3D model retrieval based on representative views was proposed. On the view representation of the 3D model, in order to fully represent the model and reduce redundant information, we firstly adopt Light Field Descriptor (LFD) to generate 2D views, and then use K-MEANS to get representative views from the 2D views. Next, a Convolution Neural Network (CNN) is adopted to extract the view feature and classify. At the same time, a similarity metrics supporting multiple query method is proposed to realize model retrieval with sketches, pictures or 3D models as input. Results on ModelNet40 showed that the proposed method could achieve an accuracy of 100% for part of models with distinct features

传感器失灵、通讯迟滞或数据离散化等多种不确定因素会导致行星齿轮箱故障诊断信息不完备情况的发生,而现有的故障诊断方法已难于适用.为此,提出一种基于数据驱动量化特征多粒度模型的行星齿轮箱故障诊断方法.首先,采用数据驱动量化特征关系对行星齿轮箱的不完备故障诊断信息进行分析;其次,利用基于悲观数据驱动量化特征多粒度模型的属性约简算法提取故障诊断决策规则;最后,使用朴素贝叶斯分类器(Naive Bayesianclassifier,NBC)推断待诊行星齿轮箱状态.实验研究表明,该方法可准确地判断实例间的不可分辨关系,降低计算复杂度,提高故障诊断准确率.

Aimed at the problems of complex diagnosis processes of planetary gearboxes and low accuracy of diagnosis results,a fault diagnosis method of planetary gearboxes was proposed based on flow graph.In this method,a construction algorithm of flow graph was utilized to intuitively represent the fault diagnosis knowledges of planetary gearboxes. Then,a reduction algorithm of flow graph was applied to removing unnecessary condition attribute nodes. The consequent relationship among attributes was represented in the simplest manner. Finally,a classification decision algorithm of flow graph was employed to determine fault types of test samples. The simulation and experimental results confirm the accuracy and intuitiveness of this method,which provide a novel way for fault diagnosis of planetary gearboxes.

In order to discover decision rules from incomplete fault diagnosis information containing multiple unknown attribute values,a knowledge discovery method of incomplete fault diagnosis information by using valued characteristic relation is proposed.Firstly,the unknown attribute value types are determined according to the reasons for incomplete information.Secondly,the incomplete fault diagnosis information is analyzed by using the valued characteristic relation.Finally,the decision rules for fault diagnosis are discovered according to the attribute reduction algorithm based on the valued characteristic relation.The effectiveness of the method is demonstrated with the diagnosis case of faulty gearbox.The experimental results show that the present method can directly discover the accurate decision rules from incomplete fault diagnosis information with three categories of unknown attribute values.