A method for preventing false welding of ultrasonic welding control and ultrasonic welding machine thereof

Abstract

The present application relates to the technical field of welding quality, in particular to an ultrasonic welding control method for preventing false welding and an ultrasonic welding machine thereof, comprising: acquiring welding pressure data and welding vibration data at each moment; obtaining a risk index of each suspected abnormal moment according to the deviation of the welding pressure data from the reference, the difference distribution and time distribution of the welding pressure data under pressure mutation, and the deviation of the welding vibration data from the reference, the difference distribution and time distribution of the welding vibration data under vibration mutation; screening the suspected abnormal moment according to the consistency and correlation of the data change trend of the welding pressure data and the welding vibration data in the time neighborhood to obtain a false welding moment; and obtaining a false welding abnormal evaluation value according to the risk index, the time distribution and the number distribution of the false welding moment, which is used to determine whether to regulate the control parameters of the ultrasonic welding. The present application can effectively reduce the probability of batch false welding.

Description

Technical Field

[0001] This invention relates to the field of welding quality technology, specifically to an ultrasonic welding control method and ultrasonic welding machine for preventing incomplete welds. Background Technology

[0002] An ultrasonic welding machine is an industrial device that uses ultrasonic vibration energy to weld materials. It converts high-voltage, high-frequency signals into high-frequency mechanical vibrations, which are applied to the workpieces to be welded. The temperature of the interface is increased through friction between the workpiece surface and molecules. When the temperature reaches the melting point of the workpiece itself, the interface melts rapidly and fills the gap between the interfaces. When the vibration stops, the workpiece cools and solidifies under a certain pressure, thus achieving welding. Due to its fast welding speed, cleanliness and environmental friendliness, and low scrap rate, it is widely used in industries such as automotive parts, consumer electronics, home appliances, and aerospace. It can weld the same metal materials such as copper, silver, aluminum, and nickel, as well as dissimilar metal materials such as copper and aluminum foil.

[0003] In practical applications, existing ultrasonic welding technologies suffer from inherent equipment instability, making it difficult to maintain precise and stable welding parameters (such as pressure, amplitude, and welding time). For example, unstable ultrasonic amplitude of the welding head can prevent sufficient fusion between workpieces, leading to incomplete welds. Furthermore, during frequent vibration welding, the connection between some mechanical components and the equipment body may gradually loosen, causing positional shifts during welding and resulting in uneven pressure distribution on the workpiece, thus affecting welding quality and increasing the risk of incomplete welds. Current technologies, which detect anomalies in the ultrasonic welding process by measuring data deviations from a reference range, struggle to distinguish between anomalies caused by incomplete welds and other interference, leading to inaccurate identification of incomplete welds and impacting welding control. Summary of the Invention

[0004] To address the technical problem that existing methods cannot accurately identify cold weld abnormalities, the present invention aims to provide an ultrasonic welding control method and ultrasonic welding machine for preventing cold welds. The specific technical solution adopted is as follows:

[0005] In a first aspect, the present invention provides an ultrasonic welding control method for preventing incomplete welds, comprising:

[0006] Acquire welding pressure and welding vibration data at every moment during the welding process;

[0007] Based on the deviation of welding pressure data from the baseline at each moment, as well as the difference distribution and time distribution of welding pressure data under pressure mutation, and combined with the deviation of welding vibration data from the baseline at each moment and the difference distribution and time distribution of welding vibration data under vibration mutation, risk indicators for each suspected abnormal moment are obtained.

[0008] Based on the consistency and correlation of the data change trends of welding pressure data and welding vibration data at each suspected abnormal moment in the time neighborhood, the suspected abnormal moments are screened to obtain the moments of false welding.

[0009] Based on the distribution of risk indicators at each moment of poor welding, as well as the time and quantity distribution of poor welding moments, a poor welding anomaly evaluation value is obtained for the welding process. This poor welding anomaly evaluation value is used to determine whether the control parameters of ultrasonic welding should be adjusted.

[0010] Preferably, the step of obtaining a risk index for each suspected abnormal moment based on the deviation of welding pressure data from the baseline at each moment, the difference distribution and time distribution of welding pressure data under pressure mutations, and the deviation of welding vibration data from the baseline at each moment, and the difference distribution and time distribution of welding vibration data under vibration mutations, specifically includes:

[0011] Based on the deviation of welding pressure data from the baseline at each moment, and the deviation of welding vibration data from the baseline at each moment, all moments are filtered to obtain suspected abnormal moments and normal moments; the suspected abnormal moments represent moments where there are sudden changes in pressure and vibration.

[0012] Based on the difference distribution and time distribution of welding pressure data between each suspected abnormal moment and the normal moment, the pressure change characteristic value of each suspected abnormal moment is obtained;

[0013] Based on the difference distribution and time distribution of welding vibration data between each suspected abnormal moment and the normal moment, the vibration mutation characteristic value of each suspected abnormal moment is obtained.

[0014] By combining the characteristic values ​​of pressure mutation and vibration mutation at each suspected abnormal moment, a risk index is obtained for each suspected abnormal moment.

[0015] Preferably, the step of filtering all moments based on the deviation of welding pressure data from the benchmark at each moment and the deviation of welding vibration data from the benchmark at each moment to obtain suspected abnormal moments and normal moments specifically includes:

[0016] For each moment, the moment when the difference between the welding pressure data and the preset pressure reference value exceeds the preset pressure fluctuation range is taken as the pressure change moment; for each moment, the moment when the difference between the welding vibration data and the preset vibration reference value exceeds the preset vibration fluctuation range is taken as the vibration change moment.

[0017] Identify suspected abnormal moments, which are simultaneously moments of pressure and vibration mutation.

[0018] All moments except for suspected abnormal moments, moments of sudden pressure changes, and moments of sudden vibration changes are considered normal moments.

[0019] Preferably, obtaining the pressure mutation characteristic value for each suspected abnormal moment based on the difference distribution and time distribution of welding pressure data between each suspected abnormal moment and a normal moment specifically includes:

[0020] Obtain the time interval between each suspected abnormal moment and the nearest normal moment in the time series; obtain the first difference value of the welding pressure data between each suspected abnormal moment and the nearest normal moment in the time series; obtain the first difference between each suspected abnormal moment and the preset pressure reference value;

[0021] Based on the reciprocal of the time interval, the first difference value, and the first difference value, the pressure mutation characteristic value of each suspected abnormal moment is determined.

[0022] Preferably, the step of obtaining the vibration abrupt change characteristic value for each suspected abnormal moment based on the difference distribution and time distribution of welding vibration data between each suspected abnormal moment and a normal moment specifically includes:

[0023] Obtain the second difference value of welding vibration data between each suspected abnormal moment and the nearest normal moment in the time sequence; obtain the second difference between each suspected abnormal moment and the preset vibration reference value;

[0024] Based on the reciprocal of the time interval, the second difference value, and the second difference value, the vibration mutation characteristic value of each suspected abnormal moment is determined.

