An intelligent group control system for welding equipment based on multi-source acquisition information

Abstract

The present invention relates to the technical field of intelligent control of welding equipment, and specifically to an intelligent group control system for welding equipment based on multi-source acquisition information, including a target welding equipment confirmation module, a status confirmation module, a prediction and adjustment module. When in a normal state, multi-source feature information is fused to form welding feature information; the welding feature information is analyzed to predict the welding result; when the predicted welding result does not meet the preset quality standard, an adaptive adjustment algorithm is used to analyze the multi-source feature information to generate an adjustment strategy, and the welding method is optimized based on the adjustment strategy. By the above method, the present invention can solve the problem that the quality of the welded seams is uneven due to the difficulty in ensuring the consistency of welding quality caused by the performance fluctuations of welding equipment, the quality of welding materials, etc.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent control of welding equipment, and particularly relates to an intelligent group control system for welding equipment based on multi-source collected information. Background Art

[0002] At present, with the rapid development of artificial intelligence and Internet of Things technologies, intelligent welding equipment has gradually replaced traditional welding equipment and has gradually become the mainstream equipment in the welding industry. During the production of new energy vehicles, intelligent welding equipment plays an important role. The intelligent welding equipment uses various sensors, industrial cameras and other auxiliary devices to collect data during the welding process in real time, and performs fusion analysis on the data through the background to achieve all-round control of the welding process between various parts of new energy vehicles. It can execute welding tasks throughout the day according to the preset welding program, reduce manual intervention in the welding process, reduce labor costs, and improve the overall quality of welding at the same time.

[0003] Due to the large number of intelligent welding equipment managed at the same time, when determining the target welding equipment according to the task list, the existing task allocation methods often have the situation of task interruption or repeated task arrangement. In addition, intelligent welding equipment often continuously executes welding tasks, resulting in early failures or abnormalities of intelligent welding equipment, thereby shortening the service life of the welding equipment.

[0004] And when controlling a large number of intelligent welding equipment to execute welding tasks at the same time, in order to ensure welding time synchronization, the intelligent welding equipment will be controlled to complete the welding work according to the preset time; due to the performance fluctuations of the welding equipment, the quality of the welding materials, etc., it is difficult to ensure the consistency of welding quality, resulting in uneven quality of the welded seams.

[0005] Therefore, the present invention provides an intelligent group control system for welding equipment based on multi-source collected information to solve the above problems. Summary of the Invention

[0006] In view of the above situation, in order to overcome the deficiencies of the prior art, the present invention provides an intelligent group control system for welding equipment based on multi-source collected information to solve the problem that it is difficult to ensure the consistency of welding quality due to the performance fluctuations of the welding equipment, the quality of the welding materials, etc., resulting in uneven quality of the welded seams.

[0007] In order to achieve the above object, the technical solution adopted by the present invention is as follows:

[0008] An intelligent group control system for welding equipment based on multi-source collected information, comprising:

[0009] The target welding equipment confirmation module obtains the welding tasks to be arranged and the working status of the welding equipment; uses the gradient boosting decision tree algorithm and the random forest algorithm to process the task list of the welding equipment and the requirement information of the welding tasks to determine the target welding equipment;

[0010] The status confirmation module, when the target welding equipment executes the welding task, collects multivariate information during the welding process in real time; extracts features from the multivariate information to generate multivariate feature information; conducts an initial analysis of the multivariate feature information to determine the health status of the welding equipment;

[0011] The prediction and adjustment module, when in a normal state, fuses the multivariate feature information to form welding feature information; analyzes the welding feature information to predict the welding result; when the predicted welding result does not meet the preset quality standard, uses the adaptive adjustment algorithm to analyze the multivariate feature information to generate an adjustment strategy, and optimizes the welding method based on the adjustment strategy.

[0012] Preferably, the use of the gradient boosting decision tree algorithm and the random forest algorithm to process the task list of the welding equipment and the requirement information of the welding tasks to determine the target welding equipment includes: analyzing the task list and requirement information to determine the input feature information; according to the feature importance evaluation mechanisms built into the gradient boosting decision tree algorithm and the random forest algorithm, calculating the relative importance scores of each input feature information in the two algorithms respectively, and determining the top-ranked, preset number of features as the first feature information according to the relative importance scores; using the gradient boosting decision tree model and the random forest model to process the first feature information respectively to obtain the decision tree prediction result and the random forest prediction result; using the logistic regression model to process the decision tree prediction result and the random forest prediction result to determine the target welding equipment.

[0013] Preferably, the extraction of features from the multivariate information to generate multivariate feature information includes: using the feature extraction model constructed by the convolutional neural network to extract features from the multivariate information to obtain the initial feature information; using the Lasso regression algorithm to perform feature selection processing on the initial feature information to obtain the intermediate feature information, and adjusting the intermediate feature information according to the key feature library to obtain the multivariate feature information.

