报告会第一部分

  时间：2016年6月30日

  地点：广东石油化工学院图书馆学术报告厅208

  报告嘉宾及报告会信息：

1.     周长兵

（1）    嘉宾简介

中国地质大学(北京)信息工程学院教授，博士生导师。法国国立电信学院南巴黎分校计算机系客座副教授，广东石油化工学院客座教授。曾在法国国立电信大学南巴黎分校从事博士后研究工作，博士毕业于爱尔兰DERI研究所。曾在朗讯科技贝尔实验室(北京)工作过五年，担任高级研究员职位和项目负责人的职位；曾在华为公司北京研究所工作过一年，担任软件工程师的职位。已发表国际学术论文100余篇。

（2）    报告题目

《Similarity Assessment for Scientific Workflow Clustering and Recommendation》

（3）    报告摘要

With an increasing number of scientific workflows published at online repositories and publicly accessible through the Web, reuse and repurposing of best practices is of core importance when a novel scientific workflow is required to be developed for satisfying the requirement of scientists. To address this challenge, this article proposes an approach for identifying and recommending workflows for reference. Specifically, a scientific workflow is represented as a layer hierarchy, which specifies hierarchical relations between this workflow, its sub-workflows, and activities. Semantic similarity is calculated between layer hierarchies of workflows. A graph-skeleton based clustering technique is adopted for grouping layer hierarchies into clusters. Barycenters in each cluster are identified, which refer to core workflows in this cluster, for facilitating cluster identification and workflow ranking and recommendation. Experimental evaluation shows that our technique is efficient and accurate on ranking and recommending appropriate clusters and workflows.

2.     吴晓鸰

（1）    嘉宾简介

吴晓鸰，广东工业大学副教授，于2001和2003年获得哈尔滨工业大学的学士和硕士学位，于2008年获得韩国庆熙大学博士学位。博士毕业后作为高级工程师就职于韩国ATLab公司。2012作为海外人才引进深圳先进技术研究院/广州中国科学院先进技术研究所智能控制研究中心担任副研究员。共受理国内发明专利近20项、作为第一作者/通讯作者发表SCI期刊论文、EI期刊论文和EI国际会议论文共40余篇。吴晓鸰博士现担任广州市科技和信息局专家库入库专家，广州骏望科技专家，华南理工大学创新班合作导师。她也是IEEE Member以及多个SCI期刊的guest editor和审稿人、多个国际会议的session chair。

（2）    报告题目

《Data compression for IWSN》

（3）    报告摘要

Data compression on sensing data is one of the issues in Industrial Wireless Sensor Networks (IWSNs).Especially, the video and image data that need to be transmitted are relatively larger than common data. However IWSNs usually have limited power supply and constrained communication bandwidth............................................................................................................................................................................................................................................., it is significant to reduce the video and image data without any distortion before transmission to lower the energy consumption and the transmission delay. In this talk a Horizontal Hierarchy Slicing (HHS) method based on Mathematical Morphology technology for compressing the data is introduced. The results show its effectiveness in data compression for video and image data in IWSNs.

  报告会第二部分

  时间：2016年7月1日

  地点：广东石油化工学院图书馆学术报告2厅

3.     Tim Gordon

（1）    嘉宾简介

    Tim Gordon, 林肯大学工程学院院长 (2013年 - 至今) 。

    Tim Gordon毕业于剑桥大学应用数学系，博士学位。1999年至2003年，担任拉夫堡大学航空与汽车工程系的系主任，是汽车工程学的福特教授。2003年至2010年，担任美国密歇根大学交通研究所的教授和工程研究主任。他的研究方向是智能车辆动力学和控制，尤其注重积极安全、人车互动和车辆自适应的自主权。他是车辆系统动力学国际协会（IAVSD）的副主席，也是IEEE SMC智能车辆系统和控制技术委员会的联合主席。他最近在瑞典哥登堡主持了一场关于公路车辆自动驾驶和自动安全系统的国际研讨会。

（2）    报告题目

    Automation and Guidance of Highway Vehicles

（3）    报告摘要

    This presentation includes a review of the evolution of the intelligent highway vehicle, prompted by the current interest about ‘self-driving cars’. There is a focus on the supporting technologies and the recent progress and critical challenges for delivering useful automation in a safe way. The intensive development of intelligent vehicles is founded on the longer-term evolution made in mechatronic vehicle design – the mainstream industry development of electromechanical and electronic systems on modern cars; combined with recent advances in artificial intelligence, advanced sensing, and computational intelligence, new opportunities are being realized. While a try ‘self-driving’ capability does not properly exist at present, some useful aspects of driving automation are already in the mainstream, for example in automatic emergency braking (AEB) systems; these sense the environment with radar and cameras, and make critical decisions for braking without requiring any action from the driver. A recent study has shown an example where the added protection from AEB has reduced occupant injuries by one third. As cars are given more computational intelligence, they can be expected to provide benefits for traffic flow and reduced carbon emissions, as well as in crash avoidance and in taking over some of the more tedious aspects of driving. To achieve all of this, future intelligent vehicles will need networked support from an intelligent infrastructure. The talk will discuss the technological successes so far, the safety issues, and the challenges ahead – in particular the need for a multi-level redundant control architecture.

4.     张宇

（1）    嘉宾简介

    张宇博士, 林肯大学工程学院的高级讲师和国际合作负责人。

    张宇博士于诺丁汉大学土木工程系2011年博士毕业。她在2011年加入林肯大学工程学院, 作为一名研究员, 研究工业燃气涡轮系统机器和传感器故障检测的项目。在2014年，她成为工程学院的讲师，2015年晋升为高级讲师。她的主要研究领域在数据分析，模糊建模，人工智能， 智能系统。她近期在研的项目包括2项创新英国基金项目（合作负责人）。除了教学和研究，她主管工程学院国际项目合作工作。她近期的国际项目包括: 广东省教育厅颁发的广东省国际重大合作项目平台的国际联合负责人, 和德国西门子颁发的西门子燃气轮机远程监控的主要项目负责人。

（2）    报告题目

    Machine fault diagnosis of industrial gas turbines

（3）报告摘要

    Machine fault detection and diagnosis has always been an essential part in industrial control systems to assure operational reliability, quality and safety. Through automatic early warning of machine faults, with faulted component localized, dramatic component failure can be avoided, and costs of initial manual investigations can also be saved.

    This presentation includes a review of various techniques for machine fault detection and diagnosis based on real applications on Siemens gas turbines. Firstly, an overview of basic underlying principles of gas turbines is given. Then, data driven techniques for pattern recognition, including empirical mode decomposition, principal component analysis, Gaussian mixture models, etc. are reviewed. Furthermore, model based techniques for system identification, such as neural networks, fuzzy systems, etc. are introduced. To conclude, through the use of experimental trials on a fleet of Siemens gas turbines during commissioning, it is demonstrated an integrated intelligent agent for machine fault detection and diagnosis. Finally, knowledge gaps and further work are discussed.