Wave-soldering process monitoring method, electronic device, and storage medium

Abstract

The application relates to the technical field of electronic manufacturing welding process monitoring, and specifically provides a wave-soldering process monitoring method, an electronic device and a storage medium. The method comprises the following steps: obtaining the unique identity of a workpiece to be processed, and synchronously and continuously collecting multi-source heterogeneous physical field parameters of the workpiece to be processed; calculating a tin wave flatness parameter according to real-time flat wave area tin wave height and real-time disturbance wave area tin wave height; associating and binding the unique identity with the tin wave flatness parameter and parameters in the multi-source heterogeneous physical field parameters except the real-time flat wave area tin wave height and the real-time disturbance wave area tin wave height, so as to obtain a process data set corresponding to the workpiece to be processed; and based on a preset multi-parameter coupling judgment logic, the process data set is associated and analyzed to identify a compound process risk. The method can effectively solve the problems that in the prior art, parameter collection is not synchronous, data cannot be traced back, and multi-parameter linkage analysis is lacking, so that a compound process degradation cannot be identified.

Description

Technical Field

[0001] This application relates to the field of electronic manufacturing welding process monitoring technology, and more specifically, to a wave soldering process monitoring method, electronic device and storage medium. Background Technology

[0002] In the electronics manufacturing industry, wave soldering is a key process for soldering components onto printed circuit boards. Existing wave soldering process monitoring solutions are mainly divided into two categories, but both have significant technical limitations.

[0003] The first type of solution uses a single-dimensional temperature measurement device, which can only collect temperature data of the printed circuit board, and the resulting curves are usually generated after the fact, making it impossible to achieve real-time monitoring during the soldering process. In addition, this type of solution only focuses on thermal parameters and lacks the ability to simultaneously collect key process variables such as solder wave height, track vibration, flux flow rate, and oxygen content, resulting in the inability to perform linked analysis on complex soldering defects such as insufficient solder, component misalignment, and solder joint oxidation.

[0004] The second type of solution uses discrete single-parameter monitoring equipment. While it can separately perform laser measurement of solder wave, chain speed acquisition, vibration detection, oxygen content monitoring, and flux flow metering, it has several shortcomings. First, the independent acquisition by each sensor and the lack of a unified high-precision clock synchronization lead to misaligned timestamps for multiple parameters, making it impossible to correspond to the complete soldering cycle of the same printed circuit board. Second, the lack of a hardware-linked barcode scanning module prevents the binding of multi-physics data to the unique identifier of each board, making it impossible to simultaneously trace the causes of soldering defects across all dimensions, including temperature, solder wave, mechanical factors, atmosphere, and fluids. Third, solder wave height monitoring relies solely on single-point laser ranging, failing to quantify the height difference between smooth and turbulent waves, making it difficult to assess solder wave flatness and thus unable to provide early warnings of solder pump impeller wear. Furthermore, track vibration monitoring can only provide simple over-limit alarms, lacking independent quantification and split-axis process capability analysis for the X, Y, and Z axes, making it impossible to accurately distinguish between faults such as guide rail loosening or motor vibration.

[0005] Existing solutions mostly rely on single-parameter independent alarms, lacking coupled judgment logic for multi-parameter linkage. This results in the inability to identify complex process degradation issues such as chain speed fluctuations and track vibrations, and abnormal tin waves and insufficient preheating in advance.

[0006] There is currently no effective technical solution to the above problems. Summary of the Invention

[0007] The purpose of this application is to provide a wave soldering process monitoring method, electronic device and storage medium, which can effectively solve the problems of asynchronous parameter acquisition, untraceable data and lack of multi-parameter linkage analysis in the prior art, which leads to the inability to identify complex process degradation.

[0008] In a first aspect, this application provides a method for monitoring wave soldering processes, which includes the following steps: S1. Obtain the unique identifier of the workpiece to be processed, and synchronously and continuously collect the multi-source heterogeneous physical field parameters of the workpiece to be processed; the multi-source heterogeneous physical field parameters include real-time workpiece temperature, real-time chain speed, real-time solder wave height in the smoothing area, real-time solder wave height in the turbulence area, real-time X-axis vibration amplitude, real-time Y-axis vibration amplitude, real-time Z-axis vibration amplitude, real-time flux flow rate, and real-time oxygen content. S2. Calculate the tin wave smoothness parameters based on the real-time tin wave height in the smooth wave area and the real-time tin wave height in the turbulence area. S3. Associate and bind the unique identifier with the tin wave flatness parameter and the parameters of the multi-source heterogeneous physical field parameters, except for the tin wave height in the real-time flat wave region and the tin wave height in the real-time turbulence region, to obtain the process dataset corresponding to the workpiece to be processed. S4. Based on the preset multi-parameter coupling judgment logic, perform correlation analysis on the process dataset to identify composite process risks.

[0009] Optionally, composite process risks include transmission equipment malfunction risks, connection failure risks, solder joint voids and oxidation risks, and conveyor mechanism aging risks.

[0010] Optionally, step S4 includes: S41. Calculate the chain speed fluctuation amplitude based on the preset chain speed and all real-time chain speeds in the process dataset, calculate the first average flow rate based on all real-time flux flow rates in the process dataset, calculate the average oxygen content based on all real-time oxygen content in the process dataset, extract the real-time workpiece temperature of the workpiece to be processed in the preheating zone from all real-time workpiece temperatures in the process dataset, and determine the peak temperature of the preheating zone based on all extracted real-time workpiece temperatures. S42. Obtain the second average flow rate corresponding to the first preset number of workpieces, obtain the actual CPK index of the X-axis, the actual CPK index of the Y-axis, and the actual CPK index of the Z-axis corresponding to the current statistical batch and the first second preset number of statistical batches, and calculate the average chain speed based on all real-time chain speeds corresponding to the current statistical batch; the workpiece model of the second average flow rate is the same as the workpiece model of the workpiece to be processed, and the actual CPK index of the X-axis, the actual CPK index of the Y-axis, and the actual CPK index of the Z-axis are the actual CPK indices corresponding to the real-time X-axis vibration amplitude, the real-time Y-axis vibration amplitude, and the real-time Z-axis vibration amplitude, respectively; S43. When the chain speed fluctuation amplitude is greater than the preset fluctuation amplitude and all real-time Z-axis vibration amplitudes in the process data set are greater than the dynamic upper warning limit corresponding to the real-time Z-axis vibration amplitude, it is determined that there is a risk of abnormality in the transmission equipment. S44. When the tin wave flatness parameter is greater than or equal to the preset level 2 alarm threshold and the peak temperature of the preheating zone is less than the preset fixed specification lower limit corresponding to the preheating zone, it is determined that there is a risk of poor connection. S45. Calculate the first flux decrease ratio based on the first average flow rate and the preset flux flow rate, and calculate the second flux decrease ratio based on the second average flow rate and the preset flux flow rate. When the first flux decrease ratio and all second flux decrease ratios are greater than the preset decrease ratio and the average oxygen content is greater than the preset oxygen content upper limit, it is determined that there is a risk of solder joint void oxidation. S46. Calculate the chain speed drift ratio based on the average chain speed and the preset average chain speed. When all actual CPK indices of the X-axis, all actual CPK indices of the Y-axis, and all actual CPK indices of the Z-axis are less than the lower limit of the first preset CPK index and the chain speed drift ratio is greater than the preset drift ratio, it is determined that there is a risk of aging of the conveyor mechanism.

[0011] Optionally, the wave soldering process monitoring method further includes the following steps: A1. For each parameter in the process dataset except for the tin wave flatness parameter, obtain the corresponding dynamic upper warning limit, dynamic lower warning limit, preset fixed specification lower limit, and preset fixed specification upper limit, and calculate the parameter deviation between the parameter and the dynamic upper warning limit and the dynamic lower warning limit respectively; the dynamic upper warning limit is less than the preset fixed specification upper limit, and the dynamic lower warning limit is greater than the preset fixed specification lower limit. A2. For each parameter in the process data set except for the tin wave flatness parameter, when the parameter is between the dynamic upper warning limit and the dynamic lower warning limit and the parameter deviation is less than the preset deviation, a software pop-up warning will be issued. A3. For each parameter in the process data set except for the tin wave flatness parameter, when the parameter is between its corresponding dynamic upper warning limit and the preset fixed specification upper limit or between its corresponding dynamic lower warning limit and the preset fixed specification lower limit, an audible and visual warning will be issued. A4. For each parameter in the process data set except for the tin wave flatness parameter, when the parameter is greater than its corresponding preset fixed specification upper limit or less than its corresponding preset fixed specification lower limit, the component is locked and the machine is stopped. A5. When the tin wave flatness parameter is greater than or equal to the preset first-level alarm threshold and less than the preset second-level alarm threshold, a software pop-up warning will be issued. A6. When the tin wave flatness parameter is greater than or equal to the preset level 2 alarm threshold and less than the preset level 3 alarm threshold, an audible and visual warning will be issued. A7. When the tin wave flatness parameter is greater than or equal to the preset level 3 alarm threshold, the locking mechanism will be stopped.

[0012] Optionally, the calculation process for the dynamic upper and lower warning limits corresponding to real-time workpiece temperature, real-time chain speed, real-time X-axis vibration amplitude, real-time Y-axis vibration amplitude, real-time Z-axis vibration amplitude, real-time flux flow rate, and real-time oxygen content is the same. The calculation process for the dynamic upper and lower warning limits corresponding to real-time workpiece temperature includes: B1. Obtain the real-time workpiece temperature of the first three preset number of qualified workpieces; B2. Calculate the mean and standard deviation of the working temperature based on all real-time workpiece temperatures; B3. Calculate the dynamic upper warning limit and dynamic lower warning limit corresponding to the real-time workpiece temperature based on the average working temperature and the standard deviation of the working temperature.

[0013] Optionally, the wave soldering process monitoring method further includes the following steps: C1. For each parameter in the process dataset, obtain the set of actual CPK indices corresponding to that parameter; the set of actual CPK indices includes the actual CPK indices corresponding to multiple statistical batches. C2. For each parameter in the process dataset, generate a predicted CPK index based on the actual CPK index set corresponding to that parameter. C3. When the actual CPK index of the fourth consecutive set number of statistical batches decreases batch by batch and the actual CPK index of the current statistical batch is greater than or equal to the second set lower limit of the CPK index, or the predicted CPK index is less than the second set lower limit of the CPK index, an abnormal warning message for the acquisition component of the parameter corresponding to the actual CPK index is generated. C4. When the actual CPK index corresponding to the fifth consecutive set number of statistical batches decreases batch by batch, and the actual CPK index of the current statistical batch is less than the second set lower limit of the CPK index and greater than or equal to the third set lower limit of the CPK index, and the rate of decrease of the actual CPK index of the current statistical batch is less than or equal to the first set rate, a shutdown maintenance reminder is generated for the acquisition component corresponding to the actual CPK index. C5. When the actual CPK index of the fifth consecutive set number of statistical batches decreases batch by batch and the actual CPK index of the current statistical batch is less than the third set lower limit of the CPK index, or when the actual CPK index of the fourth consecutive set number of statistical batches decreases batch by batch and the rate of decrease of the actual CPK index of the current statistical batch is greater than or equal to the second set rate, control the wave soldering equipment to stop; the first set rate is less than the second set rate.

