航空粉末冶金涡轮盘早期损伤的非线性超声检测机理与关键技术研究

The service and remaining life of aero engine turbine disc are mainly decided by the early damages such as tiny inclusion, closed crack and dislocation. Traditional ultrasonic testing can only be effectively detected material's macro defects, while it cannot detect early damages. The existing research work is limited to the relatively simple metal materials, and the nonlinear ultrasonic application of aero engine turbine disc was not reported. In order to meet the requirements of the development of aviation products, the nonlinear ultrasonic technique is adopted to evaluate the early damages of aero engine turbine disc in the research. The main research contents are as follows: Firstly, finite element modal of nonlinear ultrasonic testing is constructed for analyzing the response of nonlinear ultrasonic to early damages. Secondly, nonlinear ultrasonic detection platform and related experiment scheme are studied for picking testing signal effectively. Thirdly, algorithm of signal analyzing and processing are developed for abstracting the defect's signals. According to experiment study and finite element analysis, the relationship between the characteristic parameters and damage's status is observed, and the interaction mechanism between the ultrasonic wave and damages is revealed. Finally，based on the characteristic parameters, the damage's character imaging and automatic recognition are studied to enhance the evaluation ability. This study provides a new nondestructive method for aero engine turbine disk, which will play an important role in improving the safe operation of aircraft, reducing the production cost, avoiding the huge loss of life and property.

粉末冶金涡轮盘的寿命设计和剩余寿命估算主要取决于夹杂物、微裂纹、位错堆积等早期损伤，常规超声检测技术难以对早期损伤和微缺陷进行检测。科研工作者已开展了较为简单的金属材料的微损伤非线性超声检测技术研究，但其在航空粉末冶金涡轮盘中的应用还未见报道。本项目拟将非线性超声检测技术应用于涡轮盘早期损伤的无损检测中，首先，建立盘件早期损伤的非线性超声检测过程的有限元模型，理论分析损伤的非线性超声响应规律；第二，设计高效的信号激励与接收方法、并且据此搭建与之相适应的实验系统；第三，对检测信号进行深入分析，抑制干扰、增强损伤特征信息；建立各种非线性超声特征参数与损伤状态的关系，揭示材料损伤与超声波相互作用机理；第四，开展损伤成像检测与自动识别的研究工作。本研究可有效提高航空涡轮盘早期损伤的检测能力，对于提高飞行器的安全运行、降低生产成本、避免人民生命财产损失具有重要作用。

航空发动机涡轮盘的寿命设计和剩余寿命估算主要取决于夹杂物、微裂纹、位错堆积等早期损伤，常规超声检测技术难以对早期损伤和微缺陷进行检测。本项目将非线性超声检测技术应用于涡轮盘早期损伤的无损检测中，开展了检测机理、检测系统、检测方法、特征参数提取的相关研究工作，并取得了丰富的研究成果。首先，建立航空涡轮盘非线性超声检测过程的数值仿真模型，理论计算非线性超声检测的声场分布；揭示闭合裂纹及微小层片型缺陷的非线性超声响应规律，为高效激励与接收早期损伤检测信号提供坚实的理论依据。第二，基于RITEC SNAP非线性超声检测仪搭建含多个独立功能模块的检测系统，设计和定制专用探头及检测工装，在高效接收损伤相关的非线性超声检测信号的同时，避免和降低信号干扰。第三，提出高效的非线性超声检测信号激励与接收方法，基于有限幅度法和混频法并结合先进信号处理技术，从时域、频域、时频域分析和辨识非线性检测信号特征和干扰信号特征，达到抑制干扰信号、增强损伤信号特征的目的。第四，深入分析各类损伤与超声波相互作用的物理机制，提出了非线性超声检测特征参数以实现损伤状态的特征参数表征。最后，开展非线性阵列超声成像检测技术的研究工作，基本完成阵列非线性传感器的设计工作；开展焊缝超声TOFD检测图像中典型缺陷的自动分类与识别方法研究，建立了缺陷图像智能识别的深度学习模型，实现了对未熔合、夹杂、气孔、裂纹的自动和智能分类。. 项目研究成果可极大提高航空涡轮盘早期损伤及微小缺陷的检测能力，预期在军民领域均具有广泛的应用前景，对于提高飞行器的安全运行、降低生产成本、避免人民生命财产损失具有重要作用。