电容式微传声器的制备研究新进展

本文对近几年来运用单片制备方法，采用硅微机械加工技术（MEMS）工艺研制的电容式微传声器进行了详细描述。通过分析和研究指出，将微传声器与其外围电路集成在单片上可降低器件噪声，采用纹膜或复合膜结构可增加器件的灵敏度，根据具体情况优化电极形状和尺寸，可以在很大程度上提高微传声器的性能。

电磁式纳米压入仪的载荷精度研究

纳米压入仪(Nanoindenter)已成为纳米/微米力学测试的基本工具之一,广泛应用于纳米材料、薄膜、MEMS微结构的力学性能测试。目前商业化仪器有MTS NanoIndenter(incorporated into Agi lent in 2008),Hysitron TriboIndenter,CSM NanoHardness Tester,MML NanoTest,and CSIRO UMIS,其中原MTS Nano Indenter、CSMNano Hardness Tester和MML NanoTest都采用电磁驱动兼载荷计量的原理设计。这类仪器主要优点之一是载荷分辨力高,可以达到50nN甚至更小。然而仪器制造商从未明确给出此类仪器的载荷精度。目前对纳米压痕测试技术的研究也主要集中在压头形状、试样表面粗糙度、毛细力等因素对压痕测试结果的影响,尚未见针对仪器载荷精度的研究报道。国际标准ISO 14577-2:2002虽对仪器施加的试验力的允差进行了规定,但由于电磁式纳米压入仪的结构响应和试样力学行为耦合在一起,试样上的载荷并不完全等同于通过电磁转换原理计量的试验力,因此实际测试时试样上的载荷精度仍是未知。

Glass wafers bonding via Diels-Alder reaction at mild temperature

In this paper, we introduced a novel bonding method of glass wafers by Diels-Alder reaction at mild temperature. After standard hydroxylization and aminosilylation, two wafers were modified by 2-furaldehyde and maleic anhydride, respectively. Then they were brought into close contact and tightly held with a clamping fixture. A strong bonding could be achieved by annealing for 5 h at 200 degrees C. Bonding strength is as high as 1.78 MPa and sufficient for most application of microfluidic chips.

面向数据收集的无线传感器网络节点部署问题研究

随着MEMS、无线通信、自动控制和人工智能等技术的快速发展，无线传感器网络引起了学术界的极大关注。它的出现改变了人与自然界的交互方式，其应用领域将会深入到社会生活的各个方面。 节点部署问题是无线传感器网络研究的基本问题之一。在无线传感器网络的某些实际应用中，节点部署通常采用人工部署方式，为满足网络的节能性、可靠性、实时性等性能需求，需要额外部署一些节点。为限制网络布设成本，要对这些新增加的节点进行有针对性地部署。基于这种应用环境，论文对面向数据收集的无线传感器网络节点部署问题展开研究工作，主要的研究内容与成果包括以下几个方面。 论述了无线传感器网络节点部署问题的研究内容、分类和评价体系，系统地总结了前人的研究成果和不足。 为改善网络的连通性和满足网络生命期需求，提出了一种启发式的两阶段中继节点部署策略。第一阶段的部署确保了网络连通性；第二阶段部署，结合负载平衡的路由协议，满足了网络的生命期需求。仿真结果表明该方法在满足上述性能需求的前提下，最大限度地减少了需要部署的中继节点数目。 针对网络中簇首选择不合理导致数据收集成本过高的问题，提出了面向数据收集成本最优化的簇首节点选择算法。当网络状态一定时，簇首节点的优化选择问题是一类组合优化问题。论文首先提出了基于小规模网络的全局最优化的簇首节点选择算法（BFM）。针对BFM算法复杂度高的缺点，进而提出了基于大规模网络的启发式的簇首节点选择算法，该算法可以在较短时间内得到问题的次优解。 考虑网络簇首的容量限制和最大簇半径限制，把簇首节点的优化部署问题形式化为整数规划问题。提出两种启发式的簇首优化部署策略：基于K-平均的簇首部署策略和基于K-平均与模拟退火混合算法的簇首部署策略。前者适用于网络运行时的动态部署，后者适用于网络投入使用前的规划部署。仿真结果表明，分簇后的网络在满足上述限制的情况下，最大限度地延长了网络生命期。 有针对性的节点部署离不开网络状态监测，为使网络状态监测不过多占用网络资源，提出了一种高效的网络状态监测机制。通过对节点编码并在节点间建立起一个逻辑层次簇结构，利用各个子簇状态数据的相似性和编码的连续性，实现了网内无损聚合。该监测机制使得网络状态信息的收集，在不丢失数据细节信息的情况下，数据通信量大大减少。 本论文的研究仅为无线传感器网络中的规划部署问题提供了一个框架，尚未形成一个完整可行的网络规划部署系统。各项研究成果可以为无线传感器网络设计者、方案提供者以及无线传感器网络应用的系统集成，提供一些有益的指导。

