Intelligent regulation and control method and system for resistance welding process parameters of junction box

Abstract

The invention relates to the technical field of intelligent monitoring, regulation and control of welding process parameters, and particularly discloses an intelligent regulation and control method and system for resistance welding process parameters of a junction box. Abnormal pressure applying links in the welding process are recognized, and the low-risk pressure applying process is extracted for welding; after welding is completed, the strength and size indexes of welding spots are evaluated through welding spot quality detection, the welding spot quality deviation coefficient is calculated, and whether the welding quality reaches the standard or not is judged; and for the welding spots with unqualified welding quality, the system calculates the adjustment direction and range of the pressure parameters through an optimization algorithm, test welding is conducted according to the optimized parameters, and the improvement effect of the pressure applying process is verified in real time till the welding spot quality meets the preset standard.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent monitoring and control of welding process parameters, and in particular to a method and system for intelligent control of resistance welding process parameters for a junction box. Background Art

[0002] Resistance welding, as an important connection process in the field of industrial manufacturing, is widely used in the assembly of high-precision products such as junction boxes, electronic equipment, and automotive parts. During the welding process, pressure application is one of the key factors affecting welding quality, which directly determines the strength, size, and consistency of the weld. In addition, the wear of the electrode will also have an important impact on the stability of pressure transmission. In order to improve welding quality, modern welding processes have gradually introduced sensors, data analysis, and intelligent control technologies to optimize the welding process by real-time monitoring and analysis of welding parameters. However, due to the multivariate coupling effect in the welding process, how to accurately diagnose abnormalities in the pressure application process and quickly adjust parameters is still a technical problem that needs to be solved in the industrial field.

[0003] The prior art has the following deficiencies: At present, the resistance welding process relies on fixed pressure parameter settings and lacks the ability to dynamically respond to real-time changes in the welding process. Although some technologies use sensors to monitor welding pressure and electrode status, they are limited to data collection and preliminary analysis, and cannot perform closed-loop optimization of welding parameters. In addition, the existing methods lack in-depth analysis of the frequency and amplitude of pressure fluctuations, resulting in delayed identification of welding anomalies. At the same time, in order to address the problem of unstable pressure transmission caused by electrode wear, existing technologies mostly rely on manual experience or regular replacement of electrodes, which not only makes it difficult to ensure welding quality, but also increases manual intervention and testing costs. Therefore, the existing technologies have significant deficiencies in the efficiency, accuracy and intelligence level of welding parameter optimization. Summary of the invention

[0004] The purpose of the present invention is to provide a method and system for intelligently controlling the process parameters of resistance welding for a junction box, by real-time acquisition and analysis of welding pressure data, constructing a pressure anomaly coefficient model and an electrode wear risk prediction model, and accurately identifying abnormal pressure application links in the welding process. At the same time, by calculating the adjustment direction and range of the pressure parameters through an optimization algorithm, the system realizes the automatic optimization and verification of welding parameters until the strength and size of the weld meet the preset standards. This method not only effectively solves the problem of unstable welding quality caused by pressure fluctuations and electrode wear, but also significantly improves the efficiency and accuracy of welding parameter optimization, reduces manual intervention and test costs, and solves the problems in the above background.

[0005] The purpose of the present invention can be achieved through the following technical solutions: The method for intelligently controlling the process parameters of resistance welding for a junction box comprises the following steps: During the resistance welding process of the junction box, a pressure sensor is installed in the area where the electrode contacts the welding workpiece to monitor the pressure applied to the welding part in real time. The pressure data of the welding process is collected in real time through the pressure sensor, and the collected pressure data is transmitted to the data processing unit for storage and preliminary analysis; Process the real-time collected pressure data, analyze the pressure fluctuation trend, including the frequency and amplitude of pressure fluctuation, and calculate the pressure anomaly coefficient based on the comprehensive analysis results of the frequency and amplitude of pressure fluctuation to identify the pressure fluctuation during welding; According to the evaluation results, the normal pressure application process is extracted for resistance welding, and the electrode wear data is collected through the visual sensor. The electrode wear data is analyzed to determine the risk of unstable pressure transmission caused by electrode wear and evaluate the risk level of the normal pressure application process; According to the evaluation results, the low-risk pressure application process is extracted to weld the resistor. After the welding is completed, the quality of the weld is tested to evaluate whether the strength and size indicators of the weld meet the preset standards and determine whether the welding quality meets the requirements; Based on the quality inspection results, the welds with unqualified welding quality are extracted. The system will identify the relevant welding pressure application process and adjust the pressure parameters until the quality deviation coefficient of the weld is less than the preset threshold.

