機器人研磨加工之虛實整合系統與零調機參數優化及控制

Zero-tuning Grinding Process and Control Methodology of Cyber Physical Robot System

楊軒昱(Hsuan-Yu Yang)

指導教授 ： 連豊力

國立臺灣大學/電機資訊學院/電機工程學研究所/碩士(2020年)

https://doi.org/10.6342/NTU202002718

摘要

在勞力密集以及存在高危險性的產業中，機械手臂具有能扮演重要腳色的潛力。舉例來說，以機械手臂取代傳統仰賴人力的水五金研磨加工便是一個實例。然而，利用機械手臂完成研磨加工是一項非常複雜且難度相當高的工作，它仍須仰賴有豐富經驗的工程師不斷的反覆試驗調整相關的製程參數，這不僅需要耗費大量時間，也需要足夠的相關經驗才能調整得出最佳的加工品質。
此論文提出一個藉由虛實整合系統能夠根據生產者欲得到之加工品質，自動優化出相關加工參數的零調機系統。首先，藉由機械手臂研磨之虛實整合系統，能夠輕鬆快速地設計研磨路徑並且模擬整個研磨過程。並且在研磨過程中，系統裡面的雲端資料庫會同步紀錄機械手臂以及相關感測器的數據，作為後續分析的資料來源。根據這些收集來的研磨資訊，本篇論文利用了人工神經網路以及基因演算法建立了一個零調機系統，此系統會根據加工需求優化出最適合的相關製程參數。此外，由於在加工的過程中，砂帶會逐漸產生磨耗，進而影響加工品質。因此，此篇論文根據砂帶轉速、砂帶磨耗以及工件的表面粗糙度關係，利用曲線擬和的方式，建立了一個砂帶轉速控制系統模型。此模型不僅能根據砂帶磨耗的程度來調整砂帶轉速，還能評估砂帶磨耗的狀態以及判斷是否該更換新砂帶。
實驗結果顯示出本論文所提出的零調機系統之平均誤差為8.37%，且砂帶轉速控制系統模型曲線之決定係數皆大於0.878，具有高度可預測性。對比於電腦數值控制工具機，本篇利用機械手臂來進行研磨加工也更適合應用於水五金研磨產業。

關鍵字

機械手臂研磨加工 ； 虛實整合系統 ； 零調機系統 ； 製程參數最佳化 ； 砂帶速度控制 ； 砂帶磨耗監控分析

並列摘要

Industrial robots play potential and important roles on labor-intensive and high-risk jobs. For example, typical industrial robots have been used in plumbing industry. However, the automatic grinding process by robots is a complex process because it still relies on skillful engineers to adaptively adjust several key parameters. Moreover, it might take a lot of time and effort for generating desired grinding quality.
A new framework of cyber-physical robot system with zero-tuning methodology is proposed in this thesis. It can automatically optimize the process parameters of robotic grinding process according to the desired quality. To overcome the gaps between real world and simulation leading to the uncertainty, an advanced cyber-physical robot grinding system has been developed to increase efficiency, improve quality, save human effort, and the cloud database is constructed to record the relative data during the grinding process simultaneously. The proposed zero-tuning methodology combines both neural network (NN) model and genetic algorithm (GA) is designed to generate the best combination of related corresponding parameters to meet the desired quality. Moreover, due to the belt wear is gradually increased during robotic grinding, a curve fitting model is constructed to control the belt speed and compensate the wear of belt. Also, the model can detect the health condition of the belt and judge whether it needs to change a new one.
Experimental results show that the average error of the zero-tuning system is 8.37%, and the coefficient of determination values of the curve fitting results are all higher than 0.878. To compare with traditional CNC machine, the proposed solution play utilized the robot arm is more potential to apply in plumbing industry.

並列關鍵字

robotic grinding ； cyber-physical system ； zero-tuning system ； process parameters optimization ； belt speed control ； belt health condition monitor

參考文獻

Wei Ji and Lihui Wang, “Industrial Robotic Machining: a Review,” International Journal of Advanced Manufacturing Technology, Vol. 103, Issue 1-4, pp. 1239-1255, Jul. 2019.

Rubanraj Ravichandran, Nico Huebel, Sebastian Blumenthal and Erwin Prassler, “A Workbench for Quantitative Comparison of Databases in Multi-Robot Applications,” in Proceedings of IEEE/RSJ International Conference on Intelligent Robots and Systems, Madrid, Spain, Oct. 1-5, 2018.

Mica R. Endsley and David B. Kaber, “Level of Automation Effects on Performance, Situation Awareness and Workload in a Dynamic Control Task,” Ergonomics, Vol. 42, pp.462-492, 1999.

Alexandr Klimchik, Alexandre Ambiehl, Sébastien Garnier, Benoit Furet and Anatol Pashkevich, “Efficiency Evaluation of Robots in Machining Applications Using Industrial Performance Measure,” Robotics and Computer–Integrated Manufacturing, Vol. 48, pp.12-29, Dec. 2017.

Stéphane Caro, Claire Dumas, Sébastien Garnier and Benoît Furet, “Workpiece Placement Optimization for Machining Operations with a KUKA KR270-2 Robot,” in Proceedings of IEEE International Conference on Robotics and Automation, Karlsruhe, Germany, May 6-10, 2013.