[0014] Optionally, the calculation process for the CPK index corresponding to real-time workpiece temperature, real-time chain speed, real-time X-axis vibration amplitude, real-time Y-axis vibration amplitude, real-time Z-axis vibration amplitude, real-time flux flow rate, solder wave flatness parameter, and real-time oxygen content is the same. The calculation process for the CPK index corresponding to the real-time X-axis vibration amplitude includes: D1. When the number of workpieces that have completed the wave soldering process reaches the preset workpiece quantity corresponding to the statistical batch, calculate the mean value of the X-axis amplitude and the standard deviation of the X-axis amplitude based on all real-time X-axis vibration amplitude values, and obtain the preset fixed specification lower limit and preset fixed specification upper limit of the real-time X-axis vibration amplitude. D2. Calculate the CPK index corresponding to the real-time X-axis vibration amplitude based on the mean X-axis amplitude, standard deviation of X-axis amplitude, preset fixed specification lower limit and preset fixed specification upper limit; The wave soldering process monitoring method also includes the following steps: E1. When the number of workpieces that have completed the wave soldering process reaches the preset number of workpieces corresponding to the statistical batch, for each parameter in the process data set, calculate the CP index and PPK index corresponding to the parameter based on the mean X-axis amplitude, standard deviation X-axis amplitude, preset fixed specification lower limit and preset fixed specification upper limit of the parameter. E2. Integrate all CPK indices, all CP indices, and all PPK indices into a process capability matrix.

[0015] Optionally, the wave soldering process monitoring method further includes the following steps: F1. Map each parameter in the process dataset to the wave soldering 3D digital twin model.

[0016] Secondly, this application provides an electronic device including a processor and a memory, the memory storing computer-readable instructions, which, when executed by the processor, perform the steps of the method provided in the first aspect above.

[0017] Thirdly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, performs the steps of the method provided in the first aspect above.

[0018] As can be seen from the above, the wave soldering process monitoring method, electronic device and storage medium provided in this application obtain the unique identification of the workpiece to be processed and simultaneously collect multi-source heterogeneous physical field parameters. The parameters are associated and bound with the identification to form a process dataset. Based on the multi-parameter coupling judgment logic, the composite process risk is identified. This effectively solves the problems of asynchronous parameter acquisition, untraceable data and lack of multi-parameter linkage analysis in the prior art, which leads to the inability to identify composite process degradation. It has the advantages of improving the accuracy and traceability of welding process monitoring. Attached Figure Description

[0019] Figure 1 A flowchart of a wave soldering process monitoring method provided in this application embodiment.

[0020] Figure 2This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application.

[0021] Reference numerals: 101, processor; 102, memory; 103, communication bus. Detailed Implementation

[0022] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely represents selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0023] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this application, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0024] Firstly, such as Figure 1 As shown, this application provides a wave soldering process monitoring method, which includes the following steps: S1. Obtain the unique identifier of the workpiece to be processed, and synchronously and continuously collect the multi-source heterogeneous physical field parameters of the workpiece to be processed; the multi-source heterogeneous physical field parameters include real-time workpiece temperature, real-time chain speed, real-time solder wave height in the smoothing area, real-time solder wave height in the turbulence area, real-time X-axis vibration amplitude, real-time Y-axis vibration amplitude, real-time Z-axis vibration amplitude, real-time flux flow rate, and real-time oxygen content. S2. Calculate the tin wave smoothness parameters based on the real-time tin wave height in the smooth wave area and the real-time tin wave height in the turbulence area. S3. Associate and bind the unique identifier with the tin wave flatness parameter and the parameters of the multi-source heterogeneous physical field parameters, except for the tin wave height in the real-time flat wave region and the tin wave height in the real-time turbulence region, to obtain the process dataset corresponding to the workpiece to be processed. S4. Based on the preset multi-parameter coupling judgment logic, perform correlation analysis on the process dataset to identify composite process risks.

[0025] For ease of understanding, some key terms in this embodiment are explained below. The workpiece to be processed in this embodiment is preferably a PCB board. The wave soldering process monitoring method in this embodiment aims to assess process stability, identify potential process risks, and provide corresponding early warnings or intervention measures by real-time monitoring and analysis of multiple key parameters during the wave soldering process to ensure soldering quality. The unique identifier of the workpiece to be processed in this embodiment typically refers to a code used to uniquely identify each workpiece (e.g., a PCB board), such as a barcode, QR code, or RFID tag. This identifier is used to track the workpiece throughout the production process and bind it to all relevant process data. The multi-source heterogeneous physical field parameters in this embodiment refer to a set of parameters with different physical meanings and data formats collected from different types of sensors during the wave soldering process. These parameters collectively reflect various aspects of the wave soldering process. For example: real-time workpiece temperature refers to the instantaneous temperature of the workpiece in different areas of the wave soldering oven, usually measured by thermocouples or infrared sensors; real-time chain speed refers to the speed at which the conveyor chain moves, directly affecting the dwell time of the workpiece in each heating and soldering zone; real-time smoothing zone solder wave height refers to the height of the molten solder in the smooth solder wave area of ​​the wave solder bath, usually measured by a laser rangefinder; real-time turbulence zone... The solder wave height refers to the height of the molten solder in the disturbed solder wave area of ​​the wave solder bath, typically measured using a laser rangefinder. Real-time X-axis vibration amplitude, real-time Y-axis vibration amplitude, and real-time Z-axis vibration amplitude refer to the vibration intensity of the wave soldering equipment's conveyor track in the X (left / right), Y (back / forward), and Z (up / down) directions, typically measured using a triaxial accelerometer. Real-time flux flow rate refers to the amount of flux supplied per unit time in the flux spraying area, typically measured using a flow meter. Real-time oxygen content refers to the oxygen concentration in the inert gas (such as nitrogen) environment within the wave soldering furnace, typically measured using an oxygen sensor. In this embodiment, the solder wave flatness parameter is the absolute value of the difference between the solder wave height in the flat area and the solder wave height in the disturbed area. This parameter reflects the smoothness of the incoming solder wave and is an important indicator for evaluating the working condition of the solder pump and the maintenance of the solder bath. The process dataset in this embodiment refers to a data set formed by associating and binding the unique identifier of a specific workpiece to all relevant physical field parameters (including solder wave flatness parameters) collected during the wave soldering process. This dataset completely records the soldering process information of a single workpiece. The preset multi-parameter coupling judgment logic in this embodiment refers to a series of pre-defined rules or algorithms used to comprehensively analyze multiple parameters in the process dataset. These logics can identify complex process risks caused by the synergistic effect of multiple parameters that cannot be revealed by a single parameter anomaly. The composite process risk in this embodiment refers to potential production defects or equipment failures caused by the interrelationship and synergistic effect of multiple parameter anomalies in the wave soldering process. For example, chain speed fluctuations and track vibrations may jointly indicate a transmission equipment malfunction.

[0026] This application proposes a wave soldering process monitoring method, which aims to solve the technical problems existing in the monitoring of wave soldering processes, such as single parameter acquisition dimension, lack of time synchronization and identity binding of multi-source data, inability to quantify solder wave flatness, and lack of multi-parameter coupling analysis to identify complex process risks.

[0027] In step S1, the method first obtains the unique identifier of the workpiece to be processed and simultaneously and continuously collects the multi-source heterogeneous physical field parameters of the workpiece. Specifically, this embodiment can use an integrated hardware system to complete step S1. Specifically, this system includes a global temperature sensing unit, a solder wave height dual-laser monitoring unit, a chain speed acquisition unit, a three-axis track vibration monitoring unit, a flux flow monitoring unit, an oxygen content monitoring unit, a barcode scanning component (used to obtain the unique identifier by scanning the barcode on the workpiece), and an RTB multi-source time-series fusion acquisition terminal. The global temperature sensing unit (used to acquire real-time workpiece temperature), the solder wave height dual-laser monitoring unit (used to acquire real-time smoothing wave...)... The regional tin wave height and real-time scrambled regional tin wave height), chain speed acquisition unit (for acquiring real-time chain speed), three-axis track vibration monitoring unit (for acquiring real-time X-axis vibration amplitude, real-time Y-axis vibration amplitude and real-time Z-axis vibration amplitude), flux flow monitoring unit (for acquiring real-time flux flow), and oxygen content monitoring unit (for acquiring real-time oxygen content) are all electrically connected to the RTB multi-source time-series fusion acquisition terminal. This terminal has a built-in high-precision clock synchronization chip, which can achieve millisecond-level time-series alignment of six sensors and package multi-physics field raw sensor data with unified timestamps.

[0028] In step S2, the solder wave smoothness parameter is calculated based on the real-time solder wave height in the smoothing area and the real-time solder wave height in the scrambling area. One implementation method is to measure the solder wave height in the smoothing area and the scrambling area separately using two independent laser ranging sensors, and then subtract these two height values ​​to obtain the solder wave smoothness parameter. For example, a dual-laser monitoring unit for solder wave height can include two independent laser ranging sensors, corresponding to the wave soldering smoothing area and the scrambling area respectively. The terminal collects the two height values ​​in real time and automatically calculates the difference in solder wave smoothness height. This method can quantify the smoothness of the solder wave, thereby assessing the wear condition of the solder pump impeller or the maintenance status of the solder bath based on this smoothness.

[0029] In step S3, the unique identifier is associated with the solder wave flatness parameter and other parameters in the multi-source heterogeneous physical field parameters, excluding the solder wave height in the real-time smoothing region and the solder wave height in the real-time turbulence region, to obtain the process dataset corresponding to the workpiece to be processed. For example, through the multi-source time-series synchronization binding module in the monitoring software on the industrial control host computer, the PCBA single-board SN identifier (unique identifier) ​​read by the barcode scanning component can be completely bound and archived with the real-time workpiece temperature, solder wave flatness parameter, real-time chain speed, real-time X-axis vibration amplitude, real-time Y-axis vibration amplitude, real-time Z-axis vibration amplitude, real-time flux flow rate, and real-time oxygen content corresponding to the same workpiece to be processed. This binding mechanism ensures that all relevant process data for each workpiece can be traced and analyzed.

[0030] In step S4, based on preset multi-parameter coupling judgment logic, correlation analysis is performed on the process dataset to identify complex process risks. For example, the multi-parameter coupling fault inference engine in the monitoring software on the industrial control host computer has multiple sets of multi-parameter linkage judgment rules built in. This engine identifies complex process risks based on the combination of six physical field parameters synchronously collected from the same board. This method can identify complex problems that cannot be detected by single parameter monitoring. For example, chain speed fluctuations and track vibration amplitude exceeding limits may jointly indicate abnormal risks in the transmission equipment; excessive solder wave flatness parameters and low preheating zone peak temperatures may indicate poor connection risks; a continuous decrease in flux flow and an increase in oxygen content may lead to solder joint voids and oxidation risks; and a continuous decrease in triaxial vibration CPK and chain speed drift may indicate aging risks in the conveyor mechanism.

[0031] The following example will provide a more detailed explanation of the above technical solution: Suppose that a batch of workpieces needs to be welded on a wave soldering production line.

[0032] First, when each workpiece enters the wave soldering equipment, its unique identification, such as "PCBA-SN-001," is read by a barcode scanning component. Simultaneously, the global temperature sensing unit, the dual-laser solder wave height monitoring unit, the chain speed acquisition unit, the three-axis track vibration monitoring unit, the flux flow monitoring unit, and the oxygen content monitoring unit begin synchronously and continuously collecting various physical field parameters of the workpiece during the soldering process. For example, the global temperature sensing unit plots the workpiece's temperature curve (real-time workpiece temperature) in real time; the dual-laser solder wave height monitoring unit measures the solder wave height in the smoothing and turbulence areas respectively; the three-axis track vibration monitoring unit collects the vibration amplitudes of the X, Y, and Z axes in real time; and the flux flow monitoring unit and the oxygen content monitoring unit record the flux flow rate and the oxygen content in the furnace, respectively. All this data is aligned to millisecond-level timing using an RTB multi-source time-series fusion acquisition terminal and packaged with a unified timestamp.

