A worm grinding eccentric wear abnormality online monitoring method based on a ratio feature

Abstract

The present application relates to a grinding process monitoring method, in particular to a worm grinding eccentric grinding abnormality online monitoring method based on ratio characteristics, which is used to solve the problems of tooth surface morphology error and eccentric grinding in the traditional gear worm grinding process, and the long time consumption and the inability to completely avoid the eccentric grinding risk when using gear detection center to detect tooth shape and tooth direction. The worm grinding eccentric grinding abnormality online monitoring method based on ratio characteristics collects the spindle servo current data of A complete machining processes of the same workpiece as the workpiece to be monitored, constructs the left band ratio characteristic R L , the right band ratio characteristic R R , and determines the threshold value according to the state simulated by the A workpieces, and then collects the spindle servo current data of the complete machining process of the workpiece to be monitored, judges whether the worm grinding machining process has eccentric grinding according to the threshold value, and realizes the real-time monitoring and detection of the eccentric grinding phenomenon in the worm grinding process.

Description

Technical Field

[0001] This invention relates to a method for monitoring the grinding process, specifically to an online monitoring method for abnormal uneven wear in worm gear grinding based on ratio characteristics. Background Technology

[0002] Gears, as a core component of gearboxes, play a crucial role in the transmission system. Their precision directly affects transmission efficiency, load-bearing capacity, and noise levels. Grinding is one of the key processes in gear manufacturing, used for precision machining of gear tooth surfaces.

[0003] Traditional gear grinding typically uses worm gear grinding wheels. This process requires initial tooth finding to ensure the grinding wheel is aligned with the centerline of adjacent tooth surfaces, thus preventing uneven wear. However, existing technologies have several drawbacks. For instance, during worm gear grinding, heat treatment can cause gear deformation, leading to errors in tooth surface shape. Additionally, misalignment of the grinding wheel detection sensor and decreased reciprocating accuracy of the actuator can also result in uneven wear.

[0004] Chinese patent CN104759943A discloses a device for detecting the morphology and performance of gears. Its main feature is the use of a high-precision machine tool system and a robotic arm system for gear inspection, including vibration signal and noise acquisition, gear meshing inspection, and grinding burn detection. While this device improves the accuracy and automation of the inspection, its complex inspection process and system configuration result in a slow inspection speed and difficulty in comprehensively inspecting every gear, thus remaining an efficiency issue. Furthermore, this device only inspects the morphology and performance of gears after grinding, and cannot monitor and adjust them in real time during the grinding process, still relying on post-processing inspection to identify problems. This leaves the risk of uneven grinding unavoidable. Summary of the Invention

[0005] The purpose of this invention is to address the shortcomings of traditional gear and worm grinding processes, such as tooth surface morphology errors and uneven wear, as well as the time-consuming and unavoidable risks associated with tooth profile and direction detection using gear inspection centers. This invention provides an online monitoring method for uneven wear anomalies in worm grinding based on ratio characteristics.

[0006] To address the shortcomings of the existing technology, the present invention provides the following technical solution:

[0007] A method for online monitoring of abnormal wear in worm gear grinding based on ratio characteristics, characterized by the following steps:

[0008] Step 1: Based on the workpiece parameters and machining parameters of the workpiece to be monitored, calculate the frequency characteristics of the worm gear grinding process. The frequency characteristics include the main machining frequency f. DZ,as follows:

[0009] f DZ =|f D ±f N |

[0010] Where f D For AC power frequency, f N The meshing frequency is equal to the number of teeth z of the workpiece to be monitored and the workpiece shaft rotation frequency f. g The product;

[0011] Step 2: Collect spindle servo current data for the complete machining process of A workpieces identical to the workpiece to be monitored, where A is an integer ≥ 9; among the A workpieces, at least three workpieces have a tooth groove center deviation of 0 to simulate the normal state; at least three workpieces have a tooth groove center deviation of [-R, 0), and at least three workpieces have a tooth groove center deviation of (0, B] to simulate the uneven wear state; B∈[0.05, 0.15], in μm;

[0012] Step 3: From the spindle servo current data obtained in Step 2, select the continuous current data sequence during the rough grinding stage of the worm gear grinding process for each workpiece; obtain the current data envelopment spectrum for each workpiece based on this current data sequence.

