Continuous extrusion manufacturing method for copper and copper alloy microchannel flat tubes

Abstract

The present invention relates to the technical field of copper product manufacturing, in particular to a continuous extrusion manufacturing method for copper and copper alloy microchannel flat tubes. The method comprises: during extrusion welding manufacturing of a microchannel flat tube by means of an extrusion welding apparatus, acquiring real-time welding parameters of a welding interface at several preset time nodes, and performing redundancy removal processing on the acquired real-time welding parameters, so as to obtain redundancy-removed real-time welding parameters; on the basis of the redundancy-removed real-time welding parameters, making use of a performance prediction model to acquire the state condition of the welding interface of the microchannel flat tube after a future preset time period; and if the state condition of the welding interface of the microchannel flat tube after the future preset time period is an abrupt change state condition, regulating the welding parameters. The present invention can effectively improve the intelligent level of the extrusion welding manufacturing process of microchannel flat tubes, reduce product quality problems caused by abrupt changes in performance of welding interfaces, and improve the production efficiency and the stability of product quality.

Description

A method for continuous extrusion preparation of copper and copper alloy microchannel flat tubes Technical Field

[0001] This invention relates to the field of copper product manufacturing technology, and in particular to a method for continuous extrusion preparation of copper and copper alloy microchannel flat tubes. Background Technology

[0002] In recent years, with the increasing demand for high-performance heat dissipation materials in fields such as electronics, information technology, and aerospace, copper and copper alloy microchannel flat tubes have become a highly sought-after heat dissipation solution due to their excellent thermal conductivity, good processing performance, and high strength. Traditional microchannel flat tube fabrication methods, such as welding and tube expansion, suffer from problems such as difficulty in controlling weld quality and stress concentration, making them unsuitable for high-performance heat dissipation requirements. Continuous extrusion technology, however, has become a highly promising method for microchannel flat tube fabrication due to its ability to achieve efficient, continuous, and automated production and effectively control the weld interface quality. However, during continuous extrusion, the influence of welding parameters on the weld interface quality is complex; even slight errors can lead to defects such as incomplete fusion, porosity, and cracks, thus affecting the performance of the microchannel flat tube. Therefore, real-time monitoring and prediction of the weld interface state, and timely adjustment of welding parameters based on the prediction results to ensure weld interface quality, have become crucial for improving the efficiency and quality of continuous extrusion fabrication of copper and copper alloy microchannel flat tubes. Technical issues

[0003] Currently, monitoring the weld interface condition mainly relies on offline detection methods, such as metallographic microscopy and scanning electron microscopy. This results in low detection efficiency, making it difficult to meet the real-time requirements of continuous extrusion production. Furthermore, most existing methods lack the ability to predict the future state of the weld interface, failing to anticipate the impact of minute changes in welding parameters on the weld interface quality, thus hindering the implementation of effective preventative measures. Therefore, there is an urgent need to develop an intelligent control method capable of predicting the future state of the weld interface in real time and automatically adjusting welding parameters based on the prediction results. This would ensure the welding quality of copper and copper alloy microchannel flat tubes, improve production efficiency, reduce production costs, and meet the growing market demand. Technical solutions

[0004] This invention overcomes the shortcomings of the prior art and provides a method for continuous extrusion preparation of copper and copper alloy microchannel flat tubes.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0006] This invention discloses a method for continuous extrusion preparation of copper and copper alloy microchannel flat tubes, comprising the following steps:

[0007] Obtain performance change data of the weld interface under various historical welding parameter combinations, and construct a performance prediction model of the weld interface based on the performance change data of the weld interface under various historical welding parameter combinations.

[0008] When microchannel flat tubes are prepared by extrusion welding equipment, real-time welding parameters of the welding interface are collected at several preset time nodes, and the collected real-time welding parameters are deredundant to obtain the deredundant real-time welding parameters.

[0009] Based on the deredundant real-time welding parameters and combined with the performance prediction model, the state of the welding interface of the microchannel flat tube after a preset time period is obtained.

[0010] If the welding interface of the microchannel flat tube is in a normal state after a preset time period, no adjustment will be made to the welding parameters.

[0011] If the state of the welding interface of the microchannel flat tube changes abruptly after a preset time period, the welding parameters will be adjusted.

[0012] Furthermore, performance change data of the weld interface under various historical combinations of welding parameters are obtained. Based on this performance change data, a performance prediction model for the weld interface is constructed, specifically as follows:

[0013] The system acquires performance change data of the weld interface under various historical welding parameter combinations through a big data network, as well as the status of the performance change data.

[0014] A conditional random field is constructed, and the performance change data of the welded interface under various historical welding parameter combinations is imported into the conditional random field; and various historical welding parameter combinations are used as quantitative nodes, and the performance change data of each time stamp are used as variable nodes.

[0015] Based on the state of the performance change data, the conditional probability of each variable node transitioning from a normal state to a sudden change state under each quantitative node condition is calculated.

[0016] When the conditional probability is greater than a preset probability threshold, the corresponding variable node is marked as a mutation state node; when the conditional probability is not greater than the preset probability threshold, the corresponding variable node is marked as a normal state node.

[0017] Directed connections are made between each mutation state node, normal state node, and quantitative node to obtain the topology graph of the conditional random field; a graph embedding model is constructed based on a neural network, and the topology graph is embedded in the graph embedding model;

[0018] By utilizing the neuron structure of a neural network and based on the directed connections between quantitative nodes, nodes in a mutation state, and nodes in a normal state, the topological structure is encoded and learned through the forward propagation process of multiple neurons by continuously adjusting the neuron connection weights until the learning parameters meet the preset requirements, thus obtaining a performance prediction model for the welding interface.

[0019] The welding parameters include temperature, pressure, and extrusion speed; the conditions include normal and abrupt changes; and the performance change data include tensile strength, yield strength, hardness, grain size, defect concentration, thermal conductivity, and electrical conductivity.

[0020] Furthermore, the collected real-time welding parameters are deredundanted to obtain the deredundant real-time welding parameters, specifically:

[0021] A wavelet transform algorithm is introduced to perform discrete wavelet decomposition on the collected real-time welding parameters according to the preset wavelet basis and decomposition level to obtain the wavelet coefficients of each real-time welding parameter.