[0025] Preferably, the step of filtering suspected abnormal moments to obtain the moments of incomplete welding based on the consistency and correlation of the data change trends of welding pressure data and welding vibration data within the time neighborhood of each suspected abnormal moment specifically includes:

[0026] Based on the consistency between the direction of change of welding pressure data and the direction of change of welding vibration data at each suspected abnormal moment, as well as the correlation between welding pressure data and welding vibration data, the characteristic index of incomplete welding at each suspected abnormal moment is obtained.

[0027] The moment when the normalized value of the cold solder joint characteristic index is greater than or equal to the preset abnormal threshold is taken as the moment of suspected abnormality.

[0028] Preferably, the step of obtaining the incomplete weld characteristic index for each suspected abnormal moment based on the consistency between the data change direction of welding pressure data and the data change direction of welding vibration data at each suspected abnormal moment, as well as the correlation between welding pressure data and welding vibration data, specifically includes:

[0029] For any suspected anomaly moment, the first correlation coefficient is determined based on the consistency of the signs of the first difference and the second difference.

[0030] Welding pressure data at each time point within the time neighborhood of the suspected abnormal moment is used to form a first data sequence, and welding vibration data at each time point within the time neighborhood of the suspected abnormal moment is used to form a second data sequence; the correlation coefficient between the first data sequence and the second data sequence is used as the second correlation coefficient.

[0031] The product of the first correlation coefficient and the second correlation coefficient is the characteristic index of the cold solder joint at the suspected abnormal moment.

[0032] Preferably, the step of obtaining the weld defect anomaly evaluation value based on the distribution of risk indicators at each weld defect moment, as well as the temporal and quantitative distribution of weld defect moments, specifically includes:

[0033] Based on the distribution and duration characteristics of the time intervals between adjacent incomplete welds during the welding process, a time characteristic factor is obtained;

[0034] The product of the time characteristic factor, the proportion of times of poor welding, and the mean of the risk indicators of all times of poor welding is normalized to obtain the evaluation value of poor welding anomaly in the welding process.

[0035] Preferably, the step of obtaining the time characteristic factor based on the time interval distribution and duration characteristics between adjacent incomplete weld moments during the welding process specifically includes:

[0036] Other suspected abnormal moments besides the moment of poor soldering are taken as interference moments, and the average of the time interval between every two adjacent moments of poor soldering is taken as the first time coefficient.

[0037] When there is no interference time between two adjacent times of poor soldering, the time interval between two adjacent times of poor soldering is obtained as the second time coefficient.

[0038] The product of the reciprocal of the first time coefficient and the maximum value of the second time coefficient is used as the time feature factor.

[0039] In a second aspect, the present invention provides an ultrasonic welding machine for preventing cold solder joints, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the computer program, when executed by the processor, implements the steps of an ultrasonic welding control method for preventing cold solder joints.

[0040] The embodiments of the present invention have at least the following beneficial effects:

[0041] This invention first acquires welding parameter data in two dimensions during the welding process, providing a data foundation for subsequent collaborative analysis of the consistency of data characteristics under the condition of incomplete welds. Then, considering that incomplete welds often result from the coordinated anomalies of welding pressure and vibration, the abrupt change characteristics of pressure and vibration are integrated to fully reflect the threat posed by these coordinated changes to welding quality, ultimately yielding an index that accurately reflects the risk of incomplete welds. Furthermore, the correlation between pressure and vibration data at moments when data anomalies may occur is analyzed to eliminate interference from non-incomplete welds and identify moments that accurately characterize incomplete weld situations. Finally, the severity assessment result of incomplete weld risk, i.e., the incomplete weld anomaly evaluation value, is obtained through the characteristic performance of the three dimensions. This value can be used to determine whether the control parameters of ultrasonic welding need to be adjusted, providing a unified judgment standard for the incomplete weld anomaly evaluation results under different working conditions. This can effectively reduce the probability of batch incomplete welds, improve the welding quality and reliability of products, reduce product failures caused by incomplete welds, and optimize the stability of the welding process. Attached Figure Description

[0042] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0043] Figure 1 This is a flowchart of the steps of an ultrasonic welding control method for preventing cold solder joints provided by the present invention;

[0044] Figure 2 This is a flowchart of the steps for obtaining risk indicators at each suspected abnormal moment provided by the present invention;

[0045] Figure 3 This is a flowchart of the steps for obtaining suspected abnormal moments and normal moments provided by the present invention. Detailed Implementation

[0046] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of an ultrasonic welding control method and ultrasonic welding machine for preventing incomplete welds according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0047] Unless otherwise defined, 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 pertains.

[0048] The following description, in conjunction with the accompanying drawings, details the specific scheme of the ultrasonic welding control method for preventing incomplete welds and the ultrasonic welding machine provided by the present invention.

[0049] Please see Figure 1 The diagram illustrates a flowchart of an ultrasonic welding control method for preventing cold solder joints according to an embodiment of the present invention. The method includes the following steps:

[0050] Step S100: Obtain welding pressure data and welding vibration data at each moment during the welding process.

[0051] The working method and process of an ultrasonic welding machine vary depending on different welding requirements, workpiece types, and welding processes. There are also various ways the welding head moves. This embodiment takes continuous welding (integral welding) as an example. Continuous welding is commonly used for welding plastic pipes and films, as well as for products with high requirements for weld continuity and sealing. During continuous welding, as the welding head moves continuously along a preset welding path (straight line, curve, etc.), the workpiece material continues to fuse under ultrasonic vibration and pressure, forming a continuous weld. During continuous welding, the welding data exhibits a relatively stable distribution, and maintaining stable pressure ensures uniform material fusion.

[0052] In one embodiment, a high-precision piezoresistive pressure sensor is used, installed between the welding head and the pressure application mechanism to ensure accurate pressure transmission to the sensor. The output signal line of the pressure sensor is connected to the analog input channel of the data acquisition card. The pressure data acquisition frequency is set in the data acquisition software, for example, 10Hz, meaning 10 acquisitions per second, to ensure real-time capture of pressure changes. In other embodiments, the implementer can select a suitable pressure acquisition device and pressure acquisition frequency based on the specific implementation scenario. The pressure data acquired by the sensor is recorded as welding pressure data.

