A control system for testing performance of a retarder

Abstract

This invention relates to the field of speed reducer performance testing technology, specifically to a speed reducer performance testing control system, including a host computer control platform, a control module, a drive module, a load module, and a data acquisition module. The host computer control platform provides a human-machine interface, configures test parameters, monitors the test process, and receives, processes, and displays test data. The control module is signal-connected to the host computer control platform and receives control commands from the platform to control the operation of the drive module and the load module. The drive module is signal-connected to the control module and drives the input end of the speed reducer under test according to the control commands. The load module is signal-connected to the control module and applies a controllable load to the output end of the speed reducer under test. The data acquisition module is signal-connected to the host computer control platform and acquires performance signals of the speed reducer under test in real time. This invention enables coordinated closed-loop control of the drive and load.

Description

Technical Field

[0001] This invention relates to the field of speed reducer performance testing technology, and more specifically to a control system for speed reducer performance testing. Background Technology

[0002] In the research and development and production of speed reducers (especially precision speed reducers such as RV reducers and harmonic reducers), comprehensive testing of multiple key performance indicators is required, including transmission efficiency, stiffness, transmission error, starting torque, durability, and temperature rise. Currently, traditional speed reducer testing systems suffer from the following technical shortcomings:

[0003] Separation of control and data acquisition, resulting in poor synchronization: Some existing systems use independent drive control cabinets, load control cabinets, and separate data acquisition instruments (such as torque meters, tachometers, and thermometers). The lack of a unified clock synchronization mechanism between these devices leads to mismatches in the timing of data acquisition for parameters such as torque, speed, and temperature during dynamic loading, making it impossible to accurately analyze performance changes under instantaneous operating conditions.

[0004] Lack of coordinated control between drive and load: Many test benches can only achieve constant speed drive or constant torque loading, making it difficult to simulate complex scenarios where speed and load fluctuate simultaneously in actual working conditions (such as robot joint start-up or vehicle uphill gear shifting). This leads to discrepancies between test results and actual operating performance.

[0005] Low testing efficiency and high human intervention: Traditional solutions often rely on operators to manually adjust the load, record data, and switch operating points, making it difficult to achieve automated sequence testing (such as automatically running a operating condition curve containing multiple speed-torque combinations). The testing cycle is long, repeatability is poor, and it is prone to introducing human error.

[0006] Incomplete condition monitoring: Some existing simple testing systems only focus on input and output torque and speed, neglecting the synchronous acquisition and analysis of housing temperature and vibration signals. However, excessively high temperature and abnormal vibration are often key indicators of early gearbox failures (such as lubrication failure, gear pitting, and bearing wear). The lack of these data will limit in-depth diagnosis of the causes of failure.

[0007] Given the shortcomings of existing technologies, there is an urgent need for a new type of control system for testing the performance of reducers, which can realize coordinated closed-loop control of drive and load, complete synchronous high-speed acquisition of multiple physical quantity signals, and improve test accuracy, repeatability and efficiency. Summary of the Invention

[0008] To address the aforementioned problems, this invention provides a control system for testing the performance of a speed reducer, which enables coordinated closed-loop control of the drive and the load.

[0009] To achieve the above objectives, the technical solution of the present invention is as follows: a control system for testing the performance of a reducer, comprising a host computer measurement and control platform, a measurement and control module, a drive module, a load module, and a data acquisition module;

[0010] The host computer measurement and control platform is used to provide a human-machine interface, configure test parameters, monitor the test process, receive, process and display test data; the host computer measurement and control platform has built-in transmission error test module, stiffness test module and starting torque test module, which are used to automatically execute the corresponding performance test items;

[0011] The measurement and control module is connected to the host computer measurement and control platform to receive control commands from the host computer measurement and control platform to control the operation of the drive module and the load module. The measurement and control module has three built-in control modes: constant speed mode, constant torque mode and positioning mode, to realize coordinated closed-loop control of drive and load.

[0012] The drive module is signal-connected to the measurement and control module and is used to drive the input end of the reducer under test according to the control command of the measurement and control module.

[0013] The load module is connected to the measurement and control module via signal, and is used to apply a controllable load to the output terminal of the reducer under test.

[0014] The acquisition module is connected to the host computer measurement and control platform to acquire the input torque, input speed, output torque, output speed, housing temperature and vibration signals of the reducer under test in real time. The acquisition module includes at least an input angle encoder and an output angle encoder, which are used to measure the deformation at the input and output ends of the reducer under test, respectively.

[0015] Furthermore, the acquisition module also includes: an input torque and speed sensor, used to measure the real-time torque and speed at the input end of the reducer under test;

[0016] The output torque and speed sensor is used to measure the real-time torque and speed at the output end of the reducer under test.

[0017] Temperature sensors and vibration sensors are used to measure the housing temperature and vibration signals, respectively.