[0022] Construct a wavelet coefficient matrix based on the wavelet coefficients of each real-time welding parameter; calculate the coefficient difference between every two wavelet coefficients in the wavelet coefficient matrix;

[0023] A preset deviation threshold is used to compare the coefficient difference between every two wavelet coefficients in the wavelet coefficient matrix with the preset deviation threshold.

[0024] If there is a case in the wavelet coefficient matrix where the coefficient difference between two wavelet coefficients is not greater than the preset deviation threshold, it means that the real-time welding parameters corresponding to these two wavelet coefficients are redundant, and any real-time welding parameter with redundancy will be screened out.

[0025] The iteration stops once all wavelet coefficient pairs in the wavelet coefficient matrix have been analyzed and judged, and the deredundant real-time welding parameters are obtained.

[0026] Furthermore, based on the deredundant real-time welding parameters and combined with the performance prediction model, the state of the welding interface of the microchannel flat tube after a preset time period is obtained, specifically:

[0027] The deredundant real-time welding parameters are sorted based on the acquisition timestamp to obtain time-series-based real-time welding parameters.

[0028] The time-series-based real-time welding parameters are imported into the performance prediction model for prediction, and the predicted performance change data of the welding interface of the microchannel flat tube after a preset time period are obtained.

[0029] Obtain the preset fabrication process information of the microchannel flat tube, and obtain the state of the welding interface of the microchannel flat tube after a preset time period based on the preset fabrication process information.

[0030] The state of the weld interface after a preset time period includes a normal state and a sudden change state.

[0031] Furthermore, if the state of the welding interface of the microchannel flat tube changes abruptly after a preset time period, the welding parameters will be adjusted accordingly, specifically as follows:

[0032] If the state of the welding interface of the microchannel flat tube changes abruptly after a preset time period, then the preset extrusion welding process information of the microchannel flat tube is obtained.

[0033] Based on the preset extrusion welding process information, obtain preset performance change data of the welding interface of the microchannel flat tube after a preset time period in the future;

[0034] Calculate the absolute value of the data difference between the preset performance change data and the corresponding predicted performance change data of the welding interface of the microchannel flat tube after a preset time period in the future, and obtain the data deviation value between the preset performance change data and the corresponding predicted performance change data.

[0035] Preset the deviation range of various performance change data; determine whether the data deviation between each preset performance change data and the corresponding predicted performance change data is within the corresponding deviation range;

[0036] If the data deviation between a certain preset performance change data and the corresponding predicted performance change data is not within the corresponding deviation range, then the corresponding performance change data of the welding interface will be calibrated as the sudden performance change data after a preset time period in the future, and the sudden performance change data of the welding interface of the microchannel flat tube after a preset time period in the future will be obtained.

[0037] If the data deviation between a certain preset performance change data and the corresponding predicted performance change data is within the corresponding deviation range, then the corresponding performance change data of the welding interface is calibrated as the normal performance change data after a preset time period in the future.

[0038] Furthermore, if the state of the welding interface of the microchannel flat tube changes abruptly after a preset time period, the welding parameters are adjusted, which also includes the following steps:

[0039] A welding parameter control scheme corresponding to various abrupt performance changes in the pre-fabricated welding interface after a predetermined time period.

[0040] A control scheme pairing database is constructed, and the control schemes corresponding to the welding parameter control schemes corresponding to the various sudden performance changes of the welding interface after a preset time period are imported into the control scheme pairing database; and the control scheme pairing database is updated regularly.

[0041] Data on the abrupt performance changes of the welding interface of the microchannel flat tube after a preset time period are obtained. This data is then imported into the control scheme pairing database for matching, and the corresponding welding parameter control scheme is obtained.

[0042] The obtained welding parameter control scheme is sent to the control terminal of the extrusion welding equipment so that the corresponding real-time welding parameters in the welding chamber can be controlled based on the welding parameter control scheme to avoid sudden changes in the performance data of the welding interface.

[0043] Furthermore, during the adjustment of welding parameters, the temperature and pressure inside the welding chamber need to meet the following conditions: in, For the pressure inside the welding chamber, The temperature inside the welding chamber; It is a constant.

[0044] The present invention also discloses a continuous extrusion preparation system for copper and copper alloy microchannel flat tubes. The continuous extrusion preparation system for copper and copper alloy microchannel flat tubes includes a memory and a processor. The memory stores a method program for continuous extrusion preparation of copper and copper alloy microchannel flat tubes. When the method program for continuous extrusion preparation of copper and copper alloy microchannel flat tubes is executed by the processor, the steps of the continuous extrusion preparation method for copper and copper alloy microchannel flat tubes as described in any one of the inventions are implemented. Beneficial effects

[0045] This invention addresses the technical deficiencies in the prior art and offers the following advantages: It acquires performance change data of the weld interface under various historical combinations of weld parameters; constructs a performance prediction model for the weld interface based on this data; during the extrusion welding preparation of microchannel flat tubes using an extrusion welding device, it collects real-time welding parameters of the weld interface at several preset time points and performs redundancy removal on these parameters to obtain deredundant real-time welding parameters; based on these deredundant real-time welding parameters and the performance prediction model, it obtains the state of the weld interface of the microchannel flat tube after a preset future time period; if the state of the weld interface of the microchannel flat tube after the preset future time period is abruptly changed, the welding parameters are adjusted accordingly. This invention effectively improves the intelligence level of the microchannel flat tube extrusion welding preparation process, reduces product quality problems caused by abrupt changes in weld interface performance, and improves production efficiency and product quality stability. Attached Figure Description

[0046] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other embodiments can be obtained from these drawings without creative effort.

[0047] Figure 1 is a flowchart of the first method for the continuous extrusion preparation of copper and copper alloy microchannel flat tubes.

[0048] Figure 2 is a flowchart of the second method for the continuous extrusion preparation of copper and copper alloy microchannel flat tubes.

[0049] Figure 3 is a system block diagram of the continuous extrusion preparation system for copper and copper alloy microchannel flat tubes. Embodiments of the present invention

[0050] To better understand the above-mentioned objectives, features, and advantages of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in these embodiments can be combined with each other.

