Automatic welding control method and system for aluminum-based terminal blocks

Abstract

The present invention relates to the field of automatic welding technology, and in particular to an automatic welding control method and system for aluminum-based terminal blocks, comprising: using a PID control algorithm to change welding parameters from initial welding parameters to A1; obtaining the time interval of each control quantity according to the change sensitivity of each control quantity, and screening out temporary welding parameters from historical welding parameters; after the PID control algorithm sequentially changes all control quantities according to the time interval, obtaining the target welding parameter A2 according to the temporary welding parameters and the change sensitivity of the control quantities in the temporary welding parameters, using the PID control algorithm to change the welding parameters so that the welding parameters change from A1 to A2, and so on, to complete the current welding process. The present invention performs relatively accurate and reliable control on the welding process, avoiding the serious impact of the welding process at some previous moments on the subsequent welding process.

Description

Technical Field

[0001] The invention relates to the technical field of automatic welding, and in particular to an automatic welding control method and system for aluminum-based connecting terminals. Background Art

[0002] Aluminum-based terminals are widely used in power transmission and distribution systems to connect cables. Aluminum-based terminals require reliable welding (especially welding of large-diameter and high-power cables) to ensure good conductivity and long-term stability between the terminals and the cables. Ultrasonic welding technology is a common technology for welding aluminum-based terminals. However, when ultrasonically welding aluminum-based terminals, welding parameters such as ultrasonic frequency, amplitude, pressure, and welding time are often difficult to control, resulting in deterioration of the electrical properties of the aluminum-based terminals after the welding process. For example, during the welding process, the cable and the terminal have poor contact, and the oxide layer causes local overheating. The deterioration of electrical properties will increase the electrical properties of the aluminum-based terminals, such as resistance or capacitive reactance. Especially for large-diameter and high-power cables, improper control of welding parameters will cause economic losses and safety accidents. Summary of the invention

[0003] In order to solve the above problems, the present invention provides an automatic welding control method and system for aluminum-based terminal blocks.

[0004] The automatic welding control method and system of an aluminum-based terminal block of the present invention adopts the following technical solutions:

[0005] An embodiment of the present invention provides an automatic welding control method for aluminum-based terminal blocks, the method comprising the following steps:

[0006] Ultrasonic welding is performed on the terminals. The control parameters of welding parameters include pressure, amplitude, frequency and welding time.

[0007] During the historical welding process, the electrical characteristics of the terminal and the welding parameters used are recorded as historical electrical characteristics and historical welding parameters;

[0008] According to the change amount of the historical electrical characteristics when the historical welding parameters are changed, the initial welding parameters are selected from the historical welding parameters;

[0009] The current welding process uses the initial welding parameters for welding. According to the similarity between the real-time collected electrical characteristics and welding parameters and the historical electrical characteristics and historical welding parameters, and the change amount of the historical electrical characteristics when the historical welding parameters are changed, the target welding parameter A1 is selected from the historical welding parameters.

[0010] A PID control algorithm is used to change the control amount of each welding parameter so that the welding parameters change from the initial welding parameters to A1; the time interval when the PID control algorithm changes each control amount is obtained according to the change sensitivity of each control amount when the historical electrical characteristics change with the historical welding parameters; in the process of the welding parameters changing from the initial welding parameters to A1, temporary welding parameters are screened out from the historical welding parameters according to the similarity of the real-time collected electrical characteristics and welding parameters with the historical electrical characteristics and historical welding parameters, and the change amount of the historical electrical characteristics when the historical welding parameters change; after the PID control algorithm changes all the control amounts in turn according to the time interval, the target welding parameter A2 is obtained according to the temporary welding parameters and the change sensitivity of the control amounts in the temporary welding parameters, and each welding parameter is changed by the PID control algorithm so that the welding parameters change from A1 to A2, and so on, to complete the current welding process.

[0011] Preferably, the method of selecting the initial welding parameters from the historical welding parameters according to the amount of change of the historical electrical characteristics when the historical welding parameters are changed comprises the following specific steps:

[0012] For any historical welding process, the absolute value of the difference between the historical electrical characteristics before and after welding is obtained, and recorded as the welding effect of any historical welding process;

[0013] Obtain N historical welding processes with the smallest welding effect. For each of the N historical welding processes, obtain the difference x1 between the historical welding parameters at each moment in the historical welding process and the historical welding parameters at the previous moment, and obtain the absolute value x2 of the difference between the historical electrical characteristics at each moment and the historical electrical characteristics at the previous moment; obtain the welding change rate at each moment based on x1 and x2; for the welding change rates at all moments in the N historical welding processes, obtain the historical welding parameters corresponding to the moment with the largest welding change rate, and record them as the initial welding parameters; N is a preset number.

[0014] Preferably, the welding change rate at each moment is obtained according to x1 and x2, and the specific calculation formula included is:

[0015] The ratio of x1 to (x2+1) is recorded as the welding change rate at each moment.

