On‐Line Detection Of State‐Of‐Charge In Lead Acid Battery Using Radial Basis Function Neural Network

To realize a stable supply of electric power in an automobile, an accurate and reliable detection method of SOC (state‐of‐charge) in a lead acid battery is required. However the dynamics of the battery is very complicated. The characteristics of the battery greatly change due to its degradation. Moreover a automobile has many driving patterns, which are unknown beforehand. Thus it is not easy to detect the SOC analytically. In this paper, to overcome this problem, a new on‐line SOC detection method with a radial basis function neural network is proposed. In order to increase the detection accuracy of degraded batteries, physical values related to the degradation degree are used as input signal in the neural network. The detection accuracies for different sized batteries and various degradation states are investigated.     

Tree-shaped flow structures designed by minimizing path lengths

This paper outlines a direct route to the construction of effective tree-shaped flow structures. Dendritic flow structures dominate the design of natural and engineered flow systems, especially in thermal and fluid systems. The starting point is the optimization of the shape of each elemental area or volume, such that the length of the flow path housed by the element is minimized. Proceeding toward larger and more complex structures - from elements, to first constructs, second constructs, etc. - the paper develops tree-shaped flow structures between one point and a straight line, one point and a plane, a circle and its center, and a point and many points distributed uniformly over an area. In the latter, the construction method is applied to a fluid flow configuration with laminar fully developed flow. The constructions reveal several features that are supported by empirical observations of natural tree-shaped flows: asymmetry, flow rate imbalance, pairing or bifurcation, angles between branches, and Y-shaped constructs that lie in a plane. It is shown that these basic features are necessary because of “packing”, i.e., assembling optimized elements into a fixed space, and filling the space completely. For the flow between an area and one point, the best elemental shape is the regular hexagon. It is shown that the emergence of string-shaped links that connect two or more elements are necessary features, which are also required by packing. Strings cover some of the inner zones of the tree network, particularly the inner zones of large and complex trees. Dichotomous Y-shaped constructs dominate the tree structure, especially the peripheral zones of the tree canopy. The practical importance of the simplified design method is discussed.     

手势语言识别的神经网络方法

提供了一种用于人机交互（HCI）的手势语言可视化识别方法。该方法包括用于几种控制命令的手势的探测、分割、特征提取及识别，第一步的处理都用到了神经网络方法，像肤色探测、主元分析（PCA）以及在编码识别。实验结果显示正确识别率高达94％。     

GENERALIZED INVERSE GROUP OF SIGNAL AND ITS IMPLEMENTATION WITH NEURAL NETWORKS

A new concept, the generalized inverse group (GIG) of signal, is firstly proposed and its properties, leaking coefficients and implementation with neural networks are presented. Theoretical analysis and computational simulation have shown that (1) there is a group of finite length of generalized inverse signals for any given finite signal, which forms the GIG; (2) each inverse group has different leaking coefficients, thus different abnormal states; (3) each GIG can be implemented by a grouped and improved single-layer perceptron which appears with fast convergence. When used in deconvolution, the proposed GIG can form a new parallel finite length of filtering deconvolution method. On off-line processing, the computational time is reduced to O(N) from O(N2). And the less the leaking coefficient is, the more reliable the deconvolution will be.     

光学全交叉网络与SW榕树（F＝S＝2）网络拓扑等价的多样性

本文对有关互连网络拓扑等价的分析方法作了简单评述。并采用互连网络拓扑等价的图分析法分析了自由空间光学互连全交叉网络与ＳＷ榕树网络的拓扑等价及其多样性，拓展了全交叉网络在数字光计算、光子交换及多处理机系统的潜在应用。     

Energy minimization and design for testability

The problem of fault detection in general combinational circuits is NP-complete. The only previous result on identifying easily testable circuits is due to Fujiwara who gave a polynomial time algorithm for detecting any single stuck fault inK-bounded circuits. Such circuits may only contain logic blocks with no more thanK input lines and the blocks are so connected that there is no reconvergent fanout among them. We introduce a new class of combinational circuits called the (k, K)-circuits and present a polynomial time algorithm to detect any single or multiple stuck fault in such circuits. We represent the circuit as an undirected graphG with a vertex for each gate and an edge between a pair of vertices whenever the corresponding gates have a connection. For a (k, K)-circuit,G is a subgraph of ak-tree, which, by definition, cannot have a clique of size greater thank+1. Basically, this is a restriction on gate interconnections rather than on the function of gates comprising the circuit. The (k, K)-circuits are a generalization of Fujiwara'sK-bounded circuits. Using the bidirectional neural network model of the circuit and the energy function minimization formulation of the fault detection problem, we present a test generation algorithm for single and multiple faults in (k, K)-circuits. This polynomial time aggorithm minimizes the energy function by recursively eliminating the variables.     

应用改进的BP神经网络预测导弹的研制费用

结合导弹武器系统研制的特点，分析并确定了影响导弹武器系统研制费用的主要因素，并在前人研究的基础上，提出了基于改进的BP神经网络的费用预测模型。该模型不仅克服了以往由于专家主观因素而造成的负面影响，而且提高了算法的运行效率和模型的预测精度。通过具体的实例分析比较，充分证明了利用该方法的科学合理性。最后探讨了利用BP神经网络进行数值预测的方法和应该注意的问题，并给出了一些有益的建议。