[0051] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and therefore the scope of protection of the invention is not limited to the specific embodiments disclosed below.

[0052] This invention discloses a method for continuously extruding copper and copper alloy microchannel flat tubes, as shown in Figure 1, comprising the following steps:

[0053] S102: Obtain performance change data of the weld interface under various historical welding parameter combinations, and construct a performance prediction model of the weld interface based on the performance change data of the weld interface under various historical welding parameter combinations.

[0054] S104: When microchannel flat tubes are prepared by extrusion welding equipment, real-time welding parameters of the welding interface are collected at several preset time nodes, and the collected real-time welding parameters are deredundant to obtain the deredundant real-time welding parameters.

[0055] S106: Based on the deredundant real-time welding parameters and combined with the performance prediction model, obtain the state of the welding interface of the microchannel flat tube after a preset time period in the future.

[0056] S108: If the welding interface of the microchannel flat tube is in a normal state after a preset time period in the future, the welding parameters will not be adjusted.

[0057] S110: If the state of the welding interface of the microchannel flat tube changes abruptly after a preset time period, the welding parameters will be adjusted.

[0058] It should be noted that if the predicted state of the weld interface after a preset time period is normal, it indicates that the current welding process is proceeding normally within the expected range, and no adjustment of the welding parameters is required, thus maintaining the continuity and stability of the production process. Conversely, if the predicted state is abrupt, it means that the performance of the weld interface may undergo sudden and undesirable changes. In this case, adjustment of the welding parameters is necessary to avoid quality problems and ensure the welding quality of the microchannel flat tubes. This continuous extrusion preparation method for copper and copper alloy microchannel flat tubes utilizes historical data to construct a performance prediction model, collects and processes real-time welding parameters to predict the future state of the weld interface, and determines whether to adjust the welding parameters based on the prediction results. This effectively improves the intelligence level of the microchannel flat tube extrusion welding preparation process, reduces product quality problems caused by abrupt changes in weld interface performance, improves production efficiency and product quality stability, and reduces production costs, making the entire production process more controllable and efficient.

[0059] Furthermore, performance change data of the weld interface under various historical combinations of welding parameters are obtained. Based on this performance change data, a performance prediction model for the weld interface is constructed, specifically as follows:

[0060] The system acquires performance change data of the weld interface under various historical welding parameter combinations through a big data network, as well as the status of the performance change data.

[0061] A conditional random field is constructed, and the performance change data of the welded interface under various historical welding parameter combinations is imported into the conditional random field; and various historical welding parameter combinations are used as quantitative nodes, and the performance change data of each time stamp are used as variable nodes.

[0062] Based on the state of the performance change data, the conditional probability of each variable node transitioning from a normal state to a sudden change state under each quantitative node condition is calculated.

[0063] When the conditional probability is greater than a preset probability threshold, the corresponding variable node is marked as a mutation state node; when the conditional probability is not greater than the preset probability threshold, the corresponding variable node is marked as a normal state node.

[0064] Directed connections are made between each mutation state node, normal state node, and quantitative node to obtain the topology graph of the conditional random field; a graph embedding model is constructed based on a neural network, and the topology graph is embedded in the graph embedding model;

[0065] In this process, based on the association logic between the mutation state nodes and the normal state nodes under each quantitative node, starting from the quantitative node, and according to the state transition relationship of the performance change data, the mutation state nodes and the normal state nodes are respectively connected to the quantitative node in a directed manner, thereby constructing the topology diagram of the conditional random field.

[0066] By utilizing the neuron structure of a neural network and based on the directed connections between quantitative nodes, nodes in a mutation state, and nodes in a normal state, the topological structure is encoded and learned through the forward propagation process of multiple neurons by continuously adjusting the neuron connection weights until the learning parameters meet the preset requirements, thus obtaining a performance prediction model for the welding interface.

[0067] The welding parameters include temperature, pressure, and extrusion speed; the conditions include normal and abrupt changes; and the performance change data include tensile strength, yield strength, hardness, grain size, defect concentration, thermal conductivity, and electrical conductivity.

[0068] It should be noted that, firstly, a big data network is used to collect performance change data of the weld interface under different combinations of historical welding parameters (temperature, pressure, extrusion speed), including various performance data such as tensile strength and yield strength. Simultaneously, the state status (normal or abrupt change) corresponding to this performance data is acquired. This step provides foundational data for subsequent model construction; the big data network can encompass rich historical information, facilitating a comprehensive analysis of various situations during the welding process. A conditional random field (CRF) is constructed, and the performance change data is imported into it. Historical welding parameter combinations are set as quantitative nodes, and performance change data corresponding to timestamps are set as variable nodes. This setup reasonably correlates welding parameters and performance data within the model structure, reflecting the relationship between performance changes over time under different parameter combinations. Based on the state status of the performance data, the conditional probability of a variable node transitioning from a normal state to an abrupt state at each quantitative node is calculated. By comparing this probability with a preset probability threshold, variable nodes are marked as abrupt state nodes or normal state nodes. This helps identify situations where performance data may undergo abrupt changes under specific welding parameter combinations, providing a basis for the subsequent construction of the model's topology. Directed connections are made between abrupt state nodes, normal state nodes, and quantitative nodes to obtain the topology diagram of the CRF. This topology reflects the logical relationships between nodes and is an abstract representation of the relationships between various factors during the welding process. A graph embedding model is constructed based on a neural network, and the topology graph is embedded within it. This process allows the model to learn the underlying patterns in the node relationships, continuously optimizing model parameters until preset requirements are met, ultimately resulting in a performance prediction model.

[0069] Through the aforementioned steps, a weld interface performance prediction model based on big data and neural networks was constructed. This model can fully utilize parameter combinations, performance variation data, and status conditions in historical weld data to uncover the potential patterns of performance changes under different combinations of weld parameters. By constructing and learning conditional random fields and graph embedding models, the model can accurately predict the performance status of the weld interface under different weld parameters and provide early warnings of potential performance abrupt changes, thereby providing strong technical support for optimizing welding processes and improving product quality.