[0016] Preferably, the target welding parameter A1 is selected from the historical welding parameters according to the similarity between the real-time collected electrical characteristics and welding parameters and the historical electrical characteristics and welding parameters, and the change amount of the historical electrical characteristics when the historical welding parameters are changed, and the specific steps include the following:

[0017] In the current welding process, the sequence of electrical characteristics and welding parameters within a preset time before the current time is recorded as a real-time electrical characteristics sequence and a real-time welding parameter sequence respectively;

[0018] For each moment of each historical welding process, a historical welding parameter sequence consisting of historical welding parameters within a preset moment before each moment is obtained; a first similarity between the historical welding parameter sequence and the real-time welding parameter sequence is obtained, a historical electrical characteristic sequence consisting of historical electrical characteristics within a preset moment before each moment is obtained, and a second similarity between the historical electrical characteristic sequence and the real-time electrical characteristic sequence is obtained; the average of the first similarity and the second similarity is recorded as the welding matching at each moment in each historical welding process;

[0019] For each moment in the N historical welding processes, the preferred index for each moment is obtained according to the welding matching at each moment and the welding change rate at each moment, and the historical welding parameter corresponding to the moment with the largest priority index is taken as the target welding parameter A1; the preferred index is positively correlated with the welding matching and the welding change rate, respectively.

[0020] Preferably, the specific steps for obtaining the change sensitivity of each control variable are as follows:

[0021] For each historical welding parameter in the historical welding process, the historical electrical characteristics at the same time are recorded as y, each control quantity in the historical welding parameter is recorded as x, and (x, y) is used as the distribution point of each control quantity at each time;

[0022] The distribution points of each control quantity at each moment in the historical welding process are subjected to PCA dimensionality reduction to obtain the change sensitivity, and the change sensitivity is positively correlated with the variance of all dimensionality reduction results.

[0023] Preferably, the step of performing PCA dimensionality reduction on the distribution points of each control variable at each moment in the historical welding process to obtain the change sensitivity includes the following specific steps:

[0024] Perform PCA analysis on the distribution points of each control quantity at all times in each historical welding process to obtain the first direction of each historical welding process, and use the first direction to perform K-Means clustering on all historical welding processes to obtain several categories; Perform PCA analysis on the distribution points of all historical welding processes of the same category at all times to obtain the first direction of each category, and reduce the dimension of all distribution points of all historical welding processes in the same category to one dimension, and record the variance of the reduced dimension result as the electrical characteristic interference amount of each category;

[0025] The change sensitivity is obtained according to the first directions of all categories and the electrical characteristic interference amounts of all categories; the change sensitivity is positively correlated with the electrical characteristic interference amount and negatively correlated with the difference in the first directions between all categories.

[0026] Preferably, the step of obtaining the change sensitivity according to all categories of first directions and all categories of electrical characteristic interference amounts includes the following specific steps:

[0027] The first direction is represented as a unit vector, and the Euclidean distance of the unit vectors corresponding to the first directions of any two categories is obtained, and the first mean of the Euclidean distances corresponding to the first directions of all categories is obtained; the electrical characteristic interference quantities of all categories are linearly normalized to obtain the second mean of the electrical characteristic interference quantities after linear normalization; the ratio of the first mean to the second mean is denoted as m, exp(-m) is denoted as the sensitivity of change of each control quantity, and exp() represents an exponential function with a natural constant as the base.

[0028] Preferably, after the PID control algorithm sequentially changes all the control quantities according to the time interval, the target welding parameter A2 is obtained according to the temporary welding parameters and the sensitivity of the change of the control quantities in the temporary welding parameters, and the specific steps included are as follows:

[0029] In the process of the PID control algorithm sequentially changing each control amount according to the time interval, a sequence of welding parameters within a preset time before the current time is recorded as a real-time welding parameter sequence, and each welding parameter in the real-time welding parameter sequence includes an updated control amount and an unupdated control amount;

[0030] For all welding parameters in the real-time welding parameter sequence, the sum of the change sensitivities of all unupdated control quantities included in all welding parameters is obtained, recorded as y1, the sum of the change sensitivities of all updated control quantities included in all welding parameters is obtained, recorded as y2, and the ratio of y2 to (y1+1) is recorded as the control efficiency of the temporary welding parameters;

[0031] When all control variables in the welding parameters are updated and changed at least once, the target welding parameter A2 is obtained according to all the obtained temporary welding parameters and the control efficiency of the temporary welding parameters.

[0032] Preferably, the step of obtaining the target welding parameter A2 according to all the obtained temporary welding parameters and the control efficiency of the temporary welding parameters comprises the following specific steps:

[0033] All control efficiencies are normalized, and the normalized control efficiency is used as the weight to perform weighted summation on all temporary welding parameters, and the result is recorded as the target welding parameter A2.

[0034] In another embodiment of the present invention, an automatic welding control system for aluminum-based terminal blocks is provided, wherein the system includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and when the processor executes the computer program, the steps of the above-mentioned automatic welding control method for aluminum-based terminal blocks are implemented.

[0035] The beneficial effects of the technical solution of the present invention are:

[0036] The present invention uses a PID control algorithm to change the welding parameters from the initial welding parameters to A1; according to the change sensitivity of each control quantity, the time interval of each control quantity is obtained, and the PID control algorithm changes all the control quantities in turn according to the time interval. When the PID control algorithm is used for welding control in this process, different control quantities have different control time intervals. This process allows the control quantity that is more sensitive to change to have a smaller control time interval, which can improve the system response speed and enable the controller to correct errors faster for fast dynamic response; the control quantity that is not sensitive to change has a larger control time interval, which on the one hand reduces the increase in calculation load, and on the other hand can better match the response characteristics in the welding process, avoid over-regulation or unnecessary adjustment caused by too frequent sampling, and also avoid the introduction of unnecessary fluctuations, thereby improving the reliability of welding process control.

