Gearbox tiny fault detection method

Abstract

A gearbox tiny fault detection method comprises the steps that a motor of a permanent magnet alternating current servo system is directly connected with an input shaft of a gearbox to be detected, and a vibration sensor is installed at the set position around a position sensor of the motor; a first position sensor is installed on an output shaft of a tested gear box through a coupler, a second vibration sensor is installed on a gear box bearing seat close to the first position sensor, and under the condition that the rotating speed of a motor is set, a synchronous data collection board collects values of the two position sensors and the two vibration sensors at the same time in a timing mode. The transmission error is calculated according to the values of the two position sensors, the vibration difference is obtained according to the values of the two vibration sensors, and when the transmission error and the vibration difference simultaneously exceed the detection threshold value or the single transmission error exceeds the set-time detection threshold value and the single vibration difference exceeds the set-time detection threshold value, an alarm is triggered.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of servo control, and particularly relates to a gear box micro-fault detection method. BACKGROUND

[0002] There is a direct and close causal and feedback relationship between the transmission error of the gear box and the vibration of the gear box, and the transmission error is one of the main internal excitation sources of the vibration and noise of the gear system.

[0003] The transmission error refers to the deviation between the actual angular displacement of the driven gear and the ideal angular displacement (determined by the input of the driving gear and the theoretical transmission ratio) under the condition of ideal rigidity, no gap and no deformation. It reflects the instantaneous transmission inaccuracy of the gear pair during meshing. The transmission error is mainly caused by manufacturing errors (including pitch error, profile error, helix error, tooth surface roughness, etc.), assembly errors (including axis misalignment, center distance deviation, shafting misalignment, etc.), elastic deformation (referring to the deformation of the gear body, shaft and bearing under the action of load), thermal deformation (referring to the deformation caused by temperature change during operation), design factors (including contact ratio, pressure angle, modification design (or lack of modification), etc.). When a pair of gears meshes, the existence of the transmission error means the fluctuation of the instantaneous speed ratio, and even if the input speed is constant, the output speed will also fluctuate due to the TE. The transmission error (especially the acceleration component) and the meshing stiffness fluctuation jointly cause the periodic fluctuation of the dynamic force (normal force, tangential force) between the gear meshing points. These dynamic forces are the internal excitation forces directly acting on the gear box structure, and at the same time, the fluctuation of these dynamic meshing forces is transmitted to the gear box body through the shaft and bearing, forcing the box body and its internal structure to produce forced vibration, which is the main source of the gear box vibration.

[0004] The vibration also has a feedback effect on the existence of the transmission error. Although the transmission error is the main cause, strong box vibration or shafting vibration can also affect the actual meshing state of the gear in the opposite direction. Severe vibration can change the dynamic load distribution between the gears and affect the relative position of the shafting, which may cause instantaneous micro-displacement of the shaft and bearing, change the center distance or parallelism. In extreme cases (such as severe impact, excessive backlash), vibration may cause the tooth surface to separate or impact, thereby generating additional, nonlinear transmission error components. This feedback is usually nonlinear and more pronounced when the system damping is insufficient or the excitation is too large. It can exacerbate vibration, forming a vicious cycle.

[0005] The greater the amplitude of the transmission error, the greater the dynamic meshing force fluctuation, and the more intense the vibration. High-frequency errors in the frequency spectrum of the transmission error (such as tooth profile error) are prone to exciting high-order natural frequencies of the gear system, producing high-frequency noise (whistling), and low-frequency errors (such as cumulative error of the pitch, eccentricity) mainly affect the vibration near the meshing frequency and low-frequency vibration, and produce significant sidebands. If the main frequency component of the dynamic meshing force (the meshing frequency or its multiple) coincides with a certain order of the natural frequency of the gear-shaft-bearing-box system, resonance occurs, resulting in a sharp amplification of the vibration amplitude. The size of the system's damping determines how quickly the vibration energy dissipates. The smaller the damping, the higher the resonance peak, and the slower the vibration attenuation. The load affects the elastic deformation and the contact state; the rotational speed determines the excitation frequency (meshing frequency) and affects whether resonance occurs.

