Ultrasonic welding quality control system and method

Abstract

The invention provides an ultrasonic welding quality control system and method, and relates to the technical field of ultrasonic welding, and the system comprises an acquisition and recognition module which is used for collecting sensor data of welding head displacement, weldment interface temperature, welding head amplitude and welding power in real time in the ultrasonic welding process, identifying a welding head displacement peak value, an interface temperature inflection point and a power fluctuation critical characteristic value; and the parameter extraction module is used for extracting a displacement peak value characteristic value, an interface temperature inflection point characteristic value, a power fluctuation critical characteristic value, a welding head amplitude characteristic value and a welding time characteristic value in the welding process based on the sensor data to form a characteristic parameter set. By means of multi-parameter feature extraction, correlation analysis, dynamic adjustment and closed-loop feedback, accurate control over the welding process is achieved, the welding quality stability is improved, and the rate of unqualified products is reduced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of ultrasonic welding, in particular to an ultrasonic welding quality control system and method. BACKGROUND

[0002] In the production of automobile wire harness welding, the traditional ultrasonic welding quality control technology often has some defects. The traditional technology pays less attention to parameters directly related to the fusion state of the welding joint, such as the interface temperature of the welding part and the displacement peak value of the welding head, and it is difficult to completely capture the real state in the welding process. For example, when a certain automobile wire harness factory uses traditional control technology to weld AWG 22 specification copper wires and terminals, only the welding power and vibration time are monitored. When the environmental temperature fluctuation causes the thickness of the oxide layer on the surface of the wire to change, the interface temperature does not reach the ideal melting threshold and is not detected, and eventually some of the welds are not fully fused. In addition, the traditional technology is based on the pre-set fixed parameter threshold to judge the welding quality, and the correlation between multiple parameters is not established, so it is difficult to dynamically predict the quality level according to the real-time parameter changes, and it is easy to occur. For example, the traditional technology sets the welding power 500W and the time 0.8s as the qualified threshold. When the resistance of a certain batch of wires is slightly higher due to batch differences in material quality, although the power and time meet the pre-set values, the interface temperature does not reach the inflection point and is still judged to be qualified. The actual test found that 3% of the welds in this batch had virtual welding and appeared in the simulation vibration test. SUMMARY

[0003] The technical problem to be solved by the present application is to provide an ultrasonic welding quality control system and method to improve the stability and reliability of ultrasonic welding quality and reduce the incidence of virtual welding, underwelding and overwelding defects.

[0004] To solve the above technical problems, the technical solutions of the present application are as follows: In a first aspect, an ultrasonic welding quality control system comprises: A collection and identification module is used to collect sensor data of the displacement of the welding head, the interface temperature of the welding part, the amplitude of the welding head and the welding power in real time during the ultrasonic welding process, and to identify the displacement peak value of the welding head, the inflection point of the interface temperature and the critical characteristic value of the power fluctuation; A parameter extraction module is used to extract the displacement peak value characteristic value, the interface temperature inflection point characteristic value, the power fluctuation critical characteristic value, the welding head amplitude characteristic value and the welding time characteristic value in the welding process based on the sensor data, and to form a characteristic parameter set; An analysis and generation module is used to establish a multi-parameter correlation based on the characteristic parameter set, to evaluate the correlation in different regions, and to generate dynamic adjustment values combined with the distribution characteristics of the welding parameters in each region; a quality prediction module configured to input the dynamic adjustment value and the set of characteristic parameters into a prediction mechanism to obtain a welding quality level; an adjustment triggering module configured to trigger a process parameter adjustment process when the welding quality level deviates from a preset level by more than a threshold value, and dynamically adjust welding pressure, vibration amplitude, welding power and welding time; a control correction module configured to feed back the adjusted process parameters to welding control in real time, dynamically correct welding energy input according to interface temperature monitoring data, automatically reduce energy input when the interface peak temperature is greater than a threshold value, and realize closed-loop quality control of the welding process through multi-parameter correlation evaluation.

[0005] Further, based on the sensor data, the displacement peak characteristic value, the interface temperature inflection point characteristic value, the power fluctuation critical characteristic value, the welding head amplitude characteristic value and the welding time characteristic value in the welding process are extracted to form a set of characteristic parameters, including: The peak value of the welding head displacement is identified and extracted from the welding head displacement sensor data as the displacement peak characteristic value; Based on the time node corresponding to the displacement peak characteristic value, the synchronous data in the interface temperature sensor is located, the temperature value corresponding to the temperature change inflection point caused by material melting at the time node is extracted as the interface temperature inflection point characteristic value; According to the energy input stage corresponding to the interface temperature inflection point characteristic value, the power value at which the power from the stable state enters the critical point of severe fluctuation is identified from the power sensor data as the power fluctuation critical characteristic value; Based on the power fluctuation critical characteristic value, the critical state is determined as the key process node, the amplitude reading of the welding head amplitude sensor during the node is extracted as the welding head amplitude characteristic value, and the time from the start of welding to the node is extracted as the welding time characteristic value; The displacement peak characteristic value, the interface temperature inflection point characteristic value, the power fluctuation critical characteristic value, the welding head amplitude characteristic value and the welding time characteristic value are integrated to form a set of characteristic parameters for quality discrimination.

[0006] Further, based on the set of characteristic parameters, a multi-parameter correlation relationship is established, the correlation relationship is evaluated by partition, and a dynamic adjustment value is generated in combination with the distribution characteristics of the welding parameters in each region, including: Based on the set of characteristic parameters, the statistical correlation between the parameters is analyzed to construct a correlation relationship for quantitatively describing the interaction relationship between multiple ultrasonic welding parameters; Based on the parameter correlation relationship, the entire parameter correlation space is divided into multiple evaluation regions with different process states according to the distribution and clustering characteristics of the welding parameters in the relationship; For each evaluation region, the central tendency and dispersion degree of the internal welding parameter data are statistically analyzed to extract the numerical distribution characteristics corresponding to the region; Based on the numerical distribution characteristics, the compensation value of the welding parameters in each region is calculated respectively to generate a dynamic adjustment value for real-time regulation of the welding process.

[0007] Further, the dynamic adjustment value and the feature parameter set are collectively input into a prediction mechanism to obtain a welding quality level, including: The dynamic adjustment value and the feature parameter set are fused to generate a comprehensive feature vector representing the current welding process state, and the comprehensive feature vector is input into a preset welding quality classification prediction rule library; The input comprehensive feature vector is matched and logically judged by the rule library to obtain an evaluation result representing the possibility of different welding quality levels; Based on the analysis result, a quality level prediction value corresponding to the current welding state is obtained.

[0008] Further, the quality level includes virtual welding, underwelding, good welding, and overwelding.

[0009] Further, when the welding quality level deviates from the preset level by more than a threshold value, a process parameter adjustment process is triggered to dynamically adjust the welding pressure, vibration amplitude, welding power, and welding time, including: The real-time deviation value between the predicted welding quality level and the preset level is calculated, and the specific value and direction of the deviation value are determined; The deviation value is compared with the preset threshold value, and when the deviation value is greater than the preset threshold value, a parameter adjustment instruction is triggered; According to the parameter adjustment instruction, the specific adjustment amount of the welding pressure, vibration amplitude, welding power, and welding time is dynamically calculated based on the deviation value size and direction; The specific adjustment amount is applied to the current process parameters to complete real-time updating and adjustment of the welding parameters.

[0010] Further, the adjusted process parameters are fed back to the welding control in real time, and the welding energy input is dynamically corrected according to the interface temperature monitoring data. When the interface peak temperature is greater than a threshold value, the energy input is automatically reduced to achieve closed-loop quality control of the welding process through multi-parameter correlation evaluation, including: The adjusted welding pressure, amplitude, and welding time are fed back to the welding controller in real time, and an infrared temperature sensor is used to monitor the temperature change at the welding interface, and the interface peak temperature is recorded; The interface peak temperature is compared with a preset temperature threshold value, and when the interface peak temperature is greater than the preset threshold value, it is determined that there is an overheating risk, and an energy down adjustment instruction is generated; According to the energy down-regulation instruction, the welding energy input is reduced by a preset ratio, and based on the change of the welding energy input, a multi-parameter correlation evaluation mechanism is triggered synchronously to recalculate and cooperatively adjust the welding pressure, the amplitude and the welding time; The adjusted process parameters are input into the welding controller again to perform the welding operation, so as to realize closed-loop quality control of the welding process.