[0037] Furthermore, the target welding parameter A2 is obtained according to the temporary welding parameters and the sensitivity of the change of the control quantity in the temporary welding parameters, and each welding parameter is changed by the PID control algorithm so that the welding parameter changes from A1 to A2, and so on, to complete the current welding process. This process continuously updates and changes the target welding parameters, which can ensure that the welding process is adjusted in time, and to a certain extent avoid the welding process at some previous moments from having a serious impact on the subsequent welding process. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0039] Figure 1 A flowchart of the steps of an automatic welding control method for aluminum-based terminal blocks provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0040] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the automatic welding control method and system of an aluminum-based terminal block proposed by the present invention, its specific implementation, structure, features and effects, in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0041] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0042] The specific scheme of the automatic welding control method and system of aluminum-based terminal blocks provided by the present invention is described in detail below with reference to the accompanying drawings.

[0043] See also Figure 1 , which shows a flowchart of a method for automatically welding aluminum-based terminal blocks provided by an embodiment of the present invention, the method comprising the following steps:

[0044] Step S001, ultrasonically weld the terminal, and the controlled quantities of welding parameters during welding include: pressure, amplitude, frequency and welding time. In the historical welding process, the electrical characteristics of the terminal and the welding parameters used are recorded as historical electrical characteristics and historical welding parameters.

[0045] Aluminum-based terminal blocks (hereinafter referred to as terminals) require reliable welding (especially welding of large-diameter and high-power cables) to ensure good conductivity and long-term stability between the terminals and cables.

[0046] This embodiment uses ultrasonic welding technology to weld the terminal to the cable. Ultrasonic welding technology uses high-frequency mechanical vibrations of 20kHz to 40kHz to transfer energy to the surface of the contact material. When ultrasound is applied to the material (i.e., the terminal and the cable) through a welding tool (welding head), high-frequency vibrations are generated. This vibration causes friction of the internal molecules or lattice structure at the contact point, generating local heat. The heat softens the surface of the material and achieves bonding. While vibrating, the welding head applies pressure to the material, and this pressure, combined with the ultrasonic vibration, fuses the materials together at the molecular level to form a strong welded joint.

[0047] During ultrasonic welding, multiple control quantities such as the amplitude and frequency of the ultrasonic wave, the pressure applied by the welding head, and the welding time are recorded as welding parameters.

[0048] Different welding results will be obtained when the welding parameters are set to different values. If the welding parameters are not properly controlled, the materials cannot be fully contacted and fused, which will affect the electrical properties of the contact part between the cable and the terminal. For example, the resistance between the cable and the terminal will increase, resulting in power loss or local high temperature and continuous heating (especially for large-diameter and high-power cables, local high temperature and continuous heating will bring serious economic losses and safety accidents), or capacitive impedance will exist between the cable and the terminal, resulting in signal distortion.

[0049] In this embodiment, the resistance between the cable and the terminal is measured in real time while welding, for example, the resistance value is obtained every other moment, in this embodiment, one moment refers to 0.1 second, in other embodiments, one moment can be defined as other values, which are not specifically limited in this embodiment. The resistance value obtained in real time is recorded as the electrical characteristic of the aluminum-based terminal during welding.

[0050] In other embodiments, a periodic signal of known amplitude is input into the cable, and the output amplitude of the signal is measured at the terminal. The difference between the known amplitude and the output amplitude is recorded as the electrical characteristic of the aluminum-based terminal during the welding process, which is used to describe the distortion of the signal during the welding process.

[0051] In the historical welding process, each time each terminal is welded, the electrical characteristics at each moment in the welding process and the welding parameters used at each moment are obtained, and the obtained electrical characteristics and welding parameters are recorded as historical electrical characteristics and historical welding parameters.

[0052] If the terminals have not been soldered in the past, a test soldering is required, as follows:

[0053] Select a variety of different welding parameters for testing (referred to as test welding parameters) based on the experience of the welder. Alternatively, randomly generate multiple (e.g., 50) test welding parameters, such as a frequency randomly selected between 20kHz and 40kHz, an amplitude randomly selected between 10 and 50 microns, a pressure randomly selected between 20 and 100 Newtons per square centimeter, and a welding time randomly selected between 2 and 8 seconds. The specific range of welding parameters depends on the material and diameter of the cable, the requirements for electrical performance after welding, and even factors such as the power of the welding equipment, which are not specifically limited in this embodiment.

[0054] Each time the welding is tested, the PID algorithm is used to control the welding parameters to make them equal to the test welding parameters. In the test welding, the electrical characteristics at each moment in the welding process and the welding parameters used at each moment are obtained, and then the historical electrical characteristics and historical welding parameters are obtained.

[0055] Test welding refers to the welding process used for experiments, and the welded products will not be sold or used.

[0056] Step S002: based on the amount of change in the historical electrical characteristics when the historical welding parameters are changed, initial welding parameters are selected from the historical welding parameters, and the current welding process uses the initial welding parameters for welding.

[0057] As an example, according to the change amount of the historical electrical characteristics when the historical welding parameters are changed, the initial welding parameters are screened out from the historical welding parameters, including the following methods:

[0058] For any historical welding process, the absolute value of the difference between the historical electrical characteristics before and after welding is obtained and recorded as the welding effect of any historical welding process. The historical electrical characteristics before welding refer to the electrical characteristics at both ends of the cable.

