Flexible polyimide-based force sensor

We have realized a flexible force sensor, composed of four redundant capacitors, the operation of which is based on the measurement of a load-induced capacitance change. We use polyimide both as flexible substrate and as elastic dielectric between two levels of finger-shaped aluminum electrodes. In particular we have developed a technology for realization of two-level polyimide microstructures with gentle slopes to facilitate subsequent metallization processes. Thereby, we could improve step coverage and electrical contacting between the two metallization levels, as well as the mechanical stability of the sensor. The smooth polyimide slopes were obtained by combining lithographic resist-reflow techniques with dry etching procedures. We have analytically modeled the sensor's capacitance and its force sensitivity. We have electrically characterized the capacitors using an impedance analyzer and obtained capacitances in the range of 130 pF and a typical force sensitivity of 0.5-1 fF/N, in excellent agreement with our model.     

Fine-structured oxide ceramics from a novel replication method

Fine-structured oxide ceramics sintered bodies with sub-micrometer scales were fabricated by a replication method from mixed slurry of polyvinyl alcohol (PVA) and nanosized ceramic particles in water. By using this replication method and burn-out at various sintering temperatures, fine-structured line and space patterns with sizes as small as 300 nm were fabricated on surfaces of TiO₂ and ZnO. The mechanism of pattern deformation was investigated and a general expression to predict the minimum pattern size was proposed based on the relationship between pattern size and grain size. The results demonstrated the feasibility of using this simple method to produce sintered oxide ceramic materials with various compositions and structures.     

Speaking of Housing… A Reporton a consultation with victorian Women on Housing

Women in Supportive Housing (WISH) is a community-based organisation engaged in research, community development, community education and service development relevant to the housing and support needs of women in Victoria. In 1989 WISH initiated a small-scale consultation with women on their housing experiences which was funded by the State Housing Authority's Community Resourcing Programme.In 1990 WISH and the Ministerial Advisory Committee on Women and Housing (MACWH) decided to join their interests in this consultation work and, as a result, WISH was funded by the MACWH to undertake a larger state-wide consultation, resulting in the publication of a report⁹     

A new method for pattern recognition in load profiles to support decision-making in the management of the electric sector

This work presents a method for the selection, typification and clustering of load curves (STCL) capable of recognizing consumption patterns in the electricity sector. The algorithm comprises four steps that extract essential features from the load curve of residential users with an emphasis on their seasonal and temporal profile, among others. The method was successfully implemented and tested in the context of an energy efficiency program carried out by the Energy Company of Maranhão (Brazil). This program involved the replacement of refrigerators in low-income consumers’ homes in several towns located within the state of Maranhão (Brazil). The results were compared with a well known time series clustering method already established in the literature, Fuzzy CMeans (FCM). The results reveal the viability of the STCL method in recognizing patterns and in generating conclusions coherent with the reality of the electricity sector. The proposed method is also useful to support decision-making at management level.     

全球油气生产西移与消费东移的趋势及其影响

因技术进步和高油价,非常规资源正在成为战略接替,西半球油气供应在全球的地位越来越重要,将从净进口变为净出口;发展中国家油气消费进入快速增长期,世界油气需求重心由西向东转移,亚太地区成为世界最大油气消费地区。这些将改变世界油气市场和地缘政治格局,改变世界竞争形势,对中国既是机遇更多的是挑战。要坚持立足国内、增源与节流并举,坚持科学替代和技术创新,坚持增强全球供应能力,坚持参与全球能源治理。     

A filtering algorithm for k-mismatch with don't cares

We present a filtering based algorithm for the k-mismatch pattern matching problem with don't cares. Given a text t of length n and a pattern p of length m with don't care symbols in either p or t (but not both), and a bound k, our algorithm finds all the places that the pattern matches the text with at most k mismatches. The algorithm is deterministic and runs in Θ(nm1/3k1/3log2/3m) time.     

高速板材阶梯槽图形板制作研究

针对采用松下RX板材设计的高速板材阶梯槽图形板的制作进行研究。此板设计有槽底通孔、背钻孔、NPTH孔、整板的PTH阶梯孔、POFV孔。样板层数高,制作工艺复杂,难度大,对准度、阻抗以及可靠性要求高。     

Color image compression based on block truncation coding using pattern fitting principle

This paper describes a color image compression technique based on block truncation coding using pattern fitting (BTC-PF). High degree of correlation between the RGB planes of a color image is reduced by transforming them to O₁O₂O₃ planes. Each O_i plane (1?i?3) is then encoded using BTC-PF method. Size of the pattern book and the block size are selected based on the information content of the corresponding plane. The result of the proposed method is compared with that of several BTC based methods and the former is found superior. Though this method is a spatial domain technique, it is also compared with JPEG compression method, which is one of most popular frequency domain techniques. It is found that the performance of the proposed method is a little inferior to that of the JPEG in terms of quality of the reconstructed image. Decoding time is another important criterion where the compressed image is decoded frequently for various purposes. As the proposed method requires negligible decoding time compared to JPEG, the former is preferred over the latter in those cases.     

Discriminatively regularized least-squares classification

Over the past decades, regularization theory is widely applied in various areas of machine learning to derive a large family of novel algorithms. Traditionally, regularization focuses on smoothing only, and does not fully utilize the underlying discriminative knowledge which is vital for classification. In this paper, we propose a novel regularization algorithm in the least-squares sense, called discriminatively regularized least-squares classification (DRLSC) method, which is specifically designed for classification. Inspired by several new geometrically motivated methods, DRLSC directly embeds the discriminative information as well as the local geometry of the samples into the regularization term so that it can explore as much underlying knowledge inside the samples as possible and aim to maximize the margins between the samples of different classes in each local area. Furthermore, by embedding equality type constraints in the formulation, the solutions of DRLSC can follow from solving a set of linear equations and the framework naturally contains multi-class problems. Experiments on both toy and real world problems demonstrate that DRLSC is often superior in classification performance to the classical regularization algorithms, including regularization networks, support vector machines and some of the recent studied manifold regularization techniques.