[0014] Preferably, the initial analysis of the multi - feature information to determine the health status of the welding equipment includes: screening the multi - feature information to obtain the factor feature information of each health assessment index; the health assessment indexes include the operation time index, the maintenance health index, and the performance parameter health index; processing the factor feature information of each health index according to the pre - designed calculation rules to output the corresponding index value; calculating each index value by using the weighted average method to obtain the comprehensive health index; and determining the health status according to the corresponding relationship between the comprehensive health index and the health level.

[0015] Preferably, the processing of the factor feature information of each health index according to the pre - designed calculation rules to output the corresponding index value includes: the operation time index H T The calculation formula is:

[0016]

[0017] where T is the actual cumulative operation time of the welding equipment, and T b is the cumulative reference operation time;

[0018] The maintenance health index H M The calculation formula is:

[0019]

[0020] where N is the actual maintenance times of the welding equipment, N max is the maximum allowable maintenance times of the welding equipment, C is the actual maintenance cost, C b is the reference maintenance cost, R is the actual maintenance times of the same part, R b is the reference maintenance times of the same part allowed, and k c and k R are the corresponding weight coefficients respectively;

[0021] The performance parameter health index H P The calculation formula is:

[0022]

[0023] where k p and k s are the corresponding parameter coefficients respectively, n is the total number of parts and components with power in the welding equipment, P maxi is the maximum output power corresponding to the i - th part with power in the welding equipment, P i is the actual operating power corresponding to the i - th part with power in the welding equipment, m is the total number of parts and components with operating speed in the welding equipment, S maxj$S_j$ is the maximum operating speed corresponding to the $j$-th part with an operating speed in the welding equipment. j $T_j$ is the actual operating speed corresponding to the $j$-th part with an operating speed in the welding equipment, and $T$ is the actual cumulative operating time of the welding equipment. b $T_0$ is the cumulative reference operating time.

[0024] Preferably, analyzing the welding feature information to predict the welding result includes: processing the welding feature information using a multiple linear regression algorithm to obtain the predicted values of the quality indicators for each quality indicator, and determining the average quality prediction value based on the predicted values of each quality indicator; forming a predicted welding result based on the predicted values of each quality indicator, the average quality prediction value, and the factor influence value; where the calculation formula for the predicted value $Y$ of the quality indicator is:

[0025]

[0026] where $\delta$ is the influence value of the total error term, $X_i$ i is the influence value of the $i$-th factor affecting the corresponding welding quality indicator, and $X_0$ is 1, $\beta_i$ i is the corresponding regression coefficient, and the total number of factor influence values is $n + 1$.

[0027] Preferably, analyzing the multi - feature information using an adaptive adjustment algorithm to generate an adjustment strategy includes: processing the multi - feature information using an adaptive adjustment algorithm to obtain the adjustment values of the control signals of the welding equipment; generating an adjustment strategy based on the adjustment values; the calculation formula of the adaptive adjustment algorithm is:

[0028]

[0029] where $u(t)$ is the adjustment value corresponding to the control signal, $t$ is time, $K_p$ p is the proportional coefficient, $e(t)$ is the error signal, $K_i$ i is the integral coefficient, $e'(\tau)$ is the first - order derivative of the error signal in the interval $(0, t)$, $K_d$ d is the differential coefficient.

[0030] Preferably, an intelligent group control system for a welding equipment based on multi - acquisition information further includes: a parameter optimization module, which obtains the actual control information of the adjusted welding equipment and the initial control information for initially controlling the welding equipment, and uses a particle swarm optimization algorithm to perform an optimization analysis on the actual control information and the initial control information to update the initial control information.

[0031] Preferably, an intelligent group control system for welding equipment based on multi-source acquisition information further includes: obtaining multi-source information after adjusting the welding method in a loop according to a preset time, and analyzing the multi-source information to obtain the predicted welding result this time. When the predicted welding result meets the preset quality standard or the number of loops reaches the preset loop threshold, the loop is terminated; when the predicted welding result does not meet the preset quality standard after termination, an abnormal information is sent to the target welding equipment confirmation module; the target welding equipment confirmation module selects a new target welding equipment from the corresponding initial welding equipment to take over the welding task.