[0018] Furthermore, the host computer monitoring and control platform is equipped with a software alarm module, which allows users to add monitoring parameters and set three alarm levels: Level 1 warning, Level 2 protection, and Level 3 emergency stop. When the monitoring parameters exceed the set range, the corresponding level of shutdown protection action is triggered.

[0019] Furthermore, constant speed mode: based on the feedback value of the speed sensor, the actual speed of the motor in the drive module is always equal to the set speed through PID closed-loop regulation;

[0020] Constant torque mode: Based on the feedback value of the torque sensor, the actual output torque of the load module motor is always equal to the set torque through PID closed-loop regulation;

[0021] Positioning mode: Based on the feedback value of the angle encoder, a three-loop PID structure of position-speed-current is adopted to control the motor of the drive module / load module to rotate precisely to the target angle and lock it;

[0022] The positioning mode is used to control the motor of the drive module to rotate precisely to a specified angle in order to perform transmission error testing, stiffness testing and starting torque testing.

[0023] Furthermore, the transmission error test module is used to control the rotation of the output shaft of the drive module in positioning mode, and simultaneously acquire signals from the input angle encoder and the output angle encoder, calculate the actual transmission error in combination with the preset transmission ratio, and plot the curve of transmission error changing with the input angle.

[0024] Furthermore, the stiffness testing module is used to control the drive module or load module at the other end of the mechanically stalled reducer to load in positioning mode or constant torque mode, record the relationship between torque and angle changes in real time, plot the hysteresis curves of forward and reverse loading, and calculate the torsional stiffness value.

[0025] Furthermore, the starting torque test module is used in the positioning mode to divide the reducer under test into 360 degrees, first rotate it 180° to eliminate backlash, and then gradually increase the torque according to the set torque step size. The peak torque when the speed of the drive module exceeds the starting speed is monitored as the starting torque, and the process is repeated until all the set number of tests are completed.

[0026] Furthermore, it also includes an anomaly tracing and differentiation module, which includes an error generation unit and a processing unit; the error generation unit is distributed or integrated within each test module, and is used to enable the corresponding test module to generate actively superimposed errors or error operations;

[0027] The processing unit, connected to the host computer measurement and control platform and the error generation unit, is configured to execute the following control logic:

[0028] Pre-test phase: Before the gearbox is formally tested, the control error generation unit is run to obtain the test item data of each test item under the corresponding superimposed error operation, and establish the mapping relationship between the active superimposed error amount and the interference amount of the test item data. The interference amount includes the abnormal amount of the item data.

[0029] Formal testing phase: Monitoring data for each test item;

[0030] Anomaly Diagnosis Phase: When an anomaly occurs in the data of a certain test item, the test module associated with the current anomaly item is identified, and the corresponding error generation unit is controlled to independently or cumulatively perform error injection, dynamically adjusting the amount of actively superimposed error until the amount of anomaly in the real-time monitored project data approaches or equals the amount of anomaly in the project data that occurs in the formal test.

[0031] Error tracing and compensation stage: Obtain the active superposition error amount of the current output and match it with the mapping relationship established in the pre-test stage; if the match is successful, it is determined that the abnormality of the current test item data is caused by the abnormality of the system test equipment; otherwise, it is determined that the reducer itself is abnormal; when it is determined that the system test equipment is abnormal, the corresponding test module is fine-tuned to filter equipment interference, thereby obtaining the real test data of the reducer.

[0032] Furthermore, the error generation unit includes a hardware error generator and / or a system software error generator; the hardware error generator is configured to change the physical state, electrical parameters, or mechanical alignment state of the test element; the system software error generator is configured to inject digital offset, noise signal, or delay command into the control signal or acquisition signal of each test module.

[0033] Furthermore, during the anomaly diagnosis phase, the processing unit has a pre-set correlation matrix that defines the relationship between any reducer test item and one or more test modules. When the data of a certain item is abnormal, the system automatically calls the error generation unit of the corresponding component based on the correlation matrix.

[0034] The technical principles of the above solution are as follows:

[0035] The measurement and control module integrates multiple independent PID closed-loop calculation channels and has three core control algorithms preset: constant speed mode, constant torque mode, and positioning mode.

[0036] In constant speed mode, the system adjusts the drive motor speed based on the feedback value from the input speed sensor; in constant torque mode, it adjusts the load motor torque based on the feedback value from the output torque sensor. Both modes can operate independently and simultaneously, thus accurately simulating complex operating conditions where speed and load fluctuate simultaneously (such as robot startup or variable load operation).

[0037] The positioning mode adopts a three-loop PID structure of position-speed-current to control the motor to rotate precisely to the target angle and lock it, providing a position reference for precision testing of transmission error, stiffness and starting torque.

