Abstract

In this study a method of predicting the adhesion strength of Chromium aluminum nitride (CrAlN) coating on Al–Si alloy (LM28) using fuzzy logic technique was introduced. LM28 was coated with CrAlN under dissimilar coating conditions. The CrAlN coated substrates adhesion strength was determined by micro-scratch apparatus. The microstructure, topographical analysis and composition of selected coated substrates were characterized using scanning electron microscopy coupled with Energy-dispersive X-ray spectroscopy. A fuzzy logic model was applied to predict the adhesion strength of CrAlN coating on LM28. RF power, DC power, nitrogen flow rate, and temperature based on the trained data achieved from the micro scratch test were used as controllable process parameters. Then, three new experimental confirmation runs were conducted to verify the results predicted via the Fuzzy model. The predicted adhesion strength was equated with measured data. The maximum prediction error was 5.2%, while the average prediction error was 3.5%. Finally, prediction resulted in the improvement of surface hardness value from 0.9 GPa to 4.5 GPa, signifying an enhancement by 5 times.

Graphic Abstract

中文翻译：

---

Al-Si合金（LM28）上CrAlN涂层的附着强度预测：模糊建模

摘要

在这项研究中，介绍了一种使用模糊逻辑技术预测氮化铝铬合金（CrAlN）在Al-Si合金（LM28）上的附着强度的方法。LM28在不同的涂层条件下被CrAlN涂层。CrAlN涂覆的基材的粘附强度通过微划痕设备测定。使用扫描电子显微镜和能量色散X射线光谱法对所选涂层基材的微观结构，形貌分析和组成进行了表征。应用模糊逻辑模型预测Cr28N涂层在LM28上的附着强度。基于从微划痕测试获得的训练数据的RF功率，DC功率，氮气流速和温度被用作可控制的过程参数。然后，进行了三个新的实验确认运行，以验证通过Fuzzy模型预测的结果。预测的粘合强度等于测量数据。最大预测误差为5.2％，而平均预测误差为3.5％。最后，预测导致表面硬度值从0.9 GPa提高到4.5 GPa，表示提高了5倍。

图形摘要