[0059] Obtain N historical welding processes with the smallest welding effect. For each of these N historical welding processes, obtain the difference between the historical welding parameters at each moment and the historical welding parameters at the previous moment, which is recorded as the change x1 of the historical welding parameters at each moment. Obtain the absolute value of the difference between the historical electrical characteristics at each moment and the historical electrical characteristics at the previous moment, which is recorded as the change x2 of the historical electrical characteristics at each moment.

[0060] In this embodiment, N is equal to half of the number of historical welding times (rounded to the nearest integer). In other embodiments, N may be set to other values, which are not specifically limited in this embodiment.

[0061] Particularly, in this embodiment, the change amount x1 or the change amount x2 at the first moment in each historical welding process is equal to the change amount x1 or the change amount x2 at the second moment.

[0062] The ratio of x1 to (x2+1) is recorded as the welding change rate at each moment.

[0063] The larger the welding change rate, the smaller the historical electrical characteristics change rate, which means that even a slight disturbance change in the historical welding parameters will cause a significant change in the historical electrical characteristics.

[0064] For the welding change rates at all moments in the N historical welding processes, the historical welding parameters corresponding to the moment with the largest welding change rate are obtained and recorded as initial welding parameters. The current welding process is welded using the initial welding parameters.

[0065] As an example, the method to obtain x1 is as follows:

[0066] The historical welding parameters at each moment are recorded as F1, and the historical welding parameters at the previous moment are recorded as F2. F1 and F2 are represented as vectors composed of all control quantities in the historical welding parameters, and the Euclidean distance between F1 and F2 is recorded as x1.

[0067] It should be further explained that, in order to make the welding parameters in the current welding process equal to the initial welding parameters, this embodiment uses a PID control algorithm to control the applied pressure, amplitude, frequency and welding time.

[0068] The PID control algorithm aims to make the output result (that is, the welding parameters in the current welding process) stably reach the set target value (that is, the initial welding parameters). The algorithm consists of three parts: proportional controller, integral controller and differential controller. The three parts adjust the welding parameters once every time interval T0. It can smooth the response and reduce oscillation and overshoot. After every time interval T0, the PID controller gradually adjusts the welding parameters by feedback adjustment of the combined effect of these three parts, so that the output result gradually approaches and stabilizes at the target value. After multiple adjustments, accurate control effect is achieved.

[0069] This embodiment is described by taking T0 equal to 0.2 seconds as an example. In other embodiments, T0 may be set to other values, which are not specifically limited in this embodiment.

[0070] Because the electrical characteristics will not change significantly when the initial welding parameters are significantly disturbed or changed, the current welding process under PID control has a strong anti-interference ability, which can avoid to a certain extent the situation where the electrical characteristics change drastically after disturbance in the welding process based on PID control (dramatic changes in electrical characteristics may mean local poor contact, interference in the oxide layer on the aluminum surface causing local overheating, etc.).

[0071] Step S003, based on the similarity between the real-time collected electrical characteristics and welding parameters and the historical electrical characteristics and historical welding parameters, and the change amount of the historical electrical characteristics when the historical welding parameters are changed, select the target welding parameter A1 from the historical welding parameters.

[0072] The above initial welding parameters are obtained from the welding parameters of the historical welding process and may not be applicable to the entire current welding process. When the current welding process is in progress, the electrical characteristics and welding parameters at the current moment are obtained, and the sequence of electrical characteristics and welding parameters at T1 moments before the current moment (including the current moment) are recorded as the real-time electrical characteristics sequence and the real-time welding parameter sequence, respectively. This embodiment is described by taking T1=5 as an example. In other embodiments, T1 can be set to other values, which are not specifically limited in this embodiment.

[0073] For each moment of each historical welding process, a historical welding parameter sequence consisting of the historical welding parameters of the T1 moments before each moment is obtained. The similarity between the historical welding parameter sequence and the real-time welding parameter sequence is obtained, which is recorded as the first similarity. The historical electrical characteristic sequence consisting of the historical electrical characteristics of the T1 moments before each moment is obtained, and the similarity between the historical electrical characteristic sequence and the real-time electrical characteristic sequence is obtained, which is recorded as the second similarity. The average of the first similarity and the second similarity is recorded as the welding matching at each moment in each historical welding process.

[0074] For the N historical welding processes mentioned above, the product of the welding matching at each moment and the welding change rate at each moment is recorded as the preferred index at each moment, and the historical welding parameter corresponding to the moment with the largest priority index is taken as the target welding parameter A1.

[0075] In this embodiment, the method for obtaining the similarity between the historical welding parameter sequence and the real-time welding parameter sequence, and the similarity between the historical electrical characteristic sequence and the real-time electrical characteristic sequence includes:

[0076] The historical welding parameter sequence, the real-time welding parameter sequence, the historical electrical characteristic sequence and the real-time electrical characteristic sequence are regarded as vectors respectively, and the cosine similarity between the historical welding parameter sequence and the real-time welding parameter sequence, and the cosine similarity between the historical electrical characteristic sequence and the real-time electrical characteristic sequence are taken as the above similarity.