[0038] The acquisition module synchronously acquires input / output torque, input / output speed, housing temperature, and vibration signals through high-precision sensors. In particular, by using differential measurements from the input and output angle encoders, combined with a preset transmission ratio, the transmission error can be calculated in real time.

[0039] The transmission error test module, stiffness test module, and starting torque test module in the host computer measurement and control platform call the positioning mode or constant torque mode respectively, and synchronously associate the angle and torque data to draw characteristic maps such as transmission error-angle curve and stiffness hysteresis curve.

[0040] The system innovatively introduces an anomaly tracing and differentiation module, which includes an error generation unit and a processing unit, and adopts the logic of "active injection-mapping matching".

[0041] Pre-testing phase: Actively inject known amounts of error (such as offset, noise) into each test module (such as driver, load, sensor) to establish a mapping relationship between "actively superimposed error amount" and "data interference amount of the tested item".

[0042] Formal testing and diagnosis phase: When data anomalies are detected, the processing unit dynamically adjusts the error generation unit in reverse, injecting active errors of different magnitudes to make the current anomaly amount approach the interference amount in the pre-test phase.

[0043] Source tracing and compensation: The system compares the currently injected error with a pre-established mapping relationship. If a match is found, the anomaly originates from the test equipment itself; if no match is found, the anomaly is determined to be a fault in the reducer itself. For equipment anomalies, the system performs fine-tuning compensation to restore the reducer's true performance data.

[0044] The above approach has the following beneficial effects:

[0045] 1. This solution achieves real-time linkage between the drive and the load through the built-in collaborative closed-loop control of the measurement and control module, solving the problem of poor synchronization caused by the separation of control and data acquisition in traditional systems. It can accurately simulate the actual working conditions of variable speed and variable torque, making the test results closer to the real operating performance of the reducer.

[0046] 2. This solution integrates dedicated testing modules for transmission error, stiffness, and starting torque into the host computer platform, supporting automated sequential testing. Compared to traditional manual adjustment and recording methods, it significantly shortens the testing cycle, eliminates human intervention errors, and improves test repeatability and consistency.

[0047] 3. This solution introduces an anomaly source differentiation module, which can proactively distinguish whether the anomalies in the test data originate from a fault in the reducer itself (such as tooth surface pitting or lubrication failure) or from errors in the test equipment (such as sensor drift or alignment deviation). This provides a completely new approach for reducer failure analysis and equipment self-calibration. Attached Figure Description

[0048] Figure 1 This is a structural block diagram of the control system for testing the performance of the reducer according to the present invention. Detailed Implementation

[0049] The following detailed description illustrates the specific implementation method:

[0050] Example 1:

[0051] As attached Figure 1 As shown: A control system for testing the performance of a speed reducer mainly includes a host computer measurement and control platform, a measurement and control module, a drive module, a load module, and a data acquisition module.

[0052] The host computer-based measurement and control platform is specifically implemented as an industrial control computer equipped with customized testing system software. This platform provides a human-machine interface for configuring test parameters, monitoring the testing process, and receiving, processing, and displaying all test data.

[0053] Measurement and Control Module: A single measurement and control instrument is used. This instrument is developed based on an embedded PLC platform and integrates multiple independent PID closed-loop calculation channels. The instrument communicates with the host computer measurement and control platform via industrial Ethernet to receive control commands. The instrument controls the operation of the drive module and load module based on the control commands received from the host computer measurement and control platform.

[0054] The drive module integrates a drive motor, which is an asynchronous motor and its corresponding frequency converter. The monitoring and control instrument connects to the frequency converter via PROFINET communication to achieve remote real-time control of the drive motor's speed or torque.

[0055] The load module integrates a load motor, which is also an asynchronous motor and its frequency converter. The drive motor and the load motor share a common bus frequency converter system. The control instrument also controls the load motor through PN communication, applying a controllable load to the output of the reducer under test.

[0056] The data acquisition module specifically includes the following sensors:

[0057] The input torque and speed sensor (including a torque sensor and a speed sensor) is used to measure the real-time torque and speed at the input of the reducer under test.

[0058] The output torque and speed sensor (including torque sensor and speed sensor) is used to measure the real-time torque and speed at the output of the reducer under test.

[0059] Input and output angle encoders: These are grating angle encoders with an accuracy of ±1″, installed on the input and output shafts of the reducer respectively, to measure deformation and then calculate transmission error, stiffness and backlash.

[0060] Temperature and vibration sensors: installed at key locations on the reducer housing to measure housing temperature and vibration signals.

[0061] Noise analyzer: measuring range 40-130 dB, used for monitoring bench noise.

[0062] All of the above sensors are connected to the measurement and control instrument or host computer measurement and control platform via signal cables. Among them, torque, speed, and angle signals are input, while temperature, vibration, and noise signals are input to the industrial control computer measurement and control platform, realizing the synchronous acquisition of multiple physical quantity signals.