[0033] While acquiring data, the system calculates the tin wave flatness parameter based on the real-time tin wave height in the smoothing area and the real-time tin wave height in the turbulence area. For example, if the tin wave height in the smoothing area is 3.0 mm and the tin wave height in the turbulence area is 3.2 mm, then the tin wave flatness parameter is 0.2 mm.

[0034] Subsequently, the multi-source timing synchronization binding module in the industrial control host computer associates the unique identifier "PCBA-SN-001" with all real-time parameters collected throughout the entire soldering cycle, including workpiece temperature, chain speed, solder wave flatness parameters, real-time X-axis vibration amplitude, real-time Y-axis vibration amplitude, real-time Z-axis vibration amplitude, real-time flux flow rate, and real-time oxygen content, forming a complete process dataset. This dataset fully records the "health check report" of "PCBA-SN-001" during the wave soldering process.

[0035] Next, the multi-parameter coupled fault inference engine begins correlation analysis on this process dataset. For example, if the system detects that the chain speed fluctuation of "PCBA-SN-001" is greater than the preset fluctuation range (e.g., 2%), and the real-time Z-axis vibration amplitude is greater than the dynamic upper warning limit corresponding to the real-time Z-axis vibration amplitude, then according to the preset multi-parameter coupled judgment logic, the system will determine that there is a risk of abnormality in the transmission equipment. This may mean that the support rollers of the conveyor track are loose or the bearings of the conveyor motor are worn. At this time, the system will suggest checking the equipment transmission mechanism. As another example, if the system finds that the solder wave flatness parameter of "PCBA-SN-001" is greater than or equal to the preset secondary alarm threshold (e.g., 0.2mm), and its peak temperature in the preheating zone is less than the preset fixed specification lower limit corresponding to the preheating zone, then the system will determine that there is a risk of poor connection, predicting that solder bridging or bridging soldering defects may occur. At this time, the system will prompt uneven solder wave and insufficient preheating, and suggest adjusting the solder pump power and verifying the preheating heating module.

[0036] As can be seen from the above examples, the wave soldering process monitoring method of this application can comprehensively and in real time monitor the wave soldering process by synchronously collecting multi-source heterogeneous physical field parameters, calculating tin wave flatness parameters, associating and binding unique identifiers with process data, and performing correlation analysis based on preset multi-parameter coupling judgment logic.

[0037] Compared to existing single-dimensional temperature measurement devices that can only post-process temperature curves and cannot simultaneously collect other key process variables, this application achieves real-time streaming temperature curve plotting of the PCB during its movement within the furnace through a global temperature sensing unit. Combined with the synchronous acquisition of five other physical field parameters, it solves the problems of data silos and timing misalignment. Compared to discrete single-parameter monitoring devices, this application's RTB multi-source timing fusion acquisition terminal incorporates a high-precision clock synchronization chip, ensuring millisecond-level timing alignment of the six sensors. This allows for the complete binding of multi-physical field data with the single-board serial number (SN) identifier, enabling traceability of all-dimensional causes for a single workpiece. Furthermore, this application calculates tin wave flatness parameters through a dual-laser monitoring unit for tin wave height, overcoming the limitation of existing technologies that rely solely on single-point laser ranging to quantify tin wave flatness, and providing early warning of tin pump impeller wear. Most importantly, the multi-parameter coupled fault reasoning engine of this application incorporates multiple sets of multi-parameter linkage judgment rules, enabling it to identify complex process risks and output graded warnings based on the combination of six physical field parameters synchronously collected from the same board. This solves the problem of existing technologies that only alarm for a single parameter independently and cannot identify complex process degradation. Therefore, the overall technical concept of this application has significant technical contributions and innovations in the field of wave soldering process monitoring.

[0038] In some preferred embodiments, the risks of complex processes include transmission equipment malfunction, poor connection, solder joint voids and oxidation, and conveyor mechanism aging.

[0039] Complex process risks refer to potential or actual process defects or equipment failures caused by abnormal combinations or coordinated changes of multiple physical field parameters in wave soldering. These risks cannot be reflected by a single parameter exceeding its limit; rather, they are complex problems requiring comprehensive analysis of multi-dimensional data for identification. Identification can be achieved through logical judgment using pre-defined expert rules. Transmission equipment anomaly risks refer to the risk of abnormal operation of the mechanical transmission system responsible for workpiece transport in wave soldering equipment, such as loose chains, motor wear, or guide rail deformation. These anomalies may lead to unstable workpiece transport and inconsistent soldering times. This risk can be identified by monitoring chain speed fluctuations and the Z-axis vibration amplitude of transmission components. Poor connection risks refer to the risk of defects such as cold solder joints, bridging, insufficient solder, or poor soldering due to improper process parameter settings or abnormal equipment conditions during the soldering process. This risk can be identified by analyzing the flatness of the solder wave and the temperature profile of the preheating zone. The risk of solder joint voids and oxidation refers to the risk of voids appearing inside the solder joint or oxidation occurring on its surface. This severely affects the mechanical strength and electrical performance of the solder joint. This risk can be identified by monitoring flux flow and oxygen content in the welding area. The risk of conveyor mechanism aging refers to the risk of gradual performance degradation of the wave soldering equipment's conveyor system due to long-term operation, such as chain wear, bearing aging, and decreased guide rail precision. This aging can lead to reduced conveyor accuracy and stability. This risk can be identified by long-term tracking of chain speed drift and triaxial vibration amplitude across multiple statistical batches.

[0040] In some preferred embodiments, step S4 includes: S41. Calculate the chain speed fluctuation amplitude based on the preset chain speed and all real-time chain speeds in the process dataset, calculate the first average flow rate based on all real-time flux flow rates in the process dataset, calculate the average oxygen content based on all real-time oxygen content in the process dataset, extract the real-time workpiece temperature of the workpiece to be processed in the preheating zone from all real-time workpiece temperatures in the process dataset, and determine the peak temperature of the preheating zone based on all extracted real-time workpiece temperatures. S42. Obtain the second average flow rate corresponding to the first preset number of workpieces, obtain the actual CPK index of the X-axis, the actual CPK index of the Y-axis, and the actual CPK index of the Z-axis corresponding to the current statistical batch and the first second preset number of statistical batches, and calculate the average chain speed based on all real-time chain speeds corresponding to the current statistical batch; the workpiece model of the second average flow rate is the same as the workpiece model of the workpiece to be processed, and the actual CPK index of the X-axis, the actual CPK index of the Y-axis, and the actual CPK index of the Z-axis are the actual CPK indices corresponding to the real-time X-axis vibration amplitude, the real-time Y-axis vibration amplitude, and the real-time Z-axis vibration amplitude, respectively; S43. When the chain speed fluctuation amplitude is greater than the preset fluctuation amplitude and all real-time Z-axis vibration amplitudes in the process data set are greater than the dynamic upper warning limit corresponding to the real-time Z-axis vibration amplitude, it is determined that there is a risk of abnormality in the transmission equipment. S44. When the tin wave flatness parameter is greater than or equal to the preset level 2 alarm threshold and the peak temperature of the preheating zone is less than the preset fixed specification lower limit corresponding to the preheating zone, it is determined that there is a risk of poor connection. S45. Calculate the first flux decrease ratio based on the first average flow rate and the preset flux flow rate, and calculate the second flux decrease ratio based on the second average flow rate and the preset flux flow rate. When the first flux decrease ratio and all second flux decrease ratios are greater than the preset decrease ratio and the average oxygen content is greater than the preset oxygen content upper limit, it is determined that there is a risk of solder joint void oxidation. S46. Calculate the chain speed drift ratio based on the average chain speed and the preset average chain speed. When all actual CPK indices of the X-axis, all actual CPK indices of the Y-axis, and all actual CPK indices of the Z-axis are less than the lower limit of the first preset CPK index and the chain speed drift ratio is greater than the preset drift ratio, it is determined that there is a risk of aging of the conveyor mechanism.

[0041] Chain speed fluctuation is an indicator of the stability of workpiece speed during wave soldering furnace transport. It can be calculated by recording real-time chain speed data for a single workpiece throughout the entire welding cycle, identifying its maximum and minimum values, and then calculating the relative percentage fluctuation based on a preset chain speed (e.g., target set chain speed or historical stable chain speed average). For example, it can be expressed as (maximum real-time chain speed - minimum real-time chain speed) / preset chain speed. The first average flow rate refers to the average flux consumption rate during the welding process of a single workpiece. This can be obtained by summing all real-time flux flow rate data for that workpiece within the flux spraying area and dividing by the number of sampling points. The average oxygen content refers to the average oxygen concentration in the wave soldering furnace atmosphere during the welding process of a single workpiece. This can be obtained by summing all real-time oxygen content data for that workpiece within the welding area and dividing by the number of sampling points. Extracting the real-time workpiece temperature of the workpiece in the preheating zone from all real-time workpiece temperatures in the process dataset refers to filtering the temperature data experienced by the workpiece within the preheating area (e.g., preheating section 1, preheating section 2, etc.) from the real-time workpiece temperature curves throughout the entire welding process. This can be determined using preset furnace zone division timestamps or the workpiece's position information within the furnace. The peak temperature of the preheating zone refers to the highest temperature reached by the workpiece within the preheating area, which can be determined by iterating through the extracted real-time workpiece temperature data of the preheating zone and finding the maximum value.

[0042] The second average flow rate refers to the average flux consumption rate during welding of a certain number (e.g., 3, 5, or more) of workpieces of the same model prior to the current workpiece. This can be obtained by querying historical process datasets, filtering out the first preset number of workpieces of the same model that have been welded, and then averaging all real-time flux flows for these workpieces. The actual CPK index is a key indicator of process capability, reflecting process stability and yield. The CPK index is typically calculated based on real-time vibration amplitude data within a statistical batch (e.g., 100 workpieces per batch), combined with its mean, standard deviation, and preset fixed specification upper and lower limits. The first second preset number of statistical batches refers to a certain number (e.g., 2, 3, or more) of statistical batches prior to the current batch. The average chain speed refers to the average chain speed of all workpieces during welding within the current statistical batch. This can be obtained by summing the real-time chain speed data of all workpieces within the current batch and calculating its average. The workpiece model for the second average flow rate is the same as the workpiece model to be processed. This limitation ensures that the workpieces being compared have the same process requirements and flux consumption characteristics when comparing flux flow rates, thus guaranteeing the effectiveness and accuracy of the comparison. The actual CPK index for the X-axis, Y-axis, and Z-axis are the actual CPK indices corresponding to the real-time X-axis vibration amplitude, real-time Y-axis vibration amplitude, and real-time Z-axis vibration amplitude, respectively. This clarifies that the CPK index is calculated separately for the vibration amplitude of the wave soldering equipment in the X, Y, and Z axes, and the vibration characteristics of each axis are evaluated independently.

[0043] When the chain speed fluctuation exceeds a preset fluctuation range and all real-time Z-axis vibration amplitudes in the process dataset exceed the dynamic upper warning limit corresponding to the real-time Z-axis vibration amplitude, a risk of transmission equipment malfunction is identified. This judgment logic combines the stability of the chain speed with the vertical vibration situation. The preset fluctuation range is an empirical value or a threshold set according to equipment performance, such as 2%. The dynamic upper warning limit is the upper limit of Z-axis vibration amplitude dynamically calculated based on historical data. When the real-time Z-axis vibration amplitude continuously exceeds this upper limit, it indicates abnormal vibration in the vertical direction of the equipment. When both conditions are met simultaneously, it indicates that the transmission system of the wave soldering equipment may have abnormalities such as loose feed rail support rollers or worn conveyor motor bearings.

