Dynamics and genetic fuzzy neural network vibration control design of a smart flexible four-bar linkage mechanism

In this paper, a dynamic modeling method and an active vibration control scheme for a smart flexible four-bar linkage mechanism featuring piezoelectric actuators and strain gauge sensors are presented. The dynamics of this smart mechanism is described by the Discrete Time Transfer Matrix Method of Multibody System (MS-DTTMM). Then a nonlinear fuzzy neural network control is employed to suppress the vibration of this smart mechanism. For improving the dynamic performance of the fuzzy neural network, a genetic algorithm based on the MS-DTTMM is designed offline to tune the initial parameters of the fuzzy neural network. The MS-DTTMM avoids the global dynamics equations of the system, which results in the matrices involved are always very small, so the computational efficiency of the dynamic analysis and control system optimization can be greatly improved. Formulations of the method as well as a numerical simulation are given to demonstrate the proposed dynamic method and control scheme.     

基于联邦数据库的数据集成平台研究与改进

本文提出了在联邦数据库模式下增加全局映射模式来集成异构数据的方案,即通过映射规则建立自定义的全局逻辑数据库,实现逻辑数据库与关系数据库数据之间的数据映射,解决异构数据库连接平台与子数据库的数据映射抽取问题。能够根据全局映射的逻辑数据库集成操作每一个子数据库。本文所提出的方案为联邦数据库增加了全局映射模式,是联邦系统的一种改进。     

Magnetic properties and microstructural characteristics of bulk Nd–Al–Fe–Co glassy alloys

Bulk Nd–Al–Fe–Co glassy alloys with diameter up to 5 mm were investigated by magnetic measurements, magnetic force microscopy (MFM) and high resolution electron microscopy (HREM) at room temperature. The results from the measurement of vibrating sample magnetometer show that these samples with compositions Nd₆₅Al₁₀Fe_25-xCo_x (x=0–10 at.%) and Nd₆₀Al₁₀Fe₂₀Co₁₀ display hard magnetic properties with H_C of 300 kAm⁻¹, M_S of 10 Am² kg⁻¹, and M_r of 7 Am² kg⁻¹. The MFM measurements of the Nd₆₀Al₁₀Fe₂₀Co₁₀ bulk metallic glass (BMG) reveal the existence of magnetic domains with a period of about 0.36 μm, and the ordered clusters with the averaged size of about 5 nm was observed by the HREM on the sample. The domain structure or cluster is believed to be associated with the appearance of hard-magnetic properties in this alloy system. The existence of the large-size domains demonstrates that magnetic moment of a great deal of ordered atomic clusters in the BMG has been aligned by exchange-coupling.     

计及风/光/荷不确定性的综合能源站随机规划研究

传统变电站灵活性较差,可再生能源消纳能力有限,导致弃风弃光现象严重。在此背景下,提出了一种计及风/光/荷不确定性的综合能源站随机规划方法。首先,将综合能源制氢站的规划总周期划分为若干个子周期;然后,在各个子周期内采用蒙特卡罗法对风、光年利用小时数模型和电、氢负荷分布模型进行模拟,以设备全寿命周期成本最小为目标,建立包括电解水装置和储能设备在内的随机规划模型,并通过非线性规划求解器求解。最后,通过算例分析,可验证文章所提周期性随机规划方法在提升综合能源站经济效益、可再生能源利用率和供能可靠性等方面的有效性。     

基于Q学习的电网运行断面动态生成方法

电网运行断面是电力系统运行控制的重要手段.面对当前繁多的电网运行断面智能生成方法,如何合理的选择已成为电网运行断面在线生成算法领域研究的重要内容.在此背景下,提出一种基于Q学习的电网运行断面动态生成方法.该方法的主要特征在于训练得到Q学习智能体,根据电网运行特征动态选择电网运行断面生成方法,以便充分利用不同生成方法在不同场景下的算法优势.最后,基于某电网数据构造的算例表明,动态生成方法能够通过优化选择不同场景下的生成算法,提升生成结果的准确率.对于应用样本集,该方法提高准确率近5.2％.     

定向凝固的进展对高纯铝偏析法提纯工艺的影响

铝提纯工艺为高纯铝发展的关键。现在国内外通用的高纯铝制备工艺为三层液电解法和分步冷凝法。由于能耗高或提纯效率低，这两种工艺都不能完全满足当今工业发展的需要。针对当前广泛采用的凝固偏析法，结合定向凝固的新进展，讨论了改进结晶方式，提高提纯效率和生产效率的工艺方法。     

塑料注射模冷却系统的设计

分析和列举了注射模冷却系统设计的要点及注意事项；介绍了用CAE技术优化冷却系统设计的方法。     

笔式界面软件资源管理模型及其实现*

针对目前文档管理模式和WIMP界面交互范式对笔式交互的不足,提出了笔式界面软件文档资源管理的统一模型,能够有效地管理笔式界面软件所产生的文档数据,并且根据此模型实现了笔式界面的文档资源管理系统,采用了PIBG交互范式和实物界面.评估结果表明,根据此模型设计的系统,能够提高用户工作效率,降低用户的认知负担.     

Memristive neuron model with an adapting synapse and its hardware experiments

Electromagnetic induction effect caused by neuron potential can be mimicked using memristor. This paper considers a fluxcontrolled memristor to imitate the electromagnetic induction effect of adapting feedback synapse and presents a memristive neuron model with the adapting synapse. The memristive neuron model is three-dimensional and non-autonomous. It has the time-varying equilibria with multiple stabilities, which results in the global coexistence of multiple firing patterns. Multiple numerical plots are executed to uncover diverse coexisting firing patterns in the memristive neuron model. Particularly, a nonlinear fitting scheme is raised and a fitting activation function circuit is employed to implement the memristive mono-neuron model. Diverse coexisting firing patterns are observed from the hardware experiment circuit and the measured results verify the numerical simulations well.     

One-step fabrication superhydrophobic sand filter for capillary-driven separation of water-in-oil emulsions

The efficient separation of water-in-oil emulsion is of significance in environment and energy filed, and it has become a world-wide challenge. Herein, we have presented a one-step, facile and low-cost approach to prepare superhydrophobic sands for efficient separation of water-in-oil emulsion. The as-prepared sand layers possessed a water contact angle higher than 151°, demonstrating their superior superhydrophobic property. Besides, the as-prepared sand layers could separate water-in-emulsions with separation efficiency up to 99.7%, which is superior to both traditional and superwettable filtration membranes. The effect of thickness of sand layer on separation performance was also investigated. The results showed that the filtration flux decreased with the increased of filtration thickness while the separation efficiency increased. The as-prepared sand layer proposed by this study is a processing candidate for separating water-in-oil emulsion in practical industry. Additionally,the as-prepared superhydrophobic sand fabrication method also provides an alternative for desert water storage.