[0032] Preferably, an intelligent group control system for welding equipment based on multi-source acquisition information further includes: a health management module, which extracts health management feature information from the multi-source feature information; analyzes and processes the health management feature information to determine the current health degree of the welding equipment. When the health degree is less than the preset health value, a warning message is sent, and the health recovery mode is entered after completing the welding task in the current stage; the calculation formula for the health degree is:

[0033] h = 1 - (a1q te + a2q po + a3q de + a4q time + a5q wdtime ),

[0034] where h is the health degree, q te is the temperature influence index, q po is the power consumption influence index, q de is the defect rate influence index, q time is the continuous operation duration influence index, q wdtime is the influence index corresponding to the single-time weld seam completion duration, and a1, a2, a3, a4, and a5 are the weight coefficients corresponding to each index respectively.

[0035] The beneficial effects of the present invention are as follows:

[0036] 1. The present invention can analyze the welding feature information of the welding equipment through the prediction and adjustment module to predict the welding result. When the predicted welding result does not meet the preset quality standard, an adaptive adjustment algorithm is used to process the multi-source feature information to obtain the adjustment values of the control signals of the welding equipment, and the welding equipment is regulated according to the adjustment strategy generated by the adjustment values. In the above way, the present invention can solve the problem that it is difficult to ensure the consistency of welding quality due to the performance fluctuations of welding equipment, the quality of welding materials, etc., resulting in uneven quality of welded weld seams.

[0037] 2. The present invention can process the input feature information of the task list and requirement information by combining the gradient boosting algorithm and the random forest algorithm to obtain the model prediction results of both, and then use the logistic regression model to process these two model prediction results to determine the final target welding equipment. In this way, the present invention can utilize the advantages of the two algorithms to improve the accuracy and robustness of overall determining the target welding equipment, can solve the problems that the existing task allocation methods often have faults or repeat task arrangements; and saves the time cost and labor cost of traditional manual welding task arrangement.

[0038] 3. The present invention can screen and process the multivariate feature information through the status confirmation module to obtain the factor feature information of each health assessment index, and then process the respective factor feature information according to the evaluation rules of each health assessment index to determine the index value of each health assessment index; furthermore, determine the comprehensive health index of the welding equipment according to the values of each health assessment index; determine the current health status of the welding equipment according to the corresponding relationship between the comprehensive health index and the health level, monitor the status of the welding equipment in real time, and when the working status of the welding equipment is abnormal, replace the target welding equipment in time to ensure that the welding task is completed within the set time; at the same time, extend the service life of the welding equipment. Brief Description of the Drawings

[0039] Figure 1 It is a schematic module diagram of an intelligent group control system for welding equipment based on multivariate acquisition information of the present invention. Detailed Embodiments

[0040] Next, each embodiment of the present invention will be described in detail with reference to the accompanying Figure 1 Those skilled in the art should understand that these embodiments are only used to explain the technical principle of the present invention and are not intended to limit the protection scope of the present invention.

[0041] An intelligent group control system for welding equipment based on multivariate acquisition information, as shown in the accompanying Figure 1 figure, includes:

[0042] A target welding equipment confirmation module, which obtains the welding tasks to be arranged and the working status of the welding equipment; uses the gradient boosting decision tree algorithm and the random forest algorithm to process the task list of the welding equipment and the requirement information of the welding tasks to determine the target welding equipment.

[0043] Specifically, the target welding equipment confirmation module is used to process data by combining the gradient boosting decision tree algorithm and the random forest algorithm to determine the target welding equipment that meets the requirements of the welding tasks.

[0044] A status confirmation module, when the target welding equipment executes a welding task, collects multivariate information during the welding process in real time; extracts features from the multivariate information to generate multivariate feature information; and performs an initial analysis on the multivariate feature information to determine the health status of the welding equipment.

[0045] Specifically, clean the collected raw data to remove outliers, noise, and duplicate data to improve the accuracy and reliability of the obtained multivariate information. For example, identify and eliminate abnormal data by setting thresholds or using statistical methods, and use filtering algorithms to remove noise. And perform screening processing on the multivariate feature information to obtain the factor feature information of each health assessment index, then process the respective factor feature information according to the evaluation rules of each health assessment index to determine the index values of each health assessment index; furthermore, determine the comprehensive health index of the welding equipment according to the values of each health assessment index; and determine the current health status of the welding equipment according to the correspondence between the comprehensive health index and the health level.

[0046] A prediction and adjustment module, when in a normal state, fuses the multivariate feature information to form welding feature information; analyzes the welding feature information to predict the welding result; when the predicted welding result does not meet the preset quality standard, uses an adaptive adjustment algorithm to analyze the multivariate feature information to generate an adjustment strategy, and optimizes the welding method based on the adjustment strategy.