[0044] A risk of poor solder joint connection is identified when the solder wave flatness parameter is greater than or equal to the preset secondary alarm threshold and the peak temperature of the preheating zone is less than the preset fixed specification lower limit corresponding to the preheating zone. This judgment logic combines the physical morphology of the solder wave with the heat treatment status of the workpiece. The solder wave flatness parameter reflects the uniformity of the solder wave surface, and the preset secondary alarm threshold (e.g., 0.2 mm) is a key boundary for judging solder wave abnormalities. A peak temperature of the preheating zone less than the preset fixed specification lower limit indicates that the workpiece has not reached a sufficient preheating temperature before soldering. When both conditions are met simultaneously, it may lead to poor solder joint wetting, solder bridging, or other poor solder joint connection problems.

[0045] The first flux reduction ratio is the decrease in the average flux flow rate of the current workpiece relative to the standard preset flux flow rate, which can be calculated by (preset flux flow rate - first average flow rate) / preset flux flow rate. The second flux reduction ratio is the decrease in the average flux flow rate of the previous first preset number of workpieces of the same model relative to the standard preset flux flow rate, which can be calculated by (preset flux flow rate - second average flow rate) / preset flux flow rate. When both the first flux reduction ratio and all second flux reduction ratios are greater than the preset reduction ratio and the average oxygen content is greater than the preset oxygen content upper limit, a risk of solder joint void oxidation is identified. This judgment logic combines the flux supply situation with the welding environment atmosphere. The preset reduction ratio (e.g., 10%) is the threshold for judging insufficient flux supply, and the preset oxygen content upper limit (e.g., 50 ppm) is the threshold for judging excessively high oxygen concentration in the welding environment. When the flux supply is continuously insufficient and the oxygen content in the welding area is too high, the solder joint is prone to oxidation, forming voids and affecting the welding quality.

[0046] The chain speed drift ratio is the degree of deviation of the average chain speed of the current batch from the long-term stable preset chain speed average. This can be calculated by |average chain speed - preset chain speed average| / preset chain speed average. When all actual CPK indices of the X-axis, all actual CPK indices of the Y-axis, and all actual CPK indices of the Z-axis are less than the first preset CPK index lower limit and the chain speed drift ratio is greater than the preset drift ratio, it is determined that there is a risk of aging of the conveyor mechanism. This judgment logic combines the process capability of the equipment's three-axis vibration with the long-term stability of the chain speed. The first preset CPK index lower limit (e.g., 1.0) is the threshold for judging whether the process capability is qualified. The preset drift ratio (e.g., 1.5%) is the threshold for judging whether the chain speed has deviated significantly. When the three-axis vibration CPK index continues to decrease (indicating poor vibration stability) and the chain speed drifts significantly, it indicates that the conveyor mechanism (such as chain, guide rail, motor, etc.) of the wave soldering equipment may have aging phenomena. Preferably, when it is determined that there is a risk of aging in the conveyor mechanism, this embodiment can locate the wear area of ​​the equipment based on the three-axis vibration CPK index. The specific process is as follows: if the actual CPK index of the Z-axis in three consecutive statistical batches is less than 1.0, it is considered that the conveyor motor bearing is worn or the track support roller is loose; if the actual CPK index of the Y-axis in three consecutive statistical batches is less than 1.0, it is considered that the chain is stretched, the sprocket teeth are worn or the chain speed stepping is stuck; if the actual CPK index of the X-axis in three consecutive statistical batches is less than 1.0, it is considered that the guide rails on both sides of the track are worn or the track tension is inconsistent.

[0047] This solution achieves a leap from single data monitoring to complex risk diagnosis by constructing a multi-parameter coupled judgment logic. Specifically, the solution first preprocesses and statistically analyzes key parameters through S41 and S42, including calculating the chain speed fluctuation amplitude, the first average flow rate, the average oxygen content, the peak temperature of the preheating zone, obtaining the second average flow rate of the preceding workpiece, and the actual CPK index and average chain speed of the X, Y, and Z axes for the current and preceding statistical batches. These preprocessed parameters provide the basic data for subsequent risk judgment. Subsequently, the solution identifies specific complex process risks through a series of finely designed coupled judgment rules. In S43, by simultaneously monitoring the chain speed fluctuation amplitude and the real-time Z-axis vibration amplitude, when both exceed preset thresholds, it is judged as a transmission equipment abnormality risk. This coupled judgment avoids false alarms from single parameters and accurately points to mechanical transmission problems. In S44, the solder wave flatness parameter is linked with the peak temperature of the preheating zone for analysis. When the solder wave flatness is poor and the preheating is insufficient, it is judged as a connection failure risk, which reveals the defects caused by the combined physical morphology and thermodynamic conditions of the welding. In S45, by analyzing the decline rate of current and historical flux flow and combining it with the average oxygen content, a risk of solder joint void oxidation is identified when flux is consistently insufficient and oxygen content is excessively high. This comprehensively assesses the oxidation risk from two dimensions: chemical environment and flux coverage effect. Finally, in S46, by monitoring the overall downward trend of the actual CPK index of the X, Y, and Z axes and the drift rate of the chain speed, a risk of conveyor mechanism aging is identified when process capability continues to deteriorate and chain speed stability decreases, thus achieving a prediction of equipment health status. This solution combines the rich, synchronous, multi-source heterogeneous data collected and correlated in steps S1-S3 with the refined multi-parameter coupling judgment logic constructed in S4, making the identification of complex process risks such as abnormal transmission equipment, poor connection, solder joint void oxidation, and conveyor mechanism aging in wave soldering more accurate and timely. This holistic operating logic overcomes the limitations of traditional single-parameter monitoring in identifying multi-factor coupled faults, significantly improving the intelligence level of wave soldering process monitoring and the accuracy of fault diagnosis.

[0048] The following is a concrete example to illustrate this. Suppose a workpiece (e.g., model "PCB-A001") is being soldered on a wave soldering production line. First, the system continuously collects real-time data on the workpiece during the soldering process, including workpiece temperature, chain speed, solder wave height in the smoothing zone, solder wave height in the scrambling zone, X-axis vibration amplitude, Y-axis vibration amplitude, Z-axis vibration amplitude, flux flow rate, and oxygen content. Simultaneously, it calculates solder wave flatness parameters based on the solder wave heights in the smoothing and scrambling zones. All this data is linked to the unique identifier of "PCB-A001," forming its complete process dataset. When identifying complex process risks, the system performs the following judgment: In the transmission equipment anomaly risk judgment, the system calculates the chain speed fluctuation amplitude of "PCB-A001" throughout the entire soldering process. If the calculation results show that the chain speed fluctuation is 2.5% (greater than the preset fluctuation range of 2%), and the real-time Z-axis vibration amplitude is consistently higher than its corresponding dynamic upper warning limit (e.g., UCL) within the same time period, the system will immediately determine that there is a risk of abnormality in the transmission equipment. This means that the support rollers of the conveyor track are loose or the bearings of the conveyor motor are worn. In the judgment of connection failure risk, the system checks the solder wave flatness parameter of "PCB-A001". If this parameter is 0.25mm (greater than or equal to the preset secondary alarm threshold of 0.2mm), and at the same time, the system extracts the real-time workpiece temperature of "PCB-A001" in the preheating zone from the process dataset and determines that its preheating zone peak temperature is 150℃ (less than the preset fixed specification lower limit of 160℃ corresponding to the preheating zone), the system will determine that there is a risk of connection failure. This indicates that uneven solder wave and insufficient preheating together cause solder bridging or short circuits. In determining the risk of solder joint void oxidation, the system calculates the first average flow rate of "PCB-A001" and compares it with the preset flux flow rate to obtain the first flux reduction ratio. Simultaneously, the system acquires the second average flow rate of the previous three workpieces of the same model ("PCB-A001") and calculates their second flux reduction ratios. If both the first flux reduction ratio and all second flux reduction ratios are greater than the preset reduction ratio of 10% (e.g., the current workpiece reduces by 12%, and the previous three workpieces reduce by an average of 11%), and the average oxygen content of "PCB-A001" in the soldering area is 60 ppm (greater than the preset oxygen content limit of 50 ppm), the system determines that there is a risk of solder joint void oxidation, indicating that the flux supply may be insufficient and the nitrogen protection effect is poor. In determining the aging risk of the conveyor mechanism, the system continuously monitors the actual CPK index of the X, Y, and Z axes.If the actual CPK indices of the X, Y, and Z axes of the current statistical batch (e.g., the 100th workpiece) and the previous two statistical batches (a total of three batches) are all less than the first preset lower limit of CPK index 1.0 (e.g., X-axis CPK is 0.95, Y-axis CPK is 0.92, and Z-axis CPK is 0.98), and the average chain speed of the current statistical batch has a chain speed drift ratio of 1.8% compared to the preset average chain speed (greater than the preset drift ratio of 1.5%), then the system will determine that there is a risk of aging in the conveyor mechanism. This means that the chain, guide rail, or conveyor motor is aging as a whole, and the conveyor stability is continuously deteriorating. Through the above specific judgment logic, this solution can timely and accurately identify various complex process risks that may exist in the wave soldering process.

[0049] In some preferred embodiments, the wave soldering process monitoring method further includes the step of: A1. For each parameter in the process dataset except for the tin wave flatness parameter, obtain the corresponding dynamic upper warning limit, dynamic lower warning limit, preset fixed specification lower limit, and preset fixed specification upper limit, and calculate the parameter deviation between the parameter and the dynamic upper warning limit and the dynamic lower warning limit respectively; the dynamic upper warning limit is less than the preset fixed specification upper limit, and the dynamic lower warning limit is greater than the preset fixed specification lower limit. A2. For each parameter in the process data set except for the tin wave flatness parameter, when the parameter is between the dynamic upper warning limit and the dynamic lower warning limit and the parameter deviation is less than the preset deviation, a software pop-up warning will be issued. A3. For each parameter in the process data set except for the tin wave flatness parameter, when the parameter is between its corresponding dynamic upper warning limit and the preset fixed specification upper limit or between its corresponding dynamic lower warning limit and the preset fixed specification lower limit, an audible and visual warning will be issued. A4. For each parameter in the process data set except for the tin wave flatness parameter, when the parameter is greater than its corresponding preset fixed specification upper limit or less than its corresponding preset fixed specification lower limit, the component is locked and the machine is stopped. A5. When the tin wave flatness parameter is greater than or equal to the preset first-level alarm threshold and less than the preset second-level alarm threshold, a software pop-up warning will be issued. A6. When the tin wave flatness parameter is greater than or equal to the preset level 2 alarm threshold and less than the preset level 3 alarm threshold, an audible and visual warning will be issued. A7. When the tin wave flatness parameter is greater than or equal to the preset level 3 alarm threshold, the locking mechanism will be stopped.

