Printer Process Parameter Generation for Accurate Solder Paste Output
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The SMT process in electronics manufacturing relies heavily on engineers' experience and trial-and-error methods to set parameters, which is time-consuming and labor-intensive, making it difficult to achieve optimal results.
Innovation Solution
A parameter generation method and apparatus using two machine learning models to predict and update process parameters for printers, minimizing prediction errors and improving operational efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If engineers use experience and trial-and-error methods to set SMT process parameters, then the process can be completed with simple equipment, but the time consumption and labor intensity increase significantly
Solution Approach 1:
The patent replaces the mechanical trial-and-error method with an automated image processing and machine learning system. The system captures printer head images, processes them through algorithms to extract feature values, and uses these features to predict optimal parameters, substituting human experience-based mechanical adjustment with automated computational analysis.
Solution Approach 2:
The system enables self-service by automatically analyzing printer head images and generating parameter recommendations without requiring engineer intervention. The machine learning model learns from historical data and autonomously predicts optimal parameters for new printing tasks, making the system self-sufficient in parameter optimization.
2Manufacturing precision
If traditional trial-and-error methods are used for parameter setting, then equipment complexity remains low, but manufacturing precision and reliability suffer
Solution Approach 1:
The patent introduces an intermediary machine learning model that acts as a bridge between simple image capture and precise parameter determination. The model processes image features and predicts parameters, serving as an intelligent intermediary that translates visual information into actionable process settings without requiring complex direct measurement systems.
Solution Approach 2:
The system performs preliminary action by capturing and analyzing printer head images before actual printing occurs. The machine learning model predicts optimal parameters in advance based on image analysis, allowing parameter optimization to be completed beforehand rather than during the printing process, thus improving precision without adding complexity to the printing operation itself.
3Reliability
If engineers manually optimize parameters through experience, then the system remains simple to operate, but the reliability and consistency of results decrease
Solution Approach 1:
The patent implements feedback by using captured printer head images as input to the machine learning model, which then generates parameter predictions. The system continuously learns from the relationship between image features and actual printing outcomes, refining its predictions over time. This feedback loop ensures reliable and consistent results without requiring operators to have specialized expertise.
Data Source
AI summary
A parameter generation method and a parameter generation apparatus for a printer are provided. By inputting target solder paste amount data into a first machine learning model, the first machine learning model outputs multiple predicted process parameters for the printer. By inputting multiple predicted process parameters into a second machine learning model, the second machine learning model outputs predicted solder paste amount data for the printer. Weights of the first machine learning model are updated according to the minimization of a prediction error. The prediction error is an error between the target solder paste amount data and the predicted solder paste amount data. By inputting the target solder paste amount data into the updated first machine learning model, the updated first machine learning model outputs multiple new process parameters used to control the operation of the printer. Therefore, the operational efficiency can be improved.