[0011] In a second aspect, an ultrasonic welding quality control method comprises: Step 1: real-time acquisition of sensor data of the welding head displacement, the welding interface temperature, the welding head amplitude and the welding power in the ultrasonic welding process, and identification of the welding head displacement peak value, the interface temperature inflection point and the power fluctuation critical characteristic value; Step 2: based on the sensor data, extraction of the displacement peak value characteristic value, the interface temperature inflection point characteristic value, the power fluctuation critical characteristic value, the welding head amplitude characteristic value and the welding time characteristic value in the welding process to form a characteristic parameter set; Step 3: establishment of a multi-parameter correlation relationship based on the characteristic parameter set, partition evaluation of the correlation relationship, and generation of a dynamic adjustment value in combination with the distribution characteristics of the welding parameters in each region; Step 4: common input of the dynamic adjustment value and the characteristic parameter set into a prediction mechanism to obtain a welding quality grade; Step 5: when the welding quality grade deviates from a preset grade by more than a threshold value, triggering of a process parameter adjustment process to dynamically adjust the welding pressure, the vibration amplitude, the welding power and the welding time; Step 6: real-time feedback of the adjusted process parameters to the welding control, dynamic correction of the welding energy input according to the interface temperature monitoring data, automatic reduction of the energy input when the interface peak temperature is greater than a threshold value, and realization of closed-loop quality control of the welding process through the multi-parameter correlation evaluation.

[0012] In a third aspect, a computing device comprises: one or more processors; a storage device for storing one or more programs, when the one or more programs are executed by the one or more processors, the one or more processors implement the system.

[0013] In a fourth aspect, a computer readable storage medium stores a program, which is executed by a processor to implement the system.

[0014] The above-mentioned scheme of the present application has at least the following beneficial effects: By collecting core data such as welding head displacement, welding interface temperature, welding head amplitude and welding power in real time, accurately identifying key process nodes such as welding head displacement peak, interface temperature inflection point and power fluctuation critical value, the risk of common defects such as false welding, underwelding and overwelding can be predicted in advance, by establishing the multi-parameter correlation of the characteristic parameter set, the parameter distribution characteristics are evaluated in different regions and the dynamic adjustment value is generated, combined with the real-time monitoring of the interface temperature to correct the energy input, a closed-loop control system of monitoring, analysis, prediction, adjustment and correction is formed, when the quality level deviates from the preset value beyond the threshold value, the adjustment trigger mechanism can automatically calculate the adjustment amount of welding pressure, vibration amplitude, power and time and update the process parameters in real time, without manual debugging according to experience, even for different materials or different thickness of welding pieces, the process can be adjusted adaptively through multi-parameter correlation evaluation.

[0015] By dynamically adjusting the welding pressure, amplitude, power and energy input, various material combinations such as metal-plastic composite, thin-walled metal and high-hardness engineering plastic can be adapted, and different scene requirements such as precise welding of electronic equipment and high-strength welding of automobile parts can be met, and the application range is expanded; by real-time monitoring of parameters such as interface temperature and welding head displacement and dynamically correcting energy input, the interference of environmental factors on welding quality can be avoided, quality defects can be predicted in advance and process parameters can be adjusted in real time, reducing the rate of defective products and reducing the cost of labor, materials and time due to rework. BRIEF DESCRIPTION OF DRAWINGS

[0016] Figure 1 is a schematic diagram of an ultrasonic welding quality control system provided by an embodiment of the present application.

[0017] Figure 2 is a flowchart of an ultrasonic welding quality control method provided by an embodiment of the present application. DETAILED DESCRIPTION

[0018] Exemplary embodiments of the present disclosure will be described in greater detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided so that the present disclosure can be more thoroughly understood and the scope of the present disclosure can be accurately conveyed to those skilled in the art.

[0019] As Figure 1 shown, an embodiment of the present application proposes an ultrasonic welding quality control system, comprising: A collection and identification module is configured to collect sensor data of welding head displacement, welding interface temperature, welding head amplitude and welding power in real time during ultrasonic welding, and identify welding head displacement peak, interface temperature inflection point and power fluctuation critical characteristic value. A parameter extraction module is configured to extract a displacement peak characteristic value, an interface temperature inflection point characteristic value, a power fluctuation critical characteristic value, a welding head amplitude characteristic value and a welding time characteristic value in a welding process based on sensor data to form a characteristic parameter set; An analysis generation module is configured to establish a multi-parameter correlation relationship based on the characteristic parameter set, perform partition evaluation on the correlation relationship, and generate a dynamic adjustment value in combination with distribution characteristics of welding parameters in each region; A quality prediction module is configured to input the dynamic adjustment value and the characteristic parameter set into a prediction mechanism to obtain a welding quality grade; An adjustment triggering module is configured to trigger a process parameter adjustment process when a deviation between the welding quality grade and a preset grade is greater than a threshold value, and to dynamically adjust welding pressure, vibration amplitude, welding power and welding time. A control correction module is configured to feed back the adjusted process parameters to welding control in real time, dynamically correct welding energy input according to interface temperature monitoring data, automatically reduce energy input when the interface peak temperature is greater than a threshold value, and realize closed-loop quality control of the welding process through multi-parameter correlation evaluation.

[0020] In the embodiment of the present application, multi-dimensional sensor data is collected in real time and key process nodes are accurately identified to timely grasp the core dynamics in the welding process, avoid missing the final adjustment opportunity due to untimely perception of key information such as welding head displacement and welding interface temperature, reduce welding quality problems caused by improper control of key nodes, extract key real-time parameters based on sensor data and form a characteristic parameter set, convert scattered welding data into information with clear quality correlation significance, avoid inaccurate judgment of welding quality conditions due to disorganized data, establish a multi-parameter correlation relationship and perform partition evaluation to generate a dynamic adjustment value, deeply explore the internal relationship between welding parameters, develop a targeted adjustment scheme according to the actual situation of different parameter distribution regions, and avoid unstable welding quality caused by blind adjustment of parameters.

[0021] The dynamic adjustment value and the characteristic parameter set are jointly input into a prediction mechanism to obtain a welding quality grade, the welding quality result can be predicted in advance, if the quality is found to be possibly substandard, measures can be taken in time to intervene, reduce the production of unqualified products, and reduce production cost, when the welding quality grade deviates from the preset grade by more than a threshold value, the process parameter dynamic adjustment is triggered, the quality deviation problem can be quickly responded to, the welding pressure, vibration amplitude and other key parameters can be adjusted in time, the quality deviation is avoided from being further expanded, the welding quality is ensured to always approach the preset target, the adjusted process parameters are fed back to the welding control, and the energy input is corrected according to the interface temperature, a closed-loop quality control system is formed, the welding process can be monitored and the parameters can be adjusted continuously, especially when the interface peak temperature exceeds the threshold value, the energy is automatically reduced, the welding part is prevented from being damaged due to too high energy input, the stability and safety of the welding process are improved, and it is ensured that the final welding product quality meets the standard.

[0022] In a preferred embodiment of the present application, based on sensor data, the displacement peak characteristic value, the interface temperature inflection point characteristic value, the power fluctuation critical characteristic value, the welding head amplitude characteristic value and the welding time characteristic value in the welding process are extracted to form a characteristic parameter set, which can include: In the embodiment of the present application, the peak value of the welding head displacement is identified and extracted from the welding head displacement sensor data as the displacement peak characteristic value, which specifically includes: obtaining the displacement data continuously collected by the welding head displacement sensor in the whole welding process, these data are displacement values of the welding head at different time points recorded by the sensor according to a set collection frequency, for example, once per millisecond, to form an original data sequence including the corresponding relationship between time and displacement, the original data sequence is compared and analyzed point by point, starting from the first displacement data point after the welding starts, the displacement value of the current data point is compared with the displacement value of the adjacent next data point, if the displacement value of the current data point is greater than the displacement value of the next data point, the current data point is compared with the previous data point again; if the displacement value of the current data point is also greater than the displacement value of the previous data point, then the displacement value of the current data point is a potential displacement peak value. All the screened potential displacement peak values are further verified to check the distribution of the potential peak values in the whole displacement data sequence, abnormal data points caused by temporary fluctuation or interference of the sensor are excluded, for example, if the displacement change amplitude of the data points before and after a potential peak value is extremely small, and the difference between the peak value and the surrounding normal data points is too large, the potential peak value is determined to be an abnormal data point and is excluded, and finally the peak value that is most consistent with the change rule of the welding head displacement in the welding process is reserved as the displacement peak characteristic value.