面向微粒操纵的介电泳芯片系统研究

针对目前在纳米器件及传感器的制造中尚无对大量粒子进行有效操纵的方法,我们利用介电泳方法对大量微粒进行定位和传输操纵,介绍了利用MEMS工艺进行介电泳芯片加工的过程以及整个观测与实验系统的建立,通过有限元软件对传统介电泳和行波介电泳中电极阵列的电场分布进行求解,并在该实验系统下实现了对微通道中的悬浮高度和微粒的运动速度的测量.该实验系统的研究为液体环境下微纳颗粒的装配和分离提供了一条有效的技术路径.

便携式移动机器人姿态解算方法

本文根据便携式移动机器人的特点,采用四元数法解算机器人导航系统的姿态,避免了在机器人运动角度较大时出现奇异点的问题。文中应用改进的四阶龙格-库塔算法解算四元数微分方程,经仿真实验,精度完全能够达到要求。给出了合理的变换公式,在机器人运动范围内,满足了四元数与欧拉角之间转换的一一对应。

机器人化纳米刻画的建模及实时显示研究

利用AFM探针刻画方式可进行MEMS掩膜图案等的制备与修复研究,本文针对软朔性材料,建立了AFM探针刻画时的刻痕尺寸模型,依据该模型及探针的实际受力与位置信息,可得到刻痕的实时位置与尺寸,并借助虚拟现实技术将刻画过程实时显示在视觉界面上。在上述研究基础上,操作者可在线控制纳米刻画的过程及结果,对聚酯基片的纳米刻画实验初步验证了上述模型的有效性。

基于光学聚焦方法的热驱动微执行器位移测量

随着微机电系统(MEMS)设计日趋成熟,度量问题越来越成为微系统技术中的热点。使用改进的拉普拉斯求和方法(SML)和深度估计法来测量热度驱动微夹持器末端的弯曲程度。实验中,使用了10幅在42℃时微夹持器的水下工作图像来验证这种光学聚焦方法,结果证明使用光学聚焦方法可以测量出作为驱动器反馈输入的末端弯曲大小,实现对驱动器运动的精确控制。

微动机器人的致动机理研究综述

MEMS是当前研究的一个热点,微机器人对于发展MEMS具有重要意义,也是MEMS的一项不可缺少的内容。微动技术是机器人学理论的一个重要分支,也是发展微机器人及相关微技术的基础。目前,各种新型微驱动器层出不穷,极大地推动了微机器人技术的发展。对于微动原理进行分析,从本质上弄清微动产生的机理,不仅可以丰富机器人学理论,还有可能使微动技术产生质的飞跃。从这一角度出发,对各种微动原理加以详细分析和比较,以期得出有意义的结论。

微操作机器人系统及其关键技术研究

微制作机器人技术是MEMS技术的一个重要分支，也是当前机器人研究领域的一个热点。本文分析了微操作机器人集成系统的特点，并针对微制作机器人系统研制中涉及的一些关键技术，如驱动、定位、检测和控制等技术进行了论述。

纳米定位平台—基于“碰撞驱动法”