[0050] "Targeting each parameter in the process dataset except for the solder wave flatness parameter" refers to analyzing each of the multi-source heterogeneous physical field parameters collected in the wave soldering process monitoring method. These parameters include real-time workpiece temperature, real-time chain speed, real-time X-axis vibration amplitude, real-time Y-axis vibration amplitude, real-time Z-axis vibration amplitude, real-time flux flow rate, and real-time oxygen content. By independently monitoring and issuing warnings for these parameters, it can be ensured that all key aspects of the wave soldering process are under control, avoiding overall process failure due to a single parameter anomaly. "Obtaining the corresponding dynamic upper warning limit, dynamic lower warning limit, preset fixed specification lower limit, and preset fixed specification upper limit" means that the system sets two types of control limits for each monitored process parameter. The dynamic upper warning limit and dynamic lower warning limit are control limits that can be adaptively adjusted with process fluctuations, obtained from statistical analysis of historical or real-time data, and are used to capture early drift trends in the process. For example, they can be calculated based on the mean and standard deviation. The preset fixed specification lower limit and preset fixed specification upper limit are rigid quality boundaries that are pre-set according to product design requirements, process specifications, or industry standards and cannot be arbitrarily changed. They represent the final judgment standard for whether a product is qualified or not. These limits can be manually entered by process engineers or automatically loaded from the product database. "Calculate the parameter deviation between the parameter and the dynamic upper warning limit and the dynamic lower warning limit respectively" means calculating the difference between the currently collected parameter value and its corresponding dynamic upper warning limit or dynamic lower warning limit. The purpose of calculating the parameter deviation is to quantify how close the parameter is to the dynamic warning limit, so that different levels of warnings can be triggered according to the magnitude of the deviation. For example, the absolute difference between the parameter value and the dynamic control limit can be calculated, or its percentage of the distance between the dynamic control limit and the center value can be calculated. "The dynamic upper warning limit is less than the preset fixed specification upper limit, and the dynamic lower warning limit is greater than the preset fixed specification lower limit" clarifies the hierarchical relationship between the dynamic warning limit and the fixed specification limit. That is, the dynamic warning limit is within the fixed specification limit, forming a narrower "safe operating range" to ensure that the system can issue a warning in advance before the parameter reaches the rigid boundary of product quality. Step A2 refers to the system issuing a warning via a pop-up window on the software interface when the parameter value is within the dynamic control limit but close to it (i.e., the parameter deviation is less than the preset deviation, for example, less than 10% of the distance between the dynamic control limit and the center value). This non-mandatory warning reminds the operator to pay attention to slight fluctuations in the process. Step A3 refers to the system triggering an audible and visual alarm when the parameter value exceeds the dynamic control limit but remains within the fixed specification limit. This provides a stronger warning effect than a software pop-up, prompting the operator to pay immediate attention and intervene. Step A4 refers to the system immediately executing a locking and shutdown operation when the parameter value exceeds the preset fixed specification upper limit or falls below the preset fixed specification lower limit. This is a mandatory protective measure designed to prevent the continued production of non-conforming products.Step A5 refers to the system issuing a software pop-up notification when the solder wave flatness parameter reaches the preset level 1 alarm threshold but not the preset level 2 alarm threshold, indicating a slight abnormality in solder wave flatness. Step A6 refers to the system triggering an audible and visual alarm when the solder wave flatness parameter reaches the preset level 2 alarm threshold but not the preset level 3 alarm threshold, indicating an escalation of the solder wave flatness abnormality. Step A7 refers to the system executing a lock shutdown when the solder wave flatness parameter reaches or exceeds the preset level 3 alarm threshold, indicating that the solder wave flatness has severely deviated from the normal range.

[0051] This solution constructs a refined, multi-level early warning and response system by combining dynamic adaptive early warning control limits with fixed hard specification thresholds. The system first evaluates the operational status of the wave soldering process in real time based on the multi-source heterogeneous physical field parameters of the workpiece to be processed collected in the aforementioned wave soldering process monitoring method, as well as the calculated solder wave flatness parameters. For all process parameters except for the solder wave flatness parameter, the system continuously monitors their real-time values ​​and compares them with pre-set dynamic upper and lower early warning limits, as well as preset fixed specification lower and upper limits. The dynamic early warning limit serves as a buffer zone for early warning, and its range always lies within the fixed specification limits. When a parameter value is within the dynamic early warning limit, but its deviation from the dynamic early warning limit is less than a preset deviation, the system triggers a software pop-up warning, non-mandatoryly reminding the operator to pay attention to slight process drift. If the parameter value deviates further, exceeding the dynamic early warning limit but still within the fixed specification limits, the system upgrades to an audible and visual warning, using stronger sensory stimulation to prompt timely intervention by the operator. If the parameter value exceeds the fixed specification limit, it means that the process is seriously out of control, and the system will immediately execute a lock-up shutdown to forcibly interrupt production and avoid the generation of defective products. Meanwhile, for the critical indicator of solder wave flatness, this application has designed an independent three-level alarm threshold. When the solder wave flatness parameter reaches the preset first-level alarm threshold but does not reach the preset second-level alarm threshold, the system issues a software pop-up warning; when it reaches the preset second-level alarm threshold but does not reach the preset third-level alarm threshold, the system triggers an audible and visual warning; and when it reaches or exceeds the preset third-level alarm threshold, the system forcibly executes a lock-up shutdown. This hierarchical handling logic, combined with the aforementioned scheme of obtaining the unique identifier of the workpiece to be processed and simultaneously and continuously collecting multi-source heterogeneous physical field parameters, calculating solder wave flatness parameters, associating and binding process datasets, and identifying complex process risks based on preset multi-parameter coupling judgment logic, can ensure that when potential risks or process anomalies are identified, appropriate responses are immediately taken according to their severity. This forms a closed-loop management system from data acquisition and risk identification to hierarchical early warning and handling, which greatly improves the stability of wave soldering process and the reliability of product quality.

[0052] As a specific implementation method, the wave soldering process monitoring method of this application can achieve graded early warning and handling in the following way. First, after system initialization or new product changeover, process engineers will manually enter the preset fixed specification lower limit LSL and preset fixed specification upper limit USL for various process parameters (such as real-time workpiece temperature, real-time chain speed, real-time X-axis vibration amplitude, real-time Y-axis vibration amplitude, real-time Z-axis vibration amplitude, real-time flux flow rate, and real-time oxygen content) according to product standards and equipment mechanical tolerances. These fixed specification limits serve as the "hard red line" for three-level fault judgment and will not change automatically throughout the entire production cycle. At the same time, the system will initialize through a baseline sampling process, for example, by acquiring the process dataset of 50 consecutive qualified workpieces preceding the current workpiece to be processed, and calculating the baseline mean μ0 and baseline standard deviation σ0 corresponding to each parameter in the process dataset. Based on these baseline data, the system dynamically calculates the dynamic upper warning limit UCL and dynamic lower warning limit LCL for each parameter. For example, UCL can be calculated as μ0 + 2.7σ0, and LCL can be calculated as μ0 - 2.7σ0. During actual production, the system continuously monitors the real-time values ​​of various parameters. When the real-time value of a parameter is between its dynamic upper warning limit (UCL) and dynamic lower warning limit (LCL), but the difference between it and the nearest dynamic control limit (UCL or LCL) is less than a preset deviation (e.g., less than 10% of the width of the interval between the dynamic control limit and the center value), the system will trigger a "Level 1 Near-Limit Alert," which will only display a text alert window on the monitoring software interface of the industrial control computer, prompting the operator that the parameter has slight fluctuations and needs attention. If the real-time value of the parameter exceeds the dynamic upper warning limit (UCL) but is still less than the preset fixed specification upper limit (USL), or is less than the dynamic lower warning limit (LCL) but is still greater than the preset fixed specification lower limit (LSL), the system will trigger a "Level 2 Over-Limit Alarm." At this time, the system will not only display the alarm information on the software interface, but will also link the audible and visual alarm devices on the production line, emitting sound and flashing lights to more strongly alert the operator, indicating that the process has deviated significantly, but has not yet reached the level of product non-conformity, and requires immediate inspection and adjustment. At this stage, the system only issues an alert and does not lock the part, allowing the operator to intervene. However, once the real-time value of the parameter exceeds the preset fixed specification upper limit (USL) or falls below the preset fixed specification lower limit (LSL), the system will immediately trigger a "Level 3 Fault Shutdown". At this time, the system will forcibly execute a locking shutdown operation, such as stopping the conveyor chain of the wave soldering equipment and locking the current workpiece to prevent it from continuing processing or entering the next process, thereby avoiding the generation of batches of defective products. A similar tiered early warning mechanism is used for the solder wave flatness parameter. For example, when the solder wave flatness height difference ΔH is greater than or equal to the preset Level 1 alarm threshold (e.g., 0.1 mm) and less than the preset Level 2 alarm threshold (e.g., 0.2 mm), the system will issue a software pop-up warning.When ΔH is greater than or equal to the preset level 2 alarm threshold (e.g., 0.2mm) and less than the preset level 3 alarm threshold (e.g., 0.3mm), the system triggers an audible and visual alarm. When ΔH is greater than or equal to the preset level 3 alarm threshold (e.g., 0.3mm), the system will forcibly shut down the lock and indicate that there may be an abnormality in the solder pump / solder bath, requiring immediate investigation and repair.

[0053] In some preferred embodiments, the calculation processes for the dynamic upper and lower warning limits corresponding to real-time workpiece temperature, real-time chain speed, real-time X-axis vibration amplitude, real-time Y-axis vibration amplitude, real-time Z-axis vibration amplitude, real-time flux flow rate, and real-time oxygen content are the same. The calculation process for the dynamic upper and lower warning limits corresponding to real-time workpiece temperature includes: B1. Obtain the real-time workpiece temperature of the first three preset number of qualified workpieces; B2. Calculate the mean and standard deviation of the working temperature based on all real-time workpiece temperatures; B3. Calculate the dynamic upper warning limit and dynamic lower warning limit corresponding to the real-time workpiece temperature based on the average working temperature and the standard deviation of the working temperature.

[0054] The calculation process for the dynamic upper and lower warning limits corresponding to real-time workpiece temperature, real-time chain speed, real-time X-axis vibration amplitude, real-time Y-axis vibration amplitude, real-time Z-axis vibration amplitude, real-time flux flow rate, and real-time oxygen content is the same. This means that a unified and adaptive threshold setting method is used for multiple key physical parameters in wave soldering. The dynamic upper and lower warning limits are determined based on the actual statistical characteristics of the process rather than fixed values, which allows the monitoring system to better adapt to process fluctuations. The real-time workpiece temperature of the first third preset number of qualified workpieces is obtained to establish a reliable baseline dataset. The third preset number is a configurable value, for example, it can be set to 50 to ensure the statistical validity of the collected data. Qualified workpieces emphasize that these workpieces must have undergone quality verification and meet production standards, thus ensuring the accuracy and representativeness of the baseline data. This data can be obtained by querying the temperature data of historical qualified batches of workpieces from the production management system (such as the MES system).

[0055] The mean and standard deviation of the operating temperature are calculated based on all real-time workpiece temperatures to perform statistical analysis on the baseline data. The mean operating temperature represents the central tendency under acceptable process conditions, while the standard deviation quantifies the inherent fluctuation range of the process parameter.

[0056] The dynamic upper and lower warning limits for real-time workpiece temperature are calculated based on the mean and standard deviation of the operating temperature. This step is the core of dynamic threshold setting. The dynamic upper and lower warning limits are typically based on statistical process control (SPC) principles, for example, set as the mean plus or minus a certain multiple of the standard deviation (e.g., mean ± 2.7 times the standard deviation). This calculation method allows the warning thresholds to adaptively adjust according to the actual performance of the process, rather than remaining static. These calculation results can be stored in the equipment's memory and updated in real time for subsequent process parameter monitoring.

[0057] This solution introduces a dynamic early warning limit mechanism to achieve adaptive monitoring of key process parameters in wave soldering. Its core lies in transforming static specification limits into dynamic statistical intervals based on historical qualified workpiece data, thereby enabling real-time adjustment of the early warning threshold according to actual fluctuations in the production process. By acquiring the real-time workpiece temperature of the third-to-last preset number of qualified workpieces, the validity of the baseline data is ensured, utilizing samples with good performance in historical production as a reference standard. Based on this, the mean and standard deviation of the operating temperature are calculated, quantifying the distribution characteristics and dispersion of the current process parameters. Finally, dynamic upper and lower early warning limits are calculated based on the mean and standard deviation, allowing the early warning interval to automatically shift or scale with minor fluctuations in the process. This design allows the monitoring system to move beyond relying on unchanging rigid indicators, enabling it to more sensitively detect abnormal drifts in process parameters based on actual equipment performance, while ensuring product quality. This effectively improves the intelligence level of the monitoring system and its adaptability to complex production environments.