[0047] Among them, the preset quality standard includes weld appearance quality indicators, weld performance indicators, weld internal quality indicators, average quality indicators, and the standard intervals corresponding to the influence values of each factor; when there is one of the weld appearance quality indicators, weld performance indicators, weld internal quality indicators, and average quality indicators that does not meet the corresponding standard interval, it is predicted that the welding result does not meet the preset quality standard; when the weld appearance quality indicators, weld performance indicators, weld internal quality indicators, and average quality indicators all meet the corresponding standard intervals, it is predicted that the welding result meets the preset quality standard.

[0048] Preferably, the multivariate information includes but is not limited to various parameters during the welding process of the welding equipment, such as voltage, current, temperature, type of welding material, speed of feeding the welding material, weld offset, gas flow rate, image data recorded by an industrial camera during the welding process, etc.

[0049] In this embodiment, the present invention can analyze the welding characteristic information of the welding equipment through the prediction and adjustment module to predict the welding result. When the predicted welding result does not meet the preset quality standard, an adaptive adjustment algorithm is used to process the multi-feature information to obtain the adjustment values of the control signals of the welding equipment, and the welding equipment is regulated according to the adjustment strategy generated based on the adjustment values. In the above manner, the present invention can solve the problem that the quality of the welded seams is uneven due to the difficulty in ensuring the consistency of the welding quality caused by the performance fluctuations of the welding equipment, the quality of the welding materials, etc.

[0050] Furthermore, the present invention can process data by combining the gradient boosting decision tree algorithm and the random forest algorithm to determine the target welding equipment that meets the requirements of the welding task, which can solve the problems of frequent faults or repeated task arrangements in the existing task allocation methods, save the time cost of traditional manual welding task arrangements, and help save labor costs and the time for docking and communication.

[0051] Furthermore, the present invention can screen and process the multi-feature information through the status confirmation module to obtain the factor characteristic information of each health assessment index, and then process the respective factor characteristic information according to the assessment rules of each health assessment index to determine the index values of each health assessment index; furthermore, the comprehensive health index of the welding equipment is determined according to the values of each health assessment index; according to the corresponding relationship between the comprehensive health index and the health level, the current health status of the welding equipment is determined, the status of the welding equipment is monitored in real time, and when the working status of the welding equipment is abnormal, the target welding equipment is replaced in time to ensure that the welding task is completed within the set time; at the same time, the service life of the welding equipment is extended.

[0052] In an embodiment of the present invention, the task list of the welding equipment and the requirement information of the welding task are processed by using the gradient boosting decision tree algorithm and the random forest algorithm to determine the target welding equipment, including: analyzing the task list and the requirement information to determine the input feature information; according to the feature importance evaluation mechanism built in the gradient boosting decision tree algorithm and the random forest algorithm, calculating the relative importance scores of each input feature information in the two algorithms respectively, and determining the top-ranked, preset number of features as the first feature information according to the relative importance scores; using the gradient boosting decision tree model and the random forest model to process the first feature information respectively to obtain the decision tree prediction result and the random forest prediction result; using the logistic regression model to process the decision tree prediction result and the random forest prediction result to determine the target welding equipment.

[0053] Preferably, the data of the task list and the requirement information include but are not limited to voltage, current, temperature, type of welding material, speed of feeding the welding material, weld offset, gas flow rate, welding position, type of welding material, welding groove form, etc.

[0054] Through the setting method of this embodiment, the present invention can process the input feature information of the task list and requirement information by combining the gradient boosting algorithm and the random forest algorithm to obtain the model prediction results of both, and then use the logistic regression model to process these two model prediction results to determine the final target welding equipment. The present invention can utilize the advantages of the two models to improve the accuracy and robustness of overall determining the target welding equipment; and can solve the problems that the existing task allocation methods often have faults or repeated task arrangements.

[0055] In one embodiment of the present invention, the feature extraction of multivariate information to generate multivariate feature information includes: using a feature extraction model constructed by a convolutional neural network to extract features from the multivariate information to obtain initial feature information; using the Lasso regression algorithm to perform feature selection processing on the initial feature information to obtain intermediate feature information, and adjusting the intermediate feature information according to the key feature library to obtain multivariate feature information.

[0056] Specifically, after obtaining the initial feature information through the feature extraction model constructed by the convolutional neural network, a feature selection algorithm combining the Lasso regression algorithm and the key feature library is used to perform feature selection processing on the initial feature information, which specifically includes initializing the model parameters: initializing the weight coefficients and biases to 0 or some small random values; calculating the predicted value: calculating the predicted value of the model according to the current weight coefficients and biases; calculating the gradient of the loss function: calculating the gradient of the loss function with respect to the predicted value, and then according to the chain rule, calculating the gradients of the loss function with respect to the weight coefficients and biases and calculating the gradient of the L1 regularization term; updating the model parameters according to the update rule of the gradient descent method; continuously repeating the above steps until the stopping criterion is met, such as reaching a specified number of iterations or the value of the loss function is less than a certain threshold, so as to screen out the initial feature information; then, according to the verification result of comparing the key feature library with the initial feature information, adding the missing feature information to the initial feature information again to obtain the final multivariate feature information, which can avoid the key feature information being cleared in the feature selection stage; thus making the obtained multivariate feature information more in line with the actual requirements.

