. The position signal acquisition unit 200 is used to execute step S2, control the drive motor to drive the input shaft of the worm gear reducer to perform positive and negative reciprocating motion under the first test condition parameters corresponding to the current test node, and synchronously acquire the input shaft position signal of the first encoder and the output shaft position signal of the second encoder. The backlash value calculation unit 300 is used to execute step S3, calculate the dynamic angular displacement hysteresis of the output shaft position signal relative to the input shaft position signal, and determine the dynamic backlash value and gap value fluctuation characteristics under the current test node. The test dynamic adjustment unit 400 is used to execute step S4, determine whether the gap value fluctuation characteristics of the current test node meet the preset conditions, and when they meet the conditions, use the first test condition parameter as the test condition parameter of the next test node; when they do not meet the conditions, dynamically adjust the test condition parameter of the next test node. The comprehensive performance evaluation unit 500 is used to execute step S5, repeat steps S2 to S4 until all test nodes are completed, output the dynamic backlash value under each test node, and use the dynamic backlash value under each test node for the comprehensive performance evaluation of the worm gear reducer.

[0070] It is understood that the system provided in this embodiment has functions or includes modules that can be used to execute the methods described in the above method embodiments. The specific implementation can be referred to the description of the above method embodiments, and for the sake of brevity, it will not be repeated here.

[0071] The present invention also provides an electronic device including a processor and a memory, the memory being used to store computer program code, the computer program code including computer instructions, wherein when the processor executes the computer instructions, the electronic device performs a method as described in any of the above possible implementations.

[0072] The present invention also provides a computer-readable storage medium storing a computer program, the computer program including program instructions that, when executed by a processor of an electronic device, cause the processor to perform a method as described in any of the above possible implementations.

[0073] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

Claims

1. A method for comprehensive performance testing of worm gear reducers based on a comprehensive reducer test bench, characterized in that, The method includes: Step S1: Install the first encoder and the second encoder coaxially on the input shaft and output shaft of the worm gear reducer to be tested, respectively; divide the test process into multiple test nodes, and set the first test condition parameters for the current test node. The first test condition parameters include the input shaft speed, load torque, and data sampling frequency. Step S2: Control the drive motor to drive the input shaft of the worm gear reducer to perform positive and negative reciprocating motion under the first test condition parameters corresponding to the current test node, and simultaneously collect the input shaft position signal of the first encoder and the output shaft position signal of the second encoder. Step S3: Calculate the dynamic angular displacement hysteresis of the output shaft position signal relative to the input shaft position signal, and determine the dynamic backlash value and gap value fluctuation characteristics under the current test node. Step S4: Determine whether the gap value fluctuation characteristics of the current test node meet the preset conditions. If they meet the conditions, use the first test condition parameter as the test condition parameter of the next test node. If they do not meet the conditions, dynamically adjust the test condition parameter of the next test node. Step S5: Repeat steps S2 to S4 until all test nodes are completed, output the dynamic backlash value under each test node, and use the dynamic backlash value under each test node for comprehensive performance evaluation of the worm gear reducer.

2. The comprehensive performance testing method for worm gear reducers based on a comprehensive reducer test bench according to claim 1, characterized in that, The step of dynamically adjusting the test condition parameters for the next test node when the conditions are not met includes: If the standard deviation of the dynamic backlash value calculated by the current test node exceeds the first preset threshold, the data sampling frequency of subsequent test nodes will be increased. If the average value of the dynamic backlash calculated at the current test node exceeds the second preset threshold, then in the subsequent test node, a preset load torque is applied to the output shaft, and the dynamic backlash value under load is retested.

3. The comprehensive performance testing method for worm gear reducers based on a comprehensive reducer test bench according to claim 1, characterized in that, The calculation of the dynamic angular displacement hysteresis of the output shaft position signal relative to the input shaft position signal includes: Calculate the cross-correlation function between the input axis position signal and the output axis position signal, and determine the time offset corresponding to the peak value of the cross-correlation function; Multiply the time offset corresponding to the peak value by the average angular velocity in the positive and negative reciprocating motion to obtain the dynamic angular displacement hysteresis.

