Electronic product quality control method and system

Abstract

The invention discloses an electronic product quality control method and system, and the method employs a pre-control module as a core innovation point, and the pre-control module comprises a quality factor coupling modeling unit, a failure path dynamic mapping unit, and a self-adaptive pre-control parameter generation unit: firstly quantifying a technology-part-environment multi-factor risk through a coupling matrix, and obtaining a comprehensive risk index; pre-judging a dominant failure path based on a failure probability formula; and finally, a target pre-control instruction is generated according to the pre-control parameter adjustment model. And a multi-source heterogeneous data acquisition module (obtaining data such as welding temperature and pin flatness in real time), an intelligent process execution module (precisely executing pre-control parameters), a quality verification module (multi-dimensional feedback results) and a full-process data tracing module (storing tracing data) are matched to form closed-loop control.

Description

Technical Field

[0001] This invention relates to the field of electronic product quality control, specifically to an electronic product quality control method and system. Background Technology

[0002] The manufacturing process of electronic products is complex, and quality is dynamically affected by multiple factors such as processes, components, and environment. Traditional quality control has three major shortcomings: The response mode is passive: it mainly relies on final inspection to remove defective products, and cannot predict the evolution path of "factor coupling → failure" in advance. Latent failures (such as early corrosion and potential cold solder joints) are easy to enter the market. Blindly using pre-control parameters: Using fixed pre-control parameters or adjusting them based on experience without considering real-time changes in quality factors and failure risks leads to poor pre-control results and wasted resources; The coupling relationship is broken: Ignoring the mutual influence between process, components and environment (such as the need to adjust welding parameters when solder performance deteriorates under high temperature and high humidity), and only optimizing a single factor, it is impossible to avoid risks from the root.

[0003] Existing technologies mostly employ simple parameter threshold comparisons or conventional statistical process control (SPC), failing to address the core issues of "multi-factor coupling modeling" and "failure path-oriented pre-control," thus making it difficult to meet the high-precision quality control requirements of electronic products. Summary of the Invention

[0004] The purpose of this invention is to provide a method and system for quality control of electronic products to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a method for quality control of electronic products, comprising the following steps: Step S1, Multi-source quality factor acquisition: Real-time acquisition of process parameters such as welding temperature through a multi-source heterogeneous data acquisition module. Welding time Placement pressure Dispensing amount and component characteristic parameters, chip pin flatness PCB board warpage Solder alloy purity and environmental parameters, workshop temperature ,humidity Dust concentration Uploaded to the pre-control module; Step S2, Multi-factor Coupled Modeling: The quality factor coupled modeling unit of the pre-control module calls the single-factor risk sub-model to calculate the process risk sub-model. Component risk sub-model Environmental risk sub-model Combined with coupling matrix With failure amplification factor The comprehensive quality risk index is obtained. Determine the risk level; Step S3, Dynamic Failure Path Prediction: The dynamic failure path mapping unit calls the failure mode library, based on... With coupling coefficient ,pass The formula calculates the probability of occurrence of each failure mode and determines the dominant failure path and key influencing factors. Step S4, Adaptive Pre-Control Parameter Generation: The adaptive pre-control parameter generation unit, based on the key factors of the dominant failure path, adjusts the model calculation through pre-control parameters. After boundary verification, a pre-control command is generated and sent to the intelligent process execution module; Step S5, Precise Process Execution: The intelligent process execution module adjusts parameters including but not limited to welding temperature and mounting pressure according to pre-controlled instructions to complete the production of electronic products; Step S6, Multi-dimensional Quality Verification: The quality verification module verifies product quality through methods including but not limited to X-ray detection and electrical performance testing, and feeds back the verification results (pass / fail + failure mode (if any)) to the pre-control module; Step S7, Closed-Loop Optimization and Data Archiving: If the product is qualified, the full-process data traceability module stores the collected data, pre-control parameters, and verification results; if it is unqualified, the pre-control module corrects the coupling coefficient based on the feedback failure mode. With adjustment coefficient Regenerate the pre-control parameters; after each batch of production, update the model parameters to ensure compatibility.