[0006] Within the scope of the existing technical literature searched in China, domestic related research generally focuses more on the measurement and analysis of the static transmission error of the gear box alone or the collection and spectral analysis of the vibration signal alone, a small number of which involves the measurement and analysis of the dynamic transmission error of the gear box at low speed or extremely low speed, and none of which involves the joint detection of the dynamic transmission error and the vibration difference of the measured gear box from the extremely low speed (0.25% of the rated speed of the motor) to the full speed of the rated speed. Usually, the characteristic signals of the fault can only be detected when the fault of the measured gear box develops to a certain stage and the structural damage is obvious. SUMMARY

[0007] The present application proposes a gear box micro-fault detection method, which correlates the transmission error and vibration signal of the input shaft and output shaft of the gear box, realizes high-sensitivity assembly defect detection under no-load working condition, and improves the detection probability of micro-faults in the gear box.

[0008] The technical scheme for achieving the object of the present application is as follows: a gear box micro-fault detection method, comprising:

[0009] The motor of the permanent magnet AC servo system is directly connected with the input shaft of the measured gear box, a vibration sensor is arranged around the position sensor of the motor, a first position sensor is arranged on the output shaft of the measured gear box through a shaft coupling, and a second vibration sensor is arranged on the bearing seat of the gear box near the first position sensor. Under the condition of a set motor speed, the values of the two position sensors and the two vibration sensors are collected simultaneously by a synchronous data acquisition board, the transmission error is calculated according to the values of the two position sensors, and the vibration difference is obtained according to the values of the two vibration sensors. When the transmission error and the vibration difference exceed the detection threshold value at the same time or the transmission error alone exceeds the set multiple of the detection threshold value and the vibration difference alone exceeds the set multiple of the detection threshold value, an alarm is triggered.

[0010] Compared with the prior art, the present application has the following advantages:

[0011] 1) This invention highly integrates a permanent magnet AC servo motor, servo driver (controller), and feedback devices (such as rotary transformers, optical encoders, etc.) to achieve precise speed, torque, and position control. It boasts core advantages such as extremely high dynamic response, precise position / speed control, high torque / inertia ratio, high efficiency, wide speed range, strong overload capacity, small size, and high power density. The drive system of this invention employs a permanent magnet AC servo control system, and the motor is a permanent magnet AC servo motor. A vibration sensor is installed near the motor's built-in position sensor to measure motor vibration, ensuring the motor becomes an ideal power source.

[0012] 2) The values ​​of two position sensors and two vibration sensors are collected synchronously at regular intervals. The dynamic error is correlated with the vibration difference. The normal mode is used to detect the working status of the gearbox under test. When an early stage fault is detected in the gearbox under test, the excitation mode is switched to. The peak speed of the sinusoidal excitation is 25% of the rated speed of the motor. The constant speed excitation is performed for 2 seconds and the sinusoidal excitation is performed for 5 seconds to amplify the early fault state and improve the detection probability of minor faults in the gearbox under test.

[0013] 3) At each set speed, the motor runs at a constant speed for ten minutes. By comparing the sampled values, the maximum value of the transmission error and the maximum value of the vibration difference within these ten minutes are obtained. Twice the maximum value of the transmission error is taken as the detection threshold for the transmission error, and three times the maximum value of the vibration difference is taken as the detection threshold for the vibration difference. The method for determining the detection thresholds does not require extensive historical data accumulation and can effectively meet the testing needs of small-batch prototype production.

[0014] 4) Since the gear meshing impact of the gearbox under test is a transient event at the microsecond level, only by ensuring high-precision synchronization between the vibration signal of the input shaft and the position signal of the output shaft can the vibration event be accurately located to a specific gear meshing and the early fault lock be achieved. Therefore, a specially designed timed synchronous data acquisition board with FPGA chip as the core circuit is adopted to acquire four signals in hardware timed synchronous mode: position sensor (21) measurement value θ1, position sensor (22) measurement value θ2, vibration sensor (31) measurement value V1, and vibration sensor (32) measurement value V2. The sampling data rate is not less than 200KHz.

[0015] 5) A phased speed-up testing process amplifies and exposes faults of different natures at different speed bands. At extremely low speeds, inertial forces are negligible, and any tiny mechanical resistance or geometric error will lead to a huge relative transmission error. This can detect assembly geometric defects and macroscopic mechanical interference, as well as faults such as assembly eccentricity, shaft bending, severe impact / burrs, and improper bearing installation. At low speeds, the rotational speed is sufficient to generate measurable vibration signals, but not enough to mask subtle force excitations with inertial forces. This can detect mid-to-low frequency fault characteristics and initial lubrication problems, as well as faults such as gear pitch errors, slight misalignment, surface roughness, and poor lubrication. At medium speeds, the characteristic frequencies of most faults can be excited to the appropriate frequency band, and the signal-to-noise ratio is high. This can detect gear meshing and bearing operating conditions, including gear meshing defects, early bearing faults, and preliminary resonance checks. At high speeds, dynamic balance problems, gear meshing under high-speed dynamic loads, thermal effects, and high-speed bearing problems can be detected.

