Assembly quality and fault detection method, system and device for machine tool feeding system

Abstract

The invention discloses a machine tool feeding system assembly quality and fault detection method, system and device, and relates to the technical field of numerical control machine tool monitoring, the machine tool feeding system assembly quality and fault detection device comprises an upper computer, the upper computer is connected with an assembly quality and fault detection instrument of a machine tool feeding system, and a current transformer measures a current signal of a motor driven by a servo driver; the laser range finder measures the operating distance of the workbench and is arranged on the workbench. According to the assembling quality and fault detection method, system and device for the machine tool feeding system, the collected current signal data is filtered to eliminate interference signals, then the feature values are extracted, the extracted feature values are imported into the random forest model which is trained in advance, and the assembling quality and fault detection accuracy of the machine tool feeding system is improved. The assembling quality and the fault condition of the machine tool feeding system are analyzed through the model, the assembling quality and the fault condition of the machine tool feeding system can be obtained by processing and analyzing the collected current signal data, and the specific fault position is conveniently found in combination with the collected displacement data.

Description

Technical Field

[0001] The present invention relates to the technical field of numerical control machine tool monitoring, and in particular to a method, system and device for detecting assembly quality and faults of a machine tool feeding system. Background Art

[0002] The assembly quality of the machine tool feed system is crucial to the machining accuracy and service life of the CNC machine tool. Improving the assembly quality can improve the machining accuracy of the machine tool, reducing the assembly error can effectively extend the service life of the machine tool feed system components, and improving the assembly quality can reduce the maintenance time of the machine tool and improve production efficiency. Therefore, it is very important to detect the assembly quality of the machine tool feed system.

[0003] The performance indicators of CNC machine tools are mainly accuracy, reliability and life. One of the main differences between my country's high-end CNC machine tools and similar foreign machine tools is that the operating defects of functional components lead to abnormal accuracy and loss. The current detection methods for machine tool problems during operation, such as machine tool guide rail wear, bearing failure, lead screw wear and other defects, are still imperfect. Therefore, how to detect these CNC machine tool system failures caused by moving parts has become a bottleneck problem that restricts the stability of machine tool processing accuracy. Unlike the fault diagnosis of other equipment, the purpose of CNC machine tool fault diagnosis is to maintain the processing accuracy, which means that it is not possible to wait until the machine tool has serious defects for diagnosis and analysis. The traditional method of extracting mechanical operation status information is mainly to obtain vibration signals by installing external sensors, but the closed characteristics of machine tool machinery will make it difficult to install sensors. Even if it is barely installed, factors such as strong coupling of vibration information also restrict the effect of fault detection of machine tool moving parts. Summary of the invention

[0004] The purpose of the present invention is to provide a method, system and device for detecting assembly quality and faults of a machine tool feed system, so as to solve the problems raised in the above-mentioned background technology and provide a method for detecting assembly quality and faults of a machine tool feed system to perform assembly quality detection and fault diagnosis on the feed system.

[0005] To achieve the above-mentioned purpose, the present invention provides an assembly quality and fault detection device for a machine tool feeding system, comprising a host computer, the host computer is connected to an assembly quality and fault detection instrument for the machine tool feeding system, the assembly quality and fault detection instrument for the machine tool feeding system is respectively connected to a laser rangefinder and a current transformer and powered, the current transformer measures the current signal of a motor driven by a servo driver, the laser rangefinder measures the running distance of a workbench, and the laser rangefinder is arranged on the workbench.

[0006] A method for detecting assembly quality and fault detection device of a machine tool feed system, comprising the following steps:

[0007] Step 1: Connect a power line in series with the output line of the motor driven by the servo driver, connect the current transformer to the single power line, and then install and fix the current transformer and the laser rangefinder on the machine tool, so that the laser rangefinder measures the position parameters of the workbench;

[0008] Step 2: Connect the assembly quality and fault detection instrument of the machine tool to be tested and the machine tool feed system, and then connect the assembly quality and fault detection instrument of the machine tool feed system to the host computer;

[0009] Step 3: Open the host computer and adjust the assembly quality of the machine tool feed system and the acquisition parameters of the fault detection system;

[0010] Step 4: Run the machine tool workbench, collect the real-time data of the machine tool workbench through the assembly quality and fault detection system of the machine tool feed system, and form a curve of current-workbench movement displacement. After the machine tool workbench is finished running, the collection is terminated to complete the machine tool data collection;

[0011] Step 5: Process and analyze the collected real-time data of the machine tool workbench, and the faults are discovered manually by comparing with standard data and identified through random forest fault identification.