[0023] Based on the time node corresponding to the displacement peak characteristic value, the synchronous data in the interface temperature sensor is located, the temperature value corresponding to the temperature change inflection point caused by material melting at the time node is extracted as the interface temperature inflection point characteristic value, specifically including: after determining the displacement peak characteristic value, first find the specific time node corresponding to the displacement peak characteristic value in the data record, which is accurate to the time unit corresponding to the sensor collection frequency, then call the time and temperature data sequence recorded by the interface temperature sensor in the whole welding process, according to the above determined time node, locate the temperature data in a certain time range before and after the time node, for example, the temperature data from 50 milliseconds before the time node to 50 milliseconds after the time node, because the temperature change inflection point in the material melting process may not exactly appear at the time node corresponding to the displacement peak, but at a certain moment near the time node. The temperature data in the located time range is analyzed point by point, the temperature change rate between adjacent two time points is calculated, the calculation method of the temperature change rate is to subtract the temperature value of the previous time point from the temperature value of the next time point, and then divide by the time interval between the two time points, which is fixed by the sensor collection frequency, such as 1 millisecond, then the time interval is 0.001 second, the temperature change inflection point is found by observing the change of the temperature change rate, when the material does not start to melt, the temperature change rate is relatively stable and at a low level; when the material starts to melt, due to the change of the internal structure of the material, the heat absorption or heat release rate changes, the temperature change rate will appear obvious mutation, which may suddenly increase, or suddenly decrease, when it is found that the temperature change rate corresponding to a certain time point compared with the temperature change rate of adjacent time points, the change amplitude exceeds the set threshold, for example, the change amplitude exceeds 50% of the temperature change rate in the previous stable stage, the temperature value corresponding to this time point is the interface temperature inflection point characteristic value, if there are multiple suspected inflection points in the located time range, combine the welding process knowledge, select the temperature value most matched with the material melting stage as the final interface temperature inflection point characteristic value.

[0024] According to the energy input stage corresponding to the interface temperature inflection point characteristic value, the power value at which the power enters the critical point of severe fluctuation from the steady state is identified from the power sensor data as the power fluctuation critical characteristic value, specifically including: according to the time point corresponding to the interface temperature inflection point characteristic value, determining the energy input stage in which the time point is located, the energy input stage refers to the time period corresponding to the process from the start of welding to the continuous input of energy to the welding area until the interface temperature reaches the inflection point, it is necessary to determine the start time and end time of this stage, that is, the time corresponding to the interface temperature inflection point, then, the time and power data sequence recorded by the power sensor during the entire welding process is called, all power data in the above energy input stage is screened out, and the power data is analyzed by time period, first find out the time period in which the power is in a steady state, the judgment standard of the steady state is that in a continuous time period, for example, 100 milliseconds, the difference between the power value at each time point and the average value of all power values in the time period is less than the set allowed fluctuation range, for example, ±5% of the average value, then the power in this time period is in a steady state, and the average power value in the steady state is calculated. From the end point of the stable state time period, continue to analyze the power data backward, observe the change of the power value, calculate the difference between the power value at each time point and the average power value in the steady state, when it is found that the difference between the power value at a certain time point and the average power value exceeds the set fluctuation critical threshold, for example, more than 15% of the average power value, and the power values of the subsequent adjacent multiple time points, such as 5 consecutive time points, are continuously deviated from the average power value in the steady state, showing a trend of severe fluctuation, that is, the power value change amplitude between adjacent time points is more than 3 times the power change amplitude between adjacent time points in the steady state, the power value corresponding to the time point at which the power value deviates from the fluctuation critical threshold for the first time is the power fluctuation critical characteristic value.

[0025] Based on the power fluctuation critical characteristic value, the critical state is determined as the key process node, the amplitude readings of the welding head amplitude sensor during the node are extracted as the welding head amplitude characteristic value, and the time from the start of welding to the node is determined as the welding time characteristic value, specifically including: after determining the power fluctuation critical characteristic value, the time point corresponding to the characteristic value is determined as the key process node, then, the time range during the key process node is determined, which is usually set to 20 milliseconds before and after the time point corresponding to the power fluctuation critical characteristic value, this time range can be adjusted appropriately according to different welding process types and requirements; call the time and amplitude data sequence recorded by the welding head amplitude sensor during the entire welding process, screen out all amplitude readings during the key process node, and statistically analyze the amplitude readings, first eliminate the obvious abnormal data, for example, the difference between a certain amplitude reading and the average value of the surrounding amplitude readings is more than 30% of the average value of the surrounding amplitude readings, then the abnormal data is determined and eliminated. The average of the remaining effective amplitude readings is calculated by adding all the effective amplitude readings and dividing by the number of effective amplitude readings. This average is the welding head amplitude characteristic value. If a more accurate reflection of the amplitude during the node is needed, the median of all the effective amplitude readings can also be calculated. The effective amplitude readings are arranged in ascending order, and the value in the middle position is taken as the median. If the number of effective readings is even, the average of the two middle values is taken as the welding head amplitude characteristic value. The choice between the average and the median can be determined according to the accuracy requirements of the welding quality discrimination. The time point at which welding starts is determined, which is the time at which the welding equipment starts and begins to input energy to the welding area. This time is recorded by the equipment control system and synchronized to the data recording of each sensor. The time point corresponding to the key process node is determined, which is the time point corresponding to the power fluctuation critical characteristic value. The welding time characteristic value is calculated by subtracting the time point at which welding starts from the time point corresponding to the key process node. The resulting time difference is the welding time characteristic value. The time unit can be converted to seconds, milliseconds, etc. according to actual needs. For example, if the welding start time is 0 milliseconds and the key process node time is 500 milliseconds, the welding time characteristic value is 500 milliseconds, i.e. 0.5 seconds. In the calculation process, it is necessary to ensure that the time reference of the two time points is consistent, and both take the equipment start time as the time zero point to avoid errors in the calculation of the time difference due to different time references. At the same time, if there is a short pause in the equipment during the period from the start of welding to the key process node, the pause time needs to be deducted from the total time difference. For example, the equipment pauses for 100 milliseconds after running for 300 milliseconds after the start of welding, and then continues to run to 500 milliseconds to reach the key process node. The actual welding time characteristic value is 500 milliseconds - 100 milliseconds = 400 milliseconds.

[0026] The displacement peak characteristic value, the interface temperature inflection point characteristic value, the power fluctuation critical characteristic value, the welding head amplitude characteristic value and the welding time characteristic value are integrated to form a characteristic parameter set for quality discrimination, specifically including: the displacement peak characteristic value, the interface temperature inflection point characteristic value, the power fluctuation critical characteristic value, the welding head amplitude characteristic value and the welding time characteristic value extracted are respectively subjected to data format uniformity processing, for example, the numerical value units of all characteristic values are converted into industry standard units, the displacement peak characteristic value unit is unified as millimeter, the interface temperature inflection point characteristic value unit is unified as degree Celsius, the power fluctuation critical characteristic value unit is unified as watt, the welding head amplitude characteristic value unit is unified as micrometer, and the welding time characteristic value unit is unified as second; data validity check is performed on each characteristic value, and it is confirmed that the numerical value of each characteristic value is within a reasonable process range, for example, according to the requirements of different welding materials and welding processes, the normal value range of each characteristic value is set, if a certain characteristic value exceeds the set normal range, the extraction process of the characteristic value needs to be rechecked, and it is checked whether there is data acquisition error or calculation error, if it is confirmed that the extraction and calculation process are correct, the abnormal characteristic value is marked in the characteristic parameter set, and the five characteristic values after format unification and validity check are integrated in a fixed order, for example, the order of the displacement peak characteristic value, the interface temperature inflection point characteristic value, the power fluctuation critical characteristic value, the welding head amplitude characteristic value and the welding time characteristic value, to form a data set containing five characteristic parameters, and the data set is the characteristic parameter set for quality discrimination, and in the integration process, identification information corresponding to each characteristic value needs to be added, including characteristic value name, unit, extraction time, etc., each characteristic parameter can be clearly identified and used in the quality discrimination process.

[0027] By point-by-point comparison and abnormal data elimination of the welding head displacement sensor data, the real peak value of the welding head displacement can be accurately captured, and the error peak value caused by sensor fluctuation or interference is avoided, the interface temperature data is positioned based on the time node corresponding to the displacement peak value, and the temperature inflection point is found by calculating the temperature change rate, the key moment of material melting can be accurately identified, the power data in the energy input stage is analyzed, the critical point of power from the stable state to the severe fluctuation is accurately identified, the characteristic value can reflect the stability change of the welding energy input, the abnormal elimination and statistical calculation of the welding head amplitude data during the key process node are performed, and the welding head amplitude characteristic value obtained can accurately reflect the vibration of the welding head in the key stage, and the size of the welding head amplitude directly affects the energy transmission and material combination effect of the welding area, and appropriate amplitude can promote the full diffusion and fusion of the material and improve the weld strength.

[0028] In a preferred embodiment of the present application, a multi-parameter correlation relationship is established based on the characteristic parameter set, the correlation relationship is evaluated in different regions, and a dynamic adjustment value is generated in combination with the distribution characteristics of the welding parameters in each region, which can include: In the embodiment of the present application, based on the set of characteristic parameters, the statistical correlation between parameters is analyzed to construct a correlation relationship for quantitatively describing the interaction relationship between multiple ultrasonic welding parameters, specifically including: collecting at least 100 complete sets of characteristic parameter data, each set containing displacement peak characteristic value, interface temperature inflection point characteristic value, power fluctuation critical characteristic value, welding head amplitude characteristic value and welding time characteristic value, sorting these data according to parameter types, for example, all displacement peak characteristic values are classified into one group, all interface temperature inflection point characteristic values are classified into another group, etc., to ensure that each group of data corresponds to the same welding process, and performing statistical correlation analysis on each two parameters, taking displacement peak characteristic value and interface temperature inflection point characteristic value as an example, first calculating the average values of all data points of the two parameters, for example, the average value of displacement peak is 2.5 mm and the average value of interface temperature inflection point is 350°C, then calculating the deviation of each data point from the corresponding average value, for example, the displacement peak in a certain group of data is 2.6 mm with a deviation of 0.1 mm, and the interface temperature inflection point is 360°C with a deviation of 10°C, then calculating the sum of the product of all data point deviations, divided by the number of data points minus one, to obtain the covariance, then calculating the square root of the average value of the square sum of all deviations of the two parameters, i.e. the standard deviation, and finally dividing the covariance by the product of the standard deviations of the two parameters to obtain the correlation coefficient of the two parameters, to determine their correlation degree, for example, a correlation coefficient of 0.8 indicates strong positive correlation.