[0070] Furthermore, the collected real-time welding parameters are deredundanted to obtain the deredundant real-time welding parameters, as shown in Figure 2. Specifically:

[0071] S202: Introducing a wavelet transform algorithm, the collected real-time welding parameters are discretized by wavelet decomposition according to the preset wavelet basis and decomposition level to obtain the wavelet coefficients of each real-time welding parameter;

[0072] S204: Construct a wavelet coefficient matrix based on the wavelet coefficients of each real-time welding parameter; calculate the coefficient difference between every two wavelet coefficients in the wavelet coefficient matrix;

[0073] S206: Preset deviation threshold, which compares the coefficient difference between every two wavelet coefficients in the wavelet coefficient matrix with the preset deviation threshold.

[0074] S208: If there is a situation in the wavelet coefficient matrix where the coefficient difference between two wavelet coefficients is not greater than the preset deviation threshold, it means that the real-time welding parameters corresponding to these two wavelet coefficients are redundant, and any real-time welding parameter with redundancy will be screened out.

[0075] S210: Stop iterating until all wavelet coefficient pairs in the wavelet coefficient matrix have been analyzed and judged, and obtain the deredundant real-time welding parameters.

[0076] It should be noted that the selection of the wavelet basis and the setting of the decomposition level are determined based on the characteristics of the welding parameters and the needs of subsequent analysis. This decomposition transforms the welding parameters from the time domain to a multi-scale space such as the frequency domain, obtaining the wavelet coefficients of each real-time welding parameter. These wavelet coefficients contain information about the welding parameters at different scales. A wavelet coefficient matrix is ​​constructed based on the obtained wavelet coefficients of each real-time welding parameter. This matrix comprehensively reflects the relationship between all real-time welding parameters at different scales. Then, the coefficient difference between every two wavelet coefficients in the matrix is ​​calculated. These differences can be used to measure the similarity or difference between different wavelet coefficients. A preset deviation threshold is used, and the coefficient difference between every two wavelet coefficients is compared with this threshold. If the coefficient difference between two wavelet coefficients is not greater than the preset deviation threshold, it means that the real-time welding parameters corresponding to these two wavelet coefficients are similar to some extent, exhibiting redundancy. In this case, any real-time welding parameter with redundancy is removed to reduce data redundancy. The wavelet coefficient pairs in the wavelet coefficient matrix are analyzed sequentially using the method described above until all wavelet coefficient pairs have been processed. The iteration then stops, ultimately yielding the deredundant real-time welding parameters. This process ensures that all potentially redundant welding parameters are comprehensively checked and processed.

[0077] The above steps effectively remove redundant information from the acquired real-time welding parameters by deduplicated processing. Wavelet transform algorithms are used to analyze the welding parameters at multiple scales, constructing a wavelet coefficient matrix and accurately identifying and filtering out redundant parameters by comparing coefficient differences with thresholds. This helps reduce the amount of data, improves the efficiency and accuracy of subsequent analysis, prediction, or control operations based on welding parameters, and ensures that the processing results better reflect the essential characteristics of the welding process, avoiding interference and misjudgments caused by redundant data.

[0078] Furthermore, based on the deredundant real-time welding parameters and combined with the performance prediction model, the state of the welding interface of the microchannel flat tube after a preset time period is obtained, specifically:

[0079] The deredundant real-time welding parameters are sorted based on the acquisition timestamp to obtain time-series-based real-time welding parameters.

[0080] The time-series-based real-time welding parameters are imported into the performance prediction model for prediction, and the predicted performance change data of the welding interface of the microchannel flat tube after a preset time period are obtained.

[0081] Obtain the preset fabrication process information of the microchannel flat tube, and obtain the state of the welding interface of the microchannel flat tube after a preset time period based on the preset fabrication process information.

[0082] The state of the weld interface after a preset time period includes a normal state and a sudden change state.

[0083] It should be noted that, firstly, after obtaining the deredundant real-time welding parameters, these parameters are sorted according to the acquisition timestamps. The acquisition timestamps record the acquisition order of each real-time welding parameter. By sorting, these parameters can be arranged in chronological order to obtain time-series-based real-time welding parameters. This step is to ensure that the order of the data conforms to the actual welding process progress, so as to facilitate accurate predictive analysis later. The time-series-based real-time welding parameters are then imported into a pre-built performance prediction model. This performance prediction model is built based on the performance change data under historical welding parameter combinations. It can predict the predicted performance changes of the welding interface of the microchannel flat tube after a preset time period based on the input real-time welding parameters, through the algorithm and parameter relationships within the model. These predicted performance change data cover changes in various performance indicators such as tensile strength and yield strength. Simultaneously, the preset manufacturing process information of the microchannel flat tube is obtained. The preset manufacturing process information includes process parameter requirements such as ideal temperature range, pressure range, and extrusion speed during normal manufacturing. Based on this pre-set preparation process information, combined with the previously obtained predicted performance change data, we can determine the state of the welding interface of the microchannel flat tube after a pre-set time period, that is, whether it is a normal state (meets the pre-set process requirements and all performance indicators are stable) or a sudden change state (performance indicators may suddenly change and deviate from the pre-set process requirements).

[0084] By combining the deduplicated real-time welding parameters, performance prediction models, and pre-set manufacturing process information, the state of the welding interface of the microchannel flat tube can be accurately determined after a predetermined time period. This helps to know the state trend of the welding interface in advance during the continuous extrusion manufacturing process of microchannel flat tubes, so as to promptly detect possible anomalies (such as abrupt changes), thereby providing a basis for adjusting welding parameters or taking other intervention measures, ensuring the manufacturing quality and production efficiency of microchannel flat tubes.

[0085] Furthermore, if the state of the welding interface of the microchannel flat tube changes abruptly after a preset time period, the welding parameters will be adjusted accordingly, specifically as follows:

[0086] If the state of the welding interface of the microchannel flat tube changes abruptly after a preset time period, then the preset extrusion welding process information of the microchannel flat tube is obtained.

[0087] Based on the preset extrusion welding process information, obtain preset performance change data of the welding interface of the microchannel flat tube after a preset time period in the future;

[0088] Calculate the absolute value of the data difference between the preset performance change data and the corresponding predicted performance change data of the welding interface of the microchannel flat tube after a preset time period in the future, and obtain the data deviation value between the preset performance change data and the corresponding predicted performance change data.