[0063] Preferably, the measuring and control instrument has multiple built-in control modes, which mainly include:

[0064] Constant speed mode (N1 mode): Based on the feedback value of the input speed sensor, the system uses PID closed-loop regulation to ensure that the actual speed of the drive motor is always equal to the set target speed. Regardless of how the motor speed changes (within the rated speed range of the motor), the system maintains the actual output speed of the motor near the set speed through closed-loop regulation.

[0065] Constant torque mode (M mode): Based on the feedback value of the torque sensor at the output end, the actual output torque of the load motor is always equal to the set target torque through PID closed-loop regulation.

[0066] Positioning Mode (L Mode): Based on the feedback value of the angle encoder, this mode employs a three-loop PID structure (position-speed-current) to control the drive motor or load motor to precisely rotate to the specified target angle and lock it in place. This mode is crucial for achieving high-precision transmission error, stiffness, and starting torque testing. In this mode, the system controls the motor to precisely rotate to the specified angular position based on parameters such as the target position (angle) and movement speed, acceleration, and deceleration, and maintains the position lock upon arrival. The positioning mode relies on a high-precision angle encoder (in this system, a Renishaw grating angle encoder with an accuracy of ±1″) for position feedback.

[0067] Preferably, the host computer-based measurement and control platform has built-in test modules for multiple different projects, capable of automatically completing test sequences. The following example illustrates this:

[0068] 1. Transmission error test

[0069] The transmission error test module built into the host computer measurement and control platform automatically executes the following process:

[0070] a. Switch the drive motor to positioning mode (L mode) via the measurement and control module.

[0071] b. Set test parameters: Given a fixed torque limit (to ensure backlash is eliminated), set a termination angle (e.g., 360°) and a gear ratio.

[0072] c. Software begins testing: Control the drive motor to begin stepping rotation at an extremely low speed (to eliminate dynamic errors), rotating by a tiny angle each time.

[0073] d. After stabilizing at each position, synchronously acquire the real-time angle values ​​of the input angle encoder and the output angle encoder.

[0074] e. The system automatically calculates the difference between the theoretical output angle and the actual output angle, i.e., the transmission error, and plots the transmission error as a function of the input angle in real time. Since the drive and load are controlled collaboratively by the same measurement and control system, and the angle signal is synchronized with the drive command, the measurement accuracy of the transmission error is greatly improved.

[0075] 2. Stiffness test

[0076] The stiffness test module built into the host computer measurement and control platform executes automatically according to the following procedure:

[0077] a. Block the input end of the reducer under test by mechanical means (such as positioning fixtures).

[0078] b. Switch the load motor to positioning mode (L mode) or constant torque mode (M mode). In this embodiment, positioning mode is preferred to obtain more precise torsion angle control.

[0079] c. Starting from zero torque, the software controls the load motor to gradually increase the torque in the positive direction according to the set torque step size (such as 10% of the rated torque). After each step stabilizes, the current output torque and the angle change of the output end angle encoder are recorded.

[0080] d. Once the torque reaches the set rated positive torque, control the load motor to apply torque in the reverse direction, and record the torque-angle data until the set rated reverse torque is reached.

[0081] e. The system automatically plots complete forward and reverse loading hysteresis curves and calculates parameters such as torsional stiffness and hysteresis based on the curve data. The entire testing process is fully automated, avoiding the discontinuities of manual loading and human reading errors.

[0082] 3. Starting torque test

[0083] The starting torque test module built into the host computer monitoring and control platform automatically executes the following process:

[0084] a. Connect the input end of the reducer under test to the drive motor, and leave the output end free or apply slight damping.

[0085] b. Switch the drive motor to positioning mode (L mode). The software will automatically divide the 360° input shaft of the reducer into N equal parts (e.g., N=36, that is, one test point every 10°).

[0086] c. To eliminate backlash, the system first controls the drive motor to rotate 180° from zero in the positive direction.

[0087] d. Then, at the current position, a very small angle command is given, while the drive motor is controlled to gradually increase the output torque in a set torque step.

[0088] e. Real-time monitoring of the drive motor speed feedback. Once the drive motor speed exceeds the set "start speed" threshold, the system immediately records the peak torque at the current moment and uses it as the starting torque value at that angle position.

[0089] f. Stop loading. After the drive motor stops, it will automatically rotate to the next preset angle position and repeat step ce.

[0090] g. After traversing all set positions, the system generates a curve showing the starting torque as a function of angle. This automated process is not only highly efficient, but also ensures the consistency and comparability of results because a "backlash elimination" operation is performed for each test.

