汇报时间：2019年7月22日（星期一）15：00
汇报地点：兴庆校区中心三楼2333室
汇报人：熊逸伟、刘洪成.
会议信息
会议名称：SURVISHNO 2019
会议时间：July8-10, 2019
会议地点：University of Lyon,France
会议简介：Survishnois a joint organization of the conferences Surveillance, VISHNO(Vibration Shocks and Noise) and EVA(ExperimentalVibration Analysis), which aims at bringing together actors in the fields of vibrations, structural dynamics and health monitoring. The conference will provide a forum for the most recent advances in these fields as well as a unique opportunity for interacting in a collegial atmosphere.Topics of the conference include (but are not limited to): structural & machine health monitoring, modeling and identification of mechanical systems, nonlinear dynamics, vibration control, structural dynamics, modal analysis, signal processing & machine learning in mechanics, diagnosis & prognosis.
会议交流工作
Oral Presentation:Research on the Variation Mechanism of the 3-D Tip Clearance of a Cracked Blade under Multi-parameters in the Aero-engine Acceleration Process
报告人：熊逸伟
Oral Presentation:Demodulating of the 3-D tip clearance of turbine blades using BP neural network optimized by genetic algorithm
报告人：刘洪成
参会论文信息
Title:Research on the Variation Mechanism of the 3-D Tip Clearance of a Cracked Blade under Multi-parameters in the Aero-engine Acceleration Process
Author:Yiwei Xiong, Xiaodong Zhang, Siying Xie, Hongcheng Liu
Abstract:The 3-D tip clearance has some advantages over the traditional radial tip clearance in the fault diagnosis of the turbine blade crack. The research on the variation mechanism of the 3-D tip clearance is of great significance, but previous researches only focused on the steady state condition of the aero-engine, lacking considerations of the aero-engine acceleration process. In this study, a numerical model of high pressure turbine blisk was established, and the centrifugal load, thermal load and aerodynamic load of the aero-engine, varying with time, were considered in this model. Besides, the cracks with different length and location were added to the trailing edge of the blade, and the 3-D tip clearance of cracked blades was calculated. The results show that there are some obvious distinctions in the 3-D tip clearance between a normal blade and cracked blades, which can reflect blade crack information accurately and effectively. The results also indicate that the 3-D tip clearance is promising in fault diagnosis of the turbine blade crack.
Title:Demodulating of the 3-D tip clearance of turbine blades using BP neural network optimized by genetic algorithm
Author:Hongcheng Liu, Xiaodong Zhang, Yiwei Xiong, Siying Xie
Abstract:The 3-D tip clearance refers to the three-dimensional space between the probe end face and turbine blade tip surface, which contains more abundant fault information than the traditional one-dimensional tip clearance. In this paper, an approach for demodulating of the three-dimensional (3-D) tip clearance of turbine blades is presented using BP neural network optimized by genetic algorithm (GA). Through the static calibration experiments, the ratios of outer circle voltage to inner circle voltage of three units from the optical fiber probe are used as the training and test data for demodulation. The maximum errors of the radial displacementz0, the axial angleαand the circumferential angleβdemodulated by the demodulation model based on BP neural network optimized by GA are 0.1321mm, 0.9875° and 0.6456°, respectively. The mean square errors of the radial displacementz0, the axial angleαand the circumferential angleβare 0.0006mm, 0.0528° and 0.0266°, respectively. The experiment results show that this demodulation method have an excellent precision, which can fulfil the requirements of the 3-D tip clearance detection and provide a basic guarantee for the fault diagnosis of turbine blades.
欢迎有兴趣的同学届时参加。
yl23455永利
2019.7.18