[0006] Preferably, an electronic product quality control system includes a multi-source heterogeneous data acquisition module, a pre-control module, an intelligent process execution module, a quality verification module, and a full-process data traceability module; the multi-source heterogeneous data acquisition module captures process parameters, component characteristics, and environmental parameters through a combination of online sensing and offline detection, with a data acquisition frequency of up to 10Hz to ensure real-time performance and accuracy; The pre-control module includes a quality factor coupling modeling unit, a failure path dynamic mapping unit, and an adaptive pre-control parameter generation unit. The pre-control module is used to realize the pre-judgment and precise pre-control of quality risks. The pre-control module receives full-dimensional data from a multi-source heterogeneous data acquisition module. First, the quality factor coupling modeling unit quantifies the comprehensive quality risk index. Then, the failure path dynamic mapping unit locates potential failure root causes. Finally, the adaptive pre-control parameter generation unit outputs targeted pre-control instructions. At the same time, the pre-control module receives feedback results from the quality verification module and dynamically updates the model parameters to ensure that the pre-control accuracy continuously adapts to the production scenario. The intelligent process execution module includes a precision welding component, a mounting drive unit, and a dispensing control unit. It receives pre-control parameter instructions from the pre-control module and performs precise adjustments to parameters including but not limited to welding temperature and mounting pressure. The quality verification module adopts a multi-dimensional verification mode, including X-ray inspection (welding quality), infrared thermal imaging (heat dissipation characteristics), and electrical performance testing (conductivity and withstand voltage). The quality verification results are fed back to the pre-control module in real time for closed-loop optimization of pre-control parameters. If the inspection is qualified, a valid signal is fed back to the pre-control module. If a defective product is found, the specific failure mode and related factor data are fed back simultaneously, triggering the pre-control module to correct the pre-control parameters. The full-process data traceability module records the data collected by the multi-source heterogeneous data acquisition module, the pre-control parameters and failure prediction results of the pre-control module, the actual execution data of the intelligent process execution module, and the test results of the quality verification module, forming a single product's full life cycle quality archive.

[0007] Preferably, the multi-source heterogeneous data acquisition module adopts an online sensing plus offline detection fusion mode: Process sensors collect welding temperature (°C), welding time (s) Mounting pressure (N) Dispensing amount (μL); Flatness of component characteristic sensor acquisition chip pins (mm), PCB board warpage (°), Solder alloy purity (%); Environmental sensors collect workshop temperature (°C), humidity (%), dust concentration (mg / m³); sampling frequency is 10Hz, measurement accuracy , , .

[0008] Preferably, the quality factor coupling modeling unit adopts a hybrid approach of physical mechanism and data-driven method to construct a multi-dimensional quality factor coupling model and calculate the comprehensive quality risk index. The specific working steps are as follows: a1. Define core quality factors: process factors Component factors Environmental factors ; b1. Establish a single-factor risk sub-model: Process risk sub-model: ,in The process factor weights are as follows, where the weights of each factor are: , For real-time process parameters, These are standard process parameters. ; Component risk sub-model: ,in The component factor weights are as follows: , Standardized values ​​for component characteristics ( hour , hour , hour Otherwise, it will decrease proportionally. ; Environmental risk sub-model: ,in This is the environmental impact factor. ; c1. Constructing the factor coupling matrix: Based on physical mechanisms (such as high temperature and humidity accelerating solder oxidation, and uneven pins exacerbating mounting deviations), define the coupling coefficients. ,in Source factor type, To form a 3×3 coupling matrix for the target factor type. : ,in: (Process → Components) = 0.4, indicating that the soldering temperature affects the stability of the chip pins; (Process → Environment) = 0.3, indicating that high temperature exacerbates the adhesion of dust to the environment. (Component → Process) = 0.2, indicating that the pins are not flat and the mounting pressure needs to be adjusted; d1. Calculate the comprehensive quality risk index : ,in: The failure amplification factor is 1.2 (trained based on historical data). , Determined to be high risk Medium risk Low risk.

[0009] Preferably, the failure path dynamic mapping unit uses a risk factor-failure mode association method to predict potential failure paths. The specific working steps are as follows: a2. Establish a failure mode library: Based on historical failure data of electronic products from external input, define core failure modes (such as solder joint defects, chip pin corrosion, PCB board short circuits) and corresponding failure characteristics (such as solder joint defects corresponding to soldering temperature fluctuations + pin flatness deviations). b2. Calculate the probability of the failed path: In the formula: For the first The probability of occurrence of each failure mode. Failure mode weights (e.g., failure mode weights for poor solder joints: 0.4, corrosion: 0.3, short circuit: 0.3). The correlation between the quality factor and the failure mode (e.g., the correlation between welding temperature and poor weld). ); c2. Determine the dominant failure path: according to... Sort the failure paths from highest to lowest and select the top two as the dominant failure paths (e.g., "low soldering temperature + pin flatness deviation → cold solder joint"), and output the path characteristic parameters (correlation factor type, threshold requirement).