[0029] All the correlation coefficients of two parameters are sorted into a matrix form, and the value at each position in the matrix represents the correlation coefficient of the corresponding two parameters. By analyzing the matrix, it is clear which parameters have strong correlation, the absolute value of the correlation coefficient is greater than 0.7, moderate correlation, the absolute value of the correlation coefficient is between 0.3 and 0.7, or weak correlation, the absolute value of the correlation coefficient is less than 0.3. According to these correlation degrees, a correlation relationship that can quantitatively describe the interaction relationship between multiple parameters is constructed, for example, displacement peak and interface temperature inflection point have strong positive correlation, power fluctuation critical value and welding time have moderate negative correlation, etc.

[0030] Based on the parameter correlation relationship, the entire parameter correlation space is divided into multiple evaluation regions with different process states according to the distribution and clustering characteristics of the welding parameters in the relationship, specifically including: standardizing all characteristic parameter data, for each parameter, subtracting the minimum value of the parameter from the numerical value of each data point, and then dividing by the difference between the maximum and minimum values of the parameter, so that all parameter data is converted to the range of 0-1, for example, the minimum value of a parameter is 100, the maximum value is 200, and a data point is 150, then the standardized value is (150-100) ÷ (200-100) = 0.5, the standardized multi-parameter data is grouped in a step-by-step clustering manner, 5 data points are randomly selected as initial cluster centers, the distance between each data point and the 5 centers is calculated, the distance calculation method is the square root of the sum of the squares of the difference between each parameter standardization value; each data point is assigned to the group where the nearest cluster center is located, forming 5 initial cluster groups, then the center of each cluster group is recalculated, that is, the average of each parameter standardization value in the group, then the distance between all data points and the new center is recalculated and the group is reassigned, this process is repeated until the change in the cluster center before and after is less than 0.01, that is, the center change value of each parameter is less than 0.01, the clustering is stopped, and stable cluster groups are obtained. According to the final clustering result, the entire parameter correlation space is divided into evaluation regions with the same number of cluster groups, each evaluation region corresponds to a cluster group, including all parameter combinations represented by the data points in the group, at the same time, the boundary of each evaluation region is recorded, the boundary determination method is to take the minimum and maximum values of each parameter standardization value in the region, for example, the displacement peak standardization value range of a region is 0.3-0.6, which represents the boundary of the region in the displacement peak dimension.

[0031] For each evaluation area, the centralized trend and dispersion degree of the internal welding parameter data are statistically analyzed, and the numerical distribution characteristics corresponding to the area are extracted, specifically including: for each evaluation area, the centralized trend of each welding parameter is calculated, for each parameter, the numerical values of all data points in the area are arranged in ascending order, if the number of data points is odd, the numerical value at the middle position is taken as the median; if it is even, the average of the two middle numerical values is taken as the median, at the same time, the arithmetic mean of all data points of the parameter is calculated, the sum is divided by the number of data points and the mode, the numerical value with the highest frequency, for example, the power fluctuation critical characteristic value data in a certain area is 500, 510, 500, 520, 500, the median is 500, the average is (500+510+500+520+500) ÷ 5 = 506, and the mode is 500; the dispersion degree of each parameter is calculated, the deviation of each data point from the average value of the parameter is calculated first, the sum of the squares of the deviations is calculated, and then the sum is divided by the number of data points to obtain the variance, the square root of the variance is the standard deviation, which reflects the dispersion of the data, for example, the deviation of the above-mentioned power fluctuation critical characteristic value is -6, 4, -6, 14, -6, the sum of the squares of the deviations is 36+16+36+196+36=320, the variance is 320 ÷ 5 = 64, and the standard deviation is 8, at the same time, the range of the parameter is calculated, the maximum value is subtracted from the minimum value, and the range of the above-mentioned data is 520-500=20; the calculation results of the centralized trend, the median, the average, the mode and the dispersion degree, the standard deviation and the range are integrated, and the numerical distribution characteristics corresponding to each evaluation area are extracted, for example, the distribution characteristics of a certain area can be described as the average value of the displacement peak is 2.3 mm, the standard deviation is 0.2 mm, and the distribution is concentrated; the median of the interface temperature inflection point is 345℃, the range is 30℃, and the distribution is relatively uniform, etc.

[0032] Based on the numerical distribution characteristics, the compensation values of the welding parameters in each region are calculated respectively to generate dynamic adjustment values for real-time regulation of the welding process, which specifically includes: according to the characteristic parameter set of a large number of qualified welding products, the standard interval of each parameter is determined, for example, the standard interval of the displacement peak value is 2.0-2.5 mm, the standard interval of the interface temperature inflection point is 330-360℃, etc., the determination method of the standard interval is to take the 25% quantile of the parameter data of all qualified products as the lower limit and the 75% quantile as the upper limit, for each evaluation region, the average value of each parameter in the region is compared with the middle value of the standard interval, for example, the average value of the displacement peak value in a certain region is 2.6 mm, the middle value of the standard interval is 2.25 mm, the difference between the two is 0.35 mm, according to the importance of the parameter, the weight is set, such as the weight of the displacement peak value is 0.2, the preliminary compensation value is obtained by multiplying the difference value by the weight, and then it is modified combined with the standard deviation of the parameter, if the standard deviation is large, the dispersion degree is high, then the preliminary compensation value is multiplied by a correction coefficient of 1.2; if the standard deviation is small, then it is multiplied by a correction coefficient of 0.8, for example, the standard deviation of the displacement peak value is 0.2 mm, the dispersion degree is medium, and the final compensation value is 0.35*0.2*1=0.07 mm, the compensation values of each parameter are arranged according to their order of action in the welding process to form dynamic adjustment values, for example, the welding time is adjusted first, then the power is adjusted, and finally the amplitude is adjusted, at the same time, the applicable range of each adjustment value is determined, that is, the corresponding evaluation region and the adjustment opportunity, such as the first adjustment is performed at the 50th millisecond after the start of welding, to ensure that the adjustment value can be used in real time to regulate the welding process.

[0033] By statistical analysis, the parameter correlation relationship can be constructed to clearly reveal the internal relationship between the ultrasonic welding parameters, the parameter correlation space can be divided into different evaluation regions, and different process states can be accurately distinguished, each region represents a group of welding parameter combinations with similar characteristics, and corresponds to a specific welding quality level, the concentration trend and dispersion degree of each region are analyzed to deeply understand the overall performance and fluctuation of the welding parameters in the region, through these distribution characteristics, it can be judged whether the welding process in the region is stable and whether the parameter fluctuation is within an acceptable range, to reduce the welding quality fluctuation caused by unstable parameters, the dynamic adjustment values calculated based on the numerical distribution characteristics can realize real-time and accurate regulation of the welding process.

[0034] In a preferred embodiment of the present application, the dynamic adjustment values and the characteristic parameter set are jointly input into a prediction mechanism to obtain a welding quality grade, which can include: In the embodiment of the present application, the dynamic adjustment value is fused with the feature parameter set for processing to generate a comprehensive feature vector representing the current welding process state, and the comprehensive feature vector is input into a preset welding quality classification prediction rule base. Specifically, first, all data in the dynamic adjustment value and the feature parameter set are standardized. For each parameter, the value is subtracted by the minimum value of the parameter in the data, and then divided by the difference between the maximum value and the minimum value of the parameter, so that all data are converted to the range of 0-1. For example, if the historical minimum value of a dynamic adjustment value is -5, the maximum value is 10, and the current value is 2, the standardized value is (2-(-5)) / (10-(-5))=7 / 15≈0.47. According to the influence degree of each parameter on the welding quality, a weight is assigned to each parameter. For example, the displacement peak feature value has a greater influence on the welding quality, and a weight of 0.2 is assigned. A dynamic adjustment value has a smaller influence, and a weight of 0.1 is assigned. The sum of the weights of all parameters is 1. Each standardized parameter value is multiplied by the corresponding weight to obtain a weighted value. For example, the standardized displacement peak feature value is 0.6, the weight is 0.2, and the weighted value is 0.6*0.2=0.12. The standardized dynamic adjustment value is 0.47, the weight is 0.1, and the weighted value is 0.47*0.1≈0.047. The weighted values of all parameters are arranged in a fixed order to form a comprehensive feature vector. For example, the parameters are arranged in the order of displacement peak feature value, interface temperature inflection point feature value, power fluctuation critical feature value, welding head amplitude feature value, welding time feature value, and dynamic adjustment values to obtain an ordered sequence containing all weighted values, which is the comprehensive feature vector representing the current welding process state.