[0053] In one embodiment, a laser displacement sensor is used to collect the vibration displacement of the welding head in real time, serving as welding vibration data. The signal output by the laser displacement sensor is a digital signal, which can be directly connected to the digital input interface of a data acquisition card. The data acquisition frequency is consistent with the welding pressure data. The placement of the laser displacement sensor can be determined by the implementer according to the specific implementation scenario. It needs to be installed in a location where the vibration displacement of the welding head can be directly measured and the optical path is not interfered with. As a specific example, the welding head structure is a flat welding head, typically 20-50mm wide, with a flat working end face. It moves along the axial direction of the pipe, and the vibration direction is up and down (i.e., perpendicular to the workpiece surface). In this case, the laser displacement sensor is installed on the side of the non-working end of the welding head, that is, on the side away from the welding area. For example, when welding pipes, the welding head faces the outside of the equipment. This is only for illustrative purposes; the implementer can determine the more specific installation based on the specific implementation scenario.

[0054] Welding vibration data reflects the amplitude of the welding head's vibration displacement during ultrasonic welding. Stable welding head vibration displacement ensures uniform heating of the material in the welding area, achieving good fusion between workpieces; abnormal amplitude fluctuations can lead to uneven frictional heating, increasing the risk of incomplete welds. Welding pressure data is a key indicator reflecting the mechanical force exerted by the welding head on the workpiece and the effectiveness of workpiece contact and energy transfer. Maintaining stable pressure ensures uniform fusion of materials and avoids incomplete welds.

[0055] In one embodiment, the welding process up to the current moment is monitored, and welding pressure data and welding machine vibration data are collected in real time at each moment. It should be understood that a false weld usually occurs after a certain period of abnormal energy accumulation. The implementer can set the minimum time length before data analysis operations according to the specific implementation scenario, which is not limited here.

[0056] Step S200: Based on the deviation of welding pressure data from the baseline at each moment and the difference and time distribution of welding pressure data under pressure mutation, combined with the deviation of welding vibration data from the baseline at each moment and the difference and time distribution of welding vibration data under vibration mutation, a risk index is obtained for each suspected abnormal moment.

[0057] Firstly, welding pressure and vibration both have normal fluctuation ranges during the welding process, and not all parameter deviations are related to incomplete welds. For example, a brief fluctuation in the power supply may cause a slight deviation in a single pressure parameter, and slight resonance in mechanical components may cause a fluctuation in a single vibration parameter. Such deviations in a single parameter usually do not directly lead to incomplete welds. If the moments when both types of parameters change abruptly are not first identified, normal fluctuations or interference from a single parameter may be misjudged as abnormal, increasing the redundancy and misjudgment rate in subsequent risk assessments. Therefore, it is necessary to first clarify the core screening condition of simultaneous abrupt changes in pressure and vibration, and isolate the suspected abnormal moments that may be truly related to incomplete welds from all moments, laying the foundation for accurate calculation of risk indicators. The specific analysis process is described in step S201.

[0058] Secondly, the impact mechanisms and quantitative dimensions of the abnormality levels of pressure mutations and vibration mutations on welding quality differ, making them impossible to measure using a single standard. The differential and temporal distributions of welding pressure mutations reflect the stability of the mechanical pressure application system (e.g., the degree of component loosening); while the differential and temporal distributions of welding vibration mutations reflect the reliability of the ultrasonic energy transfer system (e.g., the performance of the power amplifier). Simply determining whether a mutation has occurred cannot quantify its severity. It is necessary to calculate mutation characteristic values ​​for both types of parameters separately, combining their differential and temporal distributions, to accurately characterize their respective abnormal intensity. The specific analysis process is detailed in steps S202 and S203.

[0059] Thirdly, the abrupt change characteristic value of a single parameter can only reflect the degree of abnormality of that parameter, while the occurrence of incomplete welds is mostly the result of the combined abnormality of welding pressure and welding vibration. For example, a sudden change in pressure leads to poor contact between the weld head and the workpiece, while a sudden change in vibration leads to insufficient frictional heat generation. The combined effect of the two will significantly increase the risk of incomplete welds. If only the characteristic value of pressure abrupt change is used for assessment, the superimposed effect of vibration abnormality will be ignored, leading to an underestimation of the risk. Therefore, it is necessary to integrate the abrupt change characteristic values ​​of both types of parameters to fully reflect the threat of pressure and vibration combined abrupt changes to welding quality, and finally obtain an indicator that can accurately reflect the risk of incomplete welds. The specific analysis process is shown in step S204.

[0060] In this regard, such as Figure 2 As shown, the method for obtaining the risk indicators at each suspected abnormal moment can be implemented by steps S201 to S204.

[0061] Step S201: Based on the deviation of welding pressure data from the benchmark at each moment and the deviation of welding vibration data from the benchmark at each moment, all moments are screened to obtain suspected abnormal moments and normal moments; the suspected abnormal moments represent moments where pressure changes and vibration changes occur.

[0062] The main purpose of this step is to screen out situations with pressure abrupt changes by analyzing the characteristics of welding pressure data, screen out situations with vibration abrupt changes by analyzing the characteristics of welding vibration data, and then combine the moments when both abrupt changes occur simultaneously to initially screen out suspected abnormal moments that may be related to poor welding.

[0063] As a concrete example, such as Figure 3 As shown, the method for obtaining suspected abnormal moments and normal moments can be implemented by steps S2011 to S2014.

[0064] Step S2011: For each moment, the moment when the difference between the welding pressure data and the preset pressure reference value exceeds the preset pressure fluctuation range is taken as the pressure change moment.

[0065] Welding pressure data characterizes the pressure performance at each moment during the welding process. The preset pressure benchmark value characterizes the pressure performance that should be maintained when the welding process remains stable under the current implementation scenario. The pressure fluctuation range characterizes the allowable range of pressure fluctuations when the welding process remains stable. The pressure change moment characterizes the moment when a pressure change occurs during the welding process.

[0066] During welding, if a cold weld is caused by equipment malfunction or factors related to the workpiece itself, it can easily lead to a sudden drop or irregular, large fluctuation in welding pressure. Specifically, this fluctuation manifests as a rapid jump in pressure from a high level to a low level within a short period, followed by a rapid, unannounced rebound, with no significant regularity in the amplitude or time interval of each jump; simultaneously, the frequency of these fluctuations is significantly higher than under normal welding conditions. In normal welding, minor fluctuations in welding pressure typically occur every few seconds to tens of seconds, while in a cold weld scenario, significant changes in welding pressure can occur once per second, or even once within a shorter timeframe. Based on this characteristic, by comparing welding pressure data with a preset pressure benchmark value, it is possible to reflect whether the welding pressure data deviates abruptly from the benchmark at each moment, thus identifying the time when a pressure abrupt change occurs.