[0016] 6) Advantages and value of the phased speed-up testing process: Early and efficient fault screening, which can detect 80% of assembly and manufacturing geometric defects in the first stage, greatly reducing the cost and risk of subsequent testing; Precise fault location, with fault characteristics at different speed ranges that can directly point to specific types of defects, greatly simplifying fault diagnosis and root cause analysis; Comprehensive quality assessment, evaluating the performance of the gearbox under test under different operating conditions; Building product data archives, with each prototype gearbox having a complete curve from extremely low speed to full speed.

[0017] 7) It can serve as an online condition monitoring and predictive maintenance tool to monitor the condition changes of a gearbox that has been put into use during operation, promptly detect performance degradation or sudden failures, and improve the early prediction of failures.

[0018] The present invention will now be described in further detail with reference to the accompanying drawings. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of a method for detecting minor faults in a gearbox.

[0020] Figure 2 A schematic diagram showing the motor speed settings for normal operating mode and induced operating mode. Detailed Implementation

[0021] like Figure 1As shown, a method for detecting minor faults in a gearbox involves directly connecting the motor of a permanent magnet AC servo system to the input shaft of the gearbox under test. A vibration sensor is installed near the motor's built-in position sensor. A first position sensor is installed on the output shaft of the gearbox under test via a coupling, and a second vibration sensor is installed at the gearbox bearing housing near the first position sensor. A synchronous data acquisition board periodically collects the values ​​from both position and vibration sensors. The absolute value of the two position sensor values ​​is calculated to obtain the transmission error, and the absolute value of the two vibration sensor values ​​is subtracted to obtain the vibration difference. The motor runs at a constant speed for ten minutes, and the maximum values ​​of the transmission error and vibration difference within those ten minutes are compared. Twice the maximum value of the transmission error is taken as the detection threshold for the transmission error, and three times the maximum value of the vibration difference is taken as the detection threshold for the vibration difference. An alarm is triggered when both the transmission error and vibration difference exceed the detection threshold simultaneously, or when the transmission error alone exceeds twice the detection threshold or the vibration difference alone exceeds three times the detection threshold.

[0022] A method for detecting minor faults in a gearbox involves a permanent magnet AC servo motor (10) driving the gearbox under test, a position sensor (21) built into the motor, a second position sensor (22) measuring the angle of the output shaft of the gearbox under test, a first vibration sensor (31) mounted on the motor body near the position sensor (21), a second vibration sensor (32) mounted on the bearing seat of the output shaft of the gearbox, a servo driver (11) connected to the motor, and a synchronous data acquisition board (40) periodically and synchronously acquiring signals from the position sensor (21), position sensor (22), vibration sensor (31), and vibration sensor (32) and sending them to a host computer (41). The specific steps are as follows:

[0023] Step 1: The position sensor (21) of the permanent magnet AC motor is built-in. The position sensor (22) is coaxially installed on the output shaft of the gearbox under test through a coupling. The vibration sensor (31) is attached to the motor body near the rotary transformer of the permanent magnet AC motor with a magnetic base. The vibration sensor (32) is attached to the bearing seat of the output shaft of the gearbox under test with a magnetic base.

[0024] Step 2: The timed synchronous data acquisition board (40) uses an FPGA chip as the core circuit and hardware timing. At time T, it synchronously acquires four signals: the measured value θ1 of the position sensor (21), the measured value θ2 of the position sensor (22), the measured value V1 of the vibration sensor (31), and the measured value V2 of the vibration sensor (32). Then, the four signals at sampling time T are packaged as a set of data and sent to the host computer (41).

[0025] Step 3: After unpacking in the host computer (41), the measured values ​​θ1 of the motor position sensor (21), θ2 of the second position sensor (22), and the transmission ratio i (theoretical design value) of the gearbox are calculated and the absolute values ​​are taken. The difference between the two position sensor signals is the transmission error Δθ at time T. Its expression is: Δθ(K) = |θ1(K) - θ2(K)×i|, where k is the sampling number, k=1, 2..., θ1(K) and θ2(K) are the position sensor measured values ​​θ sampled at time T, respectively, and T is the sampling time.

