作者：宋凯1, 李子璇1, 陆灵峰1, 肖树坤2, 王荣彪1
作者单位：1. 南昌航空大学 无损检测技术教育部重点实验室，江西 南昌 330063;
2. 江西洪都航空工业集团有限责任公司，江西 南昌 330096
关键词：热障涂层;机器学习;电磁无损评估;涡流探头
摘要：
热障涂层（thermal barrier coating，tbc）是一种由金属粘接层和陶瓷层构成的隔热材料，具有热导率低、抗热疲劳及耐高温氧化等优异性能，能够很好地在高温环境下保护发动机关键部位。针对某关键部位多曲面基体热障涂层厚度的涡流检测，设计r45.5 mm以及r72.5 mm两种弹压式涡流传感器，建立3种机器学习算法的反演模型，并对曲面模拟试样进行测量，最后验证3种算法的反演精度。结果表明：r45.5 mm曲面试样随机森林算法的粘接层最大相对误差为9.01%，陶瓷层最大相对误差为4.33%；r72.5 mm曲面试样随机森林算法的粘接层最大相对误差为5.23%，陶瓷层最大相对误差为6.28%，均小于工业误差要求10%。
research on inversion of thermal barrier coating thickness in keyparts of aero-engine
song kai1, li zixuan1, lu lingfeng1, xiao shukun2, wang rongbiao1
1. key laboratory of nondestructive testing(ministry of education), nanchang hangkong university, nanchang 330063, china;
2. jiangxi hongdu aviation industry group corporation limited, nanchang 330096, china
abstract: the thrust weight ratio of aero-engine is very important to improve the engine performance, so the key parts of the engine need to bear high temperature. thermal barrier coating (tbc) is a thermal insulation material composed of metal adhesive layer and ceramic layer. it has excellent properties such as low thermal conductivity, thermal fatigue resistance and high-temperature oxidation resistance, and can well protect key parts of the engine. for the eddy current testing of the thickness of the thermal barrier coating on the multi curved surface of the key parts, two kinds of elastic pressure eddy current sensors, r45.5 mm and r72.5 mm, were designed. the inversion models of three machine learning algorithms were established, and the curved surface simulation samples were measured. finally, the inversion accuracy of the three algorithms was verified. the results show that the maximum relative error of the adhesive layer in the r45.5 mm curved sample random forest algorithm is 9.01%, and the maximum relative error of the ceramic layer is 4.33%; the maximum relative error of the adhesive layer using the random forest algorithm for r72.5 mm curved samples is 5.23%, and the maximum relative error of the ceramic layer is 6.28%, both of which are less than the industrial error requirement of 10%.
keywords: thermal barrier coating (tbc);machine learning;electromagnetic non-destructive evaluation;eddy current probe
2023, 49(12):8-15 收稿日期: 2022-08-31;收到修改稿日期: 2022-10-21
基金项目: 国家自然科学基金项目（51865033）；博士启动基金（ea202208183）
作者简介: 宋凯宋(1975-),男,辽宁锦州市人,教授,博士,研究方向为无损检测新技术。
参考文献
[1] 田伟, 何爱杰, 钟燕, 等. 高推重比发动机热障涂层应用现状分析[j]. 燃气涡轮试验与研究, 2016, 29(5):52-57.
[2] 樊书琪. 热障涂层厚度脉冲涡流检测方法研究[d]. 徐州:中国矿业大学, 2020.
[3] li y, chen z m, mao y, et al. quantitative evaluation of thermal barrier coating based on eddy current technique[j]. ndt & e international, 2012, 50:29-35.
[4] 赵云松, 张迈, 戴建伟, 等. 航空发动机涡轮叶片热障涂层研究进展[j]. 材料导报, 2023(6):1-12.
[5] 王博, 刘洋, 王福德, 等. 航空发动机及燃气轮机涡轮叶片热障涂层技术研究及应用[j]. 航空发动机, 2021, 47(s1):25-31.
[6] wang l, li z d, ding k m, et al. effects of tgo growth on the stress distribution and evolution of three-dimensional cylindrical thermal barrier coatings based on finite element simulations[j]. ceramics international, 2022, 48(6):7864-7875.
[7] wei z y, liu y, cheng b, et al. influence of non-uniform feature of thermally grown oxide thickness on the local stress state and cracking behavior in tbc[j]. surface and coatings technology, 2022, 443:128607.
[8] 谢建红, 石剑, 习小文, 等. 基于bp算法的热障涂层厚度反演研究[j]. 兵器装备工程学报, 2021, 42(4):228-232 243.
[9] dou p, zhou l s, wu t h, et al. simultaneous measurement of thickness and sound velocity of porous coatings based on the ultrasonic complex reflection coefficient[j]. ndt&e international, 2022, 131:102683.
[10] li c l, niu q, ji c, et al. measurement of thermal barrier coating thickness of turbine blade based on thz-tds[c]//infrared, millimeter-wave, and terahertz technologies viii, 2021.
[11] 肖力伟. 基于ect技术的叶片热障涂层厚度测量方法研究[d]. 南昌:南昌航空大学, 2020.
[12] 黄平捷, 吴昭同, 郑建才, 等. 多层厚度电涡流检测技术及实验研究[j]. 中国机械工程, 2003(15):24-27.
[13] 黄平捷, 吴昭同, 严仍春. 多层厚度电涡流检测阻抗模型仿真及验证[j]. 仪器仪表学报, 2004(4):473-476, 480.
[14] 黄平捷, 张光新, 吴昭同, 等. 多层导电结构厚度电涡流检测正向模型及仿真器研究[j]. 传感技术学报, 2005(3):601-606.
[15] 黄平捷, 吴昭同, 郑建才, 等. 多层厚度电涡流检测反演算法及实验研究[j]. 仪器仪表学报, 2005(4):428-432.
[16] 李达, 李勇, 闫贝, 等. 航空发动机叶片热障涂层厚度的无损定量评估[j]. 空军工程大学学报(自然科学版), 2015, 16(6):22-26.
[17] zhang d j, yu y t, lai c, et al. thickness measurement of multi-layer conductive coatings using multifrequency eddy current techniques[j]. nondestructive testing and evaluation, 2016, 31(3):191-208.
[18] 高宽厚, 于亚婷, 杨姣, 等. 双层导电涂层厚度的涡流无损检测方法研究[j]. 机械工程学报, 2017, 53(14):114-119.
[19] xu j, wu j, xin w, et al. fast measurement of the coating thickness and conductivity using eddy currents and plane wave approximation[j]. ieee sensors journal, 2020, 21(1):306-314.
[20] xu j, wu j, xin w, et al. measuring ultrathin metallic coating properties using swept-frequency eddy-current technique[j]. ieee transactions on instrumentation and measurement, 2020, 69(8):5772-5781.
[21] 王亚清. 热障涂层厚度涡流检测方法研究[d]. 徐州:中国矿业大学, 2019.
[22] 来超. 多涂层厚度涡流无损检测技术及其实现方法研究[d]. 成都:电子科技大学, 2015.