[0006] As a further solution of the present invention: the process of obtaining the pressure anomaly coefficient is: Obtain real-time pressure data during the resistance welding process, analyze the frequency of pressure fluctuations, and calculate the pressure frequency anomaly index based on the analysis results; Analyze the amplitude of pressure fluctuations, and calculate the pressure amplitude abnormality index based on the analysis results; The pressure frequency anomaly index and the pressure amplitude anomaly index are comprehensively analyzed, the pressure frequency anomaly index and the pressure amplitude anomaly index are normalized and calculated, and the pressure anomaly coefficient is calculated.

[0007] As a further solution of the present invention: the process of obtaining the pressure frequency anomaly index is: During the resistance welding process of the junction box, a pressure sensor is installed in the area where the electrode contacts the welding workpiece to collect pressure data of the welding part in real time and obtain pressure data. and transmit the pressure data to a data processing unit; Pressure data Perform discrete wavelet transform and decompose into multiple layers of components, including high-frequency components and low-frequency components; The calculation expression of the discrete wavelet transform is: ; in, Indicates time; Extract detail coefficients of high frequency components , by calculating the energy of each component , the fluctuation intensity of the pressure signal at different frequency levels is obtained, and the calculation expression is: ; According to the energy distribution of high-frequency components, the pressure frequency anomaly index is calculated, and the calculation expression is: ; in, Indicates the abnormal frequency range, Represents the total frequency range, Indicates the pressure frequency anomaly index, represents the level of wavelet decomposition, Indicates The energy of layer decomposition, represents the detail coefficient, represents the total number of detail coefficients, Indicates Layer decomposition The detail coefficient, represents the wavelet basis function.

[0008] As a further solution of the present invention: the process of obtaining the pressure amplitude abnormality index is: During the resistance welding process of the junction box, a pressure sensor is installed in the area where the electrode contacts the welding workpiece to collect pressure data of the welding part in real time. and transmit the pressure data to a data processing unit for storage; in, Indicates time; The collected pressure data Perform fast Fourier transform to convert pressure data from time domain to frequency domain and obtain frequency components and amplitude components , Among them, the calculation expression of the amplitude component is: ; in, represents the real part of the spectrum, represents the imaginary part of the spectrum, Indicates frequency; According to the spectrum amplitude component Calculate the total energy in the normal frequency range and the energy in the abnormal frequency range. The calculation expression is: ; In the formula, Indicates frequency range The energy within, Indicates the frequency range; According to the energy of the normal frequency range and the abnormal frequency range, the pressure amplitude abnormal index is calculated, and the calculation expression is: ; In the formula, Indicates the pressure amplitude abnormality index, Indicates the abnormal frequency range, Represents the total frequency range, Indicates the energy in the abnormal frequency range, Represents the energy of the total frequency range.

[0009] As a further solution of the present invention: the risk of unstable pressure transmission caused by electrode wear and the risk level of the normal pressure application process are evaluated, specifically including: Extract the normal pressure application process for resistance welding, obtain electrode wear data, process the electrode wear data, and extract abnormal features of the electrode wear data; Build risk prediction models based on historical data; The abnormal characteristics of electrode wear data are input into the risk prediction model, and the risk index is output according to the prediction model; Determine whether the risk index is greater than or equal to the preset threshold. If so, it is recorded as high risk, if not, it is recorded as low risk.

[0010] As a further solution of the present invention: the process of acquiring the abnormal characteristics of the electrode wear data is as follows: Collect real-time data of electrode wear through visual sensors , remove noise through bandpass filtering, and interpolate the missing values ​​in the collected data to obtain complete electrode wear time series data; in, Indicates time; The electrode wear data after preprocessing Decompose and extract the trend component , periodic component and random noise components ; The decomposed calculation expression is: ; Based on the decomposition results, a multidimensional feature set is extracted, including: Statistical features: calculate mean, standard deviation, skewness, and kurtosis; Frequency domain features: Extract spectrum density and frequency center through Fourier transform; Deep learning features: Convolutional neural networks are used to extract high-order features, and the dimensions are reduced through autoencoders to generate compressed feature vectors; Classify and label the multidimensional feature set based on historical data, and label the feature subsets corresponding to normal and abnormal states; The multi-dimensional feature set collected in real time is input into the trained risk prediction model to identify abnormal features of electrode wear data and use it to predict the electrode wear risk index.

[0011] As a further solution of the present invention: the abnormal characteristics of the electrode wear data are input into the risk prediction model, and the risk index is output according to the prediction model, which specifically includes: Construct a feature-risk label dataset based on historical welding data; The risk prediction model is selected to optimize the model parameters using the gradient descent method, and the model is trained by minimizing the cross entropy loss function; Input the multi-dimensional feature set collected in real time into the trained risk prediction model; Calculate the risk index according to the risk index calculation expression; The risk index calculation expression is: ; in, represents the activation function, represents the weights learned by the model, Indicates Input feature values, is the bias term, represents the total number of input features, represents the input features, Represents the risk index.

