时 间：2018年4月9日，周一14:00-17:00
地 点：阳澄湖校区交通大楼二楼学术报告厅
报告一
报告人：美国凯斯西储大学 王鹏 博士
题 目：Stochastic Modeling for Performance Prognosis in Dynamical Systems 基于随机建模的动力系统性能预测
摘 要：
实时有效的动力系统状态监测和性能预测对于维持机械系统的可靠性起着十分关键的作用。该报告以航空发动机为例，介绍了最近于基于数据融合和随机建模的系统性能预测上取得的研究进展。针对于航空发动机非线性性能退化的特点，一种多模态自适应粒子滤波 (Multi-mode Adaptive Resampling Particle Filter) 被开发出来并于总变分网络（Total Variation Filter）结合用于发动机性能追踪。仿真和实验验证表明这种方法可有效追踪和预测性能缓变，并能准确检测性能突变。已进行的测试表明，和企业界里广泛应用的卡尔曼滤波方法相比，这种新方法可提高7%的预测准确率。同时这种方法可广泛用于其他机械系统，如加工刀具，中央空调系统，齿轮箱等性能和剩余寿命预测。
简 介：
2010年和2012年于北京化工大学获得学士和硕士学位，2017年于美国凯斯西储大学获得博士学位，现于美国凯斯西储大学机械与宇航工程系攻读博士后。他的研究课题和兴趣包括：随机建模和贝叶斯推理，数据融合，机械状态监测，故障诊断与寿命预测，智能制造，人机协作，云制造等等。王鹏博士的博士毕业论文课题是：基于随机建模和不确定性分析的动力系统性能预测。他在ASME Journal of Manufacturing Science and Engineering, ASME Journal of Gas Turbines and Power, SME Journal of Manufacturing Systems, 和CIRP Annals-Manufacturing Technology 等期刊上发表了13篇文章，并发表了20多篇国际会议文章。王鹏博士获得了2014年International Conference on Motion and Vibration 会议的最佳报告奖，2015年IEEE Conference on Automation Science and Engineering (CASE) 会议的最佳学生论文奖，2017年North American Manufacturing Research Conference 会议的最佳论文奖。他还得到了2016年美国数字制造和设计工程院组织的企业数据分析比赛第一名。
报告二
报告人：美国内华达大学里诺分校大学 吴建清 博士
题 目：Innovative Roadside LiDAR Sensing for Connected-Vehicles and New Traffic Applications 路侧激光雷达传感技术在车联网及新兴交通领域的应用研究
摘 要：
车联网技术的优势依赖于所有道路用户实时共享其位置、速度及行驶方向等信息。但目前联网的道路用户比例极低，造成了车辆网技术中的数据缺失。数据缺失情况下车联网技术无法充分发挥作用。本研究首创性地将激光雷达引入路侧设施中，通过激光雷达传感技术实时获取高精度微观交通数据来弥补车联网技术推广过程中的数据空缺。报告阐述了多种路侧激光雷达技术处理算法，具体包括数据融合算法、背景滤波算法、车道自动识别算法、人车检测算法及目标追踪程序。报告也介绍了从路侧激光雷达中提取的高精度数据在动物追踪，车祸识别及变道预测领域的应用。
Abstract:
The full benefits of connected-vehicle technologies rely on all road users being connected and sharing their real-time information of location, speed and direction. However, there are still limited connected-vehicles on roads. Under the mixed traffic situation, connected vehicles could not obtain the full benefits as only part of vehicle movement information can be obtained. The Light Detection and Ranging (LiDAR) sensors deployed on road side provide an effective method to obtain the high-resolution micro traffic data to serve the mixed traffic situation. This research introduced the roadside LiDAR data processing algorithms, including data integration, background filtering, lane identification, vehicle and pedestrian classification and object tracking. The new applications of the roadside LiDAR were also introduced, which covered wildlife tracking, near-crash identification and lane-change prediction.
简 介：
美国内华达大学里诺分校博士， 主要从事交通安全、交通数据库及智能交通领域的技术研究。其创新性地将激光雷达运用到了车联网技术中，安装并成功测试了世界上首个路侧激光雷达联网系统，开发了基于数据聚合的点云背景滤除算法、多传感器数据自动融合算法、车道识别算法等，构建了一套成体系的算法库。其先后获得了美国交通工程师协会 （ITE）、Transportation Research Board 等多个组织的表彰与奖励。同时担任了包括 TRR, TRF, IEEE ITSM 等多个期刊的审稿人。目前已发表学术论文12篇，其中6篇被SSCI、SCI 收录。
报告三
报告人：香港城市大学 王冬 博士
题 目：Monitoring and Health Management of Engineering Systems and Critical Components
摘 要：
Prognostics and health management is an emerging discipline to scientifically manage the health conditions of engineering systems and critical components. It mainly consists of four main aspects: construction of health indicators, fault diagnosis, remaining useful life prediction, and health management. Construction of health indicators aims to evaluate the current health conditions of engineering systems and critical components. Fault diagnosis aims to identify specific faults once any abnormal health conditions happen. Given the observations of a health indicator, prediction of remaining useful life is used to infer the time when an engineering systems or a critical component will no longer perform its intended function. Health management involves planning the optimal maintenance schedule according to the system's current and future health conditions and the replacement costs. In this presentation, Dr. Dong Wang will introduce some of his progress toward construction of health indicators, fault diagnosis and remaining useful life prediction. Teachers and students are welcome to give any questions, comments and suggestions to Dr. Dong Wang’s research works.
简 介：
Dr. Dong Wang was a recipient of Hong Kong PhD Fellowship in 2012 and received his Ph.D. at City University of Hong Kong (CityU) in 2015. He was appointed as a research associate, a senior research assistant and a post-doctoral fellow at CityU from Years 2015 to 2018. Currently, he is a research fellow at CityU.
Dr. Dong Wang’s research interests include statistical modeling, prognostics and health management, condition monitoring, fault diagnosis, signal processing, data mining and nondestructive testing. He has published 54 SCI-indexed journal papers (the first author for 37 journal papers) and his works have been cited over 1250 times (Google Scholar). 5 journal papers were selected by ESI TOP 1%. His research works appear in Mechanical Systems and Signal Processing, Journal of Sound and Vibration, ASME Transactions on Journal of Vibration and Acoustics, IEEE Transactions on Reliability, IEEE Transactions on Instrumentation and Measurement, Journal of Power Sources, Measurement Science and Technology, etc.
Dr. Dong Wang was invited to be a reviewer for 50 SCI-indexed journals and reviewed over 300 journal papers before. In recognition to his contributions to Elsevier and IEEE journals, he was awarded Outstanding Reviewer Status 13 times. He was a lead guest editor/guest editor for several SCI-indexed journals and a referee for the FONDECYT of Chile. Currently, he is an associate editor for Journal of Low Frequency Noise Vibration and Active Control and an associate editor for IEEE Access.
