大系统分散鲁棒容错控制的进展及主要成果

综述了近２０年来关联大系统分散鲁棒容错控制的研究动态及取得的主要成果。说明了统一地研究分散鲁棒容错控制的必要性。在分散控制、鲁棒控制和容错控制研究领域内，分析了各种设计方法的主导思想及实质；指出了这些研究成果的意义与存在的问题。文中着重论述了分散、鲁棒、容错控制研究方法的发展途径。     

Stabilization of inter-area and/or oscillatory modes in large-scale power system by voltage impedance power system (VIPS) apparatus based on decentralized control scheme

New technologies such as power electronics have made it possible to change continuously the impedance of a power system not only to control power flow but also to enhance stability. A power system incorporating a variable impedance apparatus such as a variable series capacitor (VSrC) and high-speed phase shifter (HSPS) is called VIPS (Variable Impedance Power System) by the authors. This paper proposes a novel control method of VIPS apparatus such as VSrC and HSPS installed at an interconnecting point for stabilizing inter-area unstable and/or oscillatory modes. The proposed design method of the control system is a kind of hierarchical decentralized control method of a large-scale power system based on a Lyapunov function. Under the proposed control scheme, each subsystem can be stabilized independently by local controllers such as AVR, speed governor and PSS, and then the whole interconnected system can be stabilized by VIPS apparatus taking into account interactions between subsystems. The effectiveness and robustness of the VIPS apparatus control are shown by numerical examples with model systems including a large-scale power system.     

Analytical Calculation of Large-Scale Deflection of Pulse-Loaded Elliptic Plates

Large deflections of pulse-loaded elliptic and round plates made of hardenable elastoplastic materials are investigated analytically using an energy-based approach. Formulas that are convenient for engineering design have been derived.     

对主要钢铁企业集团规模经营和全球化经营情况的分析

概述了在全球范围内广泛，激烈的市场竞争中，世界主要钢铁企业抓住机遇进行战略性兼并的情况；分析指出了规模经营和全球化经营是是钢铁企业做大做强，占领国际市场份额的必然趋势。     

计量秤在变角度状态的实践

介绍了物料计量秤在变角度下如何计算重量，以及变角度显示仪的工作原理和在实际中的运用。     

高炉焦炭中子测水与干焦重量补偿模型

文章叙述了高炉焦炭的中子测水原理，并给出了一种干焦重量的补偿方法，最后用一个例子说明实际的效果。     

Stability of large-scale coal-fired MHD channels

Stability of large-scale coal-fired MHD channels is studied by (1) linearized stability analysis, and (2) time-dependent 1-D analysis. The channel length is 15 m with 600 electrode pairs, and the output power ranges from 220.6 MW through 258.7 MW. Linearized stability analyses show that the Faraday channels operated with fixed loading resistance are stable, whereas the two waves of u and u–a (u, a: gas and sound velocity) become unstable in the Faraday channel with fixed loading factor. Two waves of u and u–a are unstable in the diagonal channel with fixed loading current and the u + a wave becomes unstable in the diagonal channel with fixed electrode current. Time-dependent one-dimensional analyses indicate that the Faraday channels with fixed load resistance are smooth without growth of fluctuation. The diagonal channels with fixed electrode current are smooth with no fluctuation, though the linear theory indicates that the u + a wave is unstable. The diagonal channel with fixed load current suffers large disturbance along the latter half of the channel, being consistent with the linearized analysis which indicates that the u - a and u waves are unstable.     

Adaptive fuzzy decentralized control for nonlinear large-scale systems based on high-gain observer

An adaptive fuzzy decentralized backstepping output-feedback control approach is proposed for a class of nonlinear large-scale systems with completely unknown functions,the interconnections mismatched in control inputs,and without the measurements of the states.Fuzzy logic systems are employed to approximate the unknown nonlinear functions,and an adaptive high-gain observer is developed to estimate the unmeasured states.Using the designed high-gain observer,and combining the fuzzy adaptive control theory with backstepping approach,an adaptive fuzzy decentralized backstepping output-feedback control scheme is developed.It is proved that the proposed control approach can guarantee that all the signals of the closed-loop system are semi-globally uniformly ultimately bounded(SUUB),and that the observer errors and the tracking errors converge to a small neighborhood of the origin by appropriate choice of the design parameters.Finally,a simulation example is provided to show the eectiveness of the proposed approach.     

面向大规模数据流处理的偏最小二乘法

用光谱分析鉴别生物特征,导致数据量大,而实际需要必须实时处理。偏最小二乘法是使用最广泛的鉴别算法,但是对于大规模数据流该算法无法达到实时性。为了解决这个应用矛盾,提出了一种基于NVIDIA CUDA架构下的并行计算策略,利用具有大规模并行计算特征的图形处理器（GPU）作为计算设备,结合GPU存储器的优势实现了偏最小二乘算法。实验的测试结果表明,在GPU上使用CUDA实现的偏最小二乘算法比在CPU上实现该算法快了47倍,性能得到了显著提高,从而使偏最小二乘算法应用于大规模数据流处理成为可能。     

基于GPU-CUDA的共轭斜量法实现及性能对比

偏微分方程数值解法(包括有限差分法、有限元法)以及大量的数学物理方程数值解法最终都会演变成求解大型线性方程组。因此,探讨快速、稳定、精确的大型线性方程组解法一直是数值计算领域不断深入研究的课题且具有特别重要的意义。在迭代法中,共轭斜量法(又称共轭梯度法)被公认为最好的方法之一。但是,该方法最大缺点是仅适用于线性方程组系数矩阵为对称正定矩阵的情况,而且常规的CPU算法实现非常耗时。为此,通过将线性方程组系数矩阵作转换成对称矩阵后实施基于GPU-CUDA的快速共轭斜量法来解决一般性大型线性方程组的求解问题。试验结果表明:在求解效率方面,基于GPU-CUDA的共轭斜量法运行效率高,当线性方程组阶数超过3000时,其加速比将超过14;在解的精确性与求解过程的稳定性方面,与高斯列主元消去法相当。基于GPU-CUDA的快速共轭斜量法是求解一般性大型线性方程组快速而非常有效的方法。