[0012] As a further solution of the present invention: the extraction of the low-risk pressure application process, welding the resistor, after the welding is completed, the quality inspection of the weld point is carried out to evaluate whether the strength and size indicators of the weld point meet the preset standards, and judge whether the welding quality meets the requirements, specifically including: Based on the low-risk pressure application process, the strength and size indicators of the solder joints are obtained, the strength and size indicators of the solder joints are compared with the preset standards, and the quality deviation coefficient is calculated; Determine whether the quality deviation coefficient is greater than or equal to a preset threshold value, if so, the welding quality is unqualified, if not, the welding quality is qualified; Wherein, the process of obtaining the mass deviation coefficient is: After the resistance welding of the junction box is completed, obtain the solder joint inspection data, including the strength index of the solder joint and size indicators ; Set preset standards for solder joint inspection, including target strength and target size , calculate the strength deviation of each solder joint and size deviation , the calculation expression is: ; ; The intensity deviation and size deviation Add them together to get the total deviation of the solder joint. The calculation expression is: ; An autoregressive model is established based on the total deviation value of the solder joint. The total deviation value of each solder point is used to predict the total deviation value of the current solder point. The calculation expression is: ; Calculate the deviation residual of the solder joint, the calculation expression is: ; Based on the deviation residuals of all solder joints, the quality deviation coefficient is calculated, and the calculation expression is: ; in, Indicates Solder points, Indicates the total number of solder joints, Before The number of solder joints, Indicates The total deviation of the prediction of the solder joints, represents the bias term, represents the number of time lags used in the autoregressive model, represents the weight coefficient of the autoregressive model, Indicates The random error term for each data point is Indicates Deviation residual of each solder joint, Represents the quality deviation coefficient.

[0013] The intelligent control system of resistance welding process parameters for junction boxes includes: A data acquisition module, wherein during the resistance welding process of the junction box, a pressure sensor is installed in the area where the electrode contacts the welding workpiece, the pressure applied to the welding part is monitored in real time, the pressure data of the welding process is collected in real time through the pressure sensor, and the collected pressure data is transmitted to the data processing unit for storage and preliminary analysis; A pressure fluctuation identification module processes the real-time collected pressure data, analyzes the pressure fluctuation trend, including the frequency and amplitude of the pressure fluctuation, and calculates the pressure anomaly coefficient based on the comprehensive analysis results of the frequency and amplitude of the pressure fluctuation, which is used to identify the pressure fluctuation during the welding process; A risk assessment module, wherein the risk assessment module extracts the normal pressure application process for resistance welding according to the assessment result, collects electrode wear data through a visual sensor, analyzes the electrode wear data, determines the risk of unstable pressure transmission due to electrode wear, and assesses the risk level of the normal pressure application process; A welding quality detection module, which extracts a low-risk pressure application process according to the evaluation results, welds the resistor, and after the welding is completed, performs a quality inspection on the weld point to evaluate whether the strength and size indicators of the weld point meet the preset standards, and judges whether the welding quality meets the requirements; The welding parameter adjustment module extracts the welds with unqualified welding quality according to the quality inspection results. The system will identify the relevant welding pressure application process and adjust the pressure parameters until the quality deviation coefficient of the weld is less than a preset threshold.

[0014] Beneficial effects of the present invention: (1) The present invention monitors the pressure data and electrode wear information in real time during the welding process, uses a pressure sensor to collect the dynamic pressure changes of the welding part, and combines a multi-dimensional abnormal analysis method, including the calculation and normalization of the pressure frequency abnormal index and the pressure amplitude abnormal index, to construct a pressure abnormal coefficient model to accurately identify the abnormal pressure application link in the welding process. At the same time, by introducing a risk prediction model, based on the multi-dimensional feature extraction and deep learning optimization of electrode wear data, a comprehensive evaluation and risk quantification analysis of the pressure transmission stability is achieved, ensuring that only low-risk pressure application processes are extracted for resistance welding operations. Combined with the comprehensive evaluation of weld strength, dimensional deviation and quality deviation coefficient, the pressure application parameters are further intelligently adjusted and optimized, reducing the uncertainty caused by pressure fluctuations and electrode wear on welding quality, and ultimately improving the consistency and reliability of weld strength and size, ensuring that the welding quality stably reaches the preset standard, which fully reflects the key role of intelligent control and precise optimization in improving welding quality.