[0091] Throughout the testing process, all sensor data (torque, speed, angle, temperature, vibration, and noise) from the acquisition module were transmitted in real-time to the host computer monitoring and control platform via EtherCAT bus (for signals connected to the monitoring and control instrument) or Ethernet (for signals connected to the host computer monitoring and control platform) at configurable periods ranging from 1ms to 100ms. The host computer monitoring and control platform aligned all data to the same timestamp, ensuring the accuracy of data analysis under dynamic operating conditions.

[0092] Meanwhile, the host computer monitoring and control platform is equipped with an alarm module. Users can add any monitoring parameters (such as temperature, vibration, and torque) and set three alarm levels: Level 1 warning, Level 2 protection, and Level 3 emergency stop. For example, when the shell temperature exceeds 80°C, a Level 1 warning is triggered (the interface turns yellow and a buzzer sounds); when it exceeds 90°C, a Level 2 protection is triggered (the system automatically unloads torque and stops driving); and when it exceeds 100°C, a Level 3 emergency stop is triggered (the main circuit power is cut off). When the monitored parameters exceed the set range, the system will automatically execute the corresponding level of protection action according to priority to ensure the safety of the test bench and personnel.

[0093] Example 2:

[0094] The only difference from Embodiment 1 above is that it also includes an anomaly source differentiation module, which mainly includes an error generation unit and a processing unit.

[0095] The error generation unit is distributed or integrated within conventional reducer performance test benches in existing technologies. In this embodiment, it is specifically integrated into various test modules (e.g., torque sensors, angle encoders, temperature sensors, etc. in the drive module, load module, and acquisition module), or exists as independent hardware / software. Its main purpose is to generate actively superimposed errors or erroneous operations in the corresponding test modules.

[0096] The error generation unit includes a hardware error generator and / or a system software error generator; the hardware error generator is configured to change the physical state, electrical parameters, or mechanical alignment state of the test element.

[0097] For example: a mechanism for simulating the centering error at the input end of a drive motor-reducer.

[0098] An electrically adjustable eccentric sleeve is installed at the coupling between the output shaft of the drive motor and the input shaft of the reducer. This eccentric sleeve is driven by a micro servo motor via a lead screw, and can radially translate the axis. The translation amount (0~0.5mm) is adjusted by the control and data processing unit according to instructions, thereby simulating different degrees of axis misalignment. The translation amount sensor provides real-time feedback on the position, forming a precise quantification of the alignment error.

[0099] Zero-point drift injection circuit for torque sensor:

[0100] A programmable bias power supply module is connected in parallel to the signal output of the torque sensor. This module includes an adder circuit consisting of a DA converter and an operational amplifier. The control unit generates a controllable DC bias voltage (e.g., simulating an equivalent offset of 0.001~0.1 N·m) by sending digital commands, which is then superimposed on the original torque signal to simulate the zero-point drift error of the sensor.

[0101] Temperature sensor temperature drift interference module:

[0102] For thermocouples or PT100 temperature sensors, a programmable resistor network (digital potentiometer) is connected in series in their measurement circuit. By changing the circuit resistance value, the temperature measurement deviation caused by the sensor's own temperature drift or the change in the contact resistance of the measurement channel can be simulated.

[0103] Additional excitation device for vibration sensor:

[0104] A miniature vibrator (electromagnetic type) is installed near the gearbox housing or bearing housing of the reducer under test. Its input signal is driven by a specific frequency / amplitude waveform generated by the processing unit and then amplified by power. It can generate fixed frequency or broadband simulated vibration, which interferes with the acquisition of real vibration signals.

[0105] Torque fluctuation simulation unit for load brake:

[0106] In the control circuit of a magnetic powder brake, an eddy current brake, or a load motor, a modulation signal generator is connected in series. This generator can inject low-frequency sine wave or random fluctuation components into the control current, causing the braking torque to fluctuate periodically or non-periodically, simulating the interference at the load end.

[0107] Each of the aforementioned hardware error generators is equipped with an enable switch (relay or electronic switch), which is normally in a bypass / zero state and does not affect routine testing. When an error injection command is received, the corresponding switch closes and performs quantitative adjustment.

[0108] The system software error generator is configured to inject digital offsets, noise signals, or delay instructions into the control signals or acquisition signals of each test module.

[0109] For example, the system software error injector is embedded in the data acquisition and motion control software of the industrial control computer measurement and control platform. It modifies control commands or acquired data in real time through algorithm models, specifically including:

[0110] Digital signal bias module: Before storing or analyzing the raw sensor values ​​read from the AD converter, add or subtract a variable offset (which can change slowly or step over time) to simulate long-term sensor drift.

[0111] Noise injection generator: Built-in algorithms for Gaussian white noise, impulse noise, or specific frequency noise are used to superimpose the generated noise sequence onto the actual control output signal (such as motor current command) or the acquired signal (such as speed, torque). The noise amplitude and frequency range can be configured by commands.