[0057] Through the setting method of this embodiment, the present invention can effectively achieve feature selection and parameter compression by combining the Lasso regression algorithm and the key database, thereby improving the generalization ability and interpretability of the model. In practical applications, the algorithm details can be adjusted according to the specific data set and task requirements to adjust the selected feature parameters.

[0058] In one embodiment of the present invention, the initial analysis of the multi - feature information to determine the health status of the welding equipment includes: screening the multi - feature information to obtain the factor feature information of each health assessment index; the health assessment indexes include the operation time index, the maintenance health index, and the performance parameter health index; processing the factor feature information of each health index according to the pre - designed calculation rules to output the corresponding index values; calculating the index values by using the weighted average method to obtain the comprehensive health index; and determining the health status according to the corresponding relationship between the comprehensive health index and the health level.

[0059] Further, the processing of the factor feature information of each health index according to the pre - designed calculation rules to output the corresponding index values includes:

[0060] The calculation formula of the operation time index H T is:

[0061]

[0062] where T is the actual cumulative operation time of the welding equipment, and T b is the cumulative reference operation time;

[0063] The calculation formula of the maintenance health index H M is:

[0064] ]

[0065] where N is the actual maintenance times of the welding equipment, N max is the maximum allowable maintenance times of the welding equipment, C is the actual maintenance cost, C b is the reference maintenance cost, R is the actual maintenance times of the same part, R b is the reference maintenance times of the same part allowed, k c and k R are the corresponding weight coefficients respectively;

[0066] The calculation formula of the performance parameter health index H P is:

[0067]

[0068] where k p and k s are the corresponding parameter coefficients respectively, n is the total number of parts and devices with power in the welding equipment, P maxi is the maximum output power corresponding to the i - th part with power in the welding equipment, and P i is the actual operation power corresponding to the i - th part with power in the welding equipment.m S is the total number of parts and components with operating speeds in the welding equipment. maxj S is the maximum operating speed corresponding to the j-th part with an operating speed in the welding equipment. j T is the actual cumulative operating time of the welding equipment. T is the actual operating speed corresponding to the j-th part with an operating speed in the welding equipment. b is the cumulative reference operating time.

[0069] Preferably, in this embodiment, maintenance refers to performing repairs without replacing parts. If parts are replaced, the recorded data of the corresponding parts is re-recorded.

[0070] In this embodiment, the present invention can screen and process multi-source feature information through the status confirmation module to obtain the factor feature information of each health assessment index, and then process the respective factor feature information according to the assessment rules of each health assessment index to determine the index value of each health assessment index; furthermore, determine the comprehensive health index of the welding equipment according to the values of each health assessment index; according to the correspondence between the comprehensive health index and the health level, determine the current health status of the welding equipment, monitor the status of the welding equipment in real time, and when the working status of the welding equipment is abnormal, replace the target welding equipment in time to ensure that the welding task is completed within the set time.

[0071] In an embodiment of the present invention, the analysis of welding feature information to predict welding results includes: processing the welding feature information using a multiple linear regression algorithm to determine multiple factor influence values, obtaining the quality index prediction values of each quality index according to each factor influence value; determining the average quality prediction value according to each quality index prediction value; forming a predicted welding result according to each quality index prediction value, the average quality prediction value, and the factor influence value; wherein, the calculation formula for the quality index prediction value Y is:

[0072]

[0073] wherein, δ is the influence value of the total error term, X i is the i-th factor influence value affecting the corresponding welding quality index, and X0 is 1, β i is the corresponding regression coefficient, and the total number of factor influence values is n + 1.

[0074] Preferably, the welding quality indicators include but are not limited to weld appearance quality indicators, weld performance indicators, and weld internal quality indicators; the evaluation factors of weld appearance quality indicators include but are not limited to weld shape, size, and surface defects; the evaluation factors of weld performance indicators include but are not limited to tensile strength, bending performance, impact toughness, and hardness; the weld internal quality indicators include but are not limited to metallographic structure, pores, slag inclusions, lack of penetration, and lack of fusion.

[0075] Through the above embodiments, the present invention can process welding feature information using a multiple linear regression algorithm to obtain a predicted welding result for the current welding operation, which can improve the robustness and accuracy of the prediction result.