[0015] (2) The present invention uses pressure fluctuation analysis technology and quantitative evaluation of weld quality deviation coefficient to accurately and quickly locate abnormal pressure application links in the welding process, avoiding the lag and ambiguity of abnormal link identification in traditional welding processes. Based on the real-time collected pressure data, the system uses a comprehensive model constructed by the pressure frequency abnormality index and the amplitude abnormality index to comprehensively diagnose the dynamic changes in the pressure application process, and combines the quality evaluation results of weld strength and size deviation to quickly identify key problems. Through the optimization algorithm, the system autonomously calculates the adjustment direction, range and optimization strategy of the pressure parameters, covering multi-dimensional improvement measures such as adjustment of pressure application value, suppression of fluctuation amplitude and optimization of pressure application time. The optimized welding parameters are verified in real time through test welding, and the welding quality before and after optimization is dynamically compared until the weld strength and size indicators meet the preset standards. The entire process reduces the reliance on human intervention, reduces the time and material costs caused by multiple rounds of experiments, significantly improves the efficiency and accuracy of welding parameter optimization, and realizes intelligent and closed-loop control of welding quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a flowchart of the specific steps of the intelligent control method of resistance welding process parameters for junction boxes of the present invention; Figure 2 It is a flow chart of the intelligent control system of resistance welding process parameters for junction boxes in the present invention. DETAILED DESCRIPTION

[0017] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0018] See also Figure 1 As shown, the present invention is a method for intelligently controlling resistance welding process parameters for a junction box, comprising the following steps: S1: During the resistance welding process of the junction box, a pressure sensor is installed in the area where the electrode contacts the welding workpiece to monitor the pressure applied to the welding part in real time. The pressure data of the welding process is collected in real time through the pressure sensor, and the collected pressure data is transmitted to the data processing unit for storage and preliminary analysis; S2: Process the real-time collected pressure data, analyze the pressure fluctuation trend, including the frequency and amplitude of the pressure fluctuation, and calculate the pressure anomaly coefficient based on the comprehensive analysis results of the frequency and amplitude of the pressure fluctuation, which is used to identify the pressure fluctuation during the welding process; S3: Based on the evaluation results, extract the normal pressure application process for resistance welding, collect electrode wear data through visual sensors, analyze the electrode wear data, determine the risk of unstable pressure transmission due to electrode wear, and evaluate the risk level of the normal pressure application process; S4: Based on the evaluation results, extract the low-risk pressure application process, weld the resistor, and after welding, perform quality inspection on the weld to evaluate whether the strength and size indicators of the weld meet the preset standards and determine whether the welding quality meets the requirements; S5: According to the quality inspection results, the welding spots with unqualified welding quality are extracted, and the system will identify the relevant welding pressure application process and adjust the pressure parameters until the quality deviation coefficient of the welding spot is less than the preset threshold.

[0019] In S2, the real-time collected pressure data is processed to analyze the pressure fluctuation trend, including the frequency and amplitude of the pressure fluctuation. The pressure anomaly coefficient is calculated based on the comprehensive analysis results of the frequency and amplitude of the pressure fluctuation to identify the pressure fluctuation during the welding process, including: Obtain real-time pressure data during the resistance welding process, analyze the frequency of pressure fluctuations, and calculate the pressure frequency anomaly index based on the analysis results; Analyze the amplitude of pressure fluctuations, and calculate the pressure amplitude abnormality index based on the analysis results; The pressure frequency anomaly index and the pressure amplitude anomaly index are comprehensively analyzed, the pressure frequency anomaly index and the pressure amplitude anomaly index are normalized and calculated, and the pressure anomaly coefficient is calculated.

[0020] The process of obtaining the pressure frequency anomaly index is as follows: During the resistance welding process of the junction box, a pressure sensor is installed in the area where the electrode contacts the welding workpiece to collect pressure data of the welding part in real time and obtain pressure data. and transmit the pressure data to a data processing unit; Pressure data Perform discrete wavelet transform and decompose into multiple layers of components, including high-frequency components and low-frequency components; The calculation expression of the discrete wavelet transform is: ; in, Indicates time; Extract detail coefficients of high frequency components , by calculating the energy of each component , the fluctuation intensity of the pressure signal at different frequency levels is obtained, and the calculation expression is: ; According to the energy distribution of high-frequency components, the pressure frequency anomaly index is calculated, and the calculation expression is: ; in, Indicates the abnormal frequency range, Represents the total frequency range, Indicates the pressure frequency anomaly index, represents the level of wavelet decomposition, Indicates The energy of layer decomposition, represents the detail coefficient, represents the total number of detail coefficients, Indicates Layer decomposition The detail coefficient, represents the wavelet basis function; The process of obtaining the pressure amplitude anomaly index is as follows: During the resistance welding process of the junction box, a pressure sensor is installed in the area where the electrode contacts the welding workpiece to collect pressure data of the welding part in real time. and transmit the pressure data to a data processing unit for storage; in, Indicates time; The collected pressure data Perform fast Fourier transform to convert pressure data from time domain to frequency domain and obtain frequency components and amplitude components , Among them, the calculation expression of the amplitude component is: ; in, represents the real part of the spectrum, represents the imaginary part of the spectrum, Indicates frequency; According to the spectrum amplitude component Calculate the total energy in the normal frequency range and the energy in the abnormal frequency range. The calculation expression is: ; In the formula, Indicates frequency range The energy within, Indicates the frequency range; According to the energy of the normal frequency range and the abnormal frequency range, the pressure amplitude abnormal index is calculated, and the calculation expression is: ; In the formula, Indicates the pressure amplitude abnormality index, Indicates the abnormal frequency range, Represents the total frequency range, Indicates the energy in the abnormal frequency range, Represents the energy of the total frequency range; The calculation expression of the pressure anomaly coefficient is: ; In the formula, represents the pressure anomaly coefficient, and Indicates the preset scale factor, Indicates the pressure frequency anomaly index, Indicates the pressure amplitude abnormality index; It is determined whether the pressure abnormality coefficient of the resistance welding process is greater than or equal to a preset threshold value. If so, it means that the pressure application process of the resistance welding process is abnormal. If not, it means that the pressure application process of the resistance welding process is normal.

