PLC結合機器視覺與物聯網工業平台之設計與實作

Design and Implementation of Programmable Logic Controller Combined with Machine Vision and IoT Industrial Platform

許承恩(Hsu Chen-EN)

指導教授 ： 陳金山

國立屏東科技大學/工學院/機械工程系所/碩士(2022年)

https://doi.org/10.6346/NPUST202200195

摘要

2015年起台灣推廣工業4.0來結合第二級產業提升國家整體的產業競爭力，台灣的第二級產業發展也隨著智慧化趨勢的引入了許多技術，如:物聯網、工業機器人、人工智慧等。但普通小型傳產都屬於半自動化工廠，無法追到智慧工廠的腳步。本研究主要利用機電整合機台搭配自製SCADA系統，結合機器視覺顏色感測系統判斷缺陷料，Webcam的架設可應用於機器視覺與遠端監控畫面，自走車機器人收取料件。本研究模擬小型傳產的半自動化條件，在機電整合機台手動放料讓機台作動，機器視覺判別缺陷料手動取下，手動遙控自走車機器人，機台手動取料再放料到機器人上。模擬一般小型傳產半自動化條件，另外設定自走車機器人為自動收料條件，這兩組數據因上述手動條件會有時間差，如何調快機台來彌補半自動與自動化差距，其中物聯網會把收集到的數據放在MVC網路上讓客戶端或是內部人員觀看。經由實驗結果顯示:模擬調快機台半自動化作動時間比自動化作動慢約0.6秒，經由彌補這兩者之間的時間差距來排解人為因素，讓小型傳產追上智慧工廠的腳步。

關鍵字

物聯網 ； 機器視覺 ； 遠端監控

並列摘要

In 2015, Taiwan promoted Industry 4.0 to combine secondary industries to enhance the country's overall industrial competitiveness. Taiwan 's secondary development industry has also introduced many technologies with the trend of intelligence, example Internet of Things, industrial robot and Artificial Intelligence. However, small factories are semi-automated and they're unable to follow the footsteps of smart factories. This study aims to develop the remote synchronous monitoring system for weight discrimination and arrangement system(WDAS) and connect supervisory control and data acquisition(SCADA). They can combine Machine Vision Color Sensing System to judges defective materials, setting up webcam application to machine vision and remote monitoring system, mini-car robot to be receiving materials. This study simulates semi-automated of small factories, manually unload the material to make the WDAS moving, manually unload defective materials of machine vision to judge, manually control mini-car robot when SCADA finish it's work, manually unload materials that put it on the mini car robot. Simulation semi-automated of small factories condition, additional set the mini car robot as the automatic receiving condition. There will be a time gap between these two sets of data due to the above manual conditions, how to make fast machines to make up time gap between semi-automatic and automation, the Internet of Things will put the collected data one the MVC network for clients or insiders to watch that data. shown to experimental results: simulation semi-automated action time is slower than automatic action by 0.6 seconds, eliminate human factors that the time gap between semi-automated action and automatic action, result to show small factories can catch up with factories.

並列關鍵字

Internet of Things ； Machine Vision ； Remote Monitoring

參考文獻

[1] 王怡惠，2015， 「從工業4.0看我國生產力4.0之挑戰」 ，臺灣經濟研究月刊，第38卷，第8期，第111-119頁。

[2] 黃以宜，胡曉，蔡承佐，范盛維，2017， 「工業4.0通訊技術與應用」 ，機械工業雜誌，第1卷，第413期，第17-29頁。

[3] Zannino, G. D. , Perri, R. , Salamone, G. , DiLorenzo, C. , Caltagirone C., and Carlesimo, G. A., 2010, “Manipulating color and other visual information influences picture naming at different levels of processing: evidence from Alzheimer subjects and normal controls,” Neuropsychologia, Vol. 48, No. 9, pp. 2571-2578.

[4] Li, D., Li, C., Chen, C. and Zhao, Z., 2020, “Semantic Segmentation of a Printed Circuit Board for Component Recognition Based on Depth Images, ” MDPI Sensors, Vol. 20, No. 18, pp. 1-17.

[5] Zhou, Z., Feng, Y., Rong, G., and Zhu, F., 2011, “Virtual Factory Integrated Manufacturing System for Process Simulation and Monitoring,” IFAC Proceedings Volumes, Vol. 4, No. 1 pp. 5219-5224.