[0035] The input comprehensive feature vector is matched and logically judged by the rule base to obtain an evaluation result representing the possibility of different welding quality grades. Specifically, the preset welding quality classification prediction rule base includes multiple rules, each rule consisting of a condition part and a conclusion part. The condition part is a range limitation of the elements in the comprehensive feature vector, and the conclusion part is the corresponding quality grade possibility. For example, the condition of a certain rule is that the weighted value of the displacement peak feature value is between 0.1 and 0.15, and the weighted value of the interface temperature inflection point feature value is between 0.08 and 0.12, and the conclusion is that the first quality possibility is 60% and the second quality possibility is 30%. The generated comprehensive feature vector is matched with each rule in the rule base, and the elements in the comprehensive feature vector are checked one by one to see if they meet the range limitation of the rule condition part. For example, the weighted value of the displacement peak feature value in the comprehensive feature vector is 0.12, and the weighted value of the interface temperature inflection point feature value is 0.1, which meets the condition of the above rule.

[0036] For each matched rule, a matching degree is calculated, which is the ratio of the number of elements in the comprehensive feature vector satisfying the rule conditions to the total number of rule condition elements, for example, a rule has 5 condition elements, and the comprehensive feature vector satisfies 4 of them, then the matching degree is 4 ÷ 5 = 0.8, the conclusions of each rule are weighted according to the matching degree, the higher the matching degree, the greater the weight of the corresponding rule conclusion, for example, the rule conclusion weight of the matching degree 0.8 is 0.8, and the rule conclusion weight of the matching degree 0.6 is 0.6, the quality level possibilities of all matched rules are accumulated by weight, and then divided by the total weight to obtain the evaluation result representing the possibility of different welding quality levels, for example, two matched rules, the first rule has a first-level quality possibility of 60% and a weight of 0.8, and the second rule has a first-level quality possibility of 50% and a weight of 0.6, then the comprehensive first-level quality possibility is (60% × 0.8 + 50% × 0.6) ÷ (0.8 + 0.6) = (48% + 30%) ÷ 1.4 = 78% ÷ 1.4 ≈ 55.7%.

[0037] Based on the analysis result, a quality level prediction value corresponding to the current welding state is obtained, the quality level includes virtual welding, underwelding, good welding and overwelding, and specifically includes: arranging the obtained different welding quality level possibilities in descending order, for example, the first-level quality possibility is 55.7%, the second-level quality possibility is 30%, and the third-level quality possibility is 14.3%, setting a quality level determination threshold, for example, the minimum possibility threshold of the first-level quality is 50%, and the minimum possibility threshold of the second-level quality is 30%, if the possibility of a certain quality level is higher than the corresponding threshold and is the highest among all levels, then the level is taken as a candidate level, the quality level of the similar case in the reference data to the current comprehensive feature vector is verified, the similarity between the current comprehensive feature vector and the similar case feature vector is calculated, and the calculation method is the reciprocal of the square root of the sum of squares of the difference of each corresponding element, if more than 80% of the cases with high similarity are the candidate level, then the candidate level is confirmed, after cross verification, the confirmed candidate level is taken as the quality level prediction value corresponding to the current welding state, for example, after the above process, the first-level quality is finally determined as the quality level prediction value of the current welding state.

[0038] The comprehensive feature vector is generated by standardizing and weightedly fusing the dynamic adjustment value and the feature parameter set, information in multiple dimensions can be integrated into a unified representation form, the current welding process state is comprehensively reflected, the welding quality classification prediction rule library is constructed based on experience and data, the possibility of different quality levels can be quickly and accurately obtained by matching the comprehensive feature vector with the rule library and logical judgment, the reliability of the evaluation result is improved, the final determined quality level prediction value has high accuracy and reliability, and misjudgment caused by too low possibility is avoided, the welding quality level prediction process organically combines the dynamic adjustment value and the feature parameter set, forms a complete system from data fusion to prediction judgment, the system realizes real-time and accurate prediction of the welding quality, reduces the generation of unqualified products, and reduces production cost.

[0039] In a preferred embodiment of the present application, when the welding quality level deviates from the preset level by more than a threshold value, a process parameter adjustment process is triggered, and the welding pressure, vibration amplitude, welding power and welding time are dynamically adjusted, which can include: In the embodiment of the present application, the real-time deviation value between the predicted welding quality level and the preset level is calculated, and the specific value and direction of the deviation value are determined, which specifically includes: dividing the welding quality level into more refined quantization standards, a total of 10 levels, from level 1 to level 10, each level corresponds to a specific quantization score, level 1 corresponds to 10 points, level 2 corresponds to 20 points, …, and level 10 corresponds to 100 points, and the specific indexes corresponding to each level such as weld strength and appearance flatness are also specified, for example, level 10 requires weld strength ≥ 500 MPa, and appearance without any defects; level 9 requires weld strength ≥ 480 MPa, and appearance allows 1 small scratch, etc. According to the quality level prediction value generated before, the corresponding quantization score is found, for example, the prediction result is level 7, and the corresponding quantization score is 70 points, the preset quality level and the corresponding score are determined, for example, the preset level is level 9, and the corresponding score is 90 points; the quantization score of the prediction level is subtracted from the quantization score of the preset level, and the specific value of the deviation value is obtained, for example, 70 points (prediction) - 90 points (preset) = -20 points, that is, the specific value of the deviation value is -20 points, and the absolute value is 20 points.

[0040] If the calculation result is positive, it means that the prediction level is higher than the preset level, and the deviation direction is positive; if it is negative, it means that the prediction level is lower than the preset level, and the deviation direction is negative; if it is 0, there is no deviation, and in the above example, the deviation value is -20 points, the deviation direction is negative, and it represents that the current welding quality does not meet the preset requirements.

[0041] The deviation value is compared with a preset threshold value, and when the deviation value is greater than the preset threshold value, a parameter adjustment instruction is triggered, specifically including: setting different deviation threshold values according to different welding materials and product requirements, for example, for high-strength steel welding, the threshold value is set to 15 points; for aluminum alloy welding, the threshold value is set to 10 points, and the applicable scene of the threshold value is also specified, for example, the threshold value can be appropriately relaxed when batch production, and the threshold value needs to be strictly tightened when precision component welding, first determine the threshold type corresponding to the current welding task, for example, the threshold value is 10 points when welding aluminum alloy components, then compare the calculated absolute value of the deviation value (20 points) with the threshold value (10 points), 20 points > 10 points, and it is determined that the deviation value is greater than the preset threshold value; when the triggering condition is met, the system generates a parameter adjustment instruction containing multi-dimensional information, the deviation value (-20 points), the deviation direction (negative), the current welding procedure number, such as W20230829001, the welded time, such as 30 seconds, the current welding head position coordinates, such as X100mm, Y50mm, etc.; check the integrity of the instruction information to ensure that no key parameters are missing, then verify the data accuracy, for example, check whether the current welding procedure number is consistent with the production plan, and after verification, the instruction is encrypted and packaged through a device internal communication protocol, such as Modbus protocol, and sent to the receiving port of the welding parameter adjustment mechanism, and waits for the receiving module to return an acknowledgement signal, if no acknowledgement is received within 3 seconds, the instruction is resent.

[0042] According to the parameter adjustment instruction, the specific adjustment amount of the welding pressure, the vibration amplitude, the welding power and the welding time is dynamically calculated according to the size and direction of the deviation value, specifically including: parameter adjustment weight distribution is determined according to the influence weight of each parameter on the quality based on a large amount of welding data statistical analysis, for example, through analysis of 1000 groups of aluminum alloy welding data, it is found that the quality level changes by an average of 1 level when the welding power changes by 10W, and the quality level changes by an average of 1 level when the welding time changes by 0.5 seconds, accordingly, the weight is distributed, welding power 0.3 (the largest influence), welding time 0.25, vibration amplitude 0.2, welding pressure 0.25, and the sum of the four weights is 1.

[0043] The basic adjustment amount is calculated step by step: Welding power basic adjustment amount: absolute value of deviation value (20 points) ÷ 10 points / level (points corresponding to each level) = 2 levels, multiplied by the influence coefficient of power on level (10W / level) and weight (0.3), to get 2x10x0.3=6W.

[0044] Welding time basic adjustment amount: 2 levels x 0.5 seconds / level x 0.25 = 0.25 seconds.

[0045] Vibration amplitude basic adjustment amount: 2 levels x 0.02 mm / level (empirical coefficient) x 0.2 = 0.008 mm.