[0013] Step 4: In the data envelopment spectrum of the current data of each workpiece in Step 3, select the machining main frequency f with the larger amplitude in Step 1. DZ The actual value of the processing main frequency f′ Dz To process the actual value of the main frequency f′ Dz Centered on the frequency band, preset bandwidths ξ are set on both sides, and the maximum amplitude value is searched within the frequency band as the energy characteristic F of the main frequency band to be processed. Z The maximum amplitude value is searched within the left and right frequency bands and used as the energy characteristic F of the left frequency band, respectively. L Right-side frequency band energy characteristics F R Then calculate the left-side frequency band ratio characteristic R. L Right-side frequency band ratio characteristic R R ,as follows:

[0014] R L =F Z / F L R R =F Z / F R ;

[0015] Step 5: Based on the left-side frequency band ratio characteristics R of each workpiece in Step 4 L Right-side frequency band ratio characteristic R R Based on the simulated state, determine the threshold;

[0016] Step 6: Collect spindle servo current data for the complete machining process of the workpiece to be monitored, and calculate its left-side frequency band ratio characteristic R according to Steps 3 and 4. L Right-side frequency band ratio characteristic R R Then, based on the threshold obtained in step 5, it is determined whether uneven wear occurs during the worm grinding process, thus realizing online monitoring of uneven wear abnormalities in worm grinding.

[0017] Further, in step 1, the workpiece parameters include the number of teeth z of the workpiece to be monitored, and the processing parameters include the spindle speed of the gear grinding machine and the workpiece shaft speed f. g The sensitivity and current acquisition frequency of the current sensor in the servo system used to control the spindle speed.

[0018] Furthermore, step 1 also includes:

[0019] By adding two macro commands to the NC code of the worm gear grinding CNC system as the start and end timestamps for each worm gear grinding operation, the start and end times of each worm gear grinding operation are determined by acquiring the macro commands through the OPC UA protocol.

[0020] Furthermore, it is characterized by:

[0021] In step 3, the selection of continuous current data sequence during the rough grinding stage of the worm gear grinding process of each workpiece specifically means: from the spindle servo current data obtained in step 2, select the continuous current data sequence x(t) for 10±5 seconds from the spindle servo current data 15±5 seconds after the start of the worm gear grinding process of each workpiece.

[0022] Furthermore, in step 3, obtaining the data envelopment spectrum of the current data for each workpiece based on the current data sequence specifically involves performing a Hilbert transform on the current data sequence x(t) to obtain the analytical signal. The formula is shown below:

[0023]

[0024] Where H[] represents the Hilbert transform, and τ represents the time point in the integration process;

[0025] For analytical signals After removing the mean and taking the absolute value, a new time-domain current signal X(t) is obtained. The calculation formula is as follows:

[0026]

[0027] Where n is the analytic signal The number of samples;

[0028] Performing a Fourier transform on the new time-domain current signal X(t) yields the current data envelopment spectrum, as shown below;

[0029]

[0030] Where ω = 2πf is the angular frequency, and f is the meshing frequency. N e -jωt It is a complex exponential function used to convert time-domain signals to the frequency domain.

[0031] Further, step 5 specifically involves: statistically analyzing the left-side frequency band ratio characteristic R of workpieces with non-zero tooth groove center deviation. L Right-side frequency band ratio characteristic R R The maximum value, and integers greater than the maximum value and multiples of 10, are taken as the left-side frequency band ratio characteristic R of the area where wear occurs. L Right-side frequency band ratio characteristic R R The upper limit;

[0032] The threshold value is determined as the left-side frequency band ratio characteristic R where uneven wear occurs. L Right-side frequency band ratio characteristic R R The upper limit.