[0010] Preferably, the adaptive pre-control parameter generation unit adopts a failure path-oriented method to generate targeted pre-control parameters. The specific working steps are as follows: a3. Extract the key influencing factors of the dominant failure path: for example, the key factor for the "cold solder joint" path is the soldering temperature. Placement pressure Pin flatness ; b3. Establish a pre-control parameter adjustment model: in: The adjusted pre-control parameters, To adjust the coefficient (process parameter) Component compatibility parameters ), (The extent to which the risk deviates from the safety threshold). For the sign function, where When it is 1, The time is -1; c3. Boundary verification: Pre-controlled parameters must meet the equipment capability boundaries (such as welding temperature). Placement pressure When the boundary is exceeded, the correlation factor parameters are adjusted synchronously (e.g., If the upper limit has been reached, increase the welding time. ); Generate pre-control instructions: Adjust the process parameters ( (etc.) are sent to the intelligent process execution module, while the dominant failure path and pre-control logic are marked.

[0011] Preferably, the coupling coefficient is updated based on the quality verification result of the quality verification module. With adjustment coefficient This ensures that the accuracy of pre-control continuously adapts to changes in production scenarios.

[0012] Compared with existing technologies, the beneficial effects of this invention are: This invention, through multi-factor coupling modeling and dynamic prediction of failure paths in the pre-control module, breaks the traditional passive mode of post-process defective product rejection, and proactively identifies latent failure risks (such as soldering defects and chip corrosion) caused by process-component-environment coupling. Combined with an adaptive pre-control parameter generation mechanism, it reduces the incidence of latent failures in electronic products. This invention constructs a quality factor coupling matrix to quantify the interactions between processes, components, and the environment (such as the effect of high welding temperatures on chip pin stability), avoiding the limitations of single-factor optimization. Pre-control parameters are generated based on failure path targeting, rather than relying on empirical settings, thus improving pre-control efficiency and reducing resource waste caused by ineffective parameter adjustments.

[0013] This invention updates the coupling coefficient and adjustment coefficient based on the quality verification results after each batch of production, so that the pre-control model can continuously adapt to dynamic changes such as material batch differences, equipment aging, and environmental fluctuations. It eliminates the need for frequent manual calibration, ensuring the accuracy and stability of the system in long-term operation and extending the effective service life of the production line. Attached Figure Description

[0014] Figure 1 This is a schematic diagram of the method flow of the present invention; Figure 2 This is a schematic diagram of the system structure of the present invention. Detailed Implementation

[0015] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0016] Please see Figure 1-2 This invention provides a technical solution: a method for quality control of electronic products, comprising the following steps: Step S1, Multi-source quality factor acquisition: Real-time acquisition of process parameters such as welding temperature through a multi-source heterogeneous data acquisition module. Welding time Placement pressure Dispensing amount and component characteristic parameters, chip pin flatness PCB board warpage Solder alloy purity and environmental parameters, workshop temperature ,humidity Dust concentration Uploaded to the pre-control module; Step S2, Multi-factor Coupled Modeling: The quality factor coupled modeling unit of the pre-control module calls the single-factor risk sub-model to calculate the process risk sub-model. Component risk sub-model Environmental risk sub-model Combined with coupling matrix With failure amplification factor The comprehensive quality risk index is obtained. Determine the risk level; Step S3, Dynamic Failure Path Prediction: The dynamic failure path mapping unit calls the failure mode library, based on... With coupling coefficient ,pass The formula calculates the probability of occurrence of each failure mode and determines the dominant failure path and key influencing factors. Step S4, Adaptive Pre-Control Parameter Generation: The adaptive pre-control parameter generation unit, based on the key factors of the dominant failure path, adjusts the model calculation through pre-control parameters. After boundary verification, a pre-control command is generated and sent to the intelligent process execution module; Step S5, Precise Process Execution: The intelligent process execution module adjusts parameters including but not limited to welding temperature and mounting pressure according to pre-controlled instructions to complete the production of electronic products; Step S6, Multi-dimensional Quality Verification: The quality verification module verifies product quality through methods including but not limited to X-ray detection and electrical performance testing, and feeds back the verification results (pass / fail + failure mode (if any)) to the pre-control module; Step S7, Closed-Loop Optimization and Data Archiving: If the product is qualified, the full-process data traceability module stores the collected data, pre-control parameters, and verification results; if it is unqualified, the pre-control module corrects the coupling coefficient based on the feedback failure mode. With adjustment coefficient Regenerate the pre-control parameters; after each batch of production, update the model parameters to ensure compatibility.