[0026] Step 4: Subtract the measured value V1 of the first vibration sensor (31) and the measured value V2 of the second vibration sensor (32) and take the absolute value. The difference between the two vibration signals is the vibration difference ΔV at time T. Its expression is: ΔV(K) = |V1(K) - V2(K)|, where k is the sampling number, k = 1, 2, ..., and V1(K) and V2(K) are the vibration sensor measured values ​​θ obtained at time T.

[0027] Step 5: Set the motor speed to ultra-low speed (0.25% of the motor's rated speed) and drive the gearbox under test to run at a constant speed under no-load for 10 minutes. Compare the magnitudes of the transmission error Δθ(K) sampled within 10 minutes to obtain the maximum value of the transmission error Δθmax. Take twice the maximum value of the transmission error Δθmax as the detection threshold θ of the transmission error. T ;

[0028] Step 6: Compare the magnitudes of the vibration differences ΔV(K) sampled within 10 minutes to obtain the maximum value ΔVmax of the vibration difference. Take three times the maximum value ΔVmax of the vibration difference as the detection threshold V of the vibration difference. T ;

[0029] Step 7: In normal operating mode, the motor speed is set to ultra-low speed (0.25% of the motor's rated speed). The motor runs at ultra-low speed at a constant speed for four hours. The transmission error Δθ(K) is greater than the threshold θ. T When the vibration difference ΔV(K) is greater than the threshold V at 80% of the time... T When the target is 80%, switch to incentive-based work mode;

[0030] Step 8: In normal operating mode, the motor speed is set to ultra-low speed (0.25% of the motor's rated speed). The motor runs at ultra-low speed at a constant speed for four hours. The transmission error Δθ(K) is greater than the threshold θ. T And simultaneously, the vibration difference ΔV(K) is greater than the threshold V. T If this happens, call the police immediately.

[0031] Step nine: In the excitation working mode, the motor speed changes to 25% of the rated speed within 0.5 seconds, maintains the peak speed for 2 seconds, and then performs a 10-100Hz sinusoidal sweep for 5 seconds. The transmission error Δθ(K) is greater than the threshold θ. T When it is twice the value, or the vibration difference ΔV(K) is greater than the threshold V T If the value is 3 times the normal value, immediately call the police.

[0032] Step 10: If the system does not alarm when the motor is operating at ultra-low speed, increase the motor speed to low speed (2.5% of the motor's rated speed) and repeat steps 2 to 9.

[0033] Step 11: If the system does not alarm when the motor is working at low speed, increase the motor speed to medium speed (25% of the motor's rated speed) and repeat steps 2 to 9.

[0034] Step 12: If the system does not alarm when the motor is operating at medium speed, increase the motor speed to high speed (motor rated speed) and repeat steps 2 to 9.

[0035] Step 13: If the system does not alarm when the motor is operating at high speed, the test process ends.

[0036] In a further embodiment, the position sensor (21) and the position sensor (22) are rotary transformers.

[0037] In engineering practice, the timing synchronization data acquisition board in step two typically uses hardware timing sampling, with a sampling data rate of no less than 200kHz. In steps seven and eight, the duration of normal operating conditions is determined by the specific project process; in this embodiment, the duration is no less than four hours. In this embodiment, the transmission error threshold does not exceed 0.4 degrees, and the vibration difference threshold does not exceed 0.09g.

[0038] Servo driver motor control strategy: In normal mode, the motor runs at a constant speed; in dynamic excitation mode, the motor speed changes to 500 RPM within the first 0.5 seconds, then maintains the peak speed for 2 seconds, and finally sweeps the frequency of 10-100Hz sine for 5 seconds. Normal mode and dynamic excitation mode are pre-programmed in the servo driver, and the driver enters the pre-set mode when receiving instructions from the host computer (41) during operation. The preset speed values ​​for normal working mode are: ultra-low speed (0.25% of the rated speed of the motor), low speed (2.5% of the rated speed of the motor), medium speed (25% of the rated speed of the motor), and high speed (rated speed of the motor).

[0039] In this invention, the motor is a permanent magnet AC servo motor that is compatible with the permanent magnet AC servo system, the motor position sensor is built into the motor, and the motor driver is a servo driver that is compatible with the permanent magnet AC servo system.

[0040] In this embodiment, the rated speed of the permanent magnet AC servo motor is 2000 RPM. The preset speed values ​​in steps five to twelve are: ultra-low speed 5 RPM, low speed 50 RPM, medium speed 500 RPM, and high speed 2000 RPM.

