A real-time weld deviation information extraction method based on magnetic control-temperature sensing

Abstract

The application discloses a real-time welding seam deviation extraction method based on magnetic control-temperature sensing. It mainly solves the problem that the magnetic control arc sensor cannot accurately obtain the welding seam deviation information under the unstable state of magnetic field and arc current signal collection. The technical key point is that the magnetic control-temperature sensor collects the current signal and the average temperature value of the arc at any position, and obtains the arc current change value and the temperature change rate value through data analysis and processing. The function relationship between the temperature change rate and the arc position is established, the arc current change value is input into the temperature-arc current function relationship to output the temperature sensor temperature change value. The groove welding seam deviation information is directly reflected through the temperature change value, and a novel welding seam deviation information extraction method is provided for improving the welding process, automatic welding seam tracking and other occasions.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of weld deviation information extraction, and particularly relates to a real-time weld deviation information extraction method based on magnetic control-temperature sensing. BACKGROUND

[0002] Welding has a very wide application in industrial production, especially in the fields of automobiles, construction and aerospace. However, the current ordinary magnetic control arc sensor generally works in a way of controlling and collecting arc current signals, but this way lacks stability and accuracy, and has a high requirement for the stability of the welding environment, which is not conducive to the extraction of weld deviation information under continuous and stable conditions. The magnetic control-temperature sensor uses the temperature change rate value of the temperature sensor to reflect the weld deviation information in real time through the established functional relationship, that is, the temperature sensor displays how many numerical values, which can accurately reflect the arc position, that is, the weld deviation information in real time. At the same time, since the temperature sensor is not affected by the magnetic field and the arc current signal, compared with other ordinary sensors, its adaptability is relatively stronger, and therefore it has better application and development prospects.

[0003] The patent CN105562886B applies magnetic control rotating arc to surfacing, makes the arc scan the surfacing layer weld according to the specified frequency, obtains the weld information from the magnetic control arc sensor, inputs the main control for processing and outputs to the identification controller, identifies the surface topography of the surfacing layer and determines the weld deviation signal, combines the topography signal to solve the optimal posture signal of the welding torch, and outputs to the main controller to drive the actuator and the displacement to adjust the space posture of the welding torch, so as to realize the surface weld tracking. Although the invention can effectively solve the extraction of weld deviation information in the surfacing process and realize the tracking of multi-layer and multi-pass welds, when the magnetic control is unstable, the rotating arc is also out of control, the data information obtained by the main controller from the identification controller is also unstable, and therefore the accuracy of the welding torch posture control cannot be ensured.

[0004] The patent CN109719369B uses a multi-stage capacitor sensor to collect the distance information of the capacitor to the workpiece, obtains the positions of multiple points, and then fits the arc prediction trajectory by using a cubic spline interpolation algorithm, converts the obtained height function into a current function as a standard current signal to input into a current tracking circuit, and completes real-time correction by controlling the pulse duty cycle of the magnetic control arc sensor, so as to realize automatic welding. Although the position information can be obtained in real time by using the multi-stage capacitor sensor, and the arc pre-trail is obtained by using the cubic spline interpolation algorithm to extract the weld deviation, the magnetic control of the multi-stage capacitor cannot be guaranteed to be stable, that is, the weld deviation extracted by the invention also has the problem of unstable magnetic control. SUMMARY

[0005] In order to overcome the above prior art deficiencies, the present application provides a real-time welding seam deviation information extraction method based on magnetic control-temperature sensing, which can ensure stability against magnetic field and arc current signal and effectively improve the accuracy and adaptability of welding seam deviation information extraction.

[0006] In view of the above problems, the technical solution of the present application is as follows:

[0007] The present embodiment adopts a real-time welding seam deviation information extraction method based on magnetic control-temperature sensing. The magnetic control-temperature sensor collects the current data signal I and the average temperature value Tii of the arc column at any position, and then the Hall sensor is used to obtain the welding current signal. Then, the Butterworth hardware filtering and wavelet analysis software processing methods are used respectively to obtain the arc current change value which can be directly input. The arc current change value is input into the temperature-arc function relationship to output the temperature sensor temperature change rate value. Then, the arc position deviation characteristic value is directly reflected by the temperature sensor temperature change rate value, and the slope welding seam deviation information is reflected by the arc position determined by the arc current signal value, thereby realizing a new way of obtaining the welding seam deviation information from the temperature change rate value, and providing a new welding seam deviation information extraction method for welding seam slope identification, automatic welding seam tracking, welding gun posture adjustment and other occasions.