[0076] In one embodiment of the present invention, the analysis of multi - feature information using an adaptive adjustment algorithm to generate an adjustment strategy includes: processing the multi - feature information using an adaptive adjustment algorithm to obtain adjustment values for each control signal of the welding equipment; generating an adjustment strategy based on the adjustment values; the calculation formula of the adaptive adjustment algorithm is:

[0077]

[0078] where \(u(t)\) is the adjustment value corresponding to the control signal, \(t\) is time, \(K\) p is the proportionality coefficient, \(e(t)\) is the error signal formed by the difference between the actual control signal and the reference signal, \(K\) i is the integral coefficient, \(e(\tau)\) is the first - order derivative of the error signal in the interval \((0,t)\), \(K\) d is the differential coefficient.

[0079] In this embodiment, when analyzing multi - feature information using an adaptive adjustment algorithm, the feature information considered includes but is not limited to current, voltage, temperature, wire feeding speed, welding speed, gas flow rate, and weld offset. The welding quality prediction model formed by the present invention using the adaptive adjustment algorithm can solve the problem that it is difficult to ensure the consistency of welding quality due to performance fluctuations of welding equipment, quality of welding consumables, etc., resulting in uneven quality of welded seams. And it can further improve welding quality and production efficiency, potentially reduce human intervention and errors, and lower the skill requirements for operators.

[0080] In one embodiment of the present invention, an intelligent group control system for welding equipment based on multi - acquisition information further includes: a parameter optimization module that obtains the actual control information of the adjusted welding equipment and the initial control information of the initially controlled welding equipment, and uses a particle swarm optimization algorithm to perform optimization analysis on the actual control information and the initial control information to update the initial control information.

[0081] Through the setting method of this embodiment, the present invention can, through the parameter optimization module, based on the updated initial control information, improve the welding quality when the corresponding welding equipment performs the same welding task, which helps to reduce the adjustment times of the adaptive adjustment algorithm and quickly complete finished products that meet the preset quality standards.

[0082] In an embodiment of the present invention, an intelligent group control system for welding equipment based on multi-source acquisition information further includes: cyclically acquiring multi-source information after adjusting the welding method according to a preset time, and analyzing the multi-source information to obtain the predicted welding result of this time; when the predicted welding result meets the preset quality standard or the number of cycles reaches the preset cycle threshold, terminating the cycle; when the predicted welding result does not meet the preset quality standard after termination, sending an abnormal information to the target welding equipment confirmation module; the target welding equipment confirmation module selects a new target welding equipment from the corresponding initial welding equipment to take over the welding task.

[0083] Through the setting method of this embodiment, the present invention can terminate the cycle when the predicted welding result meets the preset quality standard or the number of cycles reaches the preset cycle threshold, ensuring that the prediction stage can avoid getting into an infinite loop when the predicted welding result continuously does not meet the preset quality standard, resulting in welding equipment failure or other abnormal situations. And after termination, when the predicted welding result does not meet the preset quality standard, sending an abnormal information to the target welding equipment confirmation module; the target welding equipment confirmation module selects a new target welding equipment from the corresponding initial welding equipment to take over the welding task; thereby ensuring that the overall welding task is completed on time.

[0084] In an embodiment of the present invention, an intelligent group control system for welding equipment based on multi-source acquisition information further includes: a health management module, which extracts health management feature information from the multi-source feature information; analyzes and processes the health management feature information to determine the current health degree of the welding equipment, and when the health degree is less than the preset health value, sends out a warning information; and enters the health recovery mode after completing the welding task of the current stage; the calculation formula of the health degree is:

[0085] h = 1 - (a1q te + a2q po + a3q de + a4q time + a5q wdtime ),

[0086] where h is the health degree, q te is the temperature influence index, q po is the power consumption influence index, q de is the defect rate influence index, q time is the continuous operation duration influence index, q wdtime is the influence index corresponding to the single-time weld seam completion duration, and a1, a2, a3, a4 and a5 are the weight coefficients corresponding to each index respectively.

[0087] Specifically, when the operating temperature of the welding equipment exceeds the reference temperature range and exceeds a certain time, the calculation formula of the temperature influence index q te is: Among them, within the time interval (0, t), T te is the operating temperature at time t, and T0 is the average temperature in the reference temperature range; the power consumption impact index is the corresponding value of the power consumption change rate per unit time and the power consumption impact index value found in the comparison table; the defect rate impact index is the probability of defective products that do not meet the preset quality standards within a preset time period; the continuous operation duration impact index is the impact index value corresponding to the continuous operation duration and the continuous operation duration impact index found in the comparison table; the impact index corresponding to the single-time weld completion duration is the impact index value corresponding to the time value that exceeds the reference duration single-time found in the comparison table.