[0017] Furthermore, an electronic product quality control system includes a multi-source heterogeneous data acquisition module, a pre-control module, an intelligent process execution module, a quality verification module, and a full-process data traceability module; the multi-source heterogeneous data acquisition module captures process parameters, component characteristics, and environmental parameters through a combination of online sensing and offline detection, with a data acquisition frequency of up to 10Hz to ensure real-time performance and accuracy; The pre-control module includes a quality factor coupling modeling unit, a failure path dynamic mapping unit, and an adaptive pre-control parameter generation unit. The pre-control module is used to realize the pre-judgment and precise pre-control of quality risks. The pre-control module receives full-dimensional data from the multi-source heterogeneous data acquisition module. First, the quality factor coupling modeling unit quantifies the comprehensive quality risk index. Then, the failure path dynamic mapping unit locates the potential failure root causes. Finally, the adaptive pre-control parameter generation unit outputs targeted pre-control instructions. At the same time, the pre-control module receives feedback results from the quality verification module and dynamically updates the model parameters to ensure that the pre-control accuracy continuously adapts to the production scenario. The intelligent process execution module includes a precision welding component, a placement drive unit, and a dispensing control unit. It receives pre-control parameter commands from the pre-control module and executes precise adjustments to parameters including, but not limited to, welding temperature and placement pressure. It should be noted that the precision welding component, placement drive unit, and dispensing control unit are existing mature technologies, so their detailed structures will not be described here. The intelligent process execution module precisely executes the pre-control commands output by the creative pre-control module, translating decisions into actual production actions. Through the precision welding component and placement drive unit, it adjusts key process parameters such as welding temperature, placement pressure, and dispensing volume, achieving a control accuracy of ±1℃ for temperature and ±0.1N for pressure, ensuring the accurate implementation of the pre-control logic. During execution, it feeds back the actual process parameters to the creative pre-control module in real time, forming a short closed loop of "command → execution → feedback," avoiding quality risks caused by parameter execution deviations.

[0018] The quality verification module adopts a multi-dimensional verification mode, including X-ray inspection (welding quality), infrared thermal imaging (heat dissipation characteristics), and electrical performance testing (conductivity and withstand voltage). The quality verification results are fed back to the pre-control module in real time for closed-loop optimization of pre-control parameters. If the inspection is qualified, a valid signal is fed back to the pre-control module. If defective products are found, the specific failure mode and related factor data are fed back simultaneously, triggering the pre-control module to correct the pre-control parameters. It should be noted that the quality verification results are obtained through a four-step closed loop: "product positioning → multi-dimensional detection → result judgment → feedback transmission." This process is closely integrated with the intelligent process execution module and the creative pre-control module. The specific steps are as follows: Product Receiving and Precise Positioning: After the intelligent process execution module completes the production of electronic products, it transports the products to the quality verification station and uses a vision positioning system to lock the inspection area (such as soldering points and chip mounting areas) to ensure targeted inspection and avoid invalid scanning.

[0019] Dimensional Specialized Testing: X-ray inspection: To assess welding quality, X-rays are emitted to penetrate the welding area, and the image is then analyzed to determine whether there are defects such as cold solder joints, porosity, or insufficient solder, and to determine whether the welding meets the standards. Infrared thermal imaging detection: Based on heat dissipation characteristics, an infrared camera captures the temperature distribution of the product during operation to determine whether the heat dissipation of the chip and PCB board is uniform and whether there is a risk of local overheating. Electrical performance testing: By connecting probes to the product's test points, core electrical indicators such as conductivity (whether the circuit is unobstructed) and withstand voltage (whether there is leakage when subjected to rated voltage) are tested to verify whether the electrical functions are normal.