[0046] Welding pressure adjustment: 2 grade × 0.02MPa / grade (empirical coefficient) × 0.25 = 0.01MPa.

[0047] If the deviation is in the negative direction (insufficient quality), all parameters need to be increased to improve quality. If the deviation is in the positive direction (excess quality), the basic adjustment amount obtained by the same calculation method is used as the reduction value.

[0048] Adjustment boundary correction rules: Welding power: The current value is 300W, the maximum allowable power of the equipment is 400W, the basic adjustment is 6W, 300+6=306W<400W, no correction is needed, the final adjustment is 6W.

[0049] Welding time: The current value is 2.0 seconds, the maximum allowable time for the process is 3.0 seconds, the basic adjustment is 0.25 seconds, 2.0 + 0.25 = 2.25 seconds < 3.0 seconds, the final adjustment is 0.25 seconds.

[0050] Vibration amplitude: The current value is 0.1mm, the maximum amplitude of the equipment is 0.15mm, the foundation adjustment is 0.008mm, 0.1 + 0.008 = 0.108mm < 0.15mm, and the final adjustment is 0.008mm.

[0051] Welding pressure: The current value is 0.4MPa, the maximum allowable pressure is 0.5MPa, the basic adjustment is 0.01MPa, 0.4 + 0.01 = 0.41MPa < 0.5MPa, and the final adjustment is 0.01MPa.

[0052] If a parameter exceeds the allowable range after adjustment, the adjustment amount will be corrected to the maximum allowable increment. For example, if the current welding power is 398W and the basic adjustment amount is 6W, it will be corrected to 2W (398+2=400W).

[0053] The specific adjustment amount is applied to the current process parameter, and the real-time update and adjustment of the welding parameter are completed, which specifically includes: current process parameter accurate collection, through the high-precision sensor built-in the equipment, the accurate value of the current parameters is collected in real time, the sampling frequency is 10 times per second, for example, the welding pressure is collected for 3 times in succession as 0.401 MPa, 0.399 MPa and 0.400 MPa, and the average value 0.400 MPa is taken as the current value; the welding power continuous collection value is 300 W, 300 W and 300 W, the current value is 300 W, the target parameter value is accurately calculated, the final adjustment amount of each parameter is added to the current value to obtain the target parameter value, the welding pressure target value is 0.400 MPa+0.01 MPa=0.410 MPa; the vibration amplitude target value is 0.100 mm+0.008 mm=0.108 mm; the welding power target value is 300 W+6 W=306 W; the welding time target value is 2.0 seconds+0.25 seconds=2.25 seconds; the parameter adjustment is executed in stages, the first stage (0-2 seconds), the welding power is smoothly increased from 300 W to 303 W, the vibration amplitude is increased from 0.100 mm to 0.104 mm, and the linear adjustment mode is adopted, with a change of 3 W and 0.004 mm per second; the second stage (2-4 seconds), the welding power is continuously increased to 306 W, and the vibration amplitude is increased to 0.108 mm, and at the same time, the welding pressure is adjusted from 0.400 MPa to 0.410 MPa, with a change of 3 W, 0.004 mm and 0.005 MPa per second; the third stage (after 4 seconds), the power, amplitude and pressure are kept stable at the target value, and at the same time, the welding time timer is reset to 0 and counted for 2.25 seconds; the multi-level feedback confirmation mechanism, the primary feedback, the adjustment execution mechanism returns the actual parameter value after the end of each stage, for example, the feedback power 303 W and the amplitude 0.104 mm after the end of the first stage, the error is within ±0.5%, and it is determined to be qualified; the secondary feedback, the parameter value is collected again within 5 seconds after the adjustment is completed, to confirm the welding power 306 W (error 0), the amplitude 0.108 mm (error 0), the pressure 0.410 MPa (error 0) and the time counted for 2.25 seconds, all parameters reach the target value and are stable, the adjustment result is compared with the adjustment instruction, and after confirmation, the adjustment completion report is generated, including the parameter comparison before and after adjustment, adjustment time and other information, which is fed back to the control system.

[0054] By the determined level quantization standard and deviation calculation method, the difference between the predicted quality level and the preset level can be accurately quantified, the degree and direction of the quality deviation can be clearly reflected, the adjustment blindness caused by the fuzzy quality evaluation is avoided, the preset deviation threshold is compared and triggered for adjustment, the frequent adjustment of parameters caused by the slight deviation is avoided, the unnecessary intervention to the stability of the welding process is reduced, the adjustment amount is calculated based on the parameter weight, the adjustment intensity can be distributed according to the influence degree of each parameter on the quality, the important parameters are ensured to be adjusted enough to improve the quality, the secondary parameters are appropriately adjusted to avoid excessive intervention, the adjustment amount is corrected in combination with the deviation direction and the actual parameter range, so that the adjustment amount meets the quality improvement requirement and does not exceed the safe operation range of the equipment, and the safety and stability of the welding process are ensured.

[0055] In a preferred embodiment of the present application, the adjusted process parameters are fed back to the welding control in real time, the welding energy input is dynamically corrected according to the interface temperature monitoring data, and the energy input is automatically reduced when the interface peak temperature is greater than the threshold value. Through multi-parameter correlation evaluation, closed-loop quality control of the welding process can be realized, which can include: In the embodiment of the present application, the adjusted welding pressure, amplitude and welding time are fed back to the welding controller in real time, and an infrared temperature measurement sensor is used to monitor the temperature change at the welding interface, and the interface peak temperature is recorded. Specifically, the dynamically adjusted welding pressure, amplitude and welding time parameters are recorded in a standardized manner, accurate to three decimal places, for example, welding pressure 0.410 MPa, amplitude 0.108 mm, welding time 2.250 seconds, and the parameter effective condition is marked, such as applicable to aluminum alloy workpieces with a thickness of 2 mm and welding current of 30 A, and the parameter adjustment basis is attached, such as based on a quality level negative deviation of 20 points; The hierarchical parameter feedback mechanism sends parameters to the welding controller in a three-level transmission mode, the first level sends parameter identification and value, such as welding pressure: 0.410 MPa; The second level sends parameter accuracy requirements, such as pressure control error ≤±0.005 MPa; The third level sends parameter effective time, such as 0.5 seconds after receiving the confirmation signal; The controller needs to return confirmation information with a time stamp after receiving each level, such as 2023-08-29 10:15:30.250 pressure parameter received successfully, to ensure that the parameter transmission is complete and traceable.

[0056] Infrared temperature measurement system calibration and deployment, before welding, three-point calibration is carried out on the infrared temperature measurement sensor, and 200℃, 350℃ and 500℃ standard heat sources are used for verification respectively, to ensure that the measurement error is ≤±2℃, the sensor is fixed on an adjustable support, the distance from the welding interface is in the range of 45-55mm, the detection point is located in the center area of the weld through laser positioning confirmation, the angle between the detection angle and the normal line of the interface is ≤5°, to avoid measurement error caused by angle deviation, the sampling frequency is set to 20 times / s, that is, data is collected every 0.05 seconds to ensure the capture of instantaneous temperature change; Real-time processing and peak recording of temperature data, establishment of temperature data buffer area, real-time storage of the last 100 sampling points, 5 seconds of data, each time new data enters, compared with all data in the buffer area, if the current temperature, such as 380.2℃, is higher than all data, update the peak value record, and record the accurate time when the peak value appears, such as 10:15:32.450, and the corresponding welding stage, such as the 3rd second of welding energy input, the peak value data is backed up every 0.1 seconds to prevent data loss, and after welding, the temperature and time trend is automatically generated, and the peak value position and value are marked.

[0057] The interface peak temperature is compared with the preset temperature threshold value, and when the interface peak temperature is greater than the preset threshold value, it is determined that there is an overheating risk, and an energy down-regulation instruction is generated, which specifically includes: dynamic temperature threshold value system establishment, a three-dimensional threshold matrix is established according to the material type, thickness and welding environment temperature, for example, for aluminum alloy material, the threshold value is 340℃ when the thickness is 1mm, 350℃ when the thickness is 2mm, and 360℃ when the thickness is 3mm; When the environment temperature is higher than 25℃, the threshold value decreases by 5℃ for every increase of 5℃, for example, for the current welding of 2mm aluminum alloy and the environment temperature of 28℃, the threshold value is calculated as 350℃- [(28-25) / 5]×5℃=347℃, and the threshold value is also marked with the applicable temperature duration, such as the allowable temperature duration above 347℃≤0.5 seconds.

[0058] Multi-dimensional temperature comparison analysis, the recorded interface peak temperature (380.2℃) is compared with the dynamic threshold value (347℃) in multiple dimensions, the temperature difference is calculated, 380.2℃-347℃=33.2℃; The over-temperature duration is calculated, the time interval from 348℃ to 380.2℃ and then to 347℃ is 1.2 seconds, which exceeds the allowed 0.5 seconds; The temperature rise rate is calculated, which is (380.2-250)℃ / 3 seconds≈43.4℃ / second within 3 seconds, which exceeds the safe rate of 30℃ / second, and the three indicators are combined to determine that there is a serious overheating risk.