[0021] It should be noted that the pressure anomaly coefficient reflects the abnormal pressure fluctuation during the resistance welding process, and the larger the value of the pressure amplitude anomaly index is, the higher the degree of pressure anomaly during the corresponding resistance welding process is.

[0022] In S3, according to the evaluation results, the normal pressure application process is extracted for resistance welding, and the electrode wear data is collected through the visual sensor. The electrode wear data is analyzed to determine the risk of unstable pressure transmission due to electrode wear, and the risk level of the normal pressure application process is evaluated, including: Extract the normal pressure application process for resistance welding, obtain electrode wear data, process the electrode wear data, and extract abnormal features of the electrode wear data; Build risk prediction models based on historical data; The abnormal characteristics of electrode wear data are input into the risk prediction model, and the risk index is output according to the prediction model; Determine whether the risk index is greater than or equal to the preset threshold. If so, it is recorded as high risk; if not, it is recorded as low risk; The process of obtaining the abnormal characteristics of the electrode wear data is as follows: Collect real-time data of electrode wear through visual sensors , remove noise through bandpass filtering, and interpolate the missing values ​​in the collected data to obtain complete electrode wear time series data; in, Indicates time; The electrode wear data after preprocessing Decompose and extract the trend component , periodic component and random noise components ; The decomposed calculation expression is: ; Based on the decomposition results, a multidimensional feature set is extracted, including: Statistical features: calculate mean, standard deviation, skewness, and kurtosis; Frequency domain features: Extract spectrum density and frequency center through Fourier transform; Deep learning features: Convolutional neural networks are used to extract high-order features, and the dimensions are reduced through autoencoders to generate compressed feature vectors; Classify and label the multidimensional feature set based on historical data, and label the feature subsets corresponding to normal and abnormal states; The multi-dimensional feature set collected in real time is input into the trained risk prediction model to identify abnormal features of electrode wear data and use it to predict the electrode wear risk index; The process of obtaining the risk index is as follows: Construct a feature-risk label dataset based on historical welding data; The risk prediction model is selected to optimize the model parameters using the gradient descent method, and the model is trained by minimizing the cross entropy loss function; Input the multi-dimensional feature set collected in real time into the trained risk prediction model; Calculate the risk index according to the risk index calculation expression; The risk index calculation expression is: ; in, represents the activation function, represents the weights learned by the model, Indicates Input feature values, is the bias term, represents the total number of input features, represents the input features, Represents the risk index.

[0023] It should be noted that during the resistance welding process, if the value of the risk index is larger, it means that the risk level of the current pressure application is higher and the probability of abnormality is greater.

[0024] In S4, based on the evaluation results, the low-risk pressure application process is extracted to weld the resistor. After the welding is completed, the quality of the weld is inspected to evaluate whether the strength and size indicators of the weld meet the preset standards and determine whether the welding quality meets the requirements, including: Based on the low-risk pressure application process, the strength and size indicators of the solder joints are obtained, the strength and size indicators of the solder joints are compared with the preset standards, and the quality deviation coefficient is calculated; Determine whether the quality deviation coefficient is greater than or equal to a preset threshold value, if so, the welding quality is unqualified, if not, the welding quality is qualified; Wherein, the process of obtaining the mass deviation coefficient is: After the resistance welding of the junction box is completed, obtain the solder joint inspection data, including the strength index of the solder joint and size indicators ; Set preset standards for solder joint inspection, including target strength and target size , calculate the strength deviation of each solder joint and size deviation , the calculation expression is: ; ; The intensity deviation and size deviation Add them together to get the total deviation of the solder joint. The calculation expression is: ; An autoregressive model is established based on the total deviation value of the solder joint. The total deviation value of each solder point is used to predict the total deviation value of the current solder point. The calculation expression is: ; Calculate the deviation residual of the solder joint, the calculation expression is: ; Based on the deviation residuals of all solder joints, the quality deviation coefficient is calculated, and the calculation expression is: ; in, Indicates Solder points, Indicates the total number of solder joints, Before The number of solder joints, Indicates The total deviation of the prediction of the solder joints, represents the bias term, represents the number of time lags used in the autoregressive model, represents the weight coefficient of the autoregressive model, Indicates The random error term for each data point is Indicates Deviation residual of each solder joint, Represents the quality deviation coefficient.