<正> 纳米技术(纳米尺度:1-102mm)是21世纪全球重大的高技术。纳米定位平台是微LSI制造、MEMS制造、超精密加工和生物工程等领域高度发展的难题。高精度定位平台为实现纳米位移,国际学术界提出很多新思想、新方法。有“振动驱动法”、“尺蠖驱动法”以及“冲击驱动法”。中科院沈阳自动化研究所经过理论和试验研究,提出“碰撞驱动法”。

Wearable wireless inertial measurement for sports applications

A wearable WIMU (Wireless Inertial Measurement Unit) [1] system for sports applications based on Tyndall's 25mm mote technology [2] has been developed to identify tennis performance determining factors, giving coaches & players improved feedback [3, 4]. Multiple WIMUs transmit player motion data to a PC/laptop via a receiver unit. Internally the WIMUs consist of: an IMU layer with MEMS based sensors; a microcontroller/transceiver layer; and an interconnect layer with supplemental 70g accelerometers and a lithium-ion battery. Packaging consists of a robust ABS plastic case with internal padding, a power switch, battery charging port and status LED with Velcro-elastic straps that are used to attach the device to the player. This offers protection from impact, sweat, and movement of sensors which could cause degradation in device performance. In addition, an important requirement for this device is that it needs to be lightweight and comfortable to wear. Calibration ensures that misalignment of the accelerometer and magnetometer axes are accounted for, allowing more accurate measurements to be made.

An automated calibration tool for high performance wireless inertial measurement in professional sports

Traditional motion capture techniques, for instance, those employing optical technology, have long been used in the area of rehabilitation, sports medicine and performance analysis, where accurately capturing bio-mechanical data is of crucial importance. However their size, cost, complexity and lack of portability mean that their use is often impractical. Low cost MEMS inertial sensors when combined and assembled into a Wireless Inertial Measurement Unit (WIMU) present a possible solution for low cost and highly portable motion capture. However due to the large variability inherent to MEMS sensors, such a system would need extensive characterization to calibrate each sensor and ensure good quality data capture. A completely calibrated WIMU system would allow for motion capture in a wider range of real-world, non-laboratory based applications. Calibration can be a complex task, particularly for newer, multi-sensing range capable inertial sensors. As such we present an automated system for quickly and easily calibrating inertial sensors in a packaged WIMU, demonstrating some of the improvements in accuracy attainable.

Investigation of micro-devices for neurobiological applications

The aim of this project is to integrate neuronal cell culture with commercial or in-house built micro-electrode arrays and MEMS devices. The resulting device is intended to support neuronal cell culture on its surface, expose specific portions of a neuronal population to different environments using microfluidic gradients and stimulate/record neuronal electrical activity using micro-electrode arrays. Additionally, through integration of chemical surface patterning, such device can be used to build neuronal cell networks of specific size, conformation and composition. The design of this device takes inspiration from the nervous system because its development and regeneration are heavily influenced by surface chemistry and fluidic gradients. Hence, this device is intended to be a step forward in neuroscience research because it utilizes similar concepts to those found in nature. The large part of this research revolved around solving technical issues associated with integration of biology, surface chemistry, electrophysiology and microfluidics. Commercially available microelectrode arrays (MEAs) are mechanically and chemically brittle making them unsuitable for certain surface modification and micro-fluidic integration techniques described in the literature. In order to successfully integrate all the aspects into one device, some techniques were heavily modified to ensure that their effects on MEA were minimal. In terms of experimental work, this thesis consists of 3 parts. The first part dealt with characterization and optimization of surface patterning and micro-fluidic perfusion. Through extensive image analysis, the optimal conditions required for micro-contact printing and micro-fluidic perfusion were determined. The second part used a number of optimized techniques and successfully applied these to culturing patterned neural cells on a range of substrates including: Pyrex, cyclo-olefin and SiN coated Pyrex. The second part also described culturing neurons on MEAs and recording electrophysiological activity. The third part of the thesis described integration of MEAs with patterned neuronal culture and microfluidic devices. Although integration of all methodologies proved difficult, a large amount of data relating to biocompatibility, neuronal patterning, electrophysiology and integration was collected. Original solutions were successfully applied to solve a number of issues relating to consistency of micro printing and microfluidic integration leading to successful integration of techniques and device components.