基于案例、规则和模型推理的电站智能诊断系统

介绍了基于案例、规则和模型的诊断推理方法,讨论了相似度计算、规则表示和复合故障的诊断等问题,论述了基于案例、规则和模型的混合推理策略,并在电站设备故障诊断系统中得到了应用。     

光学榕树网络与全混洗网络之间的转换及实现

光学榕树网络的快速控制需要在全混洗网络和光学植树网络间进行快速的相互转换．分别给出了光学榕树网络转换成全混洗网络及全混洗网络转换成光学植树网络时结点、链路、输入端口、输出端口的映射规则，并用电子学进行快速实现．     

LANN AND ITS HETEROASSOCIATIVE MEMORY PROPERTIES

This paper provides a new architecture of neural network, called loop architecture neural network(LANN), and its learning rules. One of its features distinguished from other network, such as Hopfield and bidirectional associative memories, is that it can perform the associative memory among multiple categories. Analysis and simulated results have proved that it is an effective network with excellent convergence.     

多传感器多目标跟踪中的概率数据互联

通过在两维雷达和红外搜索跟踪两种不同传感器观测空间上建立多目标运动状态的投影，单传感器的ＪＰＤＡ算法被推广到此种多传感器数据融合系统之中，实现了其中的多目标数据互联和多传感器数据融合，从而提高了跟踪性能。     

机载火控雷达高重复频率线性调频测距模式目标跟踪方法研究

通过多假设的方法,把机载火控雷达高重复频率线性调频测距模式下的目标距离估计和模糊分辨问题转变为静态多模型估计和判决问题。同时给出了基于这种方法的目标跟踪脉冲重复频率的选择方法。通过仿真把本文方法和传统的跟踪方法进行了比较,结果表明,本文方法具有跟踪维持性能好、收敛快,跟踪精度高的特点。     

战场目标价值评判的神经网络方法

在建立目标价值评判指标体系的基础上,利用BP神经网络构建3层结构多节点模型进行战场目标的价值排序.网络以对数sigmod函数为传递函数,选取典型目标组成学习和测试样本,采用改进后的弹性BP算法完成模型的训练和测试,由此绘制出网络误差变化曲线和权值、阈值Hinton图.经学习、测试和大量泛化目标的检测,模型能成功运用于战场目标价值排序.     

固体火箭发动机可靠性增长试验规划初步研究

研究了固体火箭发动机产品总数一定情况下可靠性增长试验量的确定问题．提出了模型假设，以剩余产品成功数期望值最大作为目标函数，建立数学模型；然后，在初始故障数服从泊松分布时，推导了确定可靠性增长试验量的极大一极小准则，给出了产品研制后期故障数的替代分布．最后给出了一个应用实例，结果表明该准则有点保守，但较为稳健．     

裸眼3D显示液晶光栅贴合控制参数设计与实现

文章根据视差液晶光栅立体显示原理和莫尔条纹形成规律．提出了一种视差液晶光栅贴合控制参数的设计规则及工艺指导方式。通过对实际产品立体效果的影响分析。得出了设计控制的关键因素：贴合间隙的厚度和光栅与显示屏之间贴合允许最大倾斜角度,根据贴合允许的最大倾斜角度推导出了贴合对位标记垂直方向上的最大允许误差。实验验证和量产实践表明,文中提出的理论和方法能很好地应用于裸眼3D显示液晶光栅贴合控制的设计与量产工艺控制。     

基于模糊神经网络的传感器置信度的求取

根据模糊逻辑规则中传感器置信度与其前件的单调关系 ,构造了一种基于内插求值的模糊神经网络 (FNN)。详细描述了网络各层神经元的内部结构和层间关系 ,并给出了网络的学习算法。最后通过仿真验证了网络的可行性。     

Model of three-way decision based on the space of set pair information granule and its application

Through the analysis of the existing theory and method of three-way decision,model of three-way decision based on the space of set pair information granule was proposed,which was to explain the intension and extension of three-way decision on the view of unity of opposites.Firstly the space structure of set pair information granule can be di-vided into positive granule,negative granule,different granule,which were similar and slightly different with three re-gions of generalized three-way decision.The three kinds of information granules in set pair information granule space were built based on certain positive degree,negative degree and different degree.Secondly,according to the given threshold,set pair information granule is divided into mutually disjointing positive region,negative region and different region.Based on these three regions,the computing method of decision loss function is put forward,and the process of decision-making algorithm is obtained use the three-way decision method of set pair information granule space to evalu-ate and make decision on risk investment,then good decision results have been achieved.