[0033] Further, in step 6, determining whether uneven wear occurs during the worm gear grinding process based on the threshold obtained in step 5 specifically involves: if the left-side frequency band ratio characteristic R... L Right-side frequency band ratio characteristic R R If all values ​​are greater than 60, then the wear tolerance is within the allowable range and it is judged that no wear has occurred; otherwise, it is judged that wear has occurred.

[0034] Furthermore, in step 4, the preset bandwidth ξ is 1±0.5Hz.

[0035] Compared with the prior art, the beneficial effects of the present invention are:

[0036] (1) The present invention provides an online monitoring method for abnormal wear in worm grinding based on ratio characteristics. This method collects spindle servo current data of A workpieces with the same working material as the workpiece to be monitored during the complete machining process, and constructs the left-side frequency band ratio characteristic R. L Right-side frequency band ratio characteristic R R Based on the simulated state of A workpieces, a threshold is determined. Then, spindle servo current data of the complete machining process of the workpiece to be monitored is collected. Based on the threshold, it is determined whether uneven wear occurs during the worm gear grinding process, realizing real-time monitoring and detection of uneven wear during worm gear grinding. In addition, through real-time data analysis, it is possible to detect and adjust in a timely manner during the machining process to prevent uneven wear from occurring.

[0037] (2) Based on the real-time acquisition and analysis of spindle servo current data, this invention has low algorithm complexity and fast calculation speed, which can quickly process and judge during worm grinding, significantly improving detection efficiency.

[0038] (3) Since the present invention performs real-time monitoring during the processing, no additional inspection process is required. Therefore, it can achieve comprehensive inspection of each workpiece to be monitored, eliminate the limitations of sampling inspection, and ensure the processing quality of each workpiece to be monitored.

[0039] (4) By adding macro commands, the present invention can accurately collect servo current data of each grinding operation of the gear grinding machine.

[0040] (5) The left-side frequency band ratio feature R constructed in this invention L Right-side frequency band ratio characteristic R R It is computationally simple, has low algorithm complexity, good robustness, fast calculation speed, and good accuracy. It can replace some offline inspection programs after processing, shorten the overall production line cycle time, and improve production efficiency.

[0041] (6) The present invention collects spindle servo current data through a current sensor, which is inexpensive and has the advantages of high signal-to-noise ratio and strong stability compared with vibration signal, making it suitable for mass gear production processes. Attached Figure Description

[0042] Figure 1 This is a flowchart of an embodiment of an online monitoring method for abnormal wear in worm gear grinding based on ratio characteristics according to the present invention;

[0043] Figure 2 This is a discrete time series diagram of the spindle servo current data for the complete machining process of one of the workpieces in step 2 of this embodiment of the invention.

[0044] Figure 3 This is a sequence diagram of current data for one of the workpieces in step 3.1 of this embodiment of the invention;

[0045] Figure 4 This is the current data envelopment spectrum of one of the workpieces in step 3.4 of the embodiment of the present invention. Detailed Implementation

[0046] The present invention will be further described below with reference to the accompanying drawings and exemplary embodiments.

[0047] Reference Figure 1 This invention discloses an online monitoring method for abnormal wear in worm gear grinding based on ratio characteristics, comprising the following steps:

[0048] Step 1: Calculate the frequency characteristics of the worm gear grinding process based on the workpiece parameters and machining parameters of the workpiece to be monitored. The workpiece parameters include the number of teeth z of the workpiece to be monitored; the machining parameters include the spindle speed of the gear grinding machine, the workpiece shaft speed, the sensitivity of the current sensor in the servo system used to control the spindle speed, and the current acquisition frequency; the frequency characteristics include the meshing frequency f. N Workpiece axis rotation frequency f g Processing main frequency f DZ ;

[0049] Specifically, the workpiece to be monitored is a gearbox gear with 25 teeth (z); the gear grinding machine is a Qinchuan YKZ7230 with a spindle speed of 3000 rpm, a workpiece shaft speed of 480 rpm, a sensitivity of 100 mV / V, and a current acquisition frequency of 12800 Hz.