[0067] As a specific example, the preset pressure reference value is set to 500N, and ±5% of the preset pressure reference value is taken as the allowable fluctuation range of the welding pressure data, that is, the preset pressure fluctuation range is [-25, 25]. It should be noted that the implementer can set the preset pressure reference value and pressure fluctuation range according to the specific implementation scenario and specific implementation requirements. This embodiment is set for the pipe welding scenario.

[0068] When the difference between the welding pressure data at a certain moment and the preset pressure reference value is greater than 0, it indicates that the welding pressure data is higher than the reference value. When the difference between the welding pressure data at a certain moment and the preset pressure reference value is less than 0, it indicates that the welding pressure data is lower than the reference value. This process determines whether the difference is within the allowable fluctuation range. If the difference is not within the pressure fluctuation range, the corresponding moment is recorded as the pressure mutation moment, indicating that there is a sudden change in the welding pressure data at that moment.

[0069] It should be understood that the data fluctuation range corresponding to the pressure data can also be preset directly. The welding pressure data at each moment is compared with the data fluctuation range. If the welding pressure data is within the range, it means that the possibility of abnormality in the pressure within the allowable fluctuation range is small. If the welding pressure data is not within the range, it means that there is a sudden change in the welding pressure data. The moment corresponding to the welding pressure data is taken as the moment of pressure change. As a specific example, the data fluctuation range can be set to [475, 525].

[0070] Step S2012: For each moment, the moment when the difference between the welding vibration data and the preset vibration reference value exceeds the preset vibration fluctuation range is taken as the vibration change moment.

[0071] Welding vibration data characterizes the vibration displacement at each moment during the welding process. The preset vibration reference value characterizes the vibration displacement that should be maintained when the welding process remains stable under the current implementation scenario. The vibration fluctuation range characterizes the allowable range of vibration displacement fluctuations when the welding process remains stable. The vibration change moment characterizes the moment when a vibration change occurs during the welding process.

[0072] During welding, if a cold weld is caused by equipment failure or factors related to the workpiece itself, the energy transfer path of the welding vibration amplitude will be obstructed, thus breaking its stable state and resulting in obvious abnormal changes or irregular large fluctuations. Specifically, this fluctuation manifests as follows: welding vibration data, which should rapidly rise to the set value at the moment of welding initiation and subsequently remain within the allowable fluctuation range, may suddenly deviate from the set value, experiencing a rapid drop below the allowable range or a sudden rise above the allowable range. Subsequently, it may return to a level close to the baseline value without warning, and the amplitude and time interval of each jump are not significantly regular. Simultaneously, the frequency of these abnormal fluctuations will be significantly higher than under normal welding conditions. During normal continuous welding, minor fluctuations in welding vibration data are extremely rare, and even when they occur, they remain within the allowable range, with intervals typically far exceeding several seconds. However, in cold weld scenarios, the frequency of abnormal changes in welding vibration data exceeding the allowable range will significantly increase, potentially occurring every few seconds or even less, leading to uneven frictional heating and disordered energy distribution in the welding area, further exacerbating the risk of cold welds.

[0073] As a specific example, the preset vibration reference value is 30. The preset vibration fluctuation range is It should be noted that the preset vibration reference value and vibration fluctuation range can be set by the implementer according to the specific implementation scenario and requirements. This embodiment is set for the pipe welding scenario.

[0074] Step S2013: Identify suspected abnormal moments, which are simultaneously moments of pressure change and moments of vibration change.

[0075] Specifically, the moment that corresponds to both a pressure change moment and a vibration change moment is recorded as a suspected abnormal moment. Thus, the suspected abnormal moment represents the moment when both pressure change and vibration change phenomena exist simultaneously.

[0076] Step S2014: All times except suspected abnormal times, pressure change times, and vibration change times are considered normal times.

[0077] It should be understood that suspected abnormal moments represent moments when both pressure and vibration mutations occur simultaneously, pressure mutation moments represent moments when pressure mutations occur during the welding process, vibration mutation moments represent moments when vibration mutations occur during the welding process, and normal moments represent moments when the data is stable during the welding process.

[0078] Step S202: Based on the difference distribution and time distribution of welding pressure data between each suspected abnormal moment and the normal moment, obtain the pressure mutation characteristic value for each suspected abnormal moment.

[0079] The main purpose of this step is to analyze the characteristics of welding pressure data at suspected abnormal moments across three dimensions, and to quantify the severity of pressure abrupt changes in a single dimension.

[0080] Specifically, the first step is to obtain the time interval between each suspected abnormal moment and the nearest normal moment in the time series; to obtain the first difference value of the welding pressure data between each suspected abnormal moment and the nearest normal moment in the time series; and to obtain the first difference value between each suspected abnormal moment and the preset pressure benchmark value.

[0081] The first difference value is the absolute value of the difference between the welding pressure data at each suspected abnormal moment and the most recent normal moment in the time series. The first difference value is the difference between each suspected abnormal moment and the preset pressure reference value. This time interval represents the most recent time length between each suspected abnormal moment and the historical normal state. The first difference value represents the pressure difference between each suspected abnormal moment and the historical normal state. The first difference value represents the difference between the welding pressure data at each suspected abnormal moment and the preset pressure reference value.

[0082] It should be understood that the most recent normal moment in time sequence refers to the normal moment with the shortest time interval for any suspected abnormal moment.

[0083] The second step involves determining the vibration abrupt change characteristic value for each suspected abnormal moment based on the reciprocal of the time interval, the second difference value, and the second difference value.

[0084] Specifically, the larger the absolute value of the first difference, the greater the difference between the welding pressure data at the suspected abnormal moment and the corresponding preset pressure benchmark value, and the greater the degree of deviation from the benchmark. The larger the value of the first difference, the greater the pressure difference between the suspected abnormal moment and the historical normal state. The smaller the time interval, the greater the degree of abrupt change at the suspected abnormal moment, and the higher the degree of pressure change at the corresponding suspected abnormal moment.

[0085] More specifically, for each suspected anomaly moment, the product of the reciprocal of the time interval, the absolute value of the first difference, and the first difference value is normalized to obtain the stress mutation characteristic value for each suspected anomaly moment. The normalization method can be the minimization normalization method, which is not restricted here. The implementer can choose according to the specific implementation scenario.

[0086] Step S203: Based on the difference distribution and time distribution of welding vibration data between each suspected abnormal moment and the normal moment, obtain the vibration mutation characteristic value of each suspected abnormal moment.

[0087] The main purpose of this step is to analyze the characteristics of welding vibration data at suspected abnormal moments in three dimensions and quantify the severity of vibration abrupt changes in a single dimension.

[0088] Specifically, the first step is to obtain the time interval between each suspected abnormal moment and the nearest normal moment in the time series; to obtain the second difference value of the welding vibration data between each suspected abnormal moment and the nearest normal moment in the time series; and to obtain the second difference value between each suspected abnormal moment and the preset vibration reference value.