[0012] Preferably, the real-time data of the machine tool workbench includes current data measured by a current transformer and position parameters of the workbench measured by a laser rangefinder during the operation of the machine tool workbench.

[0013] A detection system for assembly quality and fault detection method of a machine tool feeding system comprises a setting module, a collection module, a display module, a storage module and a processing module.

[0014] Preferably, the setting module is used to select the channel interface for collection, set the upper and lower limits of input voltage and the upper and lower limits of output value, and set the type, order, and frequency of the filter;

[0015] The filter is a Butterworth low-pass filter, as follows:

[0016]

[0017] Where n is the order of the filter, ω c is the cut-off frequency, ω p is the passband edge frequency, ω is the frequency, |H(ω) 2 is the gain of the n-th order Butterworth filter.

[0018] Preferably, the acquisition module is used to acquire data measured by the sensor, and the measurement data includes X-axis, Y-axis, Z-axis motor current data and workbench displacement data.

[0019] Preferably, the display module is used to display the data measured by the sensor in real time, and the display module displays multiple curves simultaneously through multiple custom y-axes for comparison of multiple different y-axis curves.

[0020] Preferably, the storage module is used to store the sensor measurement data collected by the collection module, and store the data in a csv format.

[0021] Preferably, the processing module is used to process the data stored in the storage module, including simulating positive and negative motor currents, Fourier transform, wavelet envelope analysis, calculating mean value, root mean square value, kurtosis, kurtosis index, peak-to-peak value, and random forest fault identification;

[0022] Collect N points of time series current signal x1, x2, x3...x N , and calculate:

[0023] The mean is calculated as follows:

[0024]

[0025] The root mean square is calculated as follows:

[0026]

[0027] The calculation formula of kurtosis is as follows:

[0028]

[0029] The calculation formula of the kurtosis index is as follows:

[0030]

[0031] The peak-to-peak value is calculated as follows:

[0032] x p-p =max(x i )-min(x i );

[0033] Among them, max(x i ) is the maximum current value, min(x i ) is the minimum current value;

[0034] The time domain and frequency domain eigenvalues ​​of the collected current signal are extracted, and the random forest model is trained with standard data and fault data. The fault is discovered through the real-time display interface and identified by the random forest model.

[0035] Therefore, the present invention adopts the above-mentioned machine tool feed system assembly quality and fault detection method, system and device, filters the collected current signal data to eliminate interference signals, and then extracts eigenvalues. The extracted eigenvalues ​​are imported into a pre-trained random forest model, and the assembly quality and fault condition of the machine tool feed system are analyzed through the model. The specific fault location is conveniently found in combination with the collected displacement data; it is easy to install and implement, and is not limited to the machine tool operating system and programming software, and has a wide range of applications.

[0036] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 It is a schematic diagram of the structure of a method, system and device for detecting assembly quality and fault of a machine tool feeding system according to the present invention;

[0038] Figure 2 A schematic diagram of a method, system and device for detecting assembly quality and faults of a machine tool feed system according to the present invention;

[0039] Figure 3 An image of data collected by a system reading a method, system and device for detecting assembly quality and faults of a machine tool feed system according to the present invention;

[0040] Figure 4 An image drawn for a system of a method, system and device for detecting assembly quality and fault of a machine tool feed system according to the present invention. DETAILED DESCRIPTION

[0041] The technical solution of the present invention is further described below through the accompanying drawings and embodiments.