[0020] Overall result judgment: Summarize the test data of the three dimensions. If all dimensions meet the preset standards (such as no welding defects, uniform heat dissipation, and electrical indicators meet the standards), the verification is judged as "qualified". If any dimension fails to meet the standards, the verification is judged as "unqualified", and the specific failure mode (such as "poor soldering" or "local overheating of chip") and related test data (such as the coordinates of the poor soldering point and the overheating temperature value) are marked.

[0021] Real-time feedback of results: The "qualified / unqualified" conclusion and failure mode (if any) are transmitted to the creative pre-control module via industrial Ethernet, providing a precise basis for closed-loop optimization of pre-control parameters - qualified supports the stability of model parameters, unqualified triggers the correction of model coefficients.

[0022] The end-to-end data traceability module records the data collected by the multi-source heterogeneous data acquisition module, the pre-control parameters and failure prediction results of the pre-control module, the actual execution data of the intelligent process execution module, and the test results of the quality verification module, forming a quality archive for the entire life cycle of a single product.

[0023] Furthermore, the multi-source heterogeneous data acquisition module adopts a fusion mode of online sensing and offline detection: Process sensors collect welding temperature (°C), welding time (s) Mounting pressure (N) Dispensing amount (μL); Flatness of component characteristic sensor acquisition chip pins (mm), PCB board warpage (°), Solder alloy purity (%); Environmental sensors collect workshop temperature (°C), humidity (%), dust concentration (mg / m³); sampling frequency is 10Hz, measurement accuracy , , .

[0024] Furthermore, the quality factor coupling modeling unit employs a hybrid approach combining physical mechanisms and data-driven methods to construct a multi-dimensional quality factor coupling model and calculate the comprehensive quality risk index. The specific steps are as follows: a1. Define core quality factors: process factors Component factors Environmental factors ; b1. Establish a single-factor risk sub-model: Process risk sub-model: ,in The process factor weights are as follows, where the weights of each factor are: , For real-time process parameters, These are standard process parameters. ; Component risk sub-model: ,in The component factor weights are as follows: , Standardized values ​​for component characteristics ( hour , hour , hour Otherwise, it will decrease proportionally. ; Environmental risk sub-model: ,in This is the environmental impact factor. ; c1. Constructing the factor coupling matrix: Based on physical mechanisms (such as high temperature and humidity accelerating solder oxidation, and uneven pins exacerbating mounting deviations), define the coupling coefficients. ,in Source factor type, To form a 3×3 coupling matrix for the target factor type. : ,in: (Process → Components) = 0.4, indicating that the soldering temperature affects the stability of the chip pins; (Process → Environment) = 0.3, indicating that high temperature exacerbates the adhesion of dust to the environment. (Component → Process) = 0.2, indicating that the pins are not flat and the mounting pressure needs to be adjusted; d1. Calculate the comprehensive quality risk index : ,in: The failure amplification factor is 1.2 (trained based on historical data). , Determined to be high risk Medium risk Low risk.

[0025] Furthermore, the failure path dynamic mapping unit uses a risk factor-failure mode association method to predict potential failure paths. The specific steps are as follows: a2. Establish a failure mode library: Based on historical failure data of electronic products from external input, define core failure modes (such as solder joint defects, chip pin corrosion, PCB board short circuits) and corresponding failure characteristics (such as solder joint defects corresponding to soldering temperature fluctuations + pin flatness deviations). b2. Calculate the probability of the failed path: In the formula: For the first The probability of occurrence of each failure mode. Failure mode weights (e.g., failure mode weights for poor solder joints: 0.4, corrosion: 0.3, short circuit: 0.3). The correlation between the quality factor and the failure mode (e.g., the correlation between welding temperature and poor weld). ); c2. Determine the dominant failure path: according to... Sort the failure paths from highest to lowest and select the top two as the dominant failure paths (e.g., "low soldering temperature + pin flatness deviation → cold solder joint"), and output the path characteristic parameters (correlation factor type, threshold requirement).

[0026] Furthermore, the adaptive pre-control parameter generation unit adopts a failure path-oriented method to generate targeted pre-control parameters. The specific working steps are as follows: a3. Extract the key influencing factors of the dominant failure path: for example, the key factor for the "cold solder joint" path is the soldering temperature. Placement pressure Pin flatness ; b3. Establish a pre-control parameter adjustment model: in: The adjusted pre-control parameters, To adjust the coefficient (process parameter) Component compatibility parameters ), (The extent to which the risk deviates from the safety threshold). For the sign function, where When it is 1, The time is -1; c3. Boundary verification: Pre-controlled parameters must meet the equipment capability boundaries (such as welding temperature). Placement pressure When the boundary is exceeded, the correlation factor parameters are adjusted synchronously (e.g., If the upper limit has been reached, increase the welding time. ); Generate pre-control instructions: Adjust the process parameters ( (etc.) are sent to the intelligent process execution module, while the dominant failure path and pre-control logic are marked.