[0077] In some other embodiments, the method for obtaining the similarity between the historical welding parameter sequence and the real-time welding parameter sequence, and the similarity between the historical electrical characteristic sequence and the real-time electrical characteristic sequence includes:

[0078] The DTW distance between the historical welding parameter sequence and the real-time welding parameter sequence, as well as the DTW distance between the historical electrical characteristic sequence and the real-time electrical characteristic sequence are obtained, and the obtained DTW distances are all expressed as P, and exp(-P) is used as the above similarity. The DTW distance is obtained by the DTW algorithm. The larger the DTW distance, the greater the difference between the sequences and the smaller the similarity. exp() represents an exponential function with a natural constant as the base.

[0079] Step S004, using the PID control algorithm to change the control amount of each welding parameter so that the welding parameter changes from the initial welding parameter to A1; according to the sensitivity of each control amount to change when the historical electrical characteristics change with the historical welding parameters, the time interval when the PID control algorithm changes each control amount is obtained.

[0080] The target welding parameter A1 obtained above is obtained based on the historical welding process combined with the electrical characteristics and welding parameters in the current welding process. This embodiment uses a PID control algorithm to control each control quantity in the welding parameters (that is, to control the applied pressure, amplitude, frequency and welding time), so that the welding parameters in the current welding process change from the initial welding parameters to A1.

[0081] It should be noted that, from the PID control algorithm, PID adjusts each control quantity in the welding parameters once at a certain interval (denoted as T2) through the proportional controller, integral controller and differential controller (the initial value is the value of each control quantity in the initial welding parameters). After multiple adjustments, each control quantity in the initial welding parameters is equal to each control quantity in A1.

[0082] The time interval T2 corresponding to each control variable in the welding parameters is obtained by the sensitivity of each historical electrical characteristic in the historical welding process to the change of the historical welding parameters.

[0083] As an optional example, the method for obtaining the change sensitivity is:

[0084] For each historical welding parameter in the historical welding process, the historical electrical characteristics at the same time are recorded as y, each control quantity in the historical welding parameter is recorded as x, and (x, y) is used as the distribution point of each control quantity at each time.

[0085] For the distribution points of each control quantity pair at all times in all historical welding processes, these distribution points are reduced to one dimension using the PCA algorithm, and the variance of all dimensionality reduction results is recorded as the change sensitivity.

[0086] As a preferred example, the method for obtaining the change sensitivity is:

[0087] The distribution points of each control quantity at all times in each historical welding process are obtained, and PCA analysis is performed on these distribution points. From the PCA algorithm, it can be seen that two principal component directions can be obtained (each principal component direction is represented by a unit vector), and each principal component direction corresponds to an eigenvalue. The principal component direction with the largest eigenvalue is obtained and recorded as the first direction of each historical welding process (represented as a unit vector), which represents the main distribution direction of the electrical characteristics and each control quantity in each historical welding process, that is, along the first direction, the electrical characteristics undergo the most obvious changes with the change of each control quantity.

[0088] The first directions of all historical welding processes are obtained, and K-Means clustering is performed on all historical welding processes using the first directions to obtain K categories. In this embodiment, K is set to 5, and in other embodiments, K can be set to other values, which are not specifically limited in this embodiment. Each category includes multiple historical welding processes, and the first directions of these welding processes are the same or similar, indicating that the direction of change (or change trend) of the electrical characteristics with each control quantity in these welding processes is similar. The historical welding processes in different categories have significantly different change directions (that is, the change trends of the electrical characteristics with each control quantity are significantly different).

[0089] In this embodiment, PCA analysis is performed on the distribution points of all the historical welding processes of the same category at all times, and the principal component direction with the largest eigenvalue is obtained again, which is recorded as the first direction of each category (expressed as a unit vector), and all the distribution points of all the historical welding processes of the same category are reduced to one dimension, and the variance of the reduced dimension result is recorded as the electrical characteristic interference amount of each category. The larger the electrical characteristic interference amount, the greater the influence or interference of each control amount on the electrical characteristics of all the historical welding processes of the same category.

[0090] Obtain the Euclidean distance of the first direction of any two categories (that is, the Euclidean distance between the unit vectors corresponding to the first direction), and obtain the first mean of the Euclidean distance between the first directions of all categories. Perform linear normalization on the electrical characteristic interference of all categories, and obtain the second mean of the electrical characteristic interference after linear normalization. The ratio of the first mean to the second mean is denoted as m, and exp(-m) is denoted as the change sensitivity of each control quantity.

[0091] When m is larger, it means that the electrical characteristic interference is smaller and the difference between the first directions of the categories is larger, which means that when the electrical characteristics change with the change of each control quantity, the electrical characteristics will not show obvious change amplitude, and at the same time, the change direction (or change trend) of the electrical characteristics with each control quantity in different historical welding processes is also very different. Therefore, the less each control quantity can determine the change of the electrical characteristics, the smaller the change sensitivity of each control quantity is. When m is smaller, it means that the electrical characteristic interference is larger and the difference between the first directions of the categories is smaller, which means that when the electrical characteristics change with the change of each control quantity, the electrical characteristics have obvious change amplitude, and at the same time, the change direction of the electrical characteristics with each control quantity in different historical welding processes is the same or similar. Therefore, the more each control quantity can determine the change of the electrical characteristics, the greater the change sensitivity of each control quantity is.