[0050] The meshing frequency f N The number of teeth z of the workpiece to be monitored and the workpiece shaft rotation frequency f g The product of is expressed as follows:

[0051]

[0052] The processing main frequency f DZ The definition is as follows:

[0053] f DZ =|f D ±f N |

[0054] Where f D The AC power frequency is 50Hz; the main processing frequency is f. DZ For |f D +f N |or f DZ =|f D -f N |;

[0055] Processing main frequency f DZ 250Hz or 150Hz;

[0056] By adding two macro commands, R500 and R501, to the NC code of the worm grinding CNC system as the start and end timestamps of each worm grinding operation, the start and end times of each worm grinding operation are determined by collecting the macro commands through the OPC UA protocol.

[0057] Step 2: Collect discrete time series of spindle servo current data for the complete machining process of 9 workpieces identical to the workpiece to be monitored; the tooth groove center deviation of three workpieces is 0 to simulate the normal state; the tooth groove center deviation of three workpieces is in [-R, 0), and at least three workpieces have a tooth groove center deviation of (0, B) to simulate the uneven wear state.

[0058] B∈[0.05, 0.15], in this embodiment B is 0.15, and the unit is μm;

[0059] The discrete-time series of spindle servo current data for the complete machining process of one of the workpieces is as follows: Figure 2 As shown;

[0060] Step 3: From the spindle servo current data obtained in Step 2, select the continuous current data sequence during the rough grinding stage of the worm gear grinding process for each workpiece; obtain the current data envelopment spectrum for each workpiece based on this current data sequence, as follows:

[0061] Step 3.1: In the spindle servo current data obtained in Step 2, select the current data sequence x(t) that lasts for 10 consecutive seconds from the spindle servo current data 15±5 seconds after the start of worm gear grinding for each workpiece.

[0062] The current data sequence x(t) of one of the workpieces is as follows Figure 3 As shown;

[0063] Step 3.2: Perform a Hilbert transform on the current data sequence x(t) selected in Step 3.1 to obtain the analytic signal. The formula is shown below:

[0064]

[0065] Where H[] represents the Hilbert transform, and τ represents the time point in the integration process;

[0066] Step 3.3: Analyze the signal After removing the mean and taking the absolute value, a new time-domain current signal X(t) is obtained. The calculation formula is as follows:

[0067]

[0068] Where n is the analytic signal The number of samples;

[0069] Step 3.4: Perform a Fourier transform on the new time-domain current signal X(t) to obtain the current data envelopment spectrum, as shown below;

[0070]

[0071] Where ω = 2πf is the angular frequency, and f is the meshing frequency. N e -jωt It is a complex exponential function used to convert time-domain signals to the frequency domain;

[0072] The current data envelopment spectrum of one of the workpieces is as follows Figure 4 As shown;

[0073] Step 4: In the data envelopment spectrum of the current data of each workpiece in Step 3, select the machining main frequency f with the larger amplitude from Step 1. DZ (150Hz) is used as the actual value of the main processing frequency f′ Dz To process the actual value of the main frequency f′ Dz With a center, a preset bandwidth ξ is set on both sides, with the preset bandwidth ξ being 1±0.5Hz, and 1Hz in this embodiment;

[0074] The maximum amplitude value is searched within the frequency band and used as the energy characteristic F of the main frequency band for processing. Z The maximum amplitude value is searched within the left and right frequency bands and used as the energy characteristic F of the left frequency band, respectively. L Right-side frequency band energy characteristics F R The calculation formula is as follows:

[0075] F Z =max(F(f′) Dz -1): F(f′ Dz +1))

[0076] F L =max(F(f′) Dz -1): f′ Dz )

[0077] F R =max(f′) Dz :F(f′) Dz +1))

[0078] Then calculate the left-side frequency band ratio characteristic R. L Right-side frequency band ratio characteristic R R The calculation formula is as follows.