4. The comprehensive performance testing method for worm gear reducers based on a comprehensive reducer test bench according to claim 1, characterized in that, The dynamic angular displacement hysteresis of the output shaft position signal relative to the input shaft position signal includes: Perform Hilbert transforms on the input axis position signal and the output axis position signal respectively to construct the corresponding input axis analytic signal and output axis analytic signal; Calculate the instantaneous phase of the input axis analytic signal and the output axis analytic signal to determine the instantaneous phase difference; Extract the steady-state step difference value when the instantaneous phase difference reverses the direction of the input shaft motion, and use the steady-state step difference value as the dynamic angular displacement hysteresis.

5. The comprehensive performance testing method for worm gear reducers based on a comprehensive reducer test bench according to claim 1, characterized in that, The method further includes: Perform spectrum analysis on the output shaft position signal of the second encoder; Determine whether any abnormal peaks are found in the spectrum of the current test node at the meshing frequency or at harmonics of the meshing frequency; If not, use the first test condition parameter as the test condition parameter for the next test node; If so, reduce the input shaft speed in the next test node.

6. The comprehensive performance testing method for worm gear reducers based on a comprehensive reducer test bench according to claim 1, characterized in that, The method further includes: Plot the backlash-load torque characteristic curve of the worm gear reducer with the load torque as the abscissa and the corresponding dynamic backlash value as the ordinate. Feature extraction is performed on the back clearance-load torque characteristic curve to obtain at least one feature parameter, which includes at least one of the following: initial back clearance value, back clearance change rate, back clearance stability value, curve linearity, and hysteresis area. The characteristic parameters are compared with the preset performance level standards, and the comparison results are used for the comprehensive performance evaluation of the worm gear reducer.

7. The comprehensive performance testing method for worm gear reducers based on a comprehensive reducer test bench according to claim 1, characterized in that, The reciprocating motion is either a triangular wave or a sine wave, with a speed range of 0.1 RPM to 10 RPM.

8. A comprehensive performance testing system for worm gear reducers based on a comprehensive reducer testing bench, characterized in that, The system includes: The test node division unit is used to perform step S1, which involves coaxially installing the first encoder and the second encoder on the input shaft and the output shaft of the worm gear reducer under test, respectively; dividing the test process into multiple test nodes; and setting the first test condition parameters for the current test node, wherein the first test condition parameters include the input shaft speed, the load torque, and the data sampling frequency. The position signal acquisition unit is used to execute step S2, control the drive motor to drive the input shaft of the worm gear reducer to perform positive and negative reciprocating motion under the first test condition parameters corresponding to the current test node, and synchronously acquire the input shaft position signal of the first encoder and the output shaft position signal of the second encoder. The backlash value calculation unit is used to execute step S3, calculate the dynamic angular displacement hysteresis of the output shaft position signal relative to the input shaft position signal, and determine the dynamic backlash value and gap value fluctuation characteristics under the current test node. The test dynamic adjustment unit is used to execute step S4 and determine whether the gap value fluctuation characteristics of the current test node meet the preset conditions. When the conditions are met, the first test condition parameter is used as the test condition parameter of the next test node. When the conditions are not met, the test condition parameter of the next test node is dynamically adjusted. The comprehensive performance evaluation unit is used to execute step S5, repeat steps S2 to S4 until all test nodes are completed, output the dynamic backlash value under each test node, and use the dynamic backlash value under each test node for the comprehensive performance evaluation of the worm gear reducer.

9. An electronic device, characterized in that, include: The electronic device includes a processor and a memory, the memory being used to store computer program code, the computer program code including computer instructions, wherein when the processor executes the computer instructions, the electronic device performs the comprehensive performance testing method for a worm gear reducer based on a reducer comprehensive test bench as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, which includes program instructions. When executed by a processor of an electronic device, the program instructions cause the processor to perform the comprehensive performance testing method for a worm gear reducer based on a reducer integrated test bench as described in any one of claims 1 to 7.

Images