[0112] Delay and Phase Shift Module: Used to simulate synchronization errors or communication delays in multi-channel acquisition. For example, introducing a tiny time delay (on the order of milliseconds) into the acquisition of speed signals can cause a misalignment in the torque-speed correspondence.

[0113] All hardware error generators are connected to the processing unit via fieldbus (such as CAN or EtherCAT), while the system software error generator runs as a background service in the industrial control computer's measurement and control platform. The processing unit has a built-in error injection schedule table, which records the injectable error type, implementation method (hardware / software), and allowed input range for each test module. During pre-testing or diagnostics, simply sending the target error amount according to the schedule table allows for the superposition of independent or combined errors from any test module. For example, to simulate the combined error interference of "drive motor torque fluctuation + torque sensor zero-point drift," the processing unit simultaneously injects a noise signal (software) into the motor controller and sends a 0.05 N·m offset command (hardware) to the torque sensor's bias module, thus superimposing errors / mistakes onto the actual component operation.

[0114] The processing unit, connected to the host computer measurement and control platform and the error generation unit, is configured to execute the following control logic:

[0115] Pre-test phase: Before the gearbox is formally tested, the control error generation unit is run to obtain the test item data of each test item under the corresponding superimposed error operation, and establish the mapping relationship between the active superimposed error amount and the interference amount of the test item data. The interference amount includes the abnormal amount of the item data.

[0116] Formal testing phase: Monitoring data for each test item;

[0117] Anomaly Diagnosis Phase: When an anomaly occurs in the data of a certain test item, the test module associated with the current anomaly item is identified, and the corresponding error generation unit is controlled to independently or superimposedly perform error injection. The amount of actively superimposed error is dynamically adjusted until the amount of anomaly in the real-time monitored data of the item approaches or equals the amount of anomaly in the data of the item that appears in the formal test (test items such as speed, torque, transmission ratio, stiffness, transmission efficiency, temperature rise, etc.).

[0118] Error tracing and compensation stage: Obtain the active superposition error amount of the current output and match it with the mapping relationship established in the pre-test stage; if the match is successful, it is determined that the abnormality of the current test item data is caused by the abnormality of the system test equipment; otherwise, it is determined that the reducer itself is abnormal; when it is determined that the system test equipment is abnormal, the corresponding test module is fine-tuned to filter equipment interference, thereby obtaining the real test data of the reducer.

[0119] Preferably, during the anomaly diagnosis phase, the processing unit has a pre-set correlation matrix that defines the relationship between any reducer test item and one or more test modules; when the data of a certain item is abnormal, the system automatically calls the error generation unit of the corresponding component according to the correlation matrix.

[0120] First, during system initialization or periodic calibration, a test item-test element association matrix is ​​constructed and stored in the processing unit. A partial example is shown below:

[0121] Transmission error test Input angle encoder, output angle encoder, drive module (positioning accuracy) Encoder signal drift, code disk contamination, motor angle deviation Digital signal bias module for angle encoder, digital delay module for drive motor Transmission efficiency test Input torque sensor, output torque sensor, input speed sensor, output speed sensor Torque zero-point drift, speed channel noise, sensor channel crosstalk The torque sensor's digital signal bias module, noise injection generator, and speed channel delay and phase shift module. Starting torque test Drive module, input torque sensor, input angle encoder Drive motor torque fluctuation, torque sensor threshold drift Drive motor noise injection module, digital signal bias module of torque sensor Stiffness test Output torque sensor, output angle encoder, load module Load-side torque fluctuation, encoder quantization noise Torque fluctuation simulation unit for load motor or brake, noise injection module for angle encoder Temperature rise test Temperature sensor Temperature sensor temperature drift, measurement channel contact resistance change Temperature drift interference module for temperature sensor (programmable resistor network) Vibration monitoring Vibration sensor, speed sensor Sensor-related resonance and environmental vibration interference from the test bench Additional excitation device for vibration sensor, noise injection generator for speed channel

[0122] The pre-testing phase is initiated by the operator through the host computer monitoring and control platform, and the system automatically executes the following database creation process:

[0123] The system enters pre-test mode, at which point a standard "reference reducer" with known performance qualification is installed on the test bench.

[0124] The processing unit traverses the error injection schedule table and injects multiple known active error quantities of different magnitudes into each associated element in turn.

[0125] For example, for an input torque sensor, DC bias errors of 0.01 N·m, 0.02 N·m, 0.05 N·m, and 0.1 N·m are injected sequentially through its zero-point drift injection circuit.

[0126] Under each error injection state, a simple run of the corresponding test item is started (such as constant speed no-load run), and the evaluation index of the item is collected in real time. The deviation from the normal value when no error is injected is calculated, which is the "item data anomaly".

[0127] For example, after injecting an input torque bias error of 0.02 N·m, the calculated transmission efficiency is found to be 0.15% lower than the normal value. The mapping relationship is then recorded as: Input torque bias 0.02 N·m → Transmission efficiency anomaly -0.15%.