[0025] It should be noted that the quality deviation coefficient reflects the welding quality of the weld during the welding process, and the larger the value of the quality deviation coefficient, the worse the welding quality.

[0026] In S5, based on the quality inspection results, the welds with unqualified welding quality are extracted, and the system will identify the relevant welding pressure application process and adjust the pressure parameters until the quality deviation coefficient of the weld is less than the preset threshold, including: Based on the pressure application records of unqualified welds, the system will further evaluate the abnormal characteristics of the pressure application process, including the amplitude of pressure fluctuations, abnormal frequency, and whether the pressure peak exceeds the normal range; Combined with the calculation results of the quality deviation coefficient of the solder joint, the system will determine the main problems in the pressure application process and calculate the optimization direction and range of the pressure parameters by adjusting the algorithm, including increasing or decreasing the pressure application value, suppressing pressure fluctuations, and extending the duration of pressure application; The system adjusts the pressure application parameters of the welding equipment according to the optimization calculation results, and performs a test welding with the optimized parameters; During the test welding process, pressure data was collected in real time and compared with the records before adjustment to verify the improvement of the pressure application process; At the same time, the quality of the solder joints after the test welding is tested, including the strength and size of the solder joints, to evaluate whether the adjusted parameters can reduce the quality deviation coefficient of the solder joints; If the quality deviation coefficient of the weld after the test welding is still higher than the preset threshold, the system will continue to optimize the pressure application parameters.

[0027] See also Figure 2 As shown, the intelligent control system of resistance welding process parameters for junction boxes includes: A data acquisition module, wherein during the resistance welding process of the junction box, a pressure sensor is installed in the area where the electrode contacts the welding workpiece, the pressure applied to the welding part is monitored in real time, the pressure data of the welding process is collected in real time through the pressure sensor, and the collected pressure data is transmitted to the data processing unit for storage and preliminary analysis; A pressure fluctuation identification module processes the real-time collected pressure data, analyzes the pressure fluctuation trend, including the frequency and amplitude of the pressure fluctuation, and calculates the pressure anomaly coefficient based on the comprehensive analysis results of the frequency and amplitude of the pressure fluctuation, which is used to identify the pressure fluctuation during the welding process; A risk assessment module, wherein the risk assessment module extracts the normal pressure application process for resistance welding according to the assessment result, collects electrode wear data through a visual sensor, analyzes the electrode wear data, determines the risk of unstable pressure transmission due to electrode wear, and assesses the risk level of the normal pressure application process; A welding quality detection module, which extracts a low-risk pressure application process according to the evaluation results, welds the resistor, and after the welding is completed, performs a quality inspection on the weld point to evaluate whether the strength and size indicators of the weld point meet the preset standards, and judges whether the welding quality meets the requirements; A welding parameter adjustment module, which extracts welds with unqualified welding quality according to the quality inspection results. The system will identify the relevant welding pressure application process and adjust the pressure parameters until the quality deviation coefficient of the weld is less than the preset threshold The working principle of the present invention is as follows: by installing a pressure sensor to collect the pressure data in the welding process in real time, and using wavelet transform and Fourier transform to analyze the frequency and amplitude of pressure fluctuations, the pressure anomaly coefficient is calculated to identify abnormal pressure application. At the same time, the electrode wear data is obtained through a visual sensor, and a risk prediction model is established in combination with historical data to evaluate the stability of pressure transmission, and extract the low-risk pressure application process to complete the welding. After welding, the strength and size of the weld are quality inspected, and the quality deviation coefficient is calculated to evaluate the welding quality. For welds with unqualified quality, the system analyzes the abnormal pressure application process, optimizes the pressure parameters, and verifies the adjustment effect through test welding until the weld quality deviation coefficient is lower than the preset threshold, thereby realizing intelligent optimization of welding process parameters and precise control of welding quality.

[0028] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0029] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0030] It should be understood that the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. A and B can be singular or plural. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship, but it may also indicate an "and / or" relationship. Please refer to the context for specific understanding.