[0079] R L =F Z / F L

[0080] R R =F Z / F R

[0081] The left-side frequency band ratio characteristics R of the nine workpieces were obtained. L The frequency band ratio (RR) characteristics on the right side are shown in Table 1:

[0082] Table 1

[0083] state <![CDATA[F Z / g]]> <![CDATA[F L / g]]> <![CDATA[F R / g]]> <![CDATA[R L ]]> <![CDATA[R R ]]> normal 21.86 0.11 0.13 197.17 162 normal 21.25 0.15 0.14 143.96 150.53 normal 19.61 0.12 0.1 164.71 194.69 Uneven wear 18.26 0.69 0.69 29.97 26.48 Uneven wear 21.17 0.81 0.83 26.2 25.64 Uneven wear 20.58 0.67 0.62 30.53 25.24 Uneven wear 22.73 0.33 0.34 59.08 53.22 Uneven wear 20.85 0.44 0.39 47.38 53.22 Uneven wear 22.82 0.57 0.62 39.96 36.54

[0084] As can be seen from the table above, the left-side frequency band ratio characteristic R during normal worm gear grinding process is... L Right-side frequency band ratio characteristic R R All are above 140, but when uneven wear occurs, the frequency band ratio characteristic R on the left side... L Right-side frequency band ratio characteristic R R All are below 60. Due to the forced vibration of the workpiece rotation during uneven wear, the workpiece shaft sideband energy is higher near the main frequency, resulting in a higher frequency band ratio characteristic R on the left side. L Right-side frequency band ratio characteristic R R Decrease;

[0085] Step 5: Based on the left-side frequency band ratio characteristics R of each workpiece in Step 4 L Right-side frequency band ratio characteristic R R Based on the simulated state, determine the threshold;

[0086] Specifically, the left-side frequency band ratio characteristic R of workpieces with non-zero tooth groove center deviation is statistically analyzed. L Right-side frequency band ratio characteristic R R The maximum value (59.08) is used as the left-side frequency band ratio characteristic R for wear, where the integer 60 is greater than the maximum value and is a multiple of 10. L Right-side frequency band ratio characteristic R R The upper limit;

[0087] The left-side frequency band ratio characteristic R was identified as the area where uneven wear occurred. L Right-side frequency band ratio characteristic R R The upper limit;

[0088] Step 6: Collect spindle servo current data for the complete machining process of the workpiece to be monitored, and calculate its left-side frequency band ratio characteristic R according to Steps 3 and 4. L Right-side frequency band ratio characteristic R R Then, it determines whether uneven wear occurs during the worm grinding process, thus achieving online monitoring of abnormal uneven wear in worm grinding;

[0089] Specifically, if the frequency band ratio characteristic R on the left side L Right-side frequency band ratio characteristic R R If all values ​​are greater than 60, then the wear tolerance is within the allowable range and it is judged that no wear has occurred; otherwise, it is judged that wear has occurred.

