Intelligent tool wear monitoring based on parallel residual and stacked bidirectional long short-term memory network

Effective tool wear monitoring (TWM) is essential for accurately assessing the degree of tool wear and for timely preventive maintenance. Existing data-driven monitoring methods mainly rely on complex feature engineering, which reduces the monitoring efficiency. This paper proposes a novel TWM model based on a parallel residual and stacked bidirectional long short-term memory (PRes–SBiLSTM) network. First, a parallel residual network (PResNet) is used to extract the multi-scale local features of sensor signals adaptively. Subsequently, a stacked bidirectional long short-term memory (SBiLSTM) network is used to obtain the time-series features related to the tool wear characteristics. Finally, the predicted tool wear value is outputted through a fully connected network. A smoothing correction method is applied to improve the prediction accuracy. The proposed model is experimentally verified to have a high prediction accuracy without sacrificing its generalization ability. A TWM system framework based on the PRes–SBiLSTM network is proposed, which has a certain reference value for TWM in actual industrial environments.     

Register allocation for write activity minimization on non-volatile main memory for embedded systems

Non-volatile memories are good candidates for DRAM replacement as main memory in embedded systems and they have many desirable characteristics. Nevertheless, the disadvantages of non-volatile memory co-exist with its advantages. First, the lifetime of some of the non-volatile memories is limited by the number of erase operations. Second, read and write operations have asymmetric speed or power consumption in non-volatile memory. This paper focuses on the embedded systems using non-volatile memory as main memory. We propose register allocation technique with re-computation to reduce the number of store instructions. When non-volatile memory is applied as the main memory, reducing store instructions will reduce write activities on non-volatile memory. To re-compute the spills effectively during register allocation, a novel potential spill selection strategy is proposed. During this process, live range splitting is utilized to split certain long live ranges such that they are more likely to be assigned into registers. In addition, techniques for re-computation overhead reduction is proposed on systems with multiple functional units. With the proposed approach, the lifetime of non-volatile memory is extended accordingly. The experimental results demonstrate that the proposed technique can efficiently reduce the number of store instructions on systems with non-volatile memory by 33% on average.     

An illusion of memory: False recognition influenced by unconscious perception.

The results of two experiments showed that an illusion of memory can be produced by unconscious perception. In a first phase of those experiments, a long list of words was presented for study. For the test of recognition memory given in the second phase of each experiment, presentation of a "context" word preceded that of most recognition test words. Ss were to judge whether or not the test words had been presented during the earlier study phase of the experiment. Effects of a context word on this recognition memory decision were opposite when Ss were aware vs. unaware of its presentation. For example, as compared to a condition in which no context word was presented, the probability of false recognition was increased when Ss were unaware but decreased when Ss were aware of the presentation of a context word that matched the recognition test word. Results are discussed in terms of unconscious influences on an attribution process. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

压力容器CAD绘图软件包的研究

本文结合压力容器计算机辅助设计软件的研制,研究了利用AUTOCAD开发计算机辅助设计专用软件的几个关键性问题;主要有程序绘图,汉字信息、绘图数据的显示和选择以及与绘图程序的连接、专用软件的控制结构设计和内存管理等问题;提出了一整套设计新颖,使用方便,运行高效的解决方法。     

Cell处理器访存特征研究

Cell处理器是一款异构多核处理器,拥有强大的计算能力。但是,在进行应用并行化时,却受到本地存储器容量、访存带宽和数据传输延时等的限制。DMA传输是隐藏长延时、提高存储带宽利用率的有效方法。本文在分析Cell处理器结构基础上,进行了一系列详细的DMA测试,并利用指数拟合技术得到DMA平均带宽模型,发现参与DMA传输的SPE数量和每次DMA传输规模是影响DMA访存带宽的主要因素。     

2000～2001年金属磁性功能材料新进展

综述了金属磁性功能材料在2000－2001年间的若干新进展，内容包括：（1）高性能稀土永磁合金；（2）高性能硅钢合金；（3）Fe-N系新磁性材料；（4）磁电子学；（5）形状记忆磁性材料。     