[0042] Unless otherwise defined, the technical terms or scientific terms used in the present invention should be understood by people with ordinary skills in the field to which the present invention belongs. The words "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0043] Example

[0044] See also Figure 1-4 The present invention provides a machine tool feed system assembly quality and fault detection device, including a host computer, the host computer is connected to the machine tool feed system assembly quality and fault detection instrument, the machine tool feed system assembly quality and fault detection instrument is respectively connected to a laser rangefinder and a current transformer and powered, the current transformer measures the current signal of the motor driven by the servo driver, the laser rangefinder measures the running distance of the workbench, and the laser rangefinder is arranged on the workbench. Figure 1 As shown, the power supply supplies power to the assembly quality and fault detection instruments of the machine tool feed system.

[0045] From the structure and working characteristics of the machine tool, it can be known that the transmission system, such as the lead screw, lead screw nut pair, and pull cover, is connected to the servo motor for drive control. The operating status of these driven components will directly affect the operation of the servo motor, thereby reflecting the operating status information in the motor's current signal. Therefore, the drive motor current signal of the feed system is collected, and the time domain and frequency domain eigenvalues ​​of the collected current signal are extracted. The random forest model is trained with standard data and fault data. Obvious faults in the machine tool feed system can be directly discovered through the real-time display interface, and other inconspicuous faults can be identified by the random forest model. Random forest recognition can greatly improve the accuracy and reduce the error in distinguishing faults by the human eye.

[0046] A method for detecting assembly quality and fault detection of a machine tool feed system, comprising the following steps:

[0047] Step 1: Connect a power line in series with the output line of the motor driven by the servo driver, connect the current transformer to the single power line, and then install and fix the current transformer and the laser rangefinder on the machine tool, so that the laser rangefinder measures the position parameters of the workbench;

[0048] Step 2: Connect the assembly quality and fault detection instrument of the machine tool to be tested and the machine tool feed system, and then connect the assembly quality and fault detection instrument of the machine tool feed system to the host computer;

[0049] Step 3: Open the host computer and adjust the assembly quality of the machine tool feed system and the acquisition parameters of the fault detection system;

[0050] Step 4: Run the machine tool workbench, collect the real-time data of the machine tool workbench through the assembly quality and fault detection system of the machine tool feed system, and form a curve of current-workbench movement displacement. After the machine tool workbench is finished running, the collection is ended to complete the machine tool data collection. The real-time data of the machine tool workbench includes the current data measured by the current transformer and the position parameters of the workbench measured by the laser rangefinder during the operation of the machine tool workbench.

[0051] Step 5: Process and analyze the collected real-time data of the machine tool workbench. Obvious faults are discovered manually by comparing with standard data, and subtle assembly quality can be identified through random forest fault recognition.

[0052] like Figure 2 As shown, a detection system for the assembly quality and fault detection method of a machine tool feed system includes a setting module, an acquisition module, a display module, a storage module, and a processing module. During the detection, the assembly quality and fault detection system of the machine tool feed system is first opened through the host computer. According to the actual situation, the channel interface for acquisition is selected in the setting module, the upper and lower limits of the input voltage and the upper and lower limits of the output value, the frequency, and the filter parameters are set. After setting the parameters, the data measured by the current transformer and the laser rangefinder are collected through the acquisition module, and displayed in real time through the display module. After the data is collected, the data just collected is saved in CSV format through the storage module. The saved file is selected for processing through the processing module, and the processed data is displayed by the display module.

[0053] The setting module is used to select the channel interface for acquisition, set the upper and lower limits of input voltage and output value, and set the type, order, and frequency of the filter;

[0054] In this embodiment, the filter is a Butterworth low-pass filter, as follows:

[0055]

[0056] Where n is the order of the filter, ω c is the cut-off frequency, ω p is the passband edge frequency, ω is the frequency, |H(ω) 2 is the gain of the n-th order Butterworth filter.

[0057] The acquisition module is used to collect data measured by the sensor, including X-axis, Y-axis, and Z-axis motor current data and workbench displacement data.

[0058] The display module is used to display the data measured by the sensor in real time. The display module displays multiple curves simultaneously through multiple custom y-axes for comparison of multiple different y-axis curves.

[0059] The storage module is used to store the sensor measurement data collected by the acquisition module and store the data in csv format.