[0128] Perform the above operations on all combinations of components and error types to generate a multidimensional mapping table or fit a mapping function curve, and store it in the non-volatile memory of the processing unit.

[0129] An example of the combined workflow of formal testing and anomaly diagnosis is as follows: Assume the system is testing the transmission efficiency of a batch of reducers, with the normal fluctuation range for transmission efficiency set at ±0.5%. When testing a particular reducer, the processing unit detects that the transmission efficiency under rated operating conditions is 0.9% lower than the standard value, exceeding the normal range, thus triggering the anomaly diagnosis process:

[0130] S1. The processing unit marks the current actual abnormal quantity: Δη1 = -0.9%.

[0131] S2. Query the correlation matrix to determine that the components strongly related to "transmission efficiency test" are the input torque sensor, output torque sensor, input speed sensor, and output speed sensor.

[0132] S3. According to the preset component weights (usually arranged from largest to smallest), the error is preferentially injected into the input torque sensor. The processing unit sends a command to the bias circuit of the input torque sensor to gradually increase the bias voltage.

[0133] S4. The system calculates the current transmission efficiency anomaly ΔE in real time and compares it with the target anomaly Δη1. A step approximation method is used here: the injected bias is increased in steps of 0.01 N·m until ΔE = -0.9% ± 0.05%.

[0134] S5. Assuming that when the injected bias reaches 0.08 N·m, the real-time anomaly is stable at -0.89%, satisfying the approximation condition, the system records the current active superposition error D = 0.08 N·m (input torque bias).

[0135] S6. Similarly, the system can test other related components individually or in combination until all possibilities have been explored. If injecting the input torque bias alone can perfectly reproduce the anomaly, then further combination is not necessary.

[0136] S7. Error Source Tracing: The processing unit retrieves the mapping table established during the pre-test phase and queries the mapping corresponding to "Input Torque Sensor Offset". The data recorded in the pre-test may be as follows: an offset of 0.05 N·m corresponds to an efficiency anomaly of -0.45%, and an offset of 0.1 N·m corresponds to an efficiency anomaly of -1.15%. Through linear interpolation, it is found that the theoretical anomaly corresponding to an offset of 0.08 N·m is approximately -0.87%. In this diagnostic test, the actual injection of 0.08 N·m reproduced the anomaly of -0.89%, and the deviation between the two is within the preset tolerance band (e.g., ±0.1%), which is considered a successful match.

[0137] S8. Judgment and Compensation: Successful matching indicates that the 0.9% efficiency anomaly observed in the current test is highly likely caused by zero-point drift (equivalent to 0.08 N·m) of the input torque sensor in the test equipment, rather than by low efficiency of the reducer itself. The system automatically marks the reducer as "Equipment Interference - Data Recovery Pending" and performs fine-tuning compensation: a fixed offset of 0.08 N·m is subtracted from subsequent input torque acquisition values, and then the transmission efficiency is recalculated using the corrected torque value. After recalculation, the reducer's efficiency value returns to the normal range, thus obtaining the true test data after equipment self-calibration.

[0138] If, in step S7, the same abnormal quantity cannot be reproduced within the allowable error range after traversing all individual and common combination injection errors of all related components, or if the injected cost function is seriously inconsistent with the pre-test mapping (e.g., it requires injecting an unusually large amount of error to barely approach it), then the matching is determined to be a failure, and the system alarm prompts "the reducer itself is abnormal," reminding the operator to disassemble and analyze the reducer.

[0139] Obviously, the above embodiments are merely illustrative examples for clear explanation and are not intended to limit the implementation. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations here. However, obvious variations or modifications derived therefrom are still within the scope of protection of this invention.

Claims

1. A control system for testing the performance of a speed reducer, characterized in that, It includes a host computer-based measurement and control platform, a measurement and control module, a drive module, a load module, and a data acquisition module; The host computer measurement and control platform is used to provide a human-machine interface, configure test parameters, monitor the test process, receive, process and display test data; the host computer measurement and control platform has built-in transmission error test module, stiffness test module and starting torque test module, which are used to automatically execute the corresponding performance test items; The measurement and control module is connected to the host computer measurement and control platform to receive control commands from the host computer measurement and control platform to control the operation of the drive module and the load module. The measurement and control module has three built-in control modes: constant speed mode, constant torque mode and positioning mode, to realize coordinated closed-loop control of drive and load. The drive module is signal-connected to the measurement and control module and is used to drive the input end of the reducer under test according to the control command of the measurement and control module. The load module is connected to the measurement and control module via signal, and is used to apply a controllable load to the output terminal of the reducer under test. The acquisition module is connected to the host computer measurement and control platform to acquire the input torque, input speed, output torque, output speed, housing temperature and vibration signals of the reducer under test in real time. The acquisition module includes at least an input angle encoder and an output angle encoder, which are used to measure the deformation at the input and output ends of the reducer under test, respectively.