[0031] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0032] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. An intelligent control method for resistance welding process parameters of a junction box, characterized in that: The following steps are involved: During the resistance welding process of the junction box, a pressure sensor is installed in the area where the electrode contacts the welding workpiece to monitor the pressure applied to the welding part in real time. The pressure data of the welding process is collected in real time through the pressure sensor, and the collected pressure data is transmitted to the data processing unit for storage and preliminary analysis; Process the real-time collected pressure data, analyze the pressure fluctuation trend, including the frequency and amplitude of pressure fluctuation, and calculate the pressure anomaly coefficient based on the comprehensive analysis results of the frequency and amplitude of pressure fluctuation to identify the pressure fluctuation during welding; According to the evaluation results, the normal pressure application process is extracted for resistance welding, and the electrode wear data is collected through the visual sensor. The electrode wear data is analyzed to determine the risk of unstable pressure transmission caused by electrode wear and evaluate the risk level of the normal pressure application process; According to the evaluation results, the low-risk pressure application process is extracted to weld the resistor. After the welding is completed, the quality of the weld is tested to evaluate whether the strength and size indicators of the weld meet the preset standards and determine whether the welding quality meets the requirements; Based on the quality inspection results, the welds with unqualified welding quality are extracted. The system will identify the relevant welding pressure application process and adjust the pressure parameters until the quality deviation coefficient of the weld is less than the preset threshold.

2. The method for intelligently controlling resistance welding process parameters for a junction box according to claim 1, characterized in that: The process of obtaining the pressure anomaly coefficient is as follows: Obtain real-time pressure data during the resistance welding process, analyze the frequency of pressure fluctuations, and calculate the pressure frequency anomaly index based on the analysis results; Analyze the amplitude of pressure fluctuations, and calculate the pressure amplitude abnormality index based on the analysis results; The pressure frequency anomaly index and the pressure amplitude anomaly index are comprehensively analyzed, the pressure frequency anomaly index and the pressure amplitude anomaly index are normalized and calculated, and the pressure anomaly coefficient is calculated.

3. The method for intelligently controlling resistance welding process parameters for a junction box according to claim 2, characterized in that: The process of obtaining the pressure frequency anomaly index is as follows: During the resistance welding process of the junction box, a pressure sensor is installed in the area where the electrode contacts the welding workpiece to collect pressure data of the welding part in real time and obtain pressure data. and transmit the pressure data to a data processing unit; Pressure data Perform discrete wavelet transform and decompose into multiple layers of components, including high-frequency components and low-frequency components; The calculation expression of the discrete wavelet transform is: ； in, Indicates time; Extract detail coefficients of high frequency components , by calculating the energy of each component , the fluctuation intensity of the pressure signal at different frequency levels is obtained, and the calculation expression is: ； According to the energy distribution of high-frequency components, the pressure frequency anomaly index is calculated, and the calculation expression is: ； in, Indicates the abnormal frequency range, Represents the total frequency range, Indicates the pressure frequency anomaly index, represents the level of wavelet decomposition, Indicates The energy of layer decomposition, represents the detail coefficient, represents the total number of detail coefficients, Indicates Layer decomposition The detail coefficient, represents the wavelet basis function.

4. The method for intelligently controlling resistance welding process parameters for a junction box according to claim 2, characterized in that: The process of obtaining the pressure amplitude anomaly index is as follows: During the resistance welding process of the junction box, a pressure sensor is installed in the area where the electrode contacts the welding workpiece to collect pressure data of the welding part in real time. and transmit the pressure data to a data processing unit for storage; in, Indicates time; The collected pressure data Perform fast Fourier transform to convert pressure data from time domain to frequency domain and obtain frequency components and amplitude components , Among them, the calculation expression of the amplitude component is: ； in, represents the real part of the spectrum, represents the imaginary part of the spectrum, Indicates frequency; According to the spectrum amplitude component Calculate the total energy in the normal frequency range and the energy in the abnormal frequency range. The calculation expression is: ； In the formula, Indicates frequency range The energy within, Indicates the frequency range; According to the energy of the normal frequency range and the abnormal frequency range, the pressure amplitude abnormal index is calculated, and the calculation expression is: ； In the formula, Indicates the pressure amplitude abnormality index, Indicates the abnormal frequency range, Represents the total frequency range, Indicates the energy in the abnormal frequency range, Represents the energy of the total frequency range.

5. The method for intelligently controlling resistance welding process parameters for a junction box according to claim 1, characterized in that: The determination of the risk of unstable pressure transmission due to electrode wear and the assessment of the risk level of the normal pressure application process specifically include: Extract the normal pressure application process for resistance welding, obtain electrode wear data, process the electrode wear data, and extract abnormal features of the electrode wear data; Build risk prediction models based on historical data; The abnormal characteristics of electrode wear data are input into the risk prediction model, and the risk index is output according to the prediction model; Determine whether the risk index is greater than or equal to the preset threshold. If so, it is recorded as high risk, if not, it is recorded as low risk.