[0008] Firstly, the magnetic control-temperature sensor includes a welding gun (1), a magnetic conducting core (2), an excitation coil (3), a multi-directional magnetic pole (4), a temperature sensor probe (5), and a temperature sensor (7). The excitation coil is wound around the magnetic conducting core to generate an alternating magnetic field perpendicular to the direction of the arc column by applying an alternating excitation current. Meanwhile, n pairs of temperature sensors (T1, T2…Tn) are arranged around the magnetic control arc sensor. n The temperature sensors are symmetrically and uniformly distributed below the multi-directional magnetic pole and fixed as a whole with the multi-directional magnetic pole, so that the temperature sensor probe placed in front can uniformly collect the average temperature value of the arc column at any position. The temperature change rate value is calculated from the average temperature value of the arc column, and the function relationship between the temperature change rate and the arc position is established. The characteristic value of the arc position change under different temperature change rates is analyzed, and the arc position deviation characteristic value information is extracted by directly replacing the arc current signal value with the temperature sensor temperature change rate.

[0009] Secondly, the function between the temperature difference change rate and the arc position is established by the temperature change rate value and the arc current signal value in the first step. The average temperature value of the arc column of the n pairs of symmetrically distributed temperature sensors is calculated to obtain the temperature change rate value. Then, the Newton interpolation method is used to determine the X and Y direction distance value of the arc at the horizontal section of the temperature sensor probe, i.e. the relative coordinates (x i ,y iIf the arc length l is equal to the sum of the length and width of the magnetic control-temperature sensor (H, L), then we have:

[0010]

[0011] In the formula, H and L are the length and width of the magnetic control-temperature sensor, respectively.

[0012] The collected arc current signal is converted into the arc length l θ and the swing offset angle θ i of the arc.

[0013] It is assumed that the arc swing is uniform and smooth, i.e., the swing offset angle θ i = kt, (where k is a proportional coefficient and t is the arc swing time).

[0014] Since the weld bevel state is mainly the change of the welding gun height, i.e., the change of the arc length. And the change of the arc length is also related to other parameters, the mathematical model of the arc length is represented as: l = KI + b, (where: I is the welding current value; K, b are proportional coefficients related to the power source characteristics and welding materials)

[0015] In the third step, a multiple data function relationship between the temperature change rate and the arc position is established through a multiple linear regression equation model, which is specifically:

[0016] Let t1 and t2 be control variables, x i and y i (i = 1, 2, 3…n, representing n pairs of temperature sensors) be a random variable, then they have the following relationship:

[0017] kx i + ly i = b0 + b1t1 + b2t2 + ε

[0018] ε ~ N(0, σ 2 )

[0019] In the above formula, the control variables t1 and t2 represent the arc length l θ and the swing offset angle θ i data.

[0020] Finally, the least squares estimation method is used to estimate the parameters b0, b1, b2, σ 2 to obtain a function relationship with better fitting effect, so as to achieve real-time response and accuracy.

[0021] From the above equation, the fitted function relationship is:

[0022]

[0023] ε ~ N(0, σ 2 )

[0024] At the same time, it is assumed that When the least square estimation is to take the parameter vector B appropriate estimated value, so that Q reaches the minimum, let the estimated value of B be Then Should meet:

[0025]

[0026] Compared with the prior art, the beneficial effects of the present application are:

[0027] The magnetic control-temperature sensor is adopted, the arc current signal data and the average temperature value of the arc center column at any position are collected, then the wavelet analysis filtering algorithm and the double de-extreme value average method between the effective values are respectively adopted to obtain the arc current change value and the average temperature value of the arc center column, then the temperature change rate value of the temperature sensor is output in the temperature-arc current function relationship established by inputting the two data, the arc deviation information is directly reflected by the temperature change rate value of the temperature sensor, and the slope groove weld deviation information is reflected by the arc position determined by the arc current signal value. Finally, the weld deviation information extraction under the unstable state of the magnetic field and the arc current signal can be realized, and a new idea is provided for the weld automatic tracking and the welding gun posture adjustment. BRIEF DESCRIPTION OF DRAWINGS

[0028] Figure 1 It is a flow chart of the real-time weld deviation information extraction method based on the magnetic control-temperature sensor.

[0029] Figure 2 It is an arc coordinate position diagram of the real-time weld deviation information extraction method based on the magnetic control-temperature sensor.

[0030] Figure 3 It is a magnetic control-temperature sensor signal collection schematic diagram of the real-time weld deviation information extraction method based on the magnetic control-temperature sensor.

[0031] Figure 3 1-welding gun; 2-magnetic conducting core; 3-excitation coil; 4-multi-directional magnetic pole; 5-temperature sensor probe; 6-workpiece; 7-temperature sensor; 8-arc. DETAILED DESCRIPTION

[0032] The present application will be further described below in combination with the drawings and examples.