[0092] In the above optional examples, on the one hand, the dimensionality reduction is performed on the distribution points in all historical welding processes together, which results in a large amount of calculation, and on the other hand, the difference in the distribution relationship between the distribution points of different historical welding processes cannot be expressed. In the above preferred examples, on the one hand, the PCA processing is performed on each or several historical welding processes separately, which results in a relatively small amount of calculation, and on the other hand, the difference in the distribution relationship between the distribution points of different historical welding processes is further described by the difference between the first directions of each category, which relatively accurately reflects the sensitivity of the historical electrical characteristics in the historical welding process to the changes in the historical welding parameters.

[0093] The method for obtaining the time interval T2 corresponding to each control amount includes:

[0094] T2=T0×Q+(1-Q)×T3, where Q represents the sensitivity of each control variable, T3 is greater than T0, and represents the maximum value of the time interval T2. This embodiment takes T3=1 second as an example for description. Other embodiments may be set to other values, which are not specifically limited in this embodiment. When Q is larger, the value of T2 is closer to T0, and when Q is smaller, the value of T2 is closer to T3.

[0095] After the time interval T2 corresponding to each control amount is obtained, each welding parameter is changed using the PID control algorithm so that the welding parameter changes from the initial welding parameter to A1.

[0096] Step S005, screen out temporary welding parameters from historical welding parameters; after the PID control algorithm changes all control quantities in sequence according to the time interval, obtain the target welding parameter A2 according to the temporary welding parameters and the sensitivity of changes in the control quantities in the temporary welding parameters, use the PID control algorithm to change each welding parameter so that the welding parameter changes from A1 to A2, and so on, to complete the current welding process.

[0097] When the PID control algorithm changes the welding parameters from the initial welding parameters to A1, each control quantity corresponds to a T2. For example, the T2 corresponding to the frequency is expressed as T2(1), and the T2 corresponding to the amplitude is expressed as T2(2). Then, after each T2(1) time, the PID control algorithm controls the ultrasonic generator to change the frequency in the welding parameters, so that the frequency gradually approaches and stabilizes at the frequency in A1. At the same time, after each T2(2) time, the PID control algorithm controls the ultrasonic generator to change the amplitude in the welding parameters, so that the amplitude gradually approaches and stabilizes at the amplitude in A1. The other control quantities are controlled in the same way.

[0098] When the PID control algorithm is used for welding control in the above process, different control quantities have different control time intervals (i.e., T2). This process makes the control quantity that is more sensitive to change have a smaller control time interval, which can improve the system response speed and enable the controller to correct errors more quickly for fast dynamic response; the control quantity that is insensitive to change has a larger control time interval, which on the one hand reduces the increase in calculation load, and on the other hand can better match the response characteristics of the welding process, avoid over-adjustment or unnecessary adjustment due to too frequent sampling, and avoid introducing unnecessary fluctuations; in summary, the above process enables the welding process to be efficiently controlled.

[0099] It should be further explained that, in the process of changing the welding parameters from the initial welding parameters to A1, it is necessary to determine the target welding parameters again according to the real-time welding parameter sequence and the real-time electrical characteristic sequence. The purpose is that when the PID control algorithm changes the welding parameters, new welding parameters are introduced. The new welding parameters correspond to new electrical characteristics, that is, the real-time welding parameter sequence and the real-time electrical characteristic sequence include the new welding situation at the current moment in the current welding process (for example, whether poor contact or local overheating caused by the oxide layer occurs after control adjustments are made to control quantities such as frequency or pressure). In order to ensure timely adjustments to the welding process, the target welding parameters need to be updated according to the current control results. If appropriate adjustments cannot be made according to the current new welding situation, it may lead to poor welding efficiency at the current moment, thereby affecting subsequent welding processes.

[0100] Further consideration is that, since the T2 of different control quantities of welding parameters may be different, the control quantities included in the welding parameters at the same time are not necessarily all changed by the PID algorithm. For example, the frequency is changed at the current moment, but other control quantities have not changed, resulting in the welding parameters at the current moment including both the changed control quantities (called updated control quantities) and the control quantities that have not changed (recorded as unupdated control quantities). Such welding parameters are recorded as abnormal welding parameters. When the real-time welding parameter sequence contains abnormal welding parameters, the target welding parameters obtained according to the real-time welding parameter sequence and the real-time electrical characteristic sequence cannot be accurately adjusted for the next step of control based on the current control results.

[0101] In the process of welding parameters changing from initial welding parameters to A1, the real-time welding parameter sequence and the real-time electrical characteristic sequence are still obtained. According to the similarity between the real-time welding parameter sequence and the historical welding parameter sequence, the similarity between the real-time electrical characteristic sequence and the historical electrical characteristic sequence, and the change amount of the historical electrical characteristic when the historical welding parameters are changed, the temporary welding parameters are screened out from the historical welding parameters (similar to step S003). For all welding parameters included in the real-time welding parameter sequence, the sum of the change sensitivity of all unupdated control quantities in all welding parameters is obtained, recorded as y1, and the sum of the change sensitivity of all updated control quantities in all welding parameters is obtained, recorded as y2. The ratio of y2 to (y1+1) is recorded as the control efficiency of the temporary welding parameters. The greater the control efficiency, the greater the change sensitivity of the updated control quantity is than the change sensitivity of the unupdated control quantity. At this time, the interference of the temporary welding parameters obtained on the subsequent control process is smaller. When there is no unupdated control quantity, y1=0, and setting y1+1 can avoid the special case where the denominator is 0.