[0059] The energy down-regulation instruction is structured, and the instruction includes seven core elements, i.e., an overheating level (severe), a temperature exceeding value (33.2 °C), an over-temperature duration (1.2 seconds), a recommended down-regulation energy type (priority welding power, auxiliary reduction of amplitude), a basic down-regulation proportion (20%), an adjustment execution deadline (completed within 1 second), and a requirement for associated parameter coordinated adjustment (synchronous adjustment of time and pressure after power down-regulation). The instruction is transmitted in an encrypted format, contains a check code to prevent tampering, and has dynamic management of instruction priority, establishing a five-level priority system, with the energy down-regulation instruction being the highest level (level five) by default. When there are other instructions in the system, such as a pressure fine-tuning instruction at level four, the low-priority instruction is automatically suspended, and the energy down-regulation is executed with priority. During the execution process, the progress is pushed to the monitoring system in real time, such as 0.3 seconds, power starts to down-regulate; 0.7 seconds, power adjustment is completed; and 1.0 second, all parameters are adjusted to the right position.

[0060] According to the energy down-regulation instruction, the welding energy input is reduced by a preset proportion, and based on the change of the welding energy input, a multi-parameter associated evaluation mechanism is triggered synchronously to recalculate and coordinately adjust the welding pressure, amplitude and welding time, specifically including: hierarchical energy down-regulation proportion calculation, using stepwise down-regulation calculation according to the overheating level, with a basic proportion of 20% for severe overheating (over-temperature > 30 °C), and an increase of 0.5% for each excess of 1 °C, so the actual down-regulation proportion is 20% + (33.2-30) x 0.5% = 20% + 1.6% = 21.6%, the current welding power is 306 W, the down-regulation value is calculated as 306 W x 21.6% ≈ 66.1 W, which is accurate to 1 W as 66 W, and the adjusted power is 306 W - 66 W = 240 W, and the amplitude down-regulation value is calculated as 0.108 mm x 8% ≈ 0.0086 mm, and the adjusted value is 0.108 - 0.0086 = 0.0994 mm ≈ 0.099 mm.

[0061] Multi-parameter coordinated adjustment coefficient determination: based on 1000 groups of data, a coordinated adjustment coefficient table is established, when the power is down-regulated by 21.6%, the corresponding time extension coefficient is 1.15, i.e., an extension of 15%, and the pressure adjustment coefficient is 0.92, i.e., a reduction of 8%, the original welding time is 2.250 seconds, the extension value is 2.250 x 15% = 0.3375 seconds, the adjusted value is 2.250 + 0.3375 = 2.5875 seconds ≈ 2.588 seconds, the original welding pressure is 0.410 MPa, the reduction value is 0.410 x 8% = 0.0328 MPa, and the adjusted value is 0.410 - 0.0328 = 0.3772 MPa ≈ 0.377 MPa.

[0062] The associated parameter constraint verification checks whether the adjusted parameters meet the process constraint conditions. The power of 240W needs to be within the device allowed range of 150-500W (compliant); the time of 2.588 seconds needs to be ≤ the maximum allowed time of 3 seconds (compliant); the pressure of 0.377 MPa needs to be ≥ the minimum pressure of 0.3 MPa (compliant); the amplitude of 0.099 mm needs to be within the range of 0.05-0.2 mm (compliant); and the matching between parameters is verified, such as the product of power 240W and time 2.588 seconds (energy index) needs to be ≥ the basic energy value (306W×2.250 seconds=688.5W·s), and the calculation of 240×2.588≈621.1W·s<688.5 needs further adjustment.

[0063] The parameter secondary correction calculation needs to extend the time while keeping the power unchanged due to insufficient energy index. The required supplementary time is calculated as (688.5-621.1)÷240≈0.281 seconds, and the final time is adjusted to 2.588+0.281=2.869 seconds. After verification, 240×2.869≈688.6W·s meets the energy requirement. The matching between pressure and amplitude is also checked, and the ratio of 0.377 MPa to 0.099 mm is within the final ratio range (3.8-4.2) (377÷99≈3.81, compliant), confirming the effectiveness of the coordinated parameters.

[0064] The adjusted process parameters are input into the welding controller again to perform the welding operation, realizing closed-loop quality control of the welding process, which specifically includes: multi-dimensional verification of adjusted parameters, three core verifications, device compatibility verification, confirming that the welding controller supports 240W power output, and verification passes; process feasibility verification, whether the weld forming can be completed within 2.869 seconds; safety compliance verification, all parameters are within the safe operating range of the device, verification passes, generating a verification report containing the verification basis and deviation range of each parameter; parameter step loading mechanism, adopting a three-step loading process of preloading, buffering, and taking effect. In the preloading stage, 0-0.2 seconds, the parameters are sent to the controller cache; in the buffering stage, 0.2-0.5 seconds, the controller performs internal parameter conversion; in the taking effect stage, after 0.5 seconds, the parameters are formally applied to the welding process. The controller returns a status code at the end of each stage, such as 0x0001 indicating preloading completion, ensuring smooth transition of parameter loading.

[0065] Real-time monitoring and dynamic fine-tuning, after the parameters take effect, the infrared sensor continues to monitor the temperature at a frequency of 20 times per second, and calculates the average temperature every 0.1 second. The temperature decreases from 380.2℃ to 360.5℃ within the first 0.5 seconds, to 350.3℃ in the first second, and to 345.2℃ in the first 1.5 seconds. The threshold value is 347℃, and the fine-tuning mechanism is triggered when the temperature is lower than the threshold value. The parameters are adjusted by 5% proportion, and the power increases from 240W to 252W. The time decreases from 2.869 seconds to 2.726 seconds, and the temperature is stabilized at about 345℃. The closed-loop control effect is confirmed, and the control effect is confirmed from five dimensions after the welding is completed. The temperature index is that the highest temperature is 380.2℃, and the adjusted temperature is stabilized at 345℃±2℃. The parameter stability is that the parameter fluctuation amplitude is ≤±2% after adjustment. The energy compliance rate is that the actual input energy and the theoretical value deviation is ≤3%. The weld appearance is free of defects such as burning loss and cracks. The strength test shows that the weld strength reaches 520MPa, which is higher than the standard 500MPa. A closed-loop control report is generated, including the complete parameter adjustment trajectory and the temperature change trend, which serves as the basis for subsequent process optimization.

[0066] The adjusted parameters are fed back to the controller in real time, ensuring that the welding process is carried out according to the optimized parameters, avoiding welding quality fluctuations caused by parameter transmission delay or errors. The high-frequency sampling and peak recording of the infrared temperature sensor can accurately capture the temperature changes at the welding interface, timely detect potential overheating problems, set different temperature thresholds for different materials and workpiece specifications, improve the accuracy and relevance of overheating judgment, and avoid misjudgment caused by using a unified threshold. Clear overheating degree calculation and instruction content enable energy adjustment to have a clear basis and target, ensuring the rationality of the adjustment. The energy reduction proportion is determined according to the overheating degree, so that the adjustment amount matches the overheating risk, avoiding under-adjustment or over-adjustment. The coordinated adjustment based on the parameter correlation ensures the coordination between parameters, such as appropriately extending the welding time when the power is reduced, which reduces the overheating risk and ensures sufficient energy input to ensure welding strength. This multi-parameter coordinated adjustment avoids the problem of mismatching other parameters caused by single-parameter adjustment, improving the stability of the welding process and the effectiveness of parameter adjustment. The verification link of the adjusted parameters ensures that the parameters are within the allowable range of the equipment, avoiding equipment failures or safety accidents caused by parameter over-limiting.

[0067] As shown in Figure 2 The embodiment of the present application also provides an ultrasonic welding quality control method, which comprises: Step 1, real-time acquisition of sensor data of welding head displacement, welding interface temperature, welding head amplitude and welding power during ultrasonic welding, and identification of welding head displacement peak, interface temperature inflection point and power fluctuation critical characteristic value; Step 2, based on sensor data, extracting displacement peak characteristic values, interface temperature inflection point characteristic values, power fluctuation critical characteristic values, welding head amplitude characteristic values and welding time characteristic values in the welding process to form a characteristic parameter set; Step 3, based on the characteristic parameter set, establishing a multi-parameter correlation relationship, partitioning and evaluating the correlation relationship, and combining the distribution characteristics of the welding parameters in each region to generate a dynamic adjustment value; Step 4, inputting the dynamic adjustment value and the characteristic parameter set into the prediction mechanism to obtain the welding quality grade; Step 5, when the welding quality grade deviates from the preset grade by more than a threshold value, triggering a process parameter adjustment process to dynamically adjust the welding pressure, vibration amplitude, welding power and welding time; Step 6, feeding back the adjusted process parameters to the welding control in real time, dynamically correcting the welding energy input according to the interface temperature monitoring data, and automatically reducing the energy input when the interface peak temperature is greater than a threshold value, thereby realizing closed-loop quality control of the welding process through multi-parameter correlation evaluation.