[0027] Furthermore, the coupling coefficient is updated based on the quality verification results of the quality verification module. With adjustment coefficient This ensures that the accuracy of pre-control continuously adapts to changes in production scenarios.

[0028] This invention discloses a method and system for quality control of electronic products, aiming to solve the problems of traditional quality control, which relies mainly on post-production detection, lacks sufficient prediction of multi-factor coupled failures, and uses blind pre-control parameters. In electronic product manufacturing, the dynamic coupling of processes, components, and environment can easily lead to latent failures such as poor soldering and chip corrosion. Traditional systems only remove defective products and cannot avoid risks in advance. This invention takes the pre-control module as its core innovation, which includes three units: quality factor coupling modeling, failure path dynamic mapping, and adaptive pre-control parameter generation. First, the multi-factor risks of process-component-environment are quantified through a coupling matrix to obtain a comprehensive risk index. Then, the dominant failure path is predicted based on the failure probability formula. Finally, the model is adjusted according to the pre-control parameters to generate targeted pre-control instructions. Combined with a multi-source heterogeneous data acquisition module (to acquire data such as soldering temperature and pin flatness in real time), an intelligent process execution module (to accurately execute pre-control parameters), a quality verification module (to provide multi-dimensional feedback results), and a full-process data traceability module (to store traceability data), a closed-loop control is formed.

[0029] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. An electronic product quality control method, characterized by, Comprise the following steps: Step S1, multi-source quality factor collection: real-time collection of process parameters welding temperature , welding time , mounting pressure , dispensing amount and component characteristic parameters chip pin flatness , PCB board warpage , solder alloy purity and environmental parameters workshop temperature , humidity , dust concentration , uploaded to the pre-control module; Step S2, multi-factor coupling modeling: the quality factor coupling modeling unit of the pre-control module calls the single-factor risk sub-model, calculates the process risk sub-model , the component risk sub-model , the environment risk sub-model , combines the coupling matrix and the failure amplification coefficient , obtains the comprehensive quality risk index , and determines the risk level; Step S3, dynamic prediction of failure path: the dynamic mapping unit of failure path calls the failure mode library, based on and coupling coefficient , through formula to calculate the probability of each failure mode, determine the dominant failure path and key influencing factors; Step S4, adaptive pre-control parameter generation: the adaptive pre-control parameter generation unit calculates the pre-control parameter adjustment model based on the key factors of the dominant failure path , and generates a pre-control instruction after boundary checking and sends it to the intelligent process execution module. Step S5, accurate process execution: the intelligent process execution module adjusts the pre-control instruction including but not limited to welding temperature, mounting pressure parameter, complete electronic product production; Step S6, multi-dimensional quality verification: the quality verification module verifies the product quality through including but not limited to X-ray detection, electrical performance test mode, and feeds back the verification result to the pre-control module; Step S7, closed-loop optimization and data archiving: if the product is qualified, the full-process data traceability module stores the collected data, pre-control parameters, and verification results; if not, the pre-control module corrects the coupling coefficients based on the feedback failure mode and the adjustment coefficient , and regenerates the pre-control parameters; after each batch of production, the model parameters are updated to ensure adaptability.