[0102] In the process of welding parameters changing from initial welding parameters to A1, after the PID control algorithm changes all control quantities in sequence according to the time interval, as the welding process proceeds, when all control quantities in the welding parameters are updated and changed at least once, all temporary welding parameters and the control efficiency of the temporary welding parameters obtained in the process are obtained, and all control efficiencies are normalized. In this embodiment, the softmax method is used for normalization, and the normalized control efficiency is used as the weight to perform weighted summation on all temporary welding parameters, and the result is recorded as the target welding parameter A2.

[0103] All the control quantities mentioned above are updated and changed at least once, and the following examples are given to illustrate:

[0104] All control quantities are respectively recorded as k1, k2, k3, and k4. For example, these control quantities are changed in the following order according to the time interval: k1, k2, k1, k4, k2, and k3, among which all other control quantities have changed at least once before k3 is changed. When k3 in this example is changed and before the next control quantity is changed, it is determined that all control quantities have been updated and changed at least once.

[0105] The target welding parameter A2 is further obtained by the sensitivity of the changes of the updated control quantity and the unupdated control quantity during the welding process, which can ensure that the welding process is adjusted in a timely manner and avoid to a certain extent the serious impact of welding problems at some previous moments on the subsequent welding process.

[0106] Further, each welding parameter is changed by using the PID control algorithm so that the welding parameter changes from A1 to A2. And in the process of the welding parameter changing from A1 to A2, the time interval of each control amount is re-obtained by the method of step S005 again, and the target welding parameter A3 is obtained, and then each welding parameter is changed by using the PID control algorithm so that the welding parameter changes from A2 to A3, and so on, until the welding process ends (when the time of the current welding process is equal to the welding time of the welding parameter, the welding process ends).

[0107] It should be noted that the A1, A2 or A3 obtained above are directly or indirectly selected from the historical welding parameters. When the welding time in the above A1, A2 or A3 is less than the time the current welding process has been carried out, the selected historical welding parameters (such as A1, and the temporary welding parameters used to indirectly obtain A2 and A3) are excluded, and the remaining historical welding parameters are re-screened (for example, A1 or temporary welding parameters are screened out). If the welding time in the A1 or temporary welding parameters screened out again is still less than the time the current welding process has been carried out, then the A1 or temporary welding parameters at this time are excluded again, and the remaining historical welding parameters are re-screened until the welding time in the A1 or temporary welding parameters screened out is not less than the time the current welding process has been carried out. In other embodiments, when the number of consecutive exclusions is greater than 5 times, the current welding process is terminated.

[0108] In another embodiment of the present invention, an automatic welding control system for aluminum-based terminal blocks is provided, wherein the system includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the above steps S001 to S005 when executing the computer program.

[0109] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the protection scope of the present invention.

Claims

1. An automatic welding control method for aluminum-based terminal blocks, characterized in that: The method comprises the following steps: Ultrasonic welding is performed on the terminals. The control parameters of welding parameters include pressure, amplitude, frequency and welding time. During the historical welding process, the electrical characteristics of the terminal and the welding parameters used are recorded as historical electrical characteristics and historical welding parameters; According to the change amount of the historical electrical characteristics when the historical welding parameters are changed, the initial welding parameters are selected from the historical welding parameters; The current welding process uses the initial welding parameters for welding. According to the similarity between the real-time collected electrical characteristics and welding parameters and the historical electrical characteristics and historical welding parameters, and the change amount of the historical electrical characteristics when the historical welding parameters are changed, the target welding parameter A1 is selected from the historical welding parameters. A PID control algorithm is used to change the control amount of each welding parameter so that the welding parameters change from the initial welding parameters to A1; the time interval when the PID control algorithm changes each control amount is obtained according to the change sensitivity of each control amount when the historical electrical characteristics change with the historical welding parameters; in the process of the welding parameters changing from the initial welding parameters to A1, temporary welding parameters are screened out from the historical welding parameters according to the similarity of the real-time collected electrical characteristics and welding parameters with the historical electrical characteristics and historical welding parameters, and the change amount of the historical electrical characteristics when the historical welding parameters change; after the PID control algorithm changes all the control amounts in turn according to the time interval, the target welding parameter A2 is obtained according to the temporary welding parameters and the change sensitivity of the control amounts in the temporary welding parameters, and each welding parameter is changed by the PID control algorithm so that the welding parameters change from A1 to A2, and so on, to complete the current welding process.

2. The automatic welding control method of aluminum-based terminal blocks according to claim 1, characterized in that: The specific steps of selecting the initial welding parameters from the historical welding parameters according to the change amount of the historical electrical characteristics when the historical welding parameters are changed are as follows: For any historical welding process, the absolute value of the difference between the historical electrical characteristics before and after welding is obtained, and recorded as the welding effect of any historical welding process; Obtain N historical welding processes with the smallest welding effect. For each of the N historical welding processes, obtain the difference x1 between the historical welding parameters at each moment in the historical welding process and the historical welding parameters at the previous moment, and obtain the absolute value x2 of the difference between the historical electrical characteristics at each moment and the historical electrical characteristics at the previous moment; obtain the welding change rate at each moment based on x1 and x2; for the welding change rates at all moments in the N historical welding processes, obtain the historical welding parameters corresponding to the moment with the largest welding change rate, and record them as the initial welding parameters; N is a preset number.