[0089] The second difference value is the absolute value of the difference between the welding vibration data at each suspected abnormal moment and the most recent normal moment in the time series. The second difference value is the difference between each suspected abnormal moment and the preset vibration reference value. This time interval represents the most recent time length between each suspected abnormal moment and the historical normal state. The second difference value represents the vibration difference between each suspected abnormal moment and the historical normal state. The second difference value represents the difference between the welding vibration data at each suspected abnormal moment and the preset vibration reference value.

[0090] The second step involves determining the vibration abrupt change characteristic value for each suspected abnormal moment based on the reciprocal of the time interval, the second difference value, and the second difference value.

[0091] Specifically, the larger the absolute value of the second difference, the greater the difference between the welding vibration data at the suspected abnormal moment and the corresponding preset vibration reference value, and the greater the degree of deviation from the reference. The larger the value of the second difference, the greater the difference between the vibration at the suspected abnormal moment and the historical normal state. The smaller the time interval, the greater the degree of abrupt change at the suspected abnormal moment, and the higher the degree of vibration abrupt change at the corresponding suspected abnormal moment.

[0092] More specifically, for each suspected abnormal moment, the product of the reciprocal of the time interval, the absolute value of the second difference, and the second difference value is normalized to obtain the vibration mutation characteristic value of each suspected abnormal moment. The normalization method can be the minimization normalization method, which is not restricted here. The implementer can choose according to the specific implementation scenario.

[0093] Step S204: Combine the pressure mutation characteristic value and vibration mutation characteristic value at each suspected abnormal moment to obtain the risk index for each suspected abnormal moment.

[0094] As a specific example, for any suspected abnormal moment, the average of the pressure mutation characteristic value and the vibration mutation characteristic value is used as the risk indicator for the corresponding suspected abnormal moment.

[0095] In other embodiments, the pressure mutation feature value and vibration mutation feature value can be fused by weighted summation, and higher weights can be assigned to mutation dimensions that require higher attention or higher precision. This will not be elaborated on here, and implementers can set it according to the specific implementation scenario.

[0096] Thus, by integrating the results of dual-dimensional feature analysis through risk indicators, the severity of the coordinated anomaly of pressure and vibration was quantified, thereby reflecting the degree of its impact on the cold weld and the magnitude of the risk.

[0097] Step S300: Based on the consistency and correlation of the data change trends of welding pressure data and welding vibration data at each suspected abnormal moment in the time neighborhood, the suspected abnormal moments are screened to obtain the moments of false welding.

[0098] Firstly, suspected abnormal moments indicate simultaneous abrupt changes in welding pressure and vibration. However, such abrupt changes can be caused by various factors, such as instantaneous power supply fluctuations in the equipment or external mechanical interference. Not all coordinated abrupt changes are directly related to cold welds. The essence of a cold weld is the failure of energy transfer and frictional heat generation due to poor contact in the welding area. This failure manifests as changes in the intrinsic correlation between pressure and vibration data. For example, when the welding head loosens its contact with the workpiece, a pressure drop is usually accompanied by an abnormal increase in vibration amplitude (due to reduced contact impedance leading to enhanced vibration reflection), or both pressure and vibration show a sudden drop (due to interruption of the energy transfer path). Coordinated abrupt changes caused by external interference (such as ground vibration) may have no regularity in the direction of pressure and vibration changes (e.g., a sudden increase in pressure followed by a sudden drop in vibration). Therefore, it is necessary to analyze the consistency of the direction of change and the correlation of data to isolate coordinated abrupt changes caused by factors other than cold welds and accurately capture the parameter change patterns unique to cold welds.

[0099] Secondly, even if there is some consistency and correlation between pressure and vibration data, the strength of this correlation may vary (e.g., weak or strong correlation). The presence or absence of a correlation alone cannot determine whether a suspected anomaly reaches the risk level of a cold weld. For example, slight contact instability may result in a weak positive correlation between pressure and vibration, but without forming a cold weld; while severe contact failure will lead to a strong negative or strong positive correlation, meeting the characteristics of a cold weld. Therefore, consistency and correlation need to be quantified as cold weld characteristic indicators, and moments with sufficiently high correlation strength should be screened using a preset anomaly threshold to ensure that the final cold weld moments have a clear process judgment basis.

[0100] To address this, the method for selecting the timing of a cold solder joint can be implemented through the following steps: first, calculate the characteristic indicators of a cold solder joint based on the consistency of the direction of change of pressure and vibration and the correlation of data, and then determine the final timing of the cold solder joint by comparing thresholds.

[0101] The first step is to obtain the incomplete weld characteristic index for each suspected abnormal moment based on the consistency between the direction of change of welding pressure data and the direction of change of welding vibration data at each suspected abnormal moment, as well as the correlation between welding pressure data and welding vibration data.

[0102] Specifically, for any suspected abnormal moment, a first correlation coefficient is determined based on the consistency of the signs of the first difference and the second difference; welding pressure data at each moment in the time neighborhood of the suspected abnormal moment are obtained to form a first data sequence, and welding vibration data at each moment in the time neighborhood of the suspected abnormal moment are obtained to form a second data sequence; the correlation coefficient between the first data sequence and the second data sequence is used as the second correlation coefficient; the product between the first correlation coefficient and the second correlation coefficient is the cold weld characteristic index of the corresponding suspected abnormal moment.

[0103] As a concrete example, let's take any suspected abnormal moment as an illustration. The first difference at the t-th suspected abnormal moment represents the degree of pressure deviation from the benchmark, and the second difference at the t-th suspected abnormal moment represents the degree of vibration deviation from the benchmark. When both the first and second differences are greater than 0, that is, when both are positive, it indicates that the direction of pressure deviation and the direction of vibration deviation are both higher than the benchmark value. When both the first and second differences are less than 0, that is, when both are negative, it indicates that the direction of pressure deviation and the direction of vibration deviation are both lower than the benchmark value.

[0104] Therefore, when the first and second differences have the same sign, the first correlation coefficient is set to 1; when the first and second differences have different signs, the first correlation coefficient is set to 0. Thus, the first correlation coefficient characterizes the degree of consistency between the pressure deviation direction and the vibration deviation direction at the t-th suspected anomaly time; a larger value indicates higher consistency.

[0105] As a concrete example, a preset number of times adjacent to the t-th suspected anomaly time are obtained as the time neighborhood of that suspected anomaly time. The preset number can be 6, for example, taking the t-th suspected anomaly time as the center time and taking 3 times before and after it as the time neighborhood. It should be noted that if it is not possible to take 3 times on one side of the suspected anomaly time, then the 6 nearest times to the suspected anomaly time can be obtained as the time neighborhood, which will not be elaborated on here.

