智慧優化快速試產PCB製造系統

Introduce into Printed Circuit Board of Rapid Pilot Run with AI Optimization System

林冠廷 ； 林文彬 ； 陳璽在

《機械工業雜誌》 498期 (2024/09) Pp. 58-66

摘要

目前PCB產業試產製程中各站多採用人工方式進行參數監控以及手動進行參數調適，而每一輪試產製程間須進行樣品檢查及參數調適，單次製程的時間約需耗時達40分鐘以上，並且單一料號調適週期約需要4～6次，故要能達成快速試產必定需要導入製程智慧化技術，本文提出建立自動化監控製程參數及量測並智慧化進行參數的調適，除了可以大幅縮短樣品檢查及參數調適時間外，利用AI預判斷優化參數系統亦可以有效減少反覆調適次數，減少整體試產時間以及人力成本。

關鍵字

參數優化 ； 預判分析 ； 自動化監控

並列摘要

In the current PCB industry most trial production processes involve manual parameter monitoring and adjustments. Each trial production cycle requires sample inspections and parameter adjustments, with a single process taking over 40 minutes. Additionally, a single part number requires 4 to 6 adjustment cycles. Therefore, to achieve rapid trial production, the implementation of intelligent process technology is essential. This article proposes the establishment of an automated system for monitoring and measuring process parameters, with intelligent parameter adjustments. This system not only significantly reduces the time for sample inspections and parameter adjustments but also utilizes AI to predict and optimize parameters, effectively reducing the number of repeated adjustments, overall trial production time, and labor costs.

並列關鍵字

Parameter optimization ； Forecast analysis ； Supervisory control and data acquisition

參考文獻

李玫樺(2017).印刷電路板產業運用資訊化監控管理節水減排系統之研究(碩士論文).國立臺灣大學生物環境系統工程學研究所.

Sharaff, A.,Gupta, H.(2019).Extra-tree classifier with metaheuristics approach for email classification.Advances in computer communication and computational sciences.189-197.

Huang, G.,Wu, L.,Ma, X.,Zhang, W.,Fan, J.,Yu, X.,Zeng, W.,Zhou, H.(2019).Evaluation of catBoost method for prediction of reference evapotranspiration in humid regions.Journal of Hydrology.574,1029-1041.

Ibrahim, I.A.,Khatib, T.(2017).A novel hybrid model for hourly global solar radiation prediction using random forests technique and firefly algorithm.Journal of Energy Convers Manag.138,413-425.

Salvador, J.,Pérez-Pellitero, E.(2015).Naive bayes super-resolution forest.IEEE International Conference on Computer Vision (ICCV).(IEEE International Conference on Computer Vision (ICCV)).