Fluctuating NMDA Receptor Subunit Levels in Perirhinal Cortex Relate to Their Dynamic Roles in Object Memory Destabilization and Reconsolidation

Reminder cues can destabilize consolidated memories, rendering them modifiable before they return to a stable state through the process of reconsolidation. Older and stronger memories resist this process and require the presentation of reminders along with salient novel information in order to destabilize. Previously, we demonstrated in rats that novelty-induced object memory destabilization requires acetylcholine (ACh) activity at M₁ muscarinic receptors. Other research predominantly has focused on glutamate, which modulates fear memory destabilization and reconsolidation through GluN2B- and GluN2A-containing NMDARs, respectively. In the current study, we demonstrate the same dissociable roles of GluN2B- and N2A-containing NMDARs in perirhinal cortex (PRh) for object memory destabilization and reconsolidation when boundary conditions are absent. However, neither GluN2 receptor subtype was required for novelty-induced destabilization of remote, resistant memories. Furthermore, GluN2B and GluN2A subunit proteins were upregulated selectively in PRh 24 h after learning, but returned to baseline by 48 h, suggesting that NMDARs, unlike muscarinic receptors, have only a temporary role in object memory destabilization. Indeed, activation of M₁ receptors in PRh at the time of reactivation effectively destabilized remote memories despite inhibition of GluN2B-containing NMDARs. These findings suggest that cholinergic activity at M₁ receptors overrides boundary conditions to destabilize resistant memories when other established mechanisms are insufficient.     

Ti-22Nb-6Zr(at%)合金的超弹性和形状记忆效应

用弯曲法和拉伸法测定Ti-22Nb-6Zr（at％）合金的超弹性和形状记忆效应，研究固溶处理温度对Ti-22Nb-6Zr合金组织结构及性能的影响。结果表明：固溶处理后Ti-22Nb-6Zr合金的室温组织为单一的β相，晶粒尺寸随固溶处理温度升高而增大；合金的静态弹性模量小于30GPa；Ti-22Nb-6Zr合金具有良好的超弹性和一定的形状记忆效应。室温下变形，合金的超弹性和形状记忆效应随固溶处理温度升高而提高。900℃固溶处理后的合金在室温下拉伸变形，应变为5％时，总的最大回复应变达4．12％，其中超弹性回复应变为3．91％，记忆回复应变为0.21％。     

MICROSTRUCTURE AND REVERSE SHAPE MEMORY EFFECT IN A Cu-Zn-Al ALLOY

The bainite structure in a Cu-Zn-A1 alloy related to the reverse shape memory effect hasbeen observed by means of TEM.The reverse memory effect may be improved by up to oneorder of magnitude under applied constraint stress.The widespread propagation of bainite wasconfirmed to be the diffusion controlled shear process by the parabolic configuration of sideinterface of bainite plate and the twisting of intersected bainite plates.     

空冷淬火Cu-Zn-Al合金马氏体结构及稳定化现象

利用Ｘ射线衍射强度分析、透射电镜观察、相变点测量等方法研究了空冷淬火态Ｃｕ１８Ｚｎ１４Ａｌ合金马氏体结构及其稳定化现象。认为该合金空冷淬火可获得长程有序完全的Ｍ１８Ｒ马氏体，其有序组态为Ａｌ原子与Ｃｕ原子、Ｚｎ原子与Ｃｕ原子两种近邻（ｎｎ）有序（Ｋ＝偶，Ｈ＋Ｋ／２＝奇），以及产生Ｋ＝奇的衍射中Ａｌ原子与Ｚｎ原子次近邻（ｎｎｎ）有序。其基面原子分布为：Ⅰ—１４２５Ａｌ＋１１２５Ｃｕ、Ⅱ—Ｃｕ、Ⅲ—１８２５Ｚｎ＋７２５Ｃｕ。与此相对应，该合金相应为Ｈｅｕｓｌｅｒ结构。该合金在充分有序化的条件下仍存在马氏体稳定化。