【作者】 李宁；

【导师】 施心陵； 李鹏；

【作者基本信息】 云南大学 ， 生物医学工程， 2010， 硕士

【摘要】 资源、环境、生态及重大工程等复杂系统所涌现出的安全问题对人类的生存安全和社会经济发展构成巨大的威胁,因此如何对这类复杂系统进行安全诊断与预测维护是一个极其重要而有意义的课题。在本文中,基于多源信息融合及智能计算理论,结合输变电系统及水电站坝体两个应用范例,建立适用于复杂系统安全监测及诊断的信息融合理论功能模型,对多源信息融合机理、智能诊断技术以及安全监测系统设计等进行了研究,为相关部门的灾害预测与防治提供支持。取得的主要研究成果如下:(1)在数据融合级异常值识别与突变特征提取研究中,结合某水电站坝体的形变监测具体应用,提出了基于BP神经网络的异常值识别算法。与现有的异常值识别算法相比较,该算法从监测量产生的物理机理模型出发,融合了多个传感器的监测数据,可以区别异常值中存在的野值和突变,而且对于孤立型和斑点型异常值均可较好的识别。(2)在特征融合级复杂系统灾变条件特征提取研究中,结合某输变电系统覆冰监测的具体应用,提出了基于自组织图的复杂系统灾变条件特征提取算法,实现了在特征融合级对输电线覆冰的形成和增长的气象条件特征进行了提取,其结果有助于在决策融合级利用微观气象信息对输变电系统的安全态势进行估计。(3)以某水电站坝体安全监测系统开发为应用背景,针对中小型电站及水库坝体安全监测的特点和需求,对传统的应用于大坝安全监测的软件系统结构进行适当的调整,设计了面向中小型水坝安全监测软件系统,对软件系统的调整使得开发周期和开发成本大为减少,较适用于中小型水坝的安全监测系统的设计和开发。更多还原

【Abstract】 Complex systems, such as resource, environment, ecology and great engineering, have come forth many safety problems, which seriously threaten the subsistence and development of human being. So it is an important and significant problem that how to monitor and diagnose the safety of complex system. In this paper, based on multi-source information fusion and combined two complications about power transmission system and hydropower station dam safety monitoring, an information fusion functional model, which is suitable for complex systems safety monitoring and diagnosing, has been established. Research on multi-source information fusion mechanism, intelligence diagnosing technology and safety monitoring system designing, for which can provide theory and technology for disaster prediction and preventive treatment to related departments.The achievements which have got are as follows:(1)Practicing in the hydropower station dam displacement monitoring, an algorithm based on BP neural network is presented. Comparing with the existing outlier identifying algorithm, this algorithm take the advantage of the mechanism model of the dam monitoring data conformation, and fuse many data from sensors,it can identify outlier and mutation exist in unusual data and can well identify both patch-type outliers and isolated outliers.(2)Practicing in the power transmission lines system monitoring, an algorithm based on Kohonen Self-Organizing Map is presented to extract the disaster features of the complex system. In feature fusion level, for power transmission lines icing and its increasing, the extract of the feature is achieved. The results are helpful for using meteorology features to estimate the safety situation of the power transmission lines system in decision-making level.(3)On the background of developing a dam safety monitoring system, aiming at small-or-medium reservoirs safety monitoring feature and demands , adjust the traditional structure of software system which apply to the dam safety monitoring and design the safety monitoring system which is suitable for the small-or-medium reservoirs. Adjusting the software system makes the system development cycle short and cost decreased and it’s more suitable for small-or-medium reservoirs safety monitoring.更多还原