[0106] Furthermore, this embodiment uses the Pearson correlation coefficient to evaluate the correlation between the first and second data sequences. The closer the Pearson correlation coefficient between the first and second data sequences at the t-th suspected anomaly time is to 1, the higher the correlation between the pressure and vibration change trends within the time neighborhood of the t-th suspected anomaly time, and the more likely the anomaly at that time is caused by a faulty weld. Thus, the second correlation coefficient characterizes the magnitude of the correlation between the pressure and vibration change trends at the t-th suspected anomaly time.

[0107] Finally, the cold weld characteristic index integrates the correlation between pressure and vibration data at suspected abnormal moments in two dimensions. The larger the value, the higher the degree of conformity to the cold weld characteristics, and the greater the possibility of it being a cold weld.

[0108] The second step is to take the suspected abnormal moment when the normalized value of the cold solder joint feature index is greater than or equal to the preset abnormal threshold as the cold solder joint moment.

[0109] It should be noted that before performing threshold comparison, since the second correlation coefficient may have negative values, the cold solder joint characteristic index needs to be normalized. The minimax normalization method can be used, and the implementer can choose according to the specific implementation scenario. There are no restrictions here.

[0110] When the normalized value of the cold weld characteristic index at a suspected abnormal moment is greater than or equal to the preset abnormal threshold, it indicates that the pressure and vibration change trends at the suspected abnormal moment are highly correlated and meet the characteristics of cold weld to a high degree, thus indicating that the suspected abnormal moment is more likely to be a cold weld moment.

[0111] When the normalized value of the cold weld characteristic index at a suspected abnormal moment is less than the abnormal threshold, it indicates that the correlation between the pressure and vibration change trends at the suspected abnormal moment is small, and the degree of conformity with the characteristics of cold weld is low. This further indicates that the possibility of the suspected abnormal moment being a cold weld moment is smaller. At this time, the credibility of the suspected abnormal moment is low, and the suspected abnormal moment is regarded as an interference moment.

[0112] It should be understood that the core characteristic of interference moments is inconsistent or low consistency of mutation patterns. Interference moments indicate that the parameter abnormalities at this time are caused by factors other than poor soldering, such as sudden power supply voltage jumps or interference from impurities on the workpiece surface, rather than continuous energy transfer abnormalities.

[0113] In this embodiment, the abnormal threshold is set to 0.6, which can be set by the implementer according to the specific implementation scenario.

[0114] Step S400: Based on the distribution of risk indicators at each moment of poor welding, as well as the time and quantity distribution of poor welding moments, a poor welding anomaly evaluation value is obtained for the welding process. The poor welding anomaly evaluation value is used to determine whether the control parameters of ultrasonic welding should be adjusted.

[0115] Firstly, the temporal distribution of cold solder joints directly reflects the persistence and concentration of welding anomalies, and is a key dimension for assessing overall welding stability. Scattered cold solder joint moments (such as adjacent cold solder joints with intervals of several minutes) may be caused by occasional interference (such as instantaneous power supply fluctuations), and have a limited impact on overall welding quality; while concentrated cold solder joint moments (such as adjacent cold solder joints with intervals of only a few seconds) indicate that there is a continuous fault in the equipment (such as aging of power amplifiers or loosening of mechanical parts), and the abnormal state has formed a chain reaction. If parameters are not adjusted in time, it will lead to a batch of cold solder joints in subsequent welds.

[0116] The presence or absence of solder joints alone cannot distinguish between occasional and persistent anomalies. Therefore, it is necessary to extract time feature factors by analyzing the time interval distribution and persistence characteristics of adjacent solder joint moments, quantifying the concentration of solder joints in the time dimension, and providing a dynamic stability basis for overall anomaly assessment.

[0117] Secondly, the distribution of the number of instances of poor welds determines the coverage of welding anomalies. This needs to be considered in conjunction with the severity of individual instances of poor welds to avoid misjudgment. If the proportion of instances of poor welds within the total welding time is high (e.g., poor welds in 5 minutes out of 10 minutes of welding), but the average risk index for each individual instance is low (e.g., average risk index of only 20), it indicates that the anomaly coverage is broad but the severity is minor, possibly due to slight parameter deviations from the baseline. Conversely, if the proportion of instances is low but the average risk index is high (e.g., average risk index of 800), it indicates that although only a few instances are abnormal, the synergistic mutations at each instance are extremely strong (severe deviations in pressure and vibration), and the poor welds have caused substantial quality risks. Looking only at the proportion of instances ignores the severity, and looking only at the average risk index ignores the coverage. Both must be used together as evaluation dimensions to comprehensively reflect the overall impact of poor welds on the welding process.

[0118] Thirdly, single-dimensional assessment results lack comparability and need to be transformed into evaluation values ​​on a unified scale through product normalization. Under different welding scenarios (such as plastic pipe welding and copper-aluminum foil welding), the absolute values ​​of time intervals, the number of incomplete welds, and risk indicators vary significantly (e.g., a single pipe welding session lasting 1 hour may have incomplete weld intervals measured in minutes; a single foil welding session lasting 10 minutes may have incomplete weld intervals measured in seconds). Directly comparing the original data (e.g., time intervals of 5 seconds versus 5 minutes) cannot determine the degree of anomaly. Normalizing the product of the time characteristic factor, the proportion of quantities, and the mean of the risk indicators to the range of 0-1 can eliminate scenario differences, providing a unified judgment standard for the evaluation results of incomplete weld anomalies under different working conditions, and providing a clear and actionable decision-making basis for subsequent parameter adjustments.

[0119] To address this, the method for obtaining the evaluation value of the weld defect during the welding process can be implemented through the following steps: first, calculate the time characteristic factor based on the time distribution of the weld defect time; then, determine the proportion of weld defect times and the average risk index; finally, normalize the product of the three factors to obtain the final evaluation value of the weld defect.

[0120] The first step is to obtain the time characteristic factor based on the distribution and duration characteristics of the time interval between adjacent incomplete welds during the welding process, which is specifically achieved through the following steps one to three.

[0121] Step 1: Obtain the average of the time intervals between any two adjacent times of poor soldering as the first time coefficient.

[0122] Specifically, a time interval corresponds to the interval between two adjacent incomplete welds during the welding process, and the average of all such time intervals is calculated as the first time coefficient.

[0123] The first time factor directly reflects the persistence and urgency of the cold solder joint anomaly. If the interval between adjacent cold solder joints is extremely short, it indicates that the anomaly has entered the chain reaction stage. If not intervened in time, it will lead to the failure of a batch of subsequent solder joints. If the interval is long, it may be an occasional interference with limited impact on the overall welding quality.