[0089] Preset the deviation range of various performance change data; determine whether the data deviation between each preset performance change data and the corresponding predicted performance change data is within the corresponding deviation range;

[0090] If the data deviation between a certain preset performance change data and the corresponding predicted performance change data is not within the corresponding deviation range, then the corresponding performance change data of the welding interface will be calibrated as the sudden performance change data after a preset time period in the future, and the sudden performance change data of the welding interface of the microchannel flat tube after a preset time period in the future will be obtained.

[0091] If the data deviation between a certain preset performance change data and the corresponding predicted performance change data is within the corresponding deviation range, then the corresponding performance change data of the welding interface is calibrated as the normal performance change data after a preset time period in the future.

[0092] It should be noted that when it is determined that the welding interface of the microchannel flat tube will undergo a sudden change after a predetermined time period, the predetermined extrusion welding process information of the microchannel flat tube is first obtained. This information includes the process parameter requirements for the welding interface of the microchannel flat tube under ideal extrusion welding, such as suitable temperature range, pressure range, extrusion speed, etc., as well as the ideal performance change data corresponding to these process parameters, i.e., the predetermined performance change data. These predetermined performance change data are expected to be obtained under normal process conditions and cover performance indicators such as tensile strength and yield strength. Next, the absolute value of the difference between each predetermined performance change data and the corresponding predicted performance change data after the predetermined time period is calculated to obtain the data deviation value. This deviation value reflects the degree of deviation between the predicted performance and the ideal performance. The predicted performance change data is obtained through the performance prediction model, while the predetermined performance change data is derived from the ideal process; the deviation between the two reflects the difference between the actual situation and the expected situation. The deviation value range of various performance change data is predetermined. Different performance change data may have different acceptable deviation ranges, depending on factors such as process requirements and product quality standards. Then, it is determined whether the data deviation between each preset performance change data and the corresponding predicted performance change data is within the corresponding deviation range. This step is to more accurately determine which performance change data deviates significantly from the normal range and which are within the acceptable range. If the data deviation between a certain preset performance change data and the corresponding predicted performance change data is not within the corresponding deviation range, the corresponding performance change data of the welding interface is marked as a sudden performance change data after a preset future time period. This indicates that the performance indicator may experience a large abnormal change after the preset future time period. Conversely, if the deviation value is within the corresponding deviation range, it is marked as normal performance change data after the preset future time period, indicating that the performance indicator is within the acceptable fluctuation range.

[0093] By following the steps above, we can accurately identify which performance data of the microchannel flat tube welding interface may undergo abrupt changes and which remain within the normal range after a predetermined time period. This helps to accurately identify the specific performance factors that cause the welding interface to be in abrupt state, providing a detailed basis for subsequent targeted adjustment of welding parameters. This allows for more effective adjustment of welding parameters, avoids undesirable performance changes at the welding interface, and ensures the extrusion welding quality of the microchannel flat tube.

[0094] Furthermore, if the state of the welding interface of the microchannel flat tube changes abruptly after a preset time period, the welding parameters are adjusted, which also includes the following steps:

[0095] A welding parameter control scheme corresponding to various abrupt performance changes in the pre-fabricated welding interface after a predetermined time period.

[0096] A control scheme pairing database is constructed, and the control schemes corresponding to the welding parameter control schemes corresponding to the various sudden performance changes of the welding interface after a preset time period are imported into the control scheme pairing database; and the control scheme pairing database is updated regularly.

[0097] Data on the abrupt performance changes of the welding interface of the microchannel flat tube after a preset time period are obtained. This data is then imported into the control scheme pairing database for matching, and the corresponding welding parameter control scheme is obtained.

[0098] The obtained welding parameter control scheme is sent to the control terminal of the extrusion welding equipment so that the corresponding real-time welding parameters in the welding chamber can be controlled based on the welding parameter control scheme to avoid sudden changes in the performance data of the welding interface.

[0099] It should be noted that, firstly, relevant technical personnel develop corresponding welding parameter control schemes in advance, based on data regarding potential abrupt changes in the performance of the weld interface over a predetermined period. This requires a deep understanding of the relationship between various performance changes during the welding process and the welding parameters. For example, if a sudden decrease in tensile strength is predicted, possible control schemes include adjusting parameters such as welding temperature, pressure, or extrusion speed to improve tensile strength. These control schemes are based on experience and theoretical knowledge of the continuous extrusion process for preparing microchannel flat tubes of copper and copper alloys.

[0100] A matching database of control schemes is constructed, and pre-made control schemes are imported into it. This database establishes a correspondence between sudden performance changes and welding parameter control schemes, facilitating rapid subsequent queries and matching. Regular updates to the matching database are necessary to adapt to changing production conditions, raw material characteristics, or more precise process requirements. For example, with the application of new copper alloy materials or upgrades to extrusion equipment, existing control schemes may need adjustment; updating the database ensures it always contains the most effective control schemes. When sudden performance changes in the welding interface of the microchannel flat tube are obtained after a predetermined time period, this data is imported into the matching database for matching. Through the pre-defined correspondences in the database, the corresponding welding parameter control schemes can be quickly and accurately obtained. This process enables a rapid transformation from problem (sudden performance changes) to solution (welding parameter control schemes), improving the efficiency of handling sudden changes in the welding interface. Finally, the obtained welding parameter control schemes are transmitted to the control terminal of the extrusion welding equipment. The extrusion welding equipment adjusts the corresponding real-time welding parameters in the welding chamber according to the received control schemes. For example, if the control scheme requires an increase in welding temperature, the equipment will adjust parameters such as the power of the heating device accordingly. This is done to avoid sudden changes in performance data at the weld interface, thereby ensuring the welding quality of the microchannel flat tube and improving product stability and reliability.

[0101] By pre-designing control schemes, constructing and updating databases, matching and acquiring control schemes, and adjusting welding parameters, this method can quickly and effectively address potential performance abrupt changes at the welding interface during the continuous extrusion process of microchannel flat tubes. This improves the controllability and stability of the production process, reduces product quality issues caused by sudden performance changes at the welding interface, increases production efficiency, reduces production costs, and ensures that the quality of the microchannel flat tubes meets requirements.