[0060] The processing module is used to process the data stored in the storage module, including simulating positive and negative motor currents, Fourier transform, wavelet envelope analysis, calculating mean value, root mean square value, kurtosis, kurtosis index, peak-to-peak value, and random forest fault identification.

[0061] The operation process of the workbench is divided into start-stop, reversing, and uniform speed movement. This embodiment mainly focuses on the current data processing and analysis when the workbench moves at a uniform speed. Collect N point time series current signals x1, x2, x3...x N , and calculate:

[0062] The mean is calculated as follows:

[0063]

[0064] The root mean square is calculated as follows:

[0065]

[0066] The strength of current vibration is evaluated by the root mean square value. When a fault occurs, the root mean square value will be greater than the standard value.

[0067] The calculation formula of kurtosis is as follows:

[0068]

[0069] Kurtosis indicates the formation of a normal distribution curve of current, and when a fault occurs, the kurtosis will be significantly greater than the standard value.

[0070] The calculation formula of the kurtosis index is as follows:

[0071]

[0072] The kurtosis index is related to the RMS value and kurtosis, and can be positive or negative. When a fault occurs, the kurtosis index will increase rapidly. r =3 is zero kurtosis, K r When it is >3, it is positive kurtosis, K r When it is less than 3, it is negative kurtosis.

[0073] The peak-to-peak value is calculated as follows:

[0074] x p-p =max(x i )-min(x i );

[0075] Among them, max(x i ) is the maximum current value, min(x i ) is the minimum current value.

[0076] The current x during uniform motion p-p A peak-to-peak value less than 0.5 is normal, and a peak-to-peak value greater than 0.5 indicates a machine tool fault.

[0077] The time domain and frequency domain eigenvalues ​​of the collected current signal are extracted, and the random forest model is trained with standard data and fault data. Obvious faults of the machine tool feed system can be directly discovered through the real-time display interface, and other less obvious assembly accuracies can be identified through the random forest model. Random forest recognition can greatly improve the accuracy and reduce the error in distinguishing faults by the human eye.

[0078] like Figure 3 As shown, the data collected by the acquisition module is stored as a CSV file through the storage module, and the current displacement data image is displayed by the display module after being processed by the processing module using simulated positive and negative motor currents. The vertical axis on the left represents the distance of the X-axis relative to the measurement point measured by the laser rangefinder in millimeters (mm). This image can feedback the relative movement displacement of the workbench. The vertical axis on the right is the motor current measured by the mutual inductor in amperes (A). The horizontal axis is the time unit in seconds (s). There is a vertical straight line in the figure. This straight line follows the mouse cursor, and the time, displacement, and current values ​​at the moment of the cursor are displayed in the upper left corner for easy reading.

[0079] like Figure 4 The thin curve shown is the distance between the X-axis and the measuring point measured by the laser rangefinder. This image can provide feedback on the relative displacement of the workbench. The thick curve is the measured X-axis motor current. As shown in the thin curve, the displacement changes from the 2000mm position to the 1200mm position and then returns to the 2000mm position. The workbench moves 800mm from the starting point to the + positive direction of the X-axis and then returns as a stroke. The thick curve in the figure shows that when the current starts to run in the + positive direction of the X-axis from the workbench, the current suddenly changes from 0 to a negative number. When the workbench moves in the + direction of the X-axis, the motor current basically tends to a straight line until the workbench runs 800mm and then starts to move in the - negative direction of the X-axis. The current value changes from negative to positive and is symmetrical about the X-axis. Then the current tends to a straight line until the workbench stops moving. At this time, the current tends to a constant value. At this time, the current is the standby current.

[0080] Therefore, the present invention adopts the above-mentioned machine tool feed system assembly quality and fault detection method, system and device, and can obtain the assembly quality and fault condition of the machine tool feed system by processing and analyzing the collected current signal data, and combine the collected displacement data to conveniently find the specific fault location; and the device is easy to install and easy to implement, is not limited to the machine tool operating system and programming software, and has a wide range of applications.

[0081] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that they can still modify or replace the technical solution of the present invention with equivalents, and these modifications or equivalent replacements cannot cause the modified technical solution to deviate from the spirit and scope of the technical solution of the present invention.