Claims

1. A method for detecting minor faults in a gearbox, characterized in that, include: The motor of the permanent magnet AC servo system is directly connected to the input shaft of the gearbox under test. A vibration sensor is installed at a set position around the motor's built-in position sensor. The first position sensor is installed on the output shaft of the gearbox under test via a coupling. At the same time, a second vibration sensor is installed on the gearbox bearing housing near the first position sensor. With the motor speed set, the synchronous data acquisition board simultaneously collects the values ​​of the two position sensors and the two vibration sensors at regular intervals. The transmission error is calculated based on the values ​​of the two position sensors, and the vibration difference is obtained based on the values ​​of the two vibration sensors. When the transmission error and the vibration difference both exceed the detection threshold, or when the transmission error alone exceeds a set multiple of the detection threshold, or when the vibration difference alone exceeds a set multiple of the detection threshold, an alarm is triggered.

2. The gearbox minor fault detection method according to claim 1, characterized in that, The transmission error Δθ at time T is obtained by taking the absolute value of the values ​​of the two position sensors. The expression is: Δθ(K) = |θ1(K) - θ2(K)×i|, where k is the sampling number, k=1, 2..., and θ1(K) and θ2(K) are the position sensor measurement values ​​θ obtained by sampling at time T.

3. The gearbox minor fault detection method according to claim 1, characterized in that, The vibration difference ΔV at time T is obtained by subtracting the values ​​from the two vibration sensors and taking the absolute value. The expression is: ΔV(K) = |V1(K) - V2(K)|, where k is the sampling number, k=1, 2, ... V1(K) and V2(K) are the vibration sensor measurements V obtained at time T.

4. The gearbox minor fault detection method according to claim 1, characterized in that, With the gearbox under test unloaded, the motor drives the gearbox to run at a constant speed of R for M minutes. The maximum transmission error Δθ(K) sampled within M minutes is obtained by comparing the magnitude of the transmission error Δθ(K) sampled within M minutes, and the maximum vibration difference ΔV(K) sampled within M minutes is obtained by comparing the magnitude of the vibration difference ΔV(K). A multiple of the maximum transmission error Δθmax is taken as the detection threshold θ of the transmission error. T Take b times the maximum vibration difference ΔVmax as the detection threshold V for the vibration difference. T。 5. The gearbox minor fault detection method according to claim 1 or 2, characterized in that, The motor runs at a constant speed for ten minutes. By comparing the sampled values, the maximum value of the transmission error and the maximum value of the vibration difference within ten minutes are obtained. Twice the maximum value of the transmission error is taken as the detection threshold of the transmission error, and three times the maximum value of the vibration difference is taken as the detection threshold of the vibration difference.

6. The gearbox minor fault detection method according to claim 1 or 2, characterized in that, In normal operating mode, the transmission error Δθ(K) is greater than the threshold θ. T When c% is reached, the system switches to the incentive working state. In normal operating mode, the vibration difference ΔV(K) is greater than the threshold V. T When c% is reached, the system switches to the incentive working state. In normal operating mode, the transmission error Δθ(K) is greater than the threshold θ. T And simultaneously, the vibration difference ΔV(K) is greater than the threshold V. T If this happens, call the police immediately. In the excitation working mode, the transmission error Δθ(K) is greater than the threshold θ. T When the value is a times greater than a, immediately sound an alarm; In the excitation working mode, the vibration difference ΔV(K) is greater than the threshold V. T When the value is b times the normal value, immediately call the alarm. If the system does not alarm, the motor speed R will be adjusted to the next speed value. If the system does not alarm after the entire motor speed test process, the test process ends.

7. The gearbox minor fault detection method according to claim 6, characterized in that, In normal operating mode, the motor runs at a constant speed of R for H hours; in excitation operating mode, the motor speed changes to 25% of the rated speed within the first 0.5 seconds, then maintains the peak speed for 2 seconds, and finally performs a 10-100Hz sinusoidal sweep for 5 seconds.

8. The gearbox minor fault detection method according to claim 6, characterized in that, The motor speed R is set to ultra-low speed, low speed, medium speed, and high speed respectively: ultra-low speed is 0.25% of the motor's rated speed, low speed is 2.5% of the motor's rated speed, medium speed is 25% of the motor's rated speed, and high speed is the motor's rated speed.

9. The gearbox minor fault detection method according to claim 6, characterized in that, The motor speed setting sequence is: ultra-low speed, low speed, medium speed, high speed.

Images