[0102] Furthermore, during the adjustment of welding parameters, the temperature and pressure inside the welding chamber need to meet the following conditions: in, For the pressure inside the welding chamber, The temperature inside the welding chamber; It is a constant.

[0103] The continuous extrusion preparation method for copper and copper alloy microchannel flat tubes may also include the following steps:

[0104] After the microchannel flat tube is extruded and welded, acoustic wave data fed back from the welding interface of the microchannel flat tube is obtained based on the acoustic wave device, and a characteristic three-dimensional model of the welding interface is constructed based on the acoustic wave data.

[0105] Obtain the preset welding engineering drawing information of the microchannel flat tube welding interface, and construct a preset three-dimensional model of the welding interface based on the preset welding engineering drawing information;

[0106] Construct a KD tree space, import the feature 3D model diagram and the preset 3D model diagram into the KD tree space, and retrieve the welding positioning reference in the feature 3D model diagram and the preset 3D model diagram;

[0107] The feature 3D model image and the preset 3D model image are integrated and paired according to the welding positioning reference to obtain an integrated 3D model image; and the integrated 3D model image is divided into several super rectangular regions based on the KD tree space.

[0108] Determine whether each super-rectangular region contains a feature 3D model image and a preset 3D model image; if a super-rectangular region contains neither a feature 3D model image nor a preset 3D model image, then mark the super-rectangular region as a blank region;

[0109] If a super-rectangular region contains both a feature 3D model map and a preset 3D model map, then the super-rectangular region is marked as a normal region.

[0110] If a certain super-rectangular region contains only a feature 3D model image, or only a preset 3D model image, then the super-rectangular region is marked as a defect region; and the relative coordinate information between the defect region and the welding positioning reference is obtained.

[0111] Repeat the above steps until all super-rectangular regions have been judged, and obtain several defect regions of the welding interface and the relative coordinate information between each defect region and the welding positioning reference.

[0112] The least squares curve fitting algorithm is introduced to fit the defect distribution trend map of the weld interface based on the relative coordinate information between each defect area and the welding positioning reference and the least squares curve fitting algorithm.

[0113] The welding parameters in the welding chamber are adjusted and optimized based on the obtained defect distribution trend map.

[0114] It should be noted that after the extrusion welding is completed, a KD tree space is constructed and the two 3D model images are imported into it. The KD tree space is a data structure used for efficient storage and retrieval of multidimensional data. Welding positioning references are retrieved from the two model images within this space; these references are key points in the model that determine position and orientation. Then, the two model images are integrated and paired based on these references to obtain an integrated 3D model image. Next, the integrated 3D model image is divided into several hyperrectangular regions based on the KD tree space. This division method facilitates local analysis of the model, allowing for the examination of each hyperrectangular region. If a region contains neither a feature 3D model image nor a preset 3D model image, it is marked as a blank region. If both exist, it is a normal region, indicating that the welding in that region meets expectations. Regions containing only one model image are marked as defective regions, and their relative coordinate information with the welding positioning reference is obtained. By examining each hyperrectangular region, defective regions and their location information on the weld interface can be comprehensively identified. Using a least squares curve fitting algorithm, a defect distribution trend map is fitted based on the relative coordinate information of the defective regions and the welding positioning reference. The least squares method finds the most suitable curve for the data by minimizing the sum of squares of the errors, thus intuitively showing the distribution trend of defects at the weld interface. Finally, the welding parameters in the welding chamber are adjusted and optimized based on this trend chart. For example, based on the defect distribution trend chart, the areas and directions of defect concentration are analyzed; if the trend chart shows that defects are concentrated in a certain area and close to the welding start point, the initial welding parameters (such as temperature and pressure) are inappropriate, and the temperature or pressure at the start point should be appropriately increased; if defects are linearly distributed along a specific direction, it may be due to uneven extrusion speed, and the extrusion speed should be adjusted to stabilize it; if defects are concentrated at the edge of the weld interface, the die pressure distribution is checked, and the edge pressure is adjusted; for different defect distribution characteristics, the welding parameters in the welding chamber are specifically adjusted and optimized in terms of temperature, pressure, and extrusion speed.

[0115] Through the above series of steps, a characteristic 3D model of the actual welding interface can be constructed using acoustic wave data, and compared and analyzed with a preset 3D model. By efficiently partitioning and analyzing the integrated model using KD tree space, defect areas of the welding interface can be accurately located and their positional information obtained. Then, a defect distribution trend map is fitted using the least squares method, providing an intuitive and accurate basis for the control and optimization of welding parameters in the welding chamber. This helps improve the welding quality during the continuous extrusion process of copper and copper alloy microchannel flat tubes, reduce defects, and improve the overall performance of the product.

[0116] The continuous extrusion preparation method for copper and copper alloy microchannel flat tubes may also include the following steps:

[0117] The real-time welding temperature information of several preset position nodes of the welding interface is obtained, and a real-time temperature distribution map of the welding interface is constructed based on the real-time welding temperature information of each preset position node.

[0118] Obtain a preset temperature distribution map of the welding interface at the current preset time node, and calculate the structural similarity index between the real-time temperature distribution map and the preset temperature distribution map;

[0119] If the structural similarity index is not greater than the preset index value, then feature extraction processing is performed on the real-time temperature distribution map and the preset temperature distribution map to obtain the real-time isotherm map and the preset isotherm map.

[0120] The real-time isotherm map is paired with the preset isotherm map for analysis to obtain the isotherm non-overlapping regions, which are defined as temperature singular regions.

[0121] The temperature control devices corresponding to the temperature anomaly regions are marked, and the real-time operating parameters of the marked temperature control devices are obtained.

[0122] A Markov model is introduced, and the real-time operating parameters of the marked temperature control equipment are imported into the Markov model for fault prediction to obtain the state transition probability value of the marked temperature control equipment.

[0123] If the state transition probability value of the marked temperature control equipment is greater than the preset probability value, the extrusion welding equipment will be shut down and a fault warning message will be generated.