[0068] It should be noted that the method is a method corresponding to the above-mentioned system, and all the implementation manners in the above-mentioned system embodiment are applicable to this embodiment and can achieve the same technical effects.

[0069] Embodiments of the application also provide a computing device, comprising a processor and a memory storing a computer program, wherein the computer program is executed by the processor to perform the method as described above. All the implementation manners in the above-mentioned method embodiment are applicable to this embodiment and can achieve the same technical effects.

[0070] Embodiments of the application also provide a computer readable storage medium storing instructions, wherein the instructions are executed on a computer to make the computer perform the method as described above. All the implementation manners in the above-mentioned method embodiment are applicable to this embodiment and can achieve the same technical effects.

[0071] The above is the preferred embodiment of the application, and it should be noted that for those skilled in the art, without departing from the principles of the application, a number of improvements and refinements can be made, and these improvements and refinements should also be considered as the protection scope of the application.

Claims

1. An ultrasonic welding quality control system, characterized by, The method comprises the following steps: Collecting and identifying module, for real-time collection of ultrasonic welding process sensor data of welding head displacement, welding interface temperature, welding head amplitude and welding power, and identification of welding head displacement peak value, interface temperature inflection point and power fluctuation critical characteristic value; Parameter extraction module, for extracting displacement peak value characteristic value, interface temperature inflection point characteristic value, power fluctuation critical characteristic value, welding head amplitude characteristic value and welding time characteristic value in the welding process based on the sensor data, and forming a characteristic parameter set; Analysis and generation module, for establishing a multi-parameter correlation relationship based on the characteristic parameter set, evaluating the correlation relationship, and generating a dynamic adjustment value combined with the distribution characteristics of the welding parameters in each region; Quality prediction module, for inputting the dynamic adjustment value and the characteristic parameter set into the prediction mechanism to obtain the welding quality grade; Adjustment trigger module, for triggering the process parameter adjustment process when the welding quality grade deviates from the preset grade by more than a threshold value, and dynamically adjusting the welding pressure, vibration amplitude, welding power and welding time; Control correction module, for real-time feedback of the adjusted process parameters to the welding control, dynamic correction of the welding energy input according to the interface temperature monitoring data, automatic reduction of the energy input when the interface peak temperature is greater than a threshold value, and realization of closed-loop quality control of the welding process through multi-parameter correlation evaluation.

2. The ultrasonic welding quality control system of claim 1, wherein, Based on the sensor data, the displacement peak value characteristic value, the interface temperature inflection point characteristic value, the power fluctuation critical characteristic value, the welding head amplitude characteristic value and the welding time characteristic value in the welding process are extracted, Forming a characteristic parameter set, including: Identifying and extracting the peak value of the welding head displacement from the welding head displacement sensor data as the displacement peak value characteristic value; Based on the time node corresponding to the displacement peak value characteristic value, positioning the synchronous data in the interface temperature sensor, extracting the temperature value corresponding to the temperature change inflection point caused by material melting at the time node as the interface temperature inflection point characteristic value; According to the energy input stage corresponding to the interface temperature inflection point characteristic value, identifying the power value at which the power enters the critical point of severe fluctuation from the stable state from the power sensor data as the power fluctuation critical characteristic value; Based on the power fluctuation critical characteristic value, determining the critical state as a key process node, extracting the amplitude reading of the welding head amplitude sensor during the node as the welding head amplitude characteristic value, and extracting the time from the start of welding to the node as the welding time characteristic value; Integrating the displacement peak value characteristic value, the interface temperature inflection point characteristic value, the power fluctuation critical characteristic value, the welding head amplitude characteristic value and the welding time characteristic value to form a characteristic parameter set for quality discrimination.

3. The ultrasonic welding quality control system of claim 2, wherein, Based on the characteristic parameter set, a multi-parameter correlation relationship is established, the correlation relationship is evaluated, and a dynamic adjustment value is generated combined with the distribution characteristics of the welding parameters in each region, including: Based on the characteristic parameter set, the statistical correlation between the parameters is analyzed to construct a correlation relationship for quantitatively describing the interaction relationship between multiple ultrasonic welding parameters; Based on the parameter correlation relationship, the entire parameter correlation space is divided into multiple evaluation regions with different process states according to the distribution and clustering characteristics of the welding parameters in the relationship; For each evaluation area, statistically analyze the centralized trend and dispersion degree of the internal welding parameter data, and extract the numerical distribution characteristics corresponding to the area; Based on the numerical distribution characteristics, calculate the compensation value of the welding parameters in each area respectively, and generate dynamic adjustment values for real-time control of the welding process.

4. The ultrasonic welding quality control system of claim 3, wherein, The dynamic adjustment values and feature parameter sets are collectively input into the prediction mechanism to obtain the welding quality level, including: Fuse the dynamic adjustment values and the feature parameter sets to generate a comprehensive feature vector representing the current welding process state, and input the comprehensive feature vector into a pre-set welding quality classification prediction rule library; Match and logically judge the input comprehensive feature vector through the rule library to obtain an evaluation result representing the possibility of different welding quality levels; Based on the analysis result, the quality level prediction value corresponding to the current welding state is obtained.

5. The ultrasonic welding quality control system of claim 4, wherein, The quality level includes virtual welding, underwelding, good welding, and overwelding.

6. The ultrasonic welding quality control system of claim 5, wherein, When the welding quality level deviates from the pre-set level by more than a threshold value, trigger the process parameter adjustment process to dynamically adjust the welding pressure, vibration amplitude, welding power, and welding time, including: Calculate the real-time deviation value between the predicted welding quality level and the pre-set level, and determine the specific value and direction of the deviation value; Compare the deviation value with the pre-set threshold value, and when the deviation value is greater than the pre-set threshold value, trigger the parameter adjustment instruction; According to the parameter adjustment instruction, dynamically calculate the specific adjustment amount of the welding pressure, vibration amplitude, welding power, and welding time based on the size and direction of the deviation value; Apply the specific adjustment amount to the current process parameters to complete the real-time update and adjustment of the welding parameters.

7. The ultrasonic welding quality control system of claim 6, wherein, The adjusted process parameters are fed back to the welding control in real time, and the welding energy input is dynamically corrected according to the interface temperature monitoring data. When the interface peak temperature is greater than a threshold value, the energy input is automatically reduced to achieve closed-loop quality control of the welding process through multi-parameter correlation evaluation, including: The adjusted welding pressure, amplitude, and welding time are fed back to the welding controller in real time, and an infrared temperature sensor is used to monitor the temperature change at the welding interface and record the interface peak temperature; Compare the interface peak temperature with the pre-set temperature threshold value. When the interface peak temperature is greater than the pre-set threshold value, it is determined that there is an overheating risk, and an energy down adjustment instruction is generated; According to the energy down adjustment instruction, reduce the welding energy input by a pre-set proportion, and based on the change of the welding energy input, trigger the multi-parameter correlation evaluation mechanism synchronously to recalculate and cooperatively adjust the welding pressure, amplitude, and welding time; The adjusted process parameters are input into the welding controller again to perform welding operations, achieving closed-loop quality control of the welding process.

8. An ultrasonic welding quality control method, the method executing the system as claimed in any one of claims 1 to 7, characterized by, Including: Step 1, real-time acquisition of sensor data of ultrasonic welding head displacement, welding interface temperature, welding head amplitude, and welding power, and identification of welding head displacement peak, interface temperature inflection point, and power fluctuation critical characteristic value; Step 2, based on the sensor data, extract the displacement peak characteristic value, interface temperature inflection point characteristic value, power fluctuation critical characteristic value, welding head amplitude characteristic value, and welding time characteristic value in the welding process to form a feature parameter set; Step 3, based on the feature parameter set, a multi-parameter correlation relationship is established, the correlation relationship is evaluated in a partition manner, and a dynamic adjustment value is generated in combination with the distribution characteristics of the welding parameters in each region; Step 4, the dynamic adjustment value and the feature parameter set are jointly input into a prediction mechanism to obtain a welding quality grade; Step 5, when the welding quality grade deviates from a preset grade by more than a threshold value, a process parameter adjustment process is triggered, and welding pressure, vibration amplitude, welding power and welding time are dynamically adjusted; Step 6, the adjusted process parameters are fed back to welding control in real time, welding energy input is dynamically corrected according to interface temperature monitoring data, when the interface peak temperature is greater than a threshold value, the energy input is automatically reduced, and closed-loop quality control of the welding process is realized through multi-parameter correlation evaluation.

9. A computing device, comprising: Comprise: One or more processors; A storage device for storing one or more programs, when the one or more programs are executed by the one or more processors, so that the one or more processors execute the system as claimed in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer readable storage medium stores a program, which is executed by the processor to execute the system as claimed in any one of claims 1 to 7.

Images