[0058] In some preferred embodiments, the wave soldering process monitoring method further includes the step of: C1. For each parameter in the process dataset, obtain the set of actual CPK indices corresponding to that parameter; the set of actual CPK indices includes the actual CPK indices corresponding to multiple statistical batches. C2. For each parameter in the process dataset, generate a predicted CPK index based on the actual CPK index set corresponding to that parameter. C3. When the actual CPK index of the fourth consecutive set number of statistical batches decreases batch by batch and the actual CPK index of the current statistical batch is greater than or equal to the second set lower limit of the CPK index, or the predicted CPK index is less than the second set lower limit of the CPK index, an abnormal warning message for the acquisition component of the parameter corresponding to the actual CPK index is generated. C4. When the actual CPK index corresponding to the fifth consecutive set number of statistical batches decreases batch by batch, and the actual CPK index of the current statistical batch is less than the second set lower limit of the CPK index and greater than or equal to the third set lower limit of the CPK index, and the rate of decrease of the actual CPK index of the current statistical batch is less than or equal to the first set rate, a shutdown maintenance reminder is generated for the acquisition component corresponding to the actual CPK index. C5. When the actual CPK index of the fifth consecutive set number of statistical batches decreases batch by batch and the actual CPK index of the current statistical batch is less than the third set lower limit of the CPK index, or when the actual CPK index of the fourth consecutive set number of statistical batches decreases batch by batch and the rate of decrease of the actual CPK index of the current statistical batch is greater than or equal to the second set rate, control the wave soldering equipment to stop; the first set rate is less than the second set rate.

[0059] The actual CPK index is a key indicator for measuring process capability. It reflects the deviation of the process average from the specification center and the magnitude of process variability. Obtaining the actual CPK index set means that the system needs to calculate, record, and store the CPK value of each parameter in different statistical batches. This can be done by calculating the CPK value and incorporating it into the set after each statistical batch, based on the parameter data collected within that batch and in conjunction with preset specification upper and lower limits. For example, a fixed number of workpieces (e.g., 100 workpieces) can be used as a statistical batch, and the CPK value of that batch can be calculated.

[0060] For each parameter in the process dataset, a predicted CPK index is generated based on the corresponding set of actual CPK indices. The predicted CPK index is the future CPK value inferred through data analysis based on historical actual CPK index sets. Various statistical forecasting methods can be used to generate the predicted CPK index. For example, a linear regression model can be used to fit the historical CPK index sequence to predict future CPK trends; alternatively, time series analysis methods, such as ARIMA models or exponential smoothing, can be used to model the changing patterns of the CPK index and predict its future value. This predictive capability enables the system to identify potential process capability degradation risks in advance.

[0061] The "Actual CPK index decreasing batch by batch for the fourth consecutive preset number of statistical batches" is a trend judgment mechanism used to identify early signs of a continuous decline in process capability. For example, the fourth preset number can be set to 3 batches; when the CPK values ​​of 3 consecutive batches show a decreasing trend, this condition is met. The second preset lower limit of the CPK index is a preset threshold used to distinguish different degrees of process capability decline. For example, the second preset lower limit of the CPK index can be set to 1.0, indicating that the process capability is still acceptable but has shown a downward trend. Predicting that the CPK index is less than the second preset lower limit is a warning condition based on future trends. Even if the current actual CPK index is still within an acceptable range, if the prediction shows that it will soon fall below the preset lower limit, an early warning should be issued. The generated abnormal warning information can be a pop-up prompt on the software interface, a log record, or a notification. Its purpose is to remind operators or maintenance personnel to pay attention to potential problems with the acquisition components of specific parameters, such as sensor drift or unstable data acquisition.

[0062] When the actual CPK index of the fifth consecutive set number of statistical batches decreases batch by batch, and the actual CPK index of the current statistical batch is less than the second preset lower limit of the CPK index but greater than or equal to the third preset lower limit of the CPK index, and the rate of decrease of the actual CPK index of the current statistical batch is less than or equal to the first preset rate, a shutdown maintenance reminder is generated for the acquisition component corresponding to the parameter of the actual CPK index. Compared with step C3 above, "the actual CPK index of the fifth consecutive set number of statistical batches decreases batch by batch" requires a longer consecutive decreasing batch (the fifth preset number), indicating that the process capability decline trend is more obvious and continuous. For example, the fifth preset number can be set to 5 batches. "The actual CPK index of the current statistical batch is less than the second preset lower limit of the CPK index but greater than or equal to the third preset lower limit of the CPK index" defines a moderate process capability decline range. For example, if the second preset lower limit of the CPK index is 1.0 and the third preset lower limit of the CPK index is 0.67, then this condition means that the CPK index of the current statistical batch is between [0.67, 1.0). The rate of decrease of the actual CPK index in the current statistical batch (the slope obtained by linear regression fitting of the actual CPK index set) is used to measure how fast the CPK index decreases. The first preset rate is a relatively small negative value, indicating that the CPK index is decreasing slowly and there is still time for planned maintenance. For example, the first preset rate can be set to -0.02. The generated downtime maintenance reminder usually generates a maintenance work order, notifying relevant personnel to arrange planned downtime for inspection, calibration, or replacement of the corresponding data acquisition components.

[0063] When the actual CPK index decreases batch by batch for the fifth consecutive preset number of statistical batches and the actual CPK index of the current statistical batch is less than the third preset lower limit of the CPK index, or when the actual CPK index decreases batch by batch for the fourth consecutive preset number of statistical batches and the rate of decrease of the actual CPK index of the current statistical batch is greater than or equal to the second preset rate, the wave soldering equipment is shut down; the first preset rate is less than the second preset rate. "The actual CPK index decreases batch by batch for the fifth consecutive preset number of statistical batches and the actual CPK index of the current statistical batch is less than the third preset lower limit of the CPK index" indicates that the process capability has severely declined, and the CPK index is below a critical threshold. For example, the third preset lower limit of the CPK index is 0.67; below this value usually means the process is unqualified. "The actual CPK index decreases batch by batch for the fourth consecutive preset number of statistical batches and the rate of decrease of the actual CPK index of the current statistical batch is greater than or equal to the second preset rate" indicates that even if the CPK value has not yet fallen below the third preset lower limit of the CPK index, if the rate of decrease is very rapid (the second preset rate is a large negative value, such as -0.05), it indicates an emergency requiring immediate shutdown. The first preset rate is less than the second preset rate, meaning the second preset rate represents a faster descent rate. Controlling the wave soldering equipment to shut down is the highest level of response, indicating that the system has determined the process capability has severely deteriorated or is about to go out of control. To avoid generating a large number of defective products or equipment damage, the wave soldering equipment must be stopped immediately.

[0064] Preferably, the following scenarios are considered temporary fluctuations, and no abnormal warning information or shutdown maintenance reminders are generated: 1. The CPK index of the first two statistical batches decreases after changing the PCB product model, changing the solder bath, or after a major equipment overhaul; 2. A small number of temporary equipment adjustments cause the CPK index of a certain statistical batch to drop, and the CPK index of the next statistical batch of that batch automatically rebounds; 3. In a batch with a continuous decline, the CPK index corresponding to any parameter rebounds. If the CPK index corresponding to the workpiece temperature continues to decline (the CPK index of multiple consecutive statistical batches is less than the preset value), the heating tube or temperature control module of the preheating zone is considered abnormal; if the CPK index corresponding to the solder wave flatness parameter continues to decline, the solder pump, solder bath outlet, or liquid level sensor is considered abnormal; if the CPK index corresponding to the real-time X-axis vibration amplitude, real-time Y-axis vibration amplitude, or real-time Z-axis vibration amplitude continues to decline, the track roller, chain, transmission motor, or transmission bearing is considered abnormal; if the CPK index corresponding to the real-time flux flow rate continues to decline, the flow valve, delivery pipeline, or pressure pump is considered abnormal; if the CPK index corresponding to the real-time oxygen content continues to decline, the furnace body seal or nitrogen supply pipeline is considered abnormal.

[0065] This application introduces a statistical process control method based on the CPK index, achieving a leap from instantaneous monitoring of a single parameter to trend analysis of process capability. By continuously acquiring and analyzing the actual CPK index set of each parameter in the process data set, time-series data reflecting the long-term evolution of process stability can be constructed, thus eliminating excessive reliance on instantaneous values ​​and effectively overcoming the limitation of traditional monitoring schemes in predicting long-term equipment performance trends. Furthermore, by generating a predicted CPK index based on the actual CPK index set, this application enables the system to have a proactive risk identification capability, identifying potential degradation trends before actual process capability deterioration. By setting multi-level early warning and control logic (including abnormal warning information, shutdown maintenance reminders, and equipment shutdown), the system can automatically trigger corresponding response measures based on the severity and rate of CPK index decline. This hierarchical processing mechanism not only ensures production continuity and avoids production interruptions due to sudden failures, but also realizes a shift from passive response to proactive preventive maintenance through quantitative assessment of the health of the acquired components, significantly improving the overall reliability of the wave soldering process and the targeted nature of equipment maintenance. By combining the comprehensive collection and correlation of multi-source heterogeneous physical field parameters in the above-mentioned wave soldering process monitoring method, this application can accurately predict and classify the performance degradation of key components of wave soldering equipment (such as transmission mechanism, flux supply system, nitrogen protection system, etc.), which greatly improves the stability of equipment operation and maintenance efficiency.

[0066] The following is a concrete example to illustrate this. Assume the process dataset contains real-time chain speed parameters. The system forms a statistical batch every 100 workpieces welded. For each batch, the system calculates the actual CPK index based on all real-time chain speed data within that batch, combined with preset upper and lower limits for chain speed specifications. These CPK indices (e.g., CPK 1.5 for batch 1, 1.48 for batch 2, and 1.45 for batch 3) are stored, forming a set of actual CPK indices for real-time chain speeds. Based on this set of actual CPK indices for real-time chain speeds, the system can use linear regression analysis to fit the trend line of CPK index changes over time, thereby predicting the real-time chain speed CPK index for the next or future batches. For example, if historical data shows a linear downward trend in the CPK index, the system can predict that the CPK index will continue to decrease in the future. Assume the second preset lower limit for the CPK index is 1.0, and the fourth preset quantity is 3 batches. If the actual CPK index of the real-time chain speed decreases batch by batch for three consecutive batches (e.g., from 1.15 to 1.12 and then to 1.09), and the CPK index of the current batch (1.09) is still greater than or equal to 1.0, or the system predicts that the CPK index of the next batch will be less than 1.0, the system will generate an abnormal warning message for the real-time chain speed acquisition component. For example, a pop-up window will appear on the monitoring software interface prompting "Chain speed CPK index continues to decrease, please pay attention to the chain speed sensor status". Assume that the third preset lower limit of the CPK index is 0.67, the fifth preset quantity is 5 batches, and the first preset rate is -0.02. If the actual CPK index of the real-time chain speed continues to decline for five consecutive batches (e.g., from 1.05 to 0.98, then to 0.92, 0.85, and 0.78), and the CPK index of the current batch (0.78) is less than 1.0 but greater than or equal to 0.67, and the CPK decline rate calculated by linear regression is -0.015 (less than or equal to -0.02), the system will generate a shutdown maintenance reminder for the real-time chain speed acquisition component. For example, it will automatically generate a work order to notify equipment maintenance personnel to check the chain speed sensor or transmission mechanism during the planned shutdown. Assume the second preset rate is -0.05. If the actual CPK index of the real-time chain speed continues to decline for 5 consecutive batches, and the CPK index of the current batch (e.g., 0.65) is less than 0.67, or if the actual CPK index of the real-time chain speed decreases batch by batch for 4 consecutive batches, and the rate of decrease is -0.06 (greater than or equal to -0.05), the system will immediately control the wave soldering equipment to stop and issue an emergency shutdown alarm, prompting the operator to immediately check for faults in the chain speed acquisition component or the transmission system.