[0124] If the state transition probability value of the marked temperature control device is not greater than the preset probability value, then the preset operating parameters of the marked temperature control device are obtained, the difference between the preset operating parameters and the real-time operating parameters is calculated, the operating parameter deviation value is obtained, and the real-time operating parameters of the marked temperature control device are adjusted according to the operating parameter deviation value.

[0125] It should be noted that, firstly, real-time welding temperature information at multiple preset node positions on the welding interface is acquired to construct a real-time temperature distribution map, which reflects the current temperature distribution of the welding interface. Simultaneously, a preset temperature distribution map at the same time point is acquired; this map is based on an ideal welding process temperature distribution pattern. Then, a structural similarity index is calculated between the two, which measures the similarity between the real-time temperature distribution and the ideal distribution. If the index is not greater than the preset index value, it indicates a significant difference between the real-time temperature distribution and the ideal situation. When the difference between the real-time and preset temperature distributions is large, feature extraction processing is performed to obtain a real-time isotherm map and a preset isotherm map. The isotherm map can more intuitively show areas with the same temperature. By performing paired analysis on these two isotherm maps, non-overlapping areas are identified; these areas are defined as temperature singularities. Temperature singularities are areas where the actual temperature differs significantly from the ideal temperature, possibly due to equipment failure or process instability. Temperature control equipment corresponding to the temperature singularities is marked, as the temperature anomalies in these areas may be related to the operating status of these devices. Real-time operating parameters of the tagged temperature control equipment are acquired, reflecting the equipment's current operating status, such as heating power and cooling rate. A Markov model is introduced, and the real-time operating parameters are imported into it for fault prediction, yielding state transition probability values. The Markov model can predict the likelihood of future failures based on the equipment's current operating parameter status. If the state transition probability value is greater than a preset probability value, it indicates a significant risk of equipment failure. In this case, the extrusion welding equipment is shut down, and a fault warning is generated to prevent potentially more serious problems. If the state transition probability value is not greater than the preset probability value, the preset operating parameters of the equipment are acquired, and the difference between these and the real-time operating parameters is calculated to obtain the operating parameter deviation value. Based on this deviation value, the real-time operating parameters of the equipment are adjusted to bring the equipment's operating status closer to the ideal state, thereby adjusting the temperature distribution at the welding interface. Through these steps, the difference between the welding interface temperature distribution and the ideal situation can be detected in a timely manner, locating temperature anomalies and identifying potentially problematic temperature control equipment. The Markov model is used to predict equipment failure risks, and corresponding measures, such as shutdown warnings or parameter adjustments, are taken based on the prediction results. This helps improve the temperature control accuracy during the continuous extrusion process of copper and copper alloy microchannel flat tubes, reduce product quality problems caused by abnormal temperatures, ensure the normal operation of equipment, and improve the stability and reliability of the production process.

[0126] The present invention also discloses a continuous extrusion preparation system for copper and copper alloy microchannel flat tubes, as shown in Figure 3. The continuous extrusion preparation system for copper and copper alloy microchannel flat tubes includes a memory 50 and a processor 80. The memory 50 stores a method program for continuous extrusion preparation of copper and copper alloy microchannel flat tubes. When the method program for continuous extrusion preparation of copper and copper alloy microchannel flat tubes is executed by the processor 80, the steps of the continuous extrusion preparation method for copper and copper alloy microchannel flat tubes described in any one of the present invention are implemented.

[0127] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods, such as: multiple units or components can be combined, or integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the various components shown or discussed can be through some interfaces, and the indirect coupling or communication connection between devices or units can be electrical, mechanical, or other forms.

[0128] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units. They may be located in one place or distributed across multiple network units. Some or all of the units may be selected to achieve the purpose of this embodiment according to actual needs.

[0129] In addition, in the various embodiments of the present invention, each functional unit can be integrated into one processing unit, or each unit can be a separate unit, or two or more units can be integrated into one unit; the integrated unit can be implemented in hardware or in the form of hardware plus software functional units.

[0130] Those skilled in the art will understand that all or part of the steps of the above method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it performs the steps of the above method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0131] Alternatively, if the integrated units of this invention are implemented as software functional modules and sold or used as independent products, they can also be stored in a computer-readable storage medium. Based on this understanding, the technical solutions of the embodiments of this invention, or the parts that contribute to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the methods of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, ROM, RAM, magnetic disks, or optical disks. Industrial applicability

[0132] The above are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for continuously extruding a copper and copper alloy microchannel flat tube, characterized by comprising: Includes the following steps: Obtain performance change data of the weld interface under various historical welding parameter combinations, and construct a performance prediction model of the weld interface based on the performance change data of the weld interface under various historical welding parameter combinations. When microchannel flat tubes are prepared by extrusion welding equipment, real-time welding parameters of the welding interface are collected at several preset time nodes, and the collected real-time welding parameters are deredundant to obtain the deredundant real-time welding parameters. Based on the deredundant real-time welding parameters and combined with the performance prediction model, the state of the welding interface of the microchannel flat tube after a preset time period is obtained. If the welding interface of the microchannel flat tube is in a normal state after a preset time period, no adjustment will be made to the welding parameters. If the state of the welding interface of the microchannel flat tube changes abruptly after a preset time period, the welding parameters will be adjusted.

2. The method of claim 1, wherein the method is a continuous extrusion method for manufacturing a copper and copper alloy micro-channel flat tube. Obtain performance variation data of the weld interface under various historical combinations of welding parameters, and construct a performance prediction model for the weld interface based on this data. Specifically: The system acquires performance change data of the weld interface under various historical welding parameter combinations through a big data network, as well as the status of the performance change data. A conditional random field is constructed, and the performance change data of the weld interface under various historical welding parameter combinations are imported into the conditional random field; And various combinations of historical welding parameters are used as quantitative nodes, and performance change data at each time stamp are used as variable nodes; Based on the state of the performance change data, the conditional probability of each variable node transitioning from a normal state to a sudden change state under each quantitative node condition is calculated. When the conditional probability is greater than a preset probability threshold, the corresponding variable node is marked as a mutation state node; when the conditional probability is not greater than the preset probability threshold, the corresponding variable node is marked as a normal state node. Directed connections are made between each mutation state node, normal state node, and quantitative node to obtain the topology graph of the conditional random field; a graph embedding model is constructed based on a neural network, and the topology graph is embedded in the graph embedding model; By utilizing the neuronal structure of a neural network and based on the directed connections between quantitative nodes, mutation state nodes, and normal state nodes, the topological structure is encoded and learned through the forward propagation process of multi-layer neurons by continuously adjusting the neuronal connection weights. The performance prediction model of the welding interface is obtained after the learning parameters meet the preset requirements. The welding parameters include temperature, pressure, and extrusion speed; the conditions include normal and abrupt changes; and the performance change data include tensile strength, yield strength, hardness, grain size, defect concentration, thermal conductivity, and electrical conductivity.