3. The automatic welding control method of aluminum-based terminal blocks according to claim 2, characterized in that: The specific calculation formula for obtaining the welding change rate at each moment according to x1 and x2 is as follows: The ratio of x1 to (x2+1) is recorded as the welding change rate at each moment.

4. The automatic welding control method of aluminum-based terminal blocks according to claim 2, characterized in that: The target welding parameter A1 is selected from the historical welding parameters according to the similarity between the real-time collected electrical characteristics and welding parameters and the historical electrical characteristics and welding parameters, and the change amount of the historical electrical characteristics when the historical welding parameters are changed. The specific steps include the following: In the current welding process, the sequence of electrical characteristics and welding parameters within a preset time before the current time is recorded as a real-time electrical characteristics sequence and a real-time welding parameter sequence respectively; For each moment of each historical welding process, a historical welding parameter sequence consisting of historical welding parameters within a preset moment before each moment is obtained; a first similarity between the historical welding parameter sequence and the real-time welding parameter sequence is obtained, a historical electrical characteristic sequence consisting of historical electrical characteristics within a preset moment before each moment is obtained, and a second similarity between the historical electrical characteristic sequence and the real-time electrical characteristic sequence is obtained; The mean of the first similarity and the second similarity is recorded as the welding matching at each moment in each historical welding process; For each moment in the N historical welding processes, the preferred index for each moment is obtained according to the welding matching at each moment and the welding change rate at each moment, and the historical welding parameter corresponding to the moment with the largest priority index is taken as the target welding parameter A1; the preferred index is positively correlated with the welding matching and the welding change rate, respectively.

5. The automatic welding control method of aluminum-based terminal blocks according to claim 1, characterized in that: The specific steps for obtaining the change sensitivity of each control variable are as follows: For each historical welding parameter in the historical welding process, the historical electrical characteristics at the same time are recorded as y, each control quantity in the historical welding parameter is recorded as x, and (x, y) is used as the distribution point of each control quantity at each time; The distribution points of each control quantity at each moment in the historical welding process are subjected to PCA dimensionality reduction to obtain the change sensitivity, and the change sensitivity is positively correlated with the variance of all dimensionality reduction results.

6. The automatic welding control method of aluminum-based terminal blocks according to claim 5, characterized in that: The PCA dimension reduction is performed on the distribution points of each control quantity at each moment in the historical welding process to obtain the change sensitivity, and the specific steps include the following: Perform PCA analysis on the distribution points of each control quantity at all times in each historical welding process to obtain the first direction of each historical welding process. Use the first direction to perform K-Means clustering on all historical welding processes to obtain several categories. Perform PCA analysis on the distribution points at all times in all historical welding processes of the same category to obtain the first direction of each category, and reduce the dimension of all distribution points in all historical welding processes of the same category into one dimension, and record the variance of the reduced dimension result as the electrical characteristic interference amount of each category; The change sensitivity is obtained according to the first directions of all categories and the electrical characteristic interference amounts of all categories; the change sensitivity is positively correlated with the electrical characteristic interference amount and negatively correlated with the difference in the first directions between all categories.

7. The automatic welding control method of aluminum-based terminal blocks according to claim 6, characterized in that: The step of obtaining the change sensitivity according to all categories of first directions and all categories of electrical characteristic interference amounts includes the following specific steps: The first direction is represented as a unit vector, the Euclidean distance of the unit vectors corresponding to the first directions of any two categories is obtained, and the first mean of the Euclidean distances corresponding to the first directions of all categories is obtained; Linear normalization is performed on all categories of electrical characteristic interference quantities to obtain the second mean of the electrical characteristic interference quantities after linear normalization. The ratio of the first mean to the second mean is denoted as m, exp(-m) is denoted as the change sensitivity of each control quantity, and exp() represents an exponential function with a natural constant as the base.

8. The automatic welding control method of aluminum-based terminal blocks according to claim 1, characterized in that: After the PID control algorithm sequentially changes all the control quantities according to the time interval, the target welding parameter A2 is obtained according to the temporary welding parameters and the sensitivity of the change of the control quantities in the temporary welding parameters, and the specific steps included are as follows: In the process of sequentially changing each control amount according to the time interval, the PID control algorithm records the sequence of welding parameters in the preset time before the current time as the real-time welding parameter sequence, and each welding parameter in the real-time welding parameter sequence includes an updated control amount and an unupdated control amount; For all welding parameters in the real-time welding parameter sequence, the sum of the change sensitivities of all unupdated control quantities included in all welding parameters is obtained, recorded as y1, the sum of the change sensitivities of all updated control quantities included in all welding parameters is obtained, recorded as y2, and the ratio of y2 to (y1+1) is recorded as the control efficiency of the temporary welding parameters; When all control quantities in the welding parameters are updated and changed at least once, the target welding parameter A2 is obtained according to all the obtained temporary welding parameters and the control efficiency of the temporary welding parameters.

9. The automatic welding control method of aluminum-based terminal blocks according to claim 8, characterized in that: The specific steps of obtaining the target welding parameter A2 according to all the temporary welding parameters obtained and the control efficiency of the temporary welding parameters are as follows: All control efficiencies are normalized, and the normalized control efficiency is used as the weight to perform weighted summation on all temporary welding parameters, and the result is recorded as the target welding parameter A2.

10. An automatic welding control system for aluminum-based terminal blocks, the system comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the automatic welding control method for aluminum-based terminal blocks according to any one of claims 1 to 9 are implemented.

Images