Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Solder paste detection threshold optimization method based on SMT big data

An optimization method and detection threshold technology, applied in the direction of gene models, genetic rules, instruments, etc., can solve the problems of SPI equipment misjudgment and missed judgment, reduce the final yield of products, manual maintenance, etc. The number of connected tins, the improvement of production efficiency and product yield, the effect of good global optimization and parallelism

Active Publication Date: 2018-12-07
XIDIAN UNIV
View PDF5 Cites 20 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Traditional manual experience and lack of theoretical guidance make SPI equipment often have misjudgments and missed judgments, and then real defects are artificially let go and flow into follow-up stations, reducing the final yield of products
In addition, the solder paste printing standard IPC-7527 also has a lack of solder paste volume detection standards
[0004] The SPI threshold setting method based on traditional manual experience lacks theoretical guidance and has a certain degree of blindness. It is difficult to obtain a reasonable detection threshold through data analysis.
If the detection threshold is set unreasonably, potential defects cannot be effectively found during the printing process. After the PCB board flows into subsequent processes such as reflow soldering, defects such as false soldering and solder joints will occur. The production efficiency and the overall yield of SMT products are greatly reduced, and it is difficult to meet the demand for production efficiency and product quality

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Solder paste detection threshold optimization method based on SMT big data
  • Solder paste detection threshold optimization method based on SMT big data
  • Solder paste detection threshold optimization method based on SMT big data

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0032] The traditional SPI threshold setting mainly relies on the professional knowledge and experience of the operator. Only by printing a PCB board first can the empirical threshold be obtained. The obtained threshold lacks theoretical guidance and greatly reduces production efficiency. This method is also contrary to the intelligence and high efficiency of the modern manufacturing industry. For this problem, the present invention proposes a SPI threshold optimization method based on SMT big data after research. See figure 1 , including the following steps:

[0033] (1) Build a threshold estimation data package:

[0034] (1a) Data collection: For pads printed by surface mount technology of the same package type, collect historical data generated during the SMT production process. The collected data includes historical data of three main stations, which are SPI detection data, AOI automatic optical inspection data and Repair maintenance data. SPI inspection data mainly incl...

Embodiment 2

[0048] The SPI threshold optimization method based on SMT big data is the same as embodiment 1, and the establishment objective function described in step (3a) includes the following steps:

[0049] (3a1) Establishing the objective function: According to the minimum error Bayesian decision theory, calculate the SPI misjudgment rate and missed judgment rate, and the expressions of the SPI misjudgment rate and missed judgment rate are shown in the following formula.

[0050] The SPI misjudgment rate is:

[0051]

[0052] The SPI missed rate is:

[0053]

[0054] In the formula, f(x|w 1 ) is the probability density function expression of normal data obtained in step (2d); f(x|w 2 ) is the probability density function expression of the abnormal data obtained in step (2d); t is the solder paste SPI detection value, and the solder paste SPI detection value includes solder paste volume, solder paste area, solder paste height in the present invention, they all They have their...

Embodiment 3

[0060] The SPI threshold optimization method based on SMT big data is the same as that in Embodiment 1-2. In step (3b), the objective function is optimized using a genetic algorithm to obtain the optimal threshold for SPI detection.

[0061] In the present invention, the optimization process involved in the genetic algorithm process optimized for detection threshold is described as follows:

[0062] (3b1) Customize the population size, crossover probability, and mutation probability. The population size is the number of initial generation thresholds; the crossover probability is the probability of individual crossover within each population; the mutation probability is the probability of gene mutation from the current generation population to the next generation.

[0063] (3b2) Initial population: Randomly generate the initial population according to the user-defined population size, and each individual represents the genotype of the chromosome. For example, if the custom pop...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The present invention provides a solder paste detection threshold optimization method based on SMT big data. The problems are mainly solved that a traditional artificial experience SPI threshold setting method cannot effectively discover potential badness in the PCB printing process and cannot reduce the cold solder joint and continuous tin electrodeposit defects caused by the solder paste amount.The method comprises the steps of: constructing a threshold estimation data packet; estimating an SPI parameter state; optimizing an SPI threshold, taking an SPI detection value as an operation variable to allow a corresponding value to be a target function when the sum of a misjudgment rate and a misdetection rate reaches the lowest, employing a genetic algorithm to perform optimization of the target function to obtain an optimization threshold, and setting the obtained optimization threshold as the SPI threshold of the surface mounting technology. The design of the solder paste detection threshold optimization method is rigorous and complete, the threshold value setting method has theoretical property and feasibility to effectively control the defects such as cold solder joint and continuous tin electrodeposit caused by the solder paste amount to flow into the surface mounting technology (SMT) later process so as to improve the whole yield of a PCB.

Description

technical field [0001] The invention belongs to the technical field of industrial big data, and mainly relates to the optimization of a detection threshold, in particular to a solder paste detection threshold optimization method based on SMT big data, which is used for optimizing the SPI threshold of the solder paste detection of surface mount technology SMT. Background technique [0002] SMT production errors are distributed in all aspects of each assembly process. As the first process of the SMT process, solder paste printing is affected by various uncertain process parameters (such as scraper, printing speed, steel mesh, width-thickness ratio and area ratio of openings, etc.), and the printing process is affected by various uncertain process parameters. Parameter setting has a great relationship with the professional knowledge and experience of the operator, so the printing process is a key link in the quality control of the SMT process. Solder paste inspection SPI can m...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G06K9/62G06N3/12
CPCG06N3/126G06F18/24155
Inventor 常建涛孔宪光李宏刘超李名昊
Owner XIDIAN UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com