3. The method of claim 1, wherein the method is a continuous extrusion method for manufacturing a copper and copper alloy micro-channel flat tube. The collected real-time welding parameters are deredundantd to obtain the deredundant real-time welding parameters, specifically: A wavelet transform algorithm is introduced to perform discrete wavelet decomposition on the collected real-time welding parameters according to the preset wavelet basis and decomposition level to obtain the wavelet coefficients of each real-time welding parameter. Construct a wavelet coefficient matrix based on the wavelet coefficients of each real-time welding parameter; Calculate the coefficient difference between every two wavelet coefficients in the wavelet coefficient matrix; A preset deviation threshold is used to compare the coefficient difference between every two wavelet coefficients in the wavelet coefficient matrix with the preset deviation threshold. If there is a case in the wavelet coefficient matrix where the coefficient difference between two wavelet coefficients is not greater than the preset deviation threshold, it means that the real-time welding parameters corresponding to these two wavelet coefficients are redundant, and any real-time welding parameter with redundancy will be screened out. The iteration stops once all wavelet coefficient pairs in the wavelet coefficient matrix have been analyzed and judged, and the deredundant real-time welding parameters are obtained.

4. The method of claim 1 wherein the method is a continuous extrusion process for the production of copper and copper alloy microchannel flat tubes. Based on the deredundant real-time welding parameters and combined with the performance prediction model, the state of the welding interface of the microchannel flat tube after a preset time period is obtained, specifically: The deredundant real-time welding parameters are sorted based on the acquisition timestamp to obtain time-series-based real-time welding parameters. The time-series-based real-time welding parameters are imported into the performance prediction model for prediction, and the predicted performance change data of the welding interface of the microchannel flat tube after a preset time period are obtained. Obtain the preset fabrication process information of the microchannel flat tube, and obtain the state of the welding interface of the microchannel flat tube after a preset time period based on the preset fabrication process information. The state of the weld interface after a preset time period includes a normal state and a sudden change state.

5. The method of claim 4, wherein the copper and copper alloy micro- channel flat tube is continuously extruded. If the state of the welding interface of the microchannel flat tube changes abruptly after a preset time period, the welding parameters will be adjusted accordingly, specifically as follows: If the state of the welding interface of the microchannel flat tube changes abruptly after a preset time period, then the preset extrusion welding process information of the microchannel flat tube is obtained. Based on the preset extrusion welding process information, obtain preset performance change data of the welding interface of the microchannel flat tube after a preset time period in the future; Calculate the absolute value of the data difference between the preset performance change data and the corresponding predicted performance change data of the welding interface of the microchannel flat tube after a preset time period in the future, and obtain the data deviation value between the preset performance change data and the corresponding predicted performance change data. Preset the deviation range of various performance change data; determine whether the data deviation between each preset performance change data and the corresponding predicted performance change data is within the corresponding deviation range; If the data deviation between a certain preset performance change data and the corresponding predicted performance change data is not within the corresponding deviation range, then the corresponding performance change data of the welding interface will be calibrated as the sudden performance change data after a preset time period in the future, and the sudden performance change data of the welding interface of the microchannel flat tube after a preset time period in the future will be obtained. If the data deviation between a certain preset performance change data and the corresponding predicted performance change data is within the corresponding deviation range, then the corresponding performance change data of the welding interface is calibrated as the normal performance change data after a preset time period in the future.

6. The method of claim 5, wherein the copper and copper alloy micro- channel flat tube is continuously extruded. If the state of the welding interface of the microchannel flat tube changes abruptly after a preset time period, the welding parameters will be adjusted, including the following steps: A welding parameter control scheme corresponding to various abrupt performance changes in the pre-fabricated welding interface after a predetermined time period. A control scheme pairing database is constructed, and the control schemes corresponding to the welding parameter control schemes corresponding to the various sudden performance changes of the welding interface after a preset time period are imported into the control scheme pairing database; and the control scheme pairing database is updated regularly. Data on the abrupt performance changes of the welding interface of the microchannel flat tube after a preset time period are obtained. This data is then imported into the control scheme pairing database for matching, and the corresponding welding parameter control scheme is obtained. The obtained welding parameter control scheme is sent to the control terminal of the extrusion welding equipment so that the corresponding real-time welding parameters in the welding chamber can be controlled based on the welding parameter control scheme to avoid sudden changes in the performance data of the welding interface.

7. The method of claim 1 wherein the microchannel flat tube is made of copper or copper alloy. During the adjustment of welding parameters, the temperature and pressure inside the welding chamber must meet the following conditions: The method for continuously extruding copper and copper alloy micro-channel flat tubes according to claim 1, wherein during the regulation of the welding parameters, the temperature and pressure in the welding chamber need to satisfy the following conditions: Wherein, P is the pressure in the welding chamber, T is the temperature in the welding chamber; K is a constant.

8. A continuous extrusion system for the production of copper and copper alloy microchannel flat tubes, characterized by, The continuous extrusion preparation system for copper and copper alloy microchannel flat tubes includes a memory and a processor. The memory stores a method program for the continuous extrusion preparation of copper and copper alloy microchannel flat tubes. When the processor executes the method program for the continuous extrusion preparation of copper and copper alloy microchannel flat tubes, it implements the steps of the continuous extrusion preparation method for copper and copper alloy microchannel flat tubes as described in any one of claims 1 to 7.

Images