[0067] Through the aforementioned technical solution, this application, by introducing a statistical process control method based on the CPK index, achieves a shift from instantaneous monitoring of a single parameter to trend analysis of process capability. By continuously acquiring and analyzing the actual CPK index set of each parameter in the process data set, time-series data reflecting the long-term evolution of process stability can be constructed, thereby eliminating excessive reliance on instantaneous values ​​and effectively overcoming the limitation of traditional monitoring schemes in predicting long-term trends in equipment performance. Furthermore, by generating a predicted CPK index based on the actual CPK index set, this application enables the system to possess a forward-looking risk identification capability, allowing it to identify potential degradation trends before actual deterioration of process capability. By setting multi-level early warning and control logic (including abnormal early warning information, shutdown maintenance reminders, and equipment shutdown), the system can automatically trigger corresponding response measures based on the severity and rate of CPK index decline. This hierarchical processing mechanism not only ensures production continuity and avoids production interruptions caused by sudden failures, but also achieves a shift from passive response to proactive preventative maintenance through quantitative assessment of the health of the acquired components, significantly improving the overall reliability of the wave soldering process and the targeted nature of equipment maintenance, thereby effectively reducing product quality fluctuations. By combining the comprehensive collection and correlation of multi-source heterogeneous physical field parameters in the above-mentioned wave soldering process monitoring method, this application can accurately predict and classify the performance degradation of key components of wave soldering equipment (such as transmission mechanism, flux supply system, nitrogen protection system, etc.), which greatly improves the stability of equipment operation and maintenance efficiency.

[0068] In some preferred embodiments, the calculation processes for real-time workpiece temperature, real-time chain speed, real-time X-axis vibration amplitude, real-time Y-axis vibration amplitude, real-time Z-axis vibration amplitude, real-time flux flow rate, solder wave flatness parameter, and real-time oxygen content corresponding to the CPK index are the same. The calculation process for the CPK index corresponding to the real-time X-axis vibration amplitude includes: D1. When the number of workpieces that have completed the wave soldering process reaches the preset workpiece quantity corresponding to the statistical batch, calculate the mean value of the X-axis amplitude and the standard deviation of the X-axis amplitude based on all real-time X-axis vibration amplitude values, and obtain the preset fixed specification lower limit and preset fixed specification upper limit of the real-time X-axis vibration amplitude. D2. Calculate the CPK index corresponding to the real-time X-axis vibration amplitude based on the mean X-axis amplitude, standard deviation of X-axis amplitude, preset fixed specification lower limit and preset fixed specification upper limit; The wave soldering process monitoring method also includes the following steps: E1. When the number of workpieces that have completed the wave soldering process reaches the preset number of workpieces corresponding to the statistical batch, for each parameter in the process data set, calculate the CP index and PPK index corresponding to the parameter based on the mean X-axis amplitude, standard deviation X-axis amplitude, preset fixed specification lower limit and preset fixed specification upper limit of the parameter. E2. Integrate all CPK indices, all CP indices, and all PPK indices into a process capability matrix.

[0069] The technical feature of "when the number of workpieces completed in the wave soldering process reaches the preset workpiece quantity corresponding to the statistical batch" aims to ensure that the amount of data for statistical analysis is sufficiently representative and statistically significant. The preset workpiece quantity can be a fixed value, for example, it can be set to perform a statistical calculation after every 50 or 100 workpieces are completed.

[0070] The technical feature of "calculating the mean and standard deviation of X-axis amplitude based on all real-time X-axis vibration amplitudes" is used to quantify the central tendency and dispersion of a specific parameter (taking real-time X-axis vibration amplitude as an example here). The mean X-axis amplitude can be calculated using the arithmetic mean, which is the sum of all real-time X-axis vibration amplitudes divided by the sample size; the standard deviation of X-axis amplitude can be calculated using the sample standard deviation formula, reflecting the degree of fluctuation of data points around the mean.

[0071] The technical feature of "obtaining the preset fixed lower limit and preset fixed upper limit of real-time X-axis vibration amplitude" defines the acceptable performance boundary of process parameters. The preset fixed lower limit and preset fixed upper limit can be manually set and entered into the system by process engineers according to product design requirements, equipment performance indicators or industry standards. These limits are usually fixed and serve as hard standards for judging whether the product is qualified.

[0072] The technical feature of "calculating the CPK index corresponding to the real-time X-axis vibration amplitude based on the mean X-axis amplitude, the standard deviation of the X-axis amplitude, the preset fixed lower limit, and the preset fixed upper limit" is used to evaluate the short-term capability of a process, that is, the degree of concentration and deviation of the process output within the specification limits. The CPK index (ProcessCapabilityKRatio) can be calculated based on the standard formula: CPK=min[(USL-μ) / (3σ),(μ-LSL) / (3σ)], where USL is the preset fixed upper limit, LSL is the preset fixed lower limit, μ is the mean X-axis amplitude, and σ is the standard deviation of the X-axis amplitude.

[0073] The Process Capability Index (CP) measures a process's capability under ideal conditions, without considering whether the process center of gravity shifts. Its formula is CP = (USL - LSL) / (6σ). The Process Performance Index (PPK) measures the actual performance of a process, taking into account the process center of gravity shift. Its formula is similar to that of CPK, but it typically uses the long-term standard deviation (calculated based on data from all statistical batches).

[0074] The technical feature of "integrating all CPK indices, all CP indices, and all PPK indices into a process capability matrix" is used to provide a structured display and management of multiple key process capability indicators. The process capability matrix can be a two-dimensional table, where rows can represent different process parameters (such as real-time workpiece temperature, real-time chain speed, real-time X-axis vibration amplitude, etc.), and columns can represent different capability indicators (CPK index, CP index, and PPK index). This matrix can be presented in a graphical interface, for example, using color coding (e.g., green for excellent, yellow for concern, and red for failure) to intuitively display the process capability status of each parameter; alternatively, it can generate exportable and analyzable data reports to facilitate further data mining and decision support.

[0075] This solution introduces statistical process control theory to transform discrete real-time monitoring data into statistically significant process capability indicators, thus achieving a leap from instantaneous monitoring to process trend assessment. The method first performs statistical analysis on parameters such as real-time X-axis vibration amplitude when the number of workpieces completing the wave soldering process reaches the preset workpiece quantity corresponding to the statistical batch, calculating the mean and standard deviation of the X-axis amplitude. By performing statistical calculations when the preset workpiece quantity is reached, random errors from single measurements can be effectively smoothed, ensuring that the calculated mean and standard deviation accurately reflect the process level of the current batch. Simultaneously, combining preset fixed specification lower and upper limits, the CPK index corresponding to the real-time X-axis vibration amplitude is calculated using the mean, standard deviation, and preset specification upper and lower limits. This accurately quantifies the distribution concentration and deviation of each physical field parameter within the specification range, providing a quantitative basis for assessing process stability. Based on this, the corresponding CP index and PPK index are further calculated for each parameter in the process dataset. The CP index assesses the potential capability of the process, while the PPK index assesses its actual performance. Combining both allows for a comprehensive analysis of the process from both its potential and actual performance dimensions, identifying whether the process is limited by its own fluctuations or by center offset. Finally, all CPK indices, all CP indices, and all PPK indices are integrated into a process capability matrix. This matrix-based management approach not only achieves unified management of multi-dimensional parameters throughout the wave soldering process but also provides structured data support for subsequent fault prediction and maintenance decisions. This enables managers to intuitively grasp the evolution of process capability for each process parameter, allowing for proactive intervention before process capability deteriorates. Through this approach, this application, building upon basic real-time monitoring and complex risk identification, further provides a quantitative assessment of the long-term stability and process capability of the wave soldering process, making it possible to predict potential equipment performance degradation and process capability loss, thereby effectively improving the stability of welding quality.

[0076] In some preferred embodiments, the wave soldering process monitoring method further includes the step of: F1. Map each parameter in the process dataset to the wave soldering 3D digital twin model.

[0077] A wave soldering 3D digital twin model can be understood as a virtual digital copy of the wave soldering equipment, capable of reflecting the real-time operating status and process parameters of the physical equipment. Mapping each parameter in the process dataset to the wave soldering 3D digital twin model involves associating the parameters collected in the process dataset with corresponding components, areas, or indicators in the 3D digital twin model through data interfaces or visualization engines. For example, the real-time workpiece temperature of the preheating zone can be mapped to the color change of the temperature field in the preheating zone; the real-time chain speed can be mapped to the dynamic speed display of the conveyor belt; the solder wave height can be mapped to the dynamic undulations of the molten solder surface in the solder pot area; and the vibration amplitude can be mapped to the vibration animation or color warning of the transmission components. The purpose of step F1 is to transform abstract process data into intuitive and visual information, facilitating user understanding and monitoring.

[0078] This application's solution maps each parameter in the process dataset to a three-dimensional digital twin model of wave soldering, enabling the previously scattered and abstract process data to be presented in a unified and intuitive way within a virtual environment. When the wave soldering equipment is running, the multi-source heterogeneous physical field parameters acquired in real time and the calculated solder wave flatness height parameters are dynamically updated into the three-dimensional digital twin model, visualized for example through color coding, dynamic charts, virtual instruments, or component status changes. Thus, operators can quickly understand the overall operating status of the current wave soldering process, the real-time values ​​of key parameters, and potential anomalies by observing the three-dimensional digital twin model, without needing to delve into complex numerical reports. This visualization method greatly enhances the ability to perceive the process status, helping operators to more quickly and accurately locate problems and understand the interactions between different parameters.

[0079] Please refer to Figure 2 , Figure 2This application provides a schematic diagram of the structure of an electronic device according to an embodiment of the present application. The electronic device includes a processor 101 and a memory 102. The processor 101 and the memory 102 are interconnected and communicate with each other via a communication bus 103 and / or other forms of connection mechanisms (not shown). The memory 102 stores computer-readable instructions executable by the processor 101. When the electronic device is running, the processor 101 executes these computer-readable instructions to perform the method in any optional implementation of the above embodiments, thereby achieving the following function: Step S1, obtaining a unique identifier of the workpiece to be processed, and synchronously and continuously collecting multi-source heterogeneous physical field parameters of the workpiece to be processed; multi-source heterogeneous physical field parameters... The data includes real-time workpiece temperature, real-time chain speed, real-time solder wave height in the smoothing area, real-time solder wave height in the scrambling area, real-time X-axis vibration amplitude, real-time Y-axis vibration amplitude, real-time Z-axis vibration amplitude, real-time flux flow rate, and real-time oxygen content; Step S2: Calculate the solder wave flatness parameter based on the real-time solder wave height in the smoothing area and the real-time solder wave height in the scrambling area; Step S3: Associate and bind the unique identifier with the solder wave flatness parameter and the parameters other than the real-time solder wave height in the smoothing area and the real-time solder wave height in the scrambling area from the multi-source heterogeneous physical field parameters to obtain the process dataset corresponding to the workpiece to be processed; Step S4: Based on the preset multi-parameter coupling judgment logic, perform correlation analysis on the process dataset to identify composite process risks.