6. The method for intelligently controlling resistance welding process parameters for a junction box according to claim 5, characterized in that: The process of acquiring the abnormal characteristics of the electrode wear data is as follows: Collect real-time data of electrode wear through visual sensors , remove noise through bandpass filtering, and interpolate the missing values ​​in the collected data to obtain complete electrode wear time series data; in, Indicates time; The electrode wear data after preprocessing Decompose and extract the trend component , periodic component and random noise components ; The decomposed calculation expression is: ； Based on the decomposition results, a multidimensional feature set is extracted, including: Statistical features: calculate mean, standard deviation, skewness, and kurtosis; Frequency domain features: Extract spectrum density and frequency center through Fourier transform; Deep learning features: Convolutional neural networks are used to extract high-order features, and the dimensions are reduced through autoencoders to generate compressed feature vectors; Classify and label the multidimensional feature set based on historical data, and label the feature subsets corresponding to normal and abnormal states; The multi-dimensional feature set collected in real time is input into the trained risk prediction model to identify abnormal features of electrode wear data and use it to predict the electrode wear risk index.

7. The method for intelligently controlling resistance welding process parameters for a junction box according to claim 5, characterized in that: The step of inputting the abnormal characteristics of the electrode wear data into the risk prediction model and outputting the risk index according to the prediction model specifically includes: Construct a feature-risk label dataset based on historical welding data; The risk prediction model is selected to optimize the model parameters using the gradient descent method, and the model is trained by minimizing the cross entropy loss function; Input the multi-dimensional feature set collected in real time into the trained risk prediction model; Calculate the risk index according to the risk index calculation expression; The risk index calculation expression is: ； in, represents the activation function, represents the weights learned by the model, Indicates Input feature values, is the bias term, represents the total number of input features, represents the input features, Represents the risk index.

8. The method for intelligently controlling resistance welding process parameters for a junction box according to claim 1, characterized in that: The extraction of low-risk pressure application process, welding the resistor, after welding is completed, the quality inspection of the weld point, assessing whether the strength and size indicators of the weld point meet the preset standards, and judging whether the welding quality meets the requirements, specifically includes: Based on the low-risk pressure application process, the strength and size indicators of the solder joints are obtained, the strength and size indicators of the solder joints are compared with the preset standards, and the quality deviation coefficient is calculated; Determine whether the quality deviation coefficient is greater than or equal to a preset threshold value, if so, the welding quality is unqualified, if not, the welding quality is qualified; Wherein, the process of obtaining the mass deviation coefficient is: After the resistance welding of the junction box is completed, obtain the weld inspection data, including the strength index of the weld and size indicators ; Set preset standards for solder joint inspection, including target strength and target size , calculate the strength deviation of each solder joint and size deviation , the calculation expression is: ； ； The intensity deviation and size deviation Add them together to get the total deviation of the solder joint. The calculation expression is: ； An autoregressive model is established based on the total deviation value of the solder joint. The total deviation value of each solder point is used to predict the total deviation value of the current solder point. The calculation expression is: ； Calculate the deviation residual of the solder joint, the calculation expression is: ； Based on the deviation residuals of all solder joints, the quality deviation coefficient is calculated, and the calculation expression is: ； in, Indicates Solder points, Indicates the total number of solder joints, Before The number of solder joints, Indicates The total deviation of the prediction of the solder joints, represents the bias term, represents the number of time lags used in the autoregressive model, represents the weight coefficient of the autoregressive model, Indicates The random error term for each data point is Indicates Deviation residual of each solder joint, Represents the quality deviation coefficient.

9. An intelligent control system for resistance welding process parameters of a junction box, characterized in that: The method for intelligently controlling resistance welding process parameters for a junction box according to any one of claims 1 to 8 comprises: A data acquisition module, wherein during the resistance welding process of the junction box, a pressure sensor is installed in the area where the electrode contacts the welding workpiece, the pressure applied to the welding part is monitored in real time, the pressure data of the welding process is collected in real time through the pressure sensor, and the collected pressure data is transmitted to the data processing unit for storage and preliminary analysis; A pressure fluctuation identification module processes the real-time collected pressure data, analyzes the pressure fluctuation trend, including the frequency and amplitude of the pressure fluctuation, and calculates the pressure anomaly coefficient based on the comprehensive analysis results of the frequency and amplitude of the pressure fluctuation, which is used to identify the pressure fluctuation during the welding process; A risk assessment module, wherein the risk assessment module extracts the normal pressure application process for resistance welding according to the assessment result, collects electrode wear data through a visual sensor, analyzes the electrode wear data, determines the risk of unstable pressure transmission due to electrode wear, and assesses the risk level of the normal pressure application process; A welding quality detection module, which extracts a low-risk pressure application process according to the evaluation results, welds the resistor, and after the welding is completed, performs a quality inspection on the weld point to evaluate whether the strength and size indicators of the weld point meet the preset standards, and judges whether the welding quality meets the requirements; The welding parameter adjustment module extracts the welds with unqualified welding quality according to the quality inspection results. The system will identify the relevant welding pressure application process and adjust the pressure parameters until the quality deviation coefficient of the weld is less than a preset threshold.

Images