Claims

1. A method for online monitoring of abnormal wear in worm gear grinding based on ratio characteristics, characterized in that, Includes the following steps: Step 1: Based on the workpiece parameters and machining parameters of the workpiece to be monitored, calculate the frequency characteristics of the worm gear grinding process. The frequency characteristics include the main machining frequency. ,as follows: ； in AC power frequency; The meshing frequency is equal to the number of teeth on the workpiece being monitored. With workpiece axis rotation frequency The product; Step 2: Collect spindle servo current data for the complete machining process of A workpieces identical to the workpiece to be monitored, where A is an integer ≥ 9; among the A workpieces, at least three workpieces must have a tooth groove center deviation of 0 to simulate a normal state; at least three workpieces must have a tooth groove center deviation of... At least three workpieces have a tooth groove center deviation of . To simulate uneven wear; The unit is μm; Step 3: From the spindle servo current data obtained in Step 2, select a sequence of current data for 10 ± 5 seconds consecutively from the spindle servo current data 15 ± 5 seconds after the start of worm gear grinding for each workpiece. ; The data envelopment spectrum of the current data for each workpiece is obtained based on the current data sequence, specifically: [The text abruptly ends here, likely due to an incomplete sentence or a formatting error.] Perform a Hilbert transform to obtain the analytic signal. The formula is shown below: ； Where H[] denotes the Hilbert transform, This represents a point in time during the integration process; For analytical signals After removing the mean and taking the absolute value, a new time-domain current signal is obtained. The calculation formula is as follows. ； in, To analyze the signal The number of samples; For the new time-domain current signal Performing a Fourier transform yields the current data envelopment spectrum, as shown below; ； in , where is the angular frequency. Meshing frequency , It is a complex exponential function used to convert time-domain signals to the frequency domain; Step 4: In the data envelopment spectrum of the current data of each workpiece in Step 3, select the machining main frequency with the larger amplitude in Step 1. As the actual value of the processing main frequency To process the actual value of the main frequency Preset bandwidths are set on both sides of the center. The maximum amplitude value is searched within the frequency band to serve as the energy characteristic of the main frequency band being processed. The maximum amplitude value is searched within the left and right frequency bands and used as the energy characteristic of the left frequency band. Energy characteristics of the right-side frequency band Then calculate the left-side frequency band ratio characteristics. Right-side frequency band ratio characteristics ,as follows: ， ； Step 5: Based on the left-side frequency band ratio characteristics of each workpiece in Step 4. Right-side frequency band ratio characteristics Based on the simulated state, determine the threshold; Step 6: Collect spindle servo current data for the complete machining process of the workpiece to be monitored, and calculate its left-side frequency band ratio characteristics according to Steps 3 and 4. Right-side frequency band ratio characteristics Then, based on the threshold obtained in step 5, it is determined whether uneven wear occurs during the worm grinding process, thus realizing online monitoring of uneven wear abnormalities in worm grinding.

2. The online monitoring method for abnormal worm gear grinding wear based on ratio characteristics according to claim 1, characterized in that: In step 1, the workpiece parameters include the number of teeth on the workpiece to be monitored. The processing parameters include the spindle speed of the gear grinding machine and the workpiece shaft speed. The sensitivity and current acquisition frequency of the current sensor in the servo system used to control the spindle speed.

3. The online monitoring method for abnormal wear in worm gear grinding based on ratio characteristics according to claim 1, characterized in that, Step 1 further includes: By adding two macro commands to the NC code of the worm gear grinding CNC system as the start and end timestamps for each worm gear grinding operation, the start and end times of each worm gear grinding operation are determined by acquiring the macro commands through the OPC UA protocol.

4. The online monitoring method for abnormal wear in worm gear grinding based on ratio characteristics according to claim 1, characterized in that: Step 5 specifically involves: statistically analyzing the left-side frequency band ratio characteristics of workpieces with non-zero tooth groove center deviation. Right-side frequency band ratio characteristics The maximum value, and integers greater than the maximum value and multiples of 10, are used as the left-side frequency band ratio characteristics where wear occurs. Right-side frequency band ratio characteristics The upper limit; The threshold was determined as the left-side frequency band ratio characteristic where uneven wear occurred. Right-side frequency band ratio characteristics The upper limit.

5. The online monitoring method for abnormal wear in worm gear grinding based on ratio characteristics according to claim 4, characterized in that: In step 6, determining whether uneven wear occurs during the worm gear grinding process based on the threshold obtained in step 5 specifically involves: if the left-side frequency band ratio characteristic Right-side frequency band ratio characteristics If all values ​​are greater than 60, then the wear tolerance is within the allowable range and it is judged that no wear has occurred; otherwise, it is judged that wear has occurred.

6. The online monitoring method for abnormal wear in worm gear grinding based on ratio characteristics according to claim 5, characterized in that: In step 4, the preset bandwidth It is 1±0.5 Hz.

Images