[0124] Step 2: Other suspected abnormal moments besides the solder joint failure moment are taken as interference moments. When there are no interference moments between two adjacent solder joint failure moments, the time interval between two adjacent solder joint failure moments is obtained as the second time coefficient.

[0125] When there is no interfering moment between two adjacent moments of poor welding, it means that there is no non-poor welding interference in the working conditions between these two adjacent poor welding anomalies. The parameter fluctuations always revolve around the core feature of abnormal capacity transfer. The two adjacent moments of poor welding are manifestations of the same poor welding cause at different times, and the risk is continuous.

[0126] For example, t1 is the moment of poor soldering, where pressure and vibration decrease simultaneously due to power fluctuations; t2 is the normal moment, where parameters are briefly stable but no interference occurs; t3 is the moment of poor soldering, where pressure and vibration also decrease simultaneously due to power fluctuations; there are no low-confidence moments in between, indicating that the risks of t1 and t3 both stem from the continuous problem of power fluctuations, which is a continuous risk of poor soldering. Therefore, t1 and t3 are continuous, and the risk is ongoing.

[0127] When there is an interfering moment between two adjacent times of poor soldering, it means that the two adjacent poor soldering anomalies are interrupted by other non-poor soldering factors, the risk between them is dispersed, that is, there is no continuous risk, and thus the severity of poor soldering risk is relatively low.

[0128] For example, if there is an interference moment t2 between two adjacent times of poor soldering t1 and t3, it means that the risk of t1 is interrupted by the non-poor soldering interference of t2, and the risk of t3 may be caused by another independent interference, rather than a continuation of the risk of t1. Therefore, t1 and t3 are not continuous, and the risk is dispersed.

[0129] Thus, the second time coefficient characterizes the magnitude of the risk of persistent cold solder joints during the cold solder joint process. The larger the value, the greater the risk of persistent cold solder joints, which in turn leads to a higher probability of batch cold solder joints.

[0130] Step 3: The product of the reciprocal of the first time coefficient and the maximum value of the second time coefficient is used as the time feature factor.

[0131] It should be understood that the smaller the value of the first time coefficient, the more frequent and severe the cold solder joints are, indicating that the cold solder joints are occurring continuously within a short period of time. The larger the maximum value of the second time coefficient, the greater the risk of continued cold solder joints, and thus the greater the severity of the cold solder joints.

[0132] The second step is to normalize the product of the time characteristic factor, the proportion of times of poor welding, and the mean of the risk indicators for all times of poor welding, to obtain the evaluation value of poor welding anomaly in the welding process.

[0133] During the welding process, the risk of incomplete welds accumulates over time, increases in quantity, and upgrades in strength. Therefore, by multiplying the analysis results from the three dimensions to achieve the risk superposition effect, the severity of incomplete welds in the currently monitored welding process can be evaluated more accurately.

[0134] As a specific example, the calculation process for the evaluation value of incomplete weld defects in the welding process can be expressed as follows:

[0135]

[0136] in, This indicates the evaluation value for abnormal welds during the welding process. This represents the average time interval between any two adjacent instances of poor soldering, which is also known as the first time coefficient. This indicates the maximum value of the second time coefficient; This indicates the percentage of times a solder joint is faulty. Indicates the number of times a solder joint fails. This represents the number of times across all moments. This represents the average risk indicator across all instances of poor soldering. This is a normalization function, which can be adjusted by the implementer in other embodiments according to the specific implementation scenario.

[0137] A higher value for the weld defect assessment value indicates a higher risk of weld defects in the current welding process, while a lower value indicates a lower risk. The ultimate purpose of the weld defect assessment value is to provide a quantitative basis for whether to adjust welding parameters.

[0138] When the abnormal evaluation value of the cold weld in the welding process is greater than or equal to the preset risk threshold, it indicates that the cold weld has shown the characteristics of high frequency, wide coverage and high intensity. It is necessary to remind relevant personnel through early warning to conduct risk investigation on the current welding process, so as to adjust the welding parameters and avoid a larger-scale batch of cold welds.

[0139] When the abnormal evaluation value of the weld defect during the welding process is less than the preset risk threshold, it indicates that the weld defect is sporadic, with low coverage and low strength, possibly caused by short-term interference. Immediate parameter adjustment is unnecessary; continuous monitoring is sufficient to avoid over-adjustment that could lead to new parameter imbalances. In this embodiment, the risk threshold is set to 0.6, but the implementer can set it according to the specific implementation scenario.

[0140] In other embodiments, in order to predict the severity of the risk of poor soldering in advance, the welding pressure data and welding vibration data of the welding process can be predicted, and then the severity of the risk can be evaluated based on the prediction results to assess the magnitude of the future risk.

[0141] More specifically, a Long Short-Term Memory (LSTM) network model is used to predict the parameters of the current welding process. LSTM can effectively process welding pressure data and welding vibration data during the welding process, capture the complex dependencies between welding parameters at different time steps, and learn from historical welding parameters to predict welding data at future moments. That is, the historical data of the current welding process (welding pressure data and welding vibration data) are input into the LSTM model to obtain two-dimensional prediction data. The model's data prediction results are added to the dataset in step S100 (a dataset consisting of welding pressure data and welding vibration data of the welding process) as the initial welding pressure data and welding vibration data. Analysis is performed according to the same steps as steps S200 to S400 in the method embodiment to obtain a false weld anomaly evaluation value based on the prediction data. The result of judging using this parameter can characterize the severity of false weld risk in the future time period, enabling early risk prediction, timely decision-making, and avoiding greater losses. It should be understood that the method of using an LSTM model to predict data is a known technology, and this part is not the focus of this invention, so it will not be elaborated further here.

[0142] This invention also provides an ultrasonic welding machine for preventing cold solder joints, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the computer program is executed by the processor, it implements the steps of an ultrasonic welding control method for preventing cold solder joints.

[0143] This invention also provides an ultrasonic welding control system for preventing cold solder joints, comprising the steps of an ultrasonic welding control method for preventing cold solder joints, specifically including:

[0144] The data acquisition module is used to acquire welding pressure data and welding vibration data at each moment during the welding process;

[0145] The anomaly analysis module is used to obtain risk indicators for each suspected anomaly moment based on the deviation of welding pressure data from the baseline at each moment, as well as the difference distribution and time distribution of welding pressure data under pressure mutation, combined with the deviation of welding vibration data from the baseline at each moment and the difference distribution and time distribution of welding vibration data under vibration mutation.

[0146] The false weld screening module is used to screen suspected abnormal moments to obtain false weld moments based on the consistency and correlation of the data change trends of welding pressure data and welding vibration data in the time neighborhood of each suspected abnormal moment.