2. The control system for testing the performance of a reducer according to claim 1, characterized in that, The acquisition module also includes: an input torque and speed sensor, used to measure the real-time torque and speed at the input of the reducer under test; The output torque and speed sensor is used to measure the real-time torque and speed at the output end of the reducer under test. Temperature sensors and vibration sensors are used to measure the housing temperature and vibration signals, respectively.

3. The control system for testing the performance of a reducer according to claim 2, characterized in that, The host computer monitoring and control platform is equipped with a software alarm module. The software alarm module allows users to add monitoring parameters and set three alarm levels: Level 1 warning, Level 2 protection, and Level 3 emergency stop. When the monitoring parameters exceed the set range, the corresponding level of shutdown protection action is triggered.

4. The control system for testing the performance of a reducer according to claim 3, characterized in that, Constant speed mode: Based on the feedback value of the speed sensor, the actual speed of the motor in the drive module is always equal to the set speed through PID closed-loop regulation; Constant torque mode: Based on the feedback value of the torque sensor, the actual output torque of the load module motor is always equal to the set torque through PID closed-loop regulation; Positioning mode: Based on the feedback value of the angle encoder, a three-loop PID structure of position-speed-current is adopted to control the motor of the drive module / load module to rotate precisely to the target angle and lock it; the positioning mode is used to control the motor of the drive module to rotate precisely to the specified angle to perform transmission error test, stiffness test and starting torque test.

5. The control system for testing the performance of a reducer according to claim 4, characterized in that, The transmission error test module is used to control the output shaft rotation of the drive module in positioning mode, and simultaneously acquire signals from the input angle encoder and the output angle encoder. It calculates the actual transmission error in combination with the preset transmission ratio and plots the curve of transmission error as a function of the input angle.

6. The control system for testing the performance of a reducer according to claim 5, characterized in that, The stiffness testing module is used to control the drive module or load module at the other end of the mechanically stalled reducer to load in positioning mode or constant torque mode, record the relationship between torque and angle changes in real time, plot the hysteresis curves of forward and reverse loading, and calculate the torsional stiffness value.

7. The control system for testing the performance of a reducer according to claim 6, characterized in that, The starting torque test module is used in the positioning mode to divide the reducer under test into 360 degrees, first rotate it 180° to eliminate backlash, and then gradually increase the torque according to the set torque step size. The peak torque when the speed of the drive module exceeds the starting speed is monitored as the starting torque, and the process is repeated until all the set number of tests are completed.

8. The control system for testing the performance of a reducer according to claim 7, characterized in that, It also includes an anomaly source differentiation module, which includes an error generation unit and a processing unit; Error generation units are distributed or integrated within each test module to enable the corresponding test module to generate actively superimposed errors or error operations. The processing unit, connected to the host computer measurement and control platform and the error generation unit, is configured to execute the following control logic: Pre-test phase: Before the gearbox is formally tested, the control error generation unit is run to obtain the test item data of each test item under the corresponding superimposed error operation, and establish the mapping relationship between the active superimposed error amount and the interference amount of the test item data. The interference amount includes the abnormal amount of the item data. Formal testing phase: Monitoring data for each test item; Anomaly Diagnosis Phase: When an anomaly occurs in the data of a certain test item, the test module associated with the current anomaly item is identified, and the corresponding error generation unit is controlled to independently or cumulatively perform error injection, dynamically adjusting the amount of actively superimposed error until the amount of anomaly in the real-time monitored project data approaches or equals the amount of anomaly in the project data that occurs in the formal test. Error source tracing and compensation stage: Obtain the amount of active superposition error in the current output and match it with the mapping relationship established in the pre-test stage; if the match is successful, it is determined that the abnormality of the current test item data is caused by the abnormality of the system test equipment. Conversely, if the problem is not found, it is determined that the reducer itself is malfunctioning. When the system testing equipment is determined to be malfunctioning, fine-tuning is performed on the corresponding testing module to filter out equipment interference, thereby obtaining the true test data of the reducer.

9. The control system for testing the performance of a reducer according to claim 8, characterized in that, The error generation unit includes a hardware error generator and / or a system software error generator; the hardware error generator is configured to change the physical state, electrical parameters or mechanical alignment state of the test element; the system software error generator is configured to inject digital offset, noise signal or delay command into the control signal or acquisition signal of each test module.

10. The control system for testing the performance of a reducer according to claim 9, characterized in that, During the anomaly diagnosis phase, the processing unit has a pre-set correlation matrix that defines the relationship between any reducer test item and one or more test modules. When the data of a certain item is abnormal, the system automatically calls the error generation unit of the corresponding component according to the correlation matrix.

Images