[0088] Through the above implementation manners, the present invention can analyze the working state of the welding equipment through the health management module and regulate the working mode of the welding equipment. For example, when the equipment temperature is too high, the welding power is reduced or the work is paused to protect the equipment.

[0089] In an embodiment of the present invention, an intelligent group control system for welding equipment based on multi-source acquisition information further includes life prediction and evaluation; a life prediction model is constructed based on the historical operation data and usage conditions of the welding equipment in combination with machine learning algorithms. This model can consider factors such as the working time, load conditions, and maintenance records of the equipment to predict the service life of the equipment; and as the welding equipment is continuously used, new data is continuously collected and the life prediction model is updated and optimized to improve the prediction accuracy. According to the life prediction model, the remaining life of the welding equipment is estimated in real time, providing a decision-making basis for the maintenance and replacement of the equipment. When the remaining life of the equipment approaches the warning value, the operator is timely reminded to perform maintenance or replacement to avoid production interruption and losses caused by sudden equipment failures.

[0090] The various implementation manners of the systems and technologies described above in this article can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementation manners can include: implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a dedicated or general programmable processor, and can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit the data and instructions to the storage system, the at least one input device, and the at least one output device.

[0091] It should be noted that in the description of the present invention, the terms "first", "second", and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0092] The program code for implementing the methods of the present disclosure can be written in any combination of one or more programming languages. These program codes can be provided to the processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing devices, such that when the program codes are executed by the processor or controller, the functions / operations specified in the flowcharts and / or block diagrams are implemented. The program codes can be executed entirely on the machine, partially on the machine, as an independent software package partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0093] In the context of the present disclosure, a machine-readable medium can be a tangible medium that can contain or store a program for use by or in connection with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0094] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or a trackball) through which the user can provide input to the computer. Other kinds of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and the input received from the user can be in any form (including acoustic input, voice input, or tactile input).

[0095] The systems and techniques described herein can be implemented in a computing system including backend components (e.g., as a data server), or a computing system including middleware components (e.g., an application server), or a computing system including frontend components (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described herein), or a computing system including any combination of such backend components, middleware components, or frontend components. The components of the system can be interconnected to each other by digital data communication in any form or medium (e.g., a communication network). Examples of communication networks include: local area network (LAN), wide area network (WAN), and the Internet.

[0096] A computer system can include clients and servers. The clients and servers are generally far from each other and typically interact through a communication network. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, can also be a server of a distributed system, or a server incorporating blockchain.

[0097] So far, the technical solutions of the present invention have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it is easily understood by those skilled in the art that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the protection scope of the present invention.

Claims

1. An intelligent group control system for welding equipment based on multi-source acquisition information, characterized in that, Including: A target welding equipment confirmation module, which obtains the welding tasks to be arranged and the working status of the welding equipment; processes the task list of the welding equipment and the requirement information of the welding tasks by using the gradient boosting decision tree algorithm and the random forest algorithm to determine the target welding equipment; A status confirmation module, which, when the target welding equipment executes the welding task, collects multivariate information in the welding process in real time; extracts features from the multivariate information to generate multivariate feature information; conducts an initial analysis of the multivariate feature information to determine the health status of the welding equipment; A prediction and adjustment module, when in a normal state, fuses the multivariate feature information to form welding feature information; analyzes the welding feature information to predict the welding result; when the predicted welding result does not meet the preset quality standard, analyzes the multivariate feature information by using an adaptive adjustment algorithm to generate an adjustment strategy, and optimizes the welding method based on the adjustment strategy.

2. The intelligent group control system for welding equipment according to claim 1, wherein The processing of the task list of the welding equipment and the requirement information of the welding tasks by using the gradient boosting decision tree algorithm and the random forest algorithm to determine the target welding equipment includes: analyzing the task list and the requirement information to determine the input feature information; respectively calculating the relative importance scores of each input feature information in the two algorithms according to the feature importance evaluation mechanism built in the gradient boosting decision tree algorithm and the random forest algorithm, and determining the top-ranked and preset-number of features as the first feature information according to the relative importance scores; processing the first feature information by using the gradient boosting decision tree model and the random forest model respectively to obtain the decision tree prediction result and the random forest prediction result; processing the decision tree prediction result and the random forest prediction result by using the logistic regression model to determine the target welding equipment.

3. The intelligent group control system of the welding equipment according to claim 1, wherein The extraction of features from the multivariate information to generate multivariate feature information includes: extracting features from the multivariate information by using a feature extraction model constructed by a convolutional neural network to obtain initial feature information; performing feature selection processing on the initial feature information by using the Lasso regression algorithm to obtain intermediate feature information, and adjusting the intermediate feature information according to the key feature library to obtain multivariate feature information.