[0033] Referring to Figure 1 and Figure 2 , the real-time weld deviation information extraction method based on the magnetic control-temperature sensor refers to that the current data signal I and the average temperature value T of the arc column at any position moment are collected by the magnetic control-temperature sensor, then the wavelet analysis filtering algorithm and the double de-extreme value average method between the effective values are respectively adopted to obtain the arc current change value and the average temperature value of the arc center column, then the temperature change rate value of the temperature sensor is output in the temperature-arc current function relationship established by inputting the two data, the arc deviation information is directly reflected by the temperature change rate value of the temperature sensor, and the slope groove weld deviation information is reflected by the arc position determined by the arc current signal value. Finally, the weld deviation information extraction under the unstable state of the magnetic field and the arc current signal can be realized, and a new idea is provided for the weld automatic tracking and the welding gun posture adjustment. iiThe welding arc current signal is obtained by using the Hall sensor, and then the change value of the arc current is obtained by using the Butterworth hardware filtering and the wavelet analysis software processing. The change value of the arc current is input into the temperature-arc current function relationship to output the change rate value of the temperature sensor, and the change rate value of the temperature sensor directly reflects the arc position deviation characteristic value, and the arc position determined by the arc current signal value reflects the groove weld deviation information, so as to realize the weld deviation information extraction method by using the change rate of the temperature sensor instead of the arc current signal. The embodiment method can provide more selection conditions for weld groove identification, automatic weld tracking, welding gun posture adjustment and the like, and has high accuracy, good real-time performance and strong adaptability.

[0034] Referring to Figure 2 The real-time weld deviation information extraction method based on the magnetic control-temperature sensing of the embodiment 1 can be applied to weld tracking. First, the alternating magnetic field controls the arc to swing left and right in the cross section, so as to facilitate the arc to quickly scan the groove and obtain the weld deviation information of the groove. Then, the temperature sensor can input the arc current signal into the temperature-arc function equation to output the change rate value of the temperature sensor, so as to obtain the real-time and accurate weld deviation information by using the temperature change value, and input the real-time change of the weld deviation information into the control system. Finally, the control system inputs the driving signal to the actuator to promote the welding gun to move left and right, and thus accurate weld tracking is realized.

[0035] Referring to Figure 2 and Figure 3 The real-time weld deviation information extraction method based on the magnetic control-temperature sensing of the embodiment 2 can be applied to improve the welding process. The swing angle of the arc affected by the magnetic control is difficult to accurately observe and record, that is, the swing angle and height of the arc will have a certain influence on the welding process. The temperature sensor is used as a real-time detector to detect and supervise the swing angle and height of the arc at any position and time, so as to have a good effect of recording and observing the arc in real time, and has a good influence on improving the welding process in the welding process. Finally, the angle and height information of the arc swing can be understood by observing the numerical information on the temperature sensor, so as to directly obtain the weld process information, and ensure the visualization of the arc swing information affected by the magnetic control.

Claims

1. A real-time weld deviation extraction method based on magnetron-temperature sensing, characterized by: The arc current signal value at any position and moment of the arc is collected by a magnetic control-temperature sensor and the average temperature value of the arc column The arc current signal value is input into a temperature-arc function relationship formula to output the temperature sensor temperature change rate value, and the temperature change rate value of the temperature sensor directly reflects the arc position deviation characteristic value, and the arc position determined by the arc current signal value reflects the groove weld deviation information, so as to obtain the weld deviation information through the temperature change rate value. The real-time welding seam deviation information extraction method based on the magnetic control-temperature sensor is characterized in that the magnetic control-temperature sensor is obtained by winding a coil around a magnetic conducting core (2), applying an alternating excitation current, generating an alternating magnetic field perpendicular to the direction of the arc column, and arranging a plurality of temperature sensors (5) in an array around the magnetic control arc sensor The temperature sensors (5) are arranged in the array around the magnetic control arc sensor , … The positions of the temperature sensors are symmetrically and uniformly distributed below the multi-directional magnetic pole (4) and fixed as a whole with the multi-directional magnetic pole, so that the temperature sensor probes (5) placed in front of the temperature sensors uniformly collect the average temperature values of the arc column at any position and moment, and the temperature change rate values are obtained by calculating the average temperature values of the arc column, so that the functional relationship between the temperature change rate and the arc position is established. The function between the temperature change rate and the arc position is established by calculating The average temperature value of the arc column of the mutually symmetrical distributed temperature sensor is obtained, and the temperature change rate value is obtained, and then the Newton interpolation method is used to determine the X and Y direction distance value of the arc in the horizontal section of the temperature sensor probe, that is, the relative coordinates of the arc , ), the collected arc current signal is converted into arc length and swing deflection angle signals, and then a data function relationship between the temperature change rate and the arc position is established through a multiple linear regression equation model, and finally a least square estimation method is used to obtain the fitted function relationship as: , wherein -N(0, ), is a proportional coefficient; , are the average temperature values of two points in the and coordinate systems, respectively; represents the arc length model; are the length and width of the magnetron-temperature sensor, respectively; , , , , are least square estimation coefficients; -N(0, ) represents a normal distribution, wherein the mean is 0 and the variance is (the standard deviation is ); , are the arc length and swing deflection angle when the variable .

Images