2. The electronic product quality control system of claim 1, wherein: Comprise multi-source heterogeneous data acquisition module, pre-control module, intelligent process execution module, quality verification module, whole process data tracing module; The multi-source heterogeneous data acquisition module captures process parameters, part characteristics, environmental parameters through the mode of online sensing combined with offline detection, the data acquisition frequency is 10Hz, ensures real-time and precision; The pre-control module comprises quality factor coupling modeling unit, failure path dynamic mapping unit and adaptive pre-control parameter generation unit; The pre-control module is used for realizing the pre-judgment and accurate pre-control of quality risk, receiving the whole dimensional data of the multi-source heterogeneous data acquisition module, quantifying the comprehensive quality risk index by the quality factor coupling modeling unit, locating the potential failure root by the failure path dynamic mapping unit, and finally outputting the targeted pre-control instruction by the adaptive pre-control parameter generation unit, simultaneously, the pre-control module receives the feedback result of the quality verification module, dynamically updates the model parameters, and ensures that the pre-control precision continuously adapts to the production scene; The intelligent process execution module comprises precision welding assembly, mounting driving unit and dispensing control unit, receives the pre-control parameter instruction of the pre-control module, and executes the accurate adjustment of including but not limited to welding temperature and mounting pressure parameter; The quality verification module adopts the multi-dimensional verification mode of X-ray detection, infrared thermal imaging and electrical performance test, feeds back the quality verification result to the pre-control module in real time, and is used for pre-control parameter closed loop optimization; If the detection is qualified, the effective signal is fed back to the pre-control module; If the defective product is found, the specific failure mode and related factor data are fed back synchronously, and the pre-control module is triggered to correct the pre-control parameter; The whole process data tracing module records the acquisition data of the multi-source heterogeneous data acquisition module, the pre-control parameter and failure prediction result of the pre-control module, the actual execution data of the intelligent process execution module, and the detection result of the quality verification module, and forms a single product whole life cycle quality file.

3. The electronic product quality control system of claim 2, wherein: The multi-source heterogeneous data acquisition module adopts the fusion mode of online sensing and offline detection: Process sensor collects welding temperature , welding time , mounting pressure , dispensing amount ; component characteristic sensor collects chip pin flatness , PCB board warping degree , solder alloy purity ; environmental sensor collects workshop temperature , humidity , dust concentration ; collection frequency is 10Hz, measurement accuracy , , .

4. The electronic product quality control system of claim 2, wherein: The quality factor coupling modeling unit adopts the hybrid method of physical mechanism and data driving, constructs a multi-dimensional quality factor coupling model, calculates the comprehensive quality risk index, and the specific working steps are: a1, define core quality factors: process factors , component factors , environmental factors ; b1, establish a single factor risk submodel: Process risk sub-model: wherein are process factor weights, wherein each factor weight is: , are real-time process parameters, are standard process parameters, ; Component risk sub-model: wherein is the component factor weight, wherein the component weight is , is the component property normalized value, ; Environmental risk submodel: wherein is the environmental impact coefficient, ; c1. Constructing the coupling matrix of factors: define the coupling coefficients based on physical mechanism wherein is the source factor type, is the target factor type, forming a 3x3 coupling matrix : wherein: (process→component) = 0.4, indicating that the soldering temperature affects the stability of the chip pin; (process→environment) = 0.3, indicating that high temperature aggravates the adhesion of environmental dust; (component→process) = 0.2, indicating that the unevenness of the pin needs to adjust the mounting pressure; d1, calculating an overall quality risk index : wherein: is a failure amplification factor, , is determined to be high risk, is determined to be medium risk, is determined to be low risk.

5. The electronic product quality control system of claim 2, wherein: The failure path dynamic mapping unit adopts the risk factor-failure mode correlation method to predict the potential failure path, and the specific working steps are: a2, establish a failure mode library: based on the external input electronic product historical failure data, define the core failure mode and the corresponding failure characteristics; b2, calculate failure path probability: In the formula: is the occurrence probability of the failure mode, is the failure mode weight, is the correlation degree of the quality factor and the failure mode;​ c2, determine the dominant failure path: according to From high to low, select the top 2 as the dominant failure path, and output the path characteristic parameters.

6. The electronic product quality control system of claim 1, wherein: The adaptive pre-control parameter generation unit adopts the failure path oriented method to generate targeted pre-control parameters, and the specific working steps are: a3, extract the key influence factors of the dominant failure path; b3, a pre-control parameter adjustment model is established: wherein: is an adjusted pre-control parameter, is an adjustment coefficient, , is a sign function, wherein is 1, is -1; c3, boundary check: pre-control parameters need to meet the device capability boundary, when exceeding the boundary, adjust the associated factor parameters (such as Increase the welding time if the upper limit has been reached ); Generating pre-control instructions: sending the adjusted process parameters to the intelligent process execution module, while marking the dominant failure path and pre-control logic.

7. The electronic product quality control system of claim 2, wherein: updating the coupling coefficient based on a quality verification result of the quality verification module with the adjustment coefficient , ensuring that the pre-control accuracy continuously adapts to changes in the production scenario.

Images