[0080] This application embodiment also provides a computer-readable storage medium storing a computer program thereon. When the computer program is executed by a processor, it executes the method in any optional implementation of the above embodiments to achieve the following functions: Step S1: Obtain the unique identifier of the workpiece to be processed, and synchronously and continuously collect the multi-source heterogeneous physical field parameters of the workpiece to be processed; the multi-source heterogeneous physical field parameters include real-time workpiece temperature, real-time chain speed, real-time solder wave height in the smoothing region, real-time solder wave height in the scrambling region, real-time X-axis vibration amplitude, real-time Y-axis vibration amplitude, real-time Z-axis vibration amplitude, real-time flux flow rate, and real-time oxygen content; Step S2: Calculate the solder wave flatness parameter based on the real-time solder wave height in the smoothing region and the real-time solder wave height in the scrambling region; Step S3: Associate and bind the unique identifier with the solder wave flatness parameter and the parameters in the multi-source heterogeneous physical field parameters other than the real-time solder wave height in the smoothing region and the real-time solder wave height in the scrambling region to obtain the process dataset corresponding to the workpiece to be processed; Step S4: Perform correlation analysis on the process dataset based on the preset multi-parameter coupling judgment logic to identify composite process risks. The computer-readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0081] As can be seen from the above, the wave soldering process monitoring method, electronic device and storage medium provided in this application obtain the unique identification of the workpiece to be processed and simultaneously collect multi-source heterogeneous physical field parameters. The parameters are associated and bound with the identification to form a process dataset. Based on the multi-parameter coupling judgment logic, the composite process risk is identified. This effectively solves the problems of asynchronous parameter acquisition, untraceable data and lack of multi-parameter linkage analysis in the prior art, which leads to the inability to identify composite process degradation. It has the advantages of improving the accuracy and traceability of welding process monitoring.

[0082] In the embodiments provided in this application, it should be understood that relational terms such as first and second are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations.

[0083] The above description is merely an embodiment of this application and is not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for monitoring wave soldering process, characterized in that, The wave soldering process monitoring method includes the following steps: S1. Obtain the unique identifier of the workpiece to be processed, and synchronously and continuously collect the multi-source heterogeneous physical field parameters of the workpiece to be processed; the multi-source heterogeneous physical field parameters include real-time workpiece temperature, real-time chain speed, real-time solder wave height in the smoothing region, real-time solder wave height in the turbulence region, real-time X-axis vibration amplitude, real-time Y-axis vibration amplitude, real-time Z-axis vibration amplitude, real-time flux flow rate, and real-time oxygen content; S2. Calculate the tin wave smoothness parameters based on the tin wave height in the real-time smoothing region and the tin wave height in the real-time turbulence region. S3. Associate and bind the unique identifier with the tin wave flatness parameter and the parameters in the multi-source heterogeneous physical field parameters other than the tin wave height in the real-time smoothing region and the tin wave height in the real-time disturbance region to obtain the process dataset corresponding to the workpiece to be processed. S4. Based on the preset multi-parameter coupling judgment logic, perform correlation analysis on the process dataset to identify composite process risks.

2. The wave soldering process monitoring method according to claim 1, characterized in that, The risks associated with the composite process include transmission equipment malfunctions, poor connections, solder joint voids and oxidation, and aging of the conveying mechanism.

3. The wave soldering process monitoring method according to claim 2, characterized in that, Step S4 includes: S41. Calculate the chain speed fluctuation amplitude based on the preset chain speed and all the real-time chain speeds in the process dataset, calculate the first average flow rate based on all the real-time flux flow rates in the process dataset, calculate the average oxygen content based on all the real-time oxygen content in the process dataset, extract the real-time workpiece temperature of the workpiece to be processed in the preheating zone from all the real-time workpiece temperatures in the process dataset, and determine the peak temperature of the preheating zone based on all the extracted real-time workpiece temperatures. S42. Obtain the second average flow rate corresponding to the first preset number of workpieces, obtain the actual CPK index of the X-axis, the actual CPK index of the Y-axis, and the actual CPK index of the Z-axis corresponding to the current statistical batch and the first second preset number of statistical batches, and calculate the average chain speed based on all real-time chain speeds corresponding to the current statistical batch; the workpiece model of the second average flow rate is the same as the workpiece model of the workpiece to be processed, and the actual CPK index of the X-axis, the actual CPK index of the Y-axis, and the actual CPK index of the Z-axis are the actual CPK indices corresponding to the real-time X-axis vibration amplitude, the real-time Y-axis vibration amplitude, and the real-time Z-axis vibration amplitude, respectively; S43. When the chain speed fluctuation amplitude is greater than the preset fluctuation amplitude and all the real-time Z-axis vibration amplitudes in the process data set are greater than the dynamic upper warning limit corresponding to the real-time Z-axis vibration amplitude, it is determined that there is an abnormal risk of the transmission equipment. S44. When the tin wave flatness parameter is greater than or equal to the preset secondary alarm threshold and the peak temperature of the preheating zone is less than the preset fixed specification lower limit corresponding to the preheating zone, it is determined that there is a risk of connection failure. S45. Calculate the first flux decrease ratio based on the first average flow rate and the preset flux flow rate, calculate the second flux decrease ratio based on the second average flow rate and the preset flux flow rate, and determine that there is a risk of solder joint void oxidation when the first flux decrease ratio and all the second flux decrease ratios are greater than the preset decrease ratio and the average oxygen content is greater than the preset oxygen content upper limit. S46. Calculate the chain speed drift ratio based on the average chain speed and the preset average chain speed. When all the actual CPK indices of the X-axis, all the actual CPK indices of the Y-axis, and all the actual CPK indices of the Z-axis are less than the lower limit of the first preset CPK index and the chain speed drift ratio is greater than the preset drift ratio, it is determined that there is a risk of aging of the conveyor mechanism.

4. The wave soldering process monitoring method according to claim 3, characterized in that, The wave soldering process monitoring method also includes the following steps: A1. For each parameter in the process dataset other than the tin wave flatness parameter, obtain the dynamic upper warning limit, dynamic lower warning limit, preset fixed specification lower limit and preset fixed specification upper limit corresponding to the parameter, and calculate the parameter deviation between the parameter and the dynamic upper warning limit and the dynamic lower warning limit respectively. The dynamic upper warning limit is less than the preset fixed specification upper limit, and the dynamic lower warning limit is greater than the preset fixed specification lower limit; A2. For each parameter in the process dataset other than the tin wave flatness parameter, when the parameter is between the dynamic upper warning limit and the dynamic lower warning limit and the parameter deviation is less than the preset deviation, a software pop-up warning is issued. A3. For each parameter in the process dataset other than the tin wave flatness parameter, when the parameter is between its corresponding dynamic upper warning limit and the preset fixed specification upper limit or between its corresponding dynamic lower warning limit and the preset fixed specification lower limit, an audible and visual warning is issued. A4. For each parameter in the process data set except for the tin wave flatness parameter, when the parameter is greater than its corresponding preset fixed specification upper limit or less than its corresponding preset fixed specification lower limit, the machine is stopped by locking the component. A5. When the tin wave flatness parameter is greater than or equal to the preset first-level alarm threshold and less than the preset second-level alarm threshold, a software pop-up warning will be issued. A6. When the tin wave flatness parameter is greater than or equal to the preset level 2 alarm threshold and less than the preset level 3 alarm threshold, an audible and visual warning will be issued. A7. When the tin wave flatness parameter is greater than or equal to the preset level three alarm threshold, the locking mechanism is stopped.

5. The wave soldering process monitoring method according to claim 1, characterized in that, The calculation processes for the dynamic upper and lower warning limits corresponding to the real-time workpiece temperature, real-time chain speed, real-time X-axis vibration amplitude, real-time Y-axis vibration amplitude, real-time Z-axis vibration amplitude, real-time flux flow rate, and real-time oxygen content are the same. The calculation process for the dynamic upper and lower warning limits corresponding to the real-time workpiece temperature includes: B1. Obtain the real-time workpiece temperature of the first three preset number of qualified workpieces; B2. Calculate the mean operating temperature and the standard deviation of the operating temperature based on all the real-time workpiece temperatures mentioned. B3. Calculate the dynamic upper warning limit and dynamic lower warning limit corresponding to the real-time workpiece temperature based on the average working temperature and the standard deviation of the working temperature.

6. The wave soldering process monitoring method according to claim 1, characterized in that, The wave soldering process monitoring method also includes the following steps: C1. For each parameter in the process dataset, obtain the set of actual CPK indices corresponding to that parameter; the set of actual CPK indices includes the actual CPK indices corresponding to multiple statistical batches; C2. For each parameter in the process dataset, generate a predicted CPK index based on the actual CPK index set corresponding to that parameter; C3. When the actual CPK index of the fourth consecutive set number of statistical batches decreases batch by batch and the actual CPK index of the current statistical batch is greater than or equal to the second set lower limit of the CPK index, or the predicted CPK index is less than the second set lower limit of the CPK index, an abnormal warning message for the acquisition component of the parameter corresponding to the actual CPK index is generated. C4. When the actual CPK index corresponding to the fifth consecutive set number of statistical batches decreases batch by batch, and the actual CPK index of the current statistical batch is less than the second set CPK index lower limit and greater than or equal to the third set CPK index lower limit, and the rate of decrease of the actual CPK index of the current statistical batch is less than or equal to the first set rate, a shutdown maintenance reminder is generated for the acquisition component corresponding to the actual CPK index. C5. When the actual CPK index of the fifth consecutive set number of statistical batches decreases batch by batch and the actual CPK index of the current statistical batch is less than the third set lower limit of the CPK index, or when the actual CPK index of the fourth consecutive set number of statistical batches decreases batch by batch and the rate of decrease of the actual CPK index of the current statistical batch is greater than or equal to the second set rate, control the wave soldering equipment to stop; the first set rate is less than the second set rate.

7. The wave soldering process monitoring method according to claim 6, characterized in that, The calculation processes for the real-time workpiece temperature, real-time chain speed, real-time X-axis vibration amplitude, real-time Y-axis vibration amplitude, real-time Z-axis vibration amplitude, real-time flux flow rate, solder wave flatness parameter, and CPK index corresponding to real-time oxygen content are the same. The calculation process for the CPK index corresponding to the real-time X-axis vibration amplitude includes: D1. When the number of workpieces that have completed the wave soldering process reaches the preset number of workpieces corresponding to the statistical batch, calculate the mean value of the X-axis amplitude and the standard deviation of the X-axis amplitude based on all the real-time X-axis vibration amplitudes, and obtain the preset fixed specification lower limit and preset fixed specification upper limit of the real-time X-axis vibration amplitude. D2. Calculate the CPK index corresponding to the real-time X-axis vibration amplitude based on the mean X-axis amplitude, the standard deviation of X-axis amplitude, the lower limit of the preset fixed specification, and the upper limit of the preset fixed specification. The wave soldering process monitoring method further includes the following steps: E1. When the number of workpieces that have completed the wave soldering process reaches the preset number of workpieces corresponding to the statistical batch, for each parameter in the process data set, calculate the CP index and PPK index corresponding to the parameter based on the mean X-axis amplitude, standard deviation of X-axis amplitude, the preset fixed specification lower limit and the preset fixed specification upper limit. E2. Integrate all the CPK indices, all the CP indices, and all the PPK indices into a process capability matrix.

8. The wave soldering process monitoring method according to claim 1, characterized in that, The wave soldering process monitoring method also includes the following steps: F1. Map each parameter in the process dataset to the wave soldering three-dimensional digital twin model.

9. An electronic device, characterized in that, It includes a processor and a memory, the memory storing computer-readable instructions that, when executed by the processor, perform the steps of the method as described in any one of claims 1-8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it performs the steps of the method as described in any one of claims 1-8.

Images