Claims

1. A machine tool feed system assembly quality and fault detection device, characterized in that: It includes a host computer, which is connected to an assembly quality and fault detection instrument of a machine tool feeding system. The assembly quality and fault detection instrument of the machine tool feeding system are respectively connected to a laser rangefinder and a current transformer and are powered. The current transformer measures the current signal of a motor driven by a servo driver, and the laser rangefinder measures the running distance of a workbench. The laser rangefinder is arranged on the workbench.

2. A detection method using the assembly quality and fault detection device for a machine tool feed system as claimed in claim 1, characterized in that: The following steps are involved: Step 1: Connect a power line in series with the output line of the motor driven by the servo driver, connect the current transformer to the single power line, and then install and fix the current transformer and the laser rangefinder on the machine tool, so that the laser rangefinder measures the position parameters of the workbench; Step 2: Connect the assembly quality and fault detection instrument of the machine tool to be tested and the machine tool feed system, and then connect the assembly quality and fault detection instrument of the machine tool feed system to the host computer; Step 3: Open the host computer and adjust the assembly quality of the machine tool feed system and the acquisition parameters of the fault detection system; Step 4: Run the machine tool workbench, collect the real-time data of the machine tool workbench through the assembly quality and fault detection system of the machine tool feed system, and form a curve of current-workbench movement displacement. After the machine tool workbench is finished running, the collection is terminated to complete the machine tool data collection; Step 5: Process and analyze the collected real-time data of the machine tool workbench, and the faults are discovered manually by comparing with standard data and identified through random forest fault identification.

3. A detection method for the assembly quality and fault detection device of a machine tool feed system according to claim 2, characterized in that: The real-time data of the machine tool workbench includes the current data measured by the current transformer and the position parameters of the workbench measured by the laser rangefinder during the operation of the machine tool workbench.

4. A detection system using the assembly quality and fault detection method of a machine tool feed system as claimed in claim 2, characterized in that: It includes a setting module, an acquisition module, a display module, a storage module, and a processing module.

5. A detection system for the assembly quality and fault detection method of a machine tool feed system according to claim 4, characterized in that: The setting module is used to select the channel interface for acquisition, set the upper and lower limits of input voltage and output value, and set the type, order, and frequency of the filter; The filter is a Butterworth low-pass filter, as follows: Where n is the order of the filter, ω c is the cut-off frequency, ω p is the passband edge frequency, ω is the frequency, |H(ω) 2 is the gain of the n-th order Butterworth filter.

6. A detection system for the assembly quality and fault detection method of a machine tool feed system according to claim 4, characterized in that: The acquisition module is used to collect data measured by the sensor, including X-axis, Y-axis, and Z-axis motor current data and workbench displacement data.

7. A detection system for the assembly quality and fault detection method of a machine tool feed system according to claim 4, characterized in that: The display module is used to display the data measured by the sensor in real time. The display module displays multiple curves simultaneously through multiple custom y-axes for comparison of multiple different y-axis curves.

8. A detection system for the assembly quality and fault detection method of a machine tool feed system according to claim 4, characterized in that: The storage module is used to store the sensor measurement data collected by the acquisition module and store the data in csv format.

9. A detection system for the assembly quality and fault detection method of a machine tool feed system according to claim 4, characterized in that: The processing module is used to process the data stored in the storage module, including simulating positive and negative motor currents, Fourier transform, wavelet envelope analysis, calculating mean value, root mean square value, kurtosis, kurtosis index, peak-to-peak value, and random forest fault identification; Collect N points of time series current signal x1, x2, x3...x N , and calculate: The mean is calculated as follows: The root mean square is calculated as follows: The calculation formula of kurtosis is as follows: The calculation formula of the kurtosis index is as follows: The peak-to-peak value is calculated as follows: x p-p =max(x i )-min(x i ); Among them, max(x i ) is the maximum current value, min(x i ) is the minimum current value; The time domain and frequency domain eigenvalues ​​of the collected current signal are extracted, and the random forest model is trained with standard data and fault data. The fault is discovered through the real-time display interface and identified by the random forest model.

Images