[0147] The cold weld evaluation module is used to obtain the cold weld anomaly evaluation value of the welding process based on the distribution of risk indicators at each cold weld moment and the time and quantity distribution of cold weld moments. The cold weld anomaly evaluation value is used to determine whether the control parameters of ultrasonic welding should be adjusted.

[0148] Since an embodiment of an ultrasonic welding control method for preventing incomplete welds has already been described in detail, it will not be repeated here.

[0149] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. An ultrasonic welding control method for preventing a false weld, characterized by, The method comprises the following steps: obtaining welding pressure data and welding vibration data at each moment in the welding process; obtaining the risk index of each suspected abnormal moment according to the deviation of the welding pressure data at each moment from the reference value, the difference distribution and time distribution of the welding pressure data under pressure mutation, and the deviation of the welding vibration data at each moment from the reference value, the difference distribution and time distribution of the welding vibration data under vibration mutation; screening the suspected abnormal moments according to the consistency and correlation of the data change trend of the welding pressure data and the welding vibration data at each suspected abnormal moment in the time neighborhood to obtain the virtual welding moment; obtaining the virtual welding abnormal evaluation value of the welding process according to the distribution of the risk index of each virtual welding moment and the time distribution and quantity distribution of the virtual welding moment, wherein the virtual welding abnormal evaluation value is used to judge whether to regulate the control parameters of ultrasonic welding; wherein the method for obtaining the risk index of each suspected abnormal moment comprises: screening all moments to obtain suspected abnormal moments and normal moments according to the deviation of the welding pressure data at each moment from the reference value and the deviation of the welding vibration data at each moment from the reference value; obtaining the pressure mutation characteristic value of each suspected abnormal moment according to the difference distribution and time distribution of the welding pressure data between each suspected abnormal moment and the normal moment; obtaining the vibration mutation characteristic value of each suspected abnormal moment according to the difference distribution and time distribution of the welding vibration data between each suspected abnormal moment and the normal moment; obtaining the risk index of each suspected abnormal moment by comprehensively considering the pressure mutation characteristic value and the vibration mutation characteristic value of each suspected abnormal moment; specifically comprising: for each moment, the moment corresponding to the difference between the welding pressure data and the preset pressure reference value exceeding the preset pressure fluctuation range is determined as the pressure mutation moment; for each moment, the moment corresponding to the difference between the welding vibration data and the preset vibration reference value exceeding the preset vibration fluctuation range is determined as the vibration mutation moment; determining the suspected abnormal moment, which belongs to both the pressure mutation moment and the vibration mutation moment; the moment other than the suspected abnormal moment, the pressure mutation moment and the vibration mutation moment is determined as the normal moment.

2. The ultrasonic welding control method for preventing false welding according to claim 1, characterized by specifically comprising: obtaining the time interval between each suspected abnormal moment and the nearest normal moment in time sequence; obtaining the first difference value of the welding pressure data between each suspected abnormal moment and the nearest normal moment in time sequence; obtaining the first difference value between each suspected abnormal moment and the preset pressure reference value; determining the pressure mutation characteristic value of each suspected abnormal moment based on the reciprocal of the time interval, the first difference value and the first difference value.

3. The ultrasonic welding control method for preventing false welding according to claim 2, characterized by The step of obtaining vibration abrupt change characteristic values ​​for each suspected abnormal moment based on the difference distribution and time distribution of welding vibration data between each suspected abnormal moment and a normal moment specifically includes: Obtain the second difference value of welding vibration data between each suspected abnormal moment and the nearest normal moment in the time sequence; obtain the second difference between each suspected abnormal moment and the preset vibration reference value; Based on the reciprocal of the time interval, the second difference value, and the second difference value, the vibration mutation characteristic value of each suspected abnormal moment is determined.

4. The ultrasonic welding control method for preventing false welding according to claim 3, characterized by The process of filtering suspected abnormal moments to obtain the moments of incomplete welding, based on the consistency and correlation of the data change trends of welding pressure data and welding vibration data within the time neighborhood of each suspected abnormal moment, specifically includes: Based on the consistency between the direction of change of welding pressure data and the direction of change of welding vibration data at each suspected abnormal moment, as well as the correlation between welding pressure data and welding vibration data, the characteristic index of incomplete welding at each suspected abnormal moment is obtained. The moment when the normalized value of the cold solder joint characteristic index is greater than or equal to the preset abnormal threshold is taken as the moment of suspected abnormality.

5. A control method for preventing false welding in ultrasonic welding according to claim 4, wherein The method involves obtaining the incomplete weld characteristic index for each suspected abnormal moment based on the consistency between the direction of change in welding pressure data and the direction of change in welding vibration data at each suspected abnormal moment, as well as the correlation between welding pressure data and welding vibration data. Specifically, this includes: For any suspected anomaly moment, the first correlation coefficient is determined based on the consistency of the signs of the first difference and the second difference. Welding pressure data at each time point within the time neighborhood of the suspected abnormal moment is used to form a first data sequence, and welding vibration data at each time point within the time neighborhood of the suspected abnormal moment is used to form a second data sequence; the correlation coefficient between the first data sequence and the second data sequence is used as the second correlation coefficient. The product of the first correlation coefficient and the second correlation coefficient is the characteristic index of the cold solder joint at the suspected abnormal moment.

6. The ultrasonic welding control method for preventing a false weld according to Claim 1, characterized by The process of obtaining an abnormal evaluation value for weld defects based on the distribution of risk indicators at each moment of weld defect, as well as the temporal and quantitative distribution of these moments, specifically includes: Based on the distribution and duration characteristics of the time intervals between adjacent incomplete welds during the welding process, a time characteristic factor is obtained; The product of the time characteristic factor, the proportion of times of poor welding, and the mean of the risk indicators of all times of poor welding is normalized to obtain the evaluation value of poor welding anomaly in the welding process.

7. A control method for preventing false welding in ultrasonic welding according to claim 6, wherein The time characteristic factor is obtained based on the distribution and duration characteristics of the time interval between adjacent incomplete weld moments during the welding process, specifically including: Other suspected abnormal moments besides the moment of poor soldering are taken as interference moments, and the average of the time interval between every two adjacent moments of poor soldering is taken as the first time coefficient. When there is no interference time between two adjacent times of poor soldering, the time interval between two adjacent times of poor soldering is obtained as the second time coefficient. The product of the reciprocal of the first time coefficient and the maximum value of the second time coefficient is used as the time feature factor.

8. An ultrasonic welder that prevents false welds, comprising a memory, a processor, and a computer program stored on the memory and running on the processor, characterized in that, The computer program, which is executed by a processor, implements the steps of a control method for ultrasonic welding for preventing false welds according to any one of claims 1 to 7.

Images