4. The intelligent group control system for a welding device according to claim 1, characterized in that, The initial analysis of the multivariate feature information to determine the health status of the welding equipment includes: Screening the multivariate feature information to obtain the factor feature information of each health evaluation index; the health evaluation indexes include the operation time index, the maintenance health index, and the performance parameter health index; processing the respective factor feature information according to the preset calculation rules of each health index to output the corresponding index values; calculating the index values by using the weighted average method to obtain the comprehensive health index; determining the health status according to the corresponding relationship between the comprehensive health index and the health level.

5. The intelligent group control system for a welding device according to claim 4, characterized in that Processing the respective factor characteristic information according to the preset calculation rules of each health indicator to output the corresponding indicator value, including: running time indicator H T The calculation formula is: Among them, T is the actual cumulative operating time of the welding equipment, and T b is the cumulative reference operating time; Maintain health indicator H M The calculation formula is as follows: Among them, N is the actual maintenance times of the welding equipment, N max is the maximum allowable maintenance times of the welding equipment, C is the actual maintenance cost, C b is the reference maintenance cost, R is the actual maintenance times of the same part, R b is the reference maintenance times allowed for the same part, k c and k R are the corresponding weight coefficients respectively; Performance parameter health indicator H P The calculation formula is as follows: where k p and k s are the corresponding parameter coefficients respectively, n is the total number of parts and components with power in the welding equipment, P maxi is the maximum output power corresponding to the i-th part with power in the welding equipment, P i is the actual operating power corresponding to the i-th part with power in the welding equipment, m is the total number of parts and components with operating speed in the welding equipment, S maxj is the maximum operating speed corresponding to the j-th part with operating speed in the welding equipment, S j is the actual operating speed corresponding to the j-th part with operating speed in the welding equipment, T is the actual cumulative operating time of the welding equipment, T b is the cumulative reference operating time.

6. The intelligent group control system of the welding equipment according to claim 1, characterized in that, Analyzing the welding feature information to predict the welding result includes: processing the welding feature information using a multiple linear regression algorithm to obtain the quality index prediction values of each quality index, and determining the average quality prediction value according to the quality index prediction values; forming a predicted welding result based on the quality index prediction values, the average quality prediction value, and the factor influence value; where the calculation formula for the quality index prediction value Y is: Among them, δ is the influence value of the total error term, and X i is the influence value of the i-th factor that affects the corresponding welding quality index, and X0 is 1, β i is the corresponding regression coefficient, and the total number of factor influence values is n + 1.

7. The intelligent group control system of the welding equipment according to claim 1, wherein Analyzing the multi - feature information using an adaptive adjustment algorithm to generate an adjustment strategy includes: Processing the multi - feature information using an adaptive adjustment algorithm to obtain the adjustment values of each control signal of the welding equipment; generating an adjustment strategy according to the adjustment values; the calculation formula of the adaptive adjustment algorithm is: Among them, u(t) is the adjustment value corresponding to the control signal, t is time, K p is the proportionality coefficient, e(t) is the error signal, K i is the integral coefficient, e(τ) is the first derivative of the error signal within the interval (0, t), K d is the differential coefficient.

8. The intelligent group control system of the welding equipment according to claim 7, wherein, It further includes: A parameter optimization module, which obtains the actual control information of the adjusted welding equipment and the initial control information of the initially controlled welding equipment, and uses a particle swarm optimization algorithm to perform optimization analysis on the actual control information and the initial control information to update the initial control information.

9. The intelligent group control system for welding equipment according to claim 1, wherein It further includes: Obtaining the multi - information after adjusting the welding method in a preset time cycle, and analyzing the multi - information to obtain the predicted welding result of this time. When the predicted welding result meets the preset quality standard or the number of cycles reaches the preset cycle threshold, the cycle is terminated; when the predicted welding result does not meet the preset quality standard after termination, an abnormal information is sent to the target welding equipment confirmation module; The target welding equipment confirmation module selects a new target welding equipment from the corresponding initial welding equipment to take over the welding task.

10. The intelligent group control system of the welding equipment according to claim 1, wherein, It further includes: A health management module, which extracts health management feature information from the multi - feature information; Analyzing and processing the health management feature information to determine the current health degree of the welding equipment. When the health degree is less than the preset health value, a warning message is sent, and it enters the health recovery mode after completing the welding task of the current stage; the calculation formula of the health degree is: h = 1 - (a1q te + a2q po + a3q de + a4q time + a5q wdtime ), Among them, h is the health degree, q te is the temperature influence index, q po is the power consumption influence index, q de is the defect rate influence index, q time is the continuous operation duration influence index, q wdtime is the influence index corresponding to the single-weld completion duration, and a1, a2, a3, a4, and a5 are the weight coefficients corresponding to each index respectively.

Images