Evolution of neural networks for classification and regression

Although Artificial Neural Networks (ANNs) are importantdata mining techniques, the search for the optimal ANN is a challenging task: the ANN should learn the input–output mapping without overfitting the data and training algorithms may get trapped in local minima. The use of Evolutionary Computation (EC) is a promising alternative for ANN optimization. This work presents two hybrid EC/ANN algorithms: the first evolves neural topologies while the latter performs simultaneous optimization of architectures and weights. Sixteen real-world tasks were used to test these strategies. Competitive results were achieved when compared with a heuristic model selection and other Data Mining algorithms.     

极紫外/软X射线多层膜衍射光栅

多层膜光栅是集多层膜的高反射率与光栅的高光谱分辨于一身的新型光学元件,它的出现使得在极紫外/软X射线(EUV/SXR)波段采用非掠入射、高光谱分辨率的分光元件成为可能.对多层膜光栅的衍射特性进行了分析,这些衍射特性使多层膜光栅能够作为核心部件,实现很多新颖的EUV/SXR光谱仪器,如窄带X-UV单色仪与多色仪.近年来,多层膜光栅的制备领域不断地出现新的思想与方法,同时多层膜光栅的实际应用也不断增多.多层膜光栅作为一种重要的光学元件,在天文学、物理学和材料科学等领域具有广泛的用途.介绍了多层膜光栅在众多研究领域中的应用.     

Fabrication of Al and Al/Ti coatings on magnesium alloy by sputtering

Multi-magnetron sputtering was applied to prepare aluminum coating and aluminum/titanium multilayer coating on AZ31 magnesium alloy. FESEM, AFM and XRD were used to investigate the morphology and phase structure of these obtained coatings. Aluminum coating presented a (111) preferred texture and this texture was strongly strengthened with the Ti(002) plane as template in Al/Ti multilayer coating. The top surface of Al/Ti-coated sample took on a round roof-like morphology compared to the pyramid-like morphology of Al-coated sample. The result of polarization tests showed that both Al coating and Al/Ti multilayer coating could improve the corrosion resistance of AZ31 magnesium alloy in 3.5 wt.% NaCl solution.     

Electroluminescent characteristics of spin-assembled multilayer films with confined layer structure

Multilayer films composed of poly(p-phenylene vinylene) (PPV) as the semiconducting polymer and poly(methacrylic acid) (PMAA) as the insulating polymer were fabricated by spin-assembly method. These films, comprising a confined layer structure, showed that the water contact angles are periodically and distinctly oscillated when the top surface layer is alternated between PPV precursor and PMAA. The turn-on voltage of the multilayer electroluminescent (EL) devices increased from 2.6 V to 9.8 V as the thickness of the PMAA layer inserted between neighboring PPV layers was increased from 0 (i.e., PPV single layer film) to 2.0 nm. Furthermore, the emission peaks in the photoluminescent and EL spectra of these devices were strongly blue-shifted due to excitons formed at the confined PPV layers. Particularly when inserting about 1.0 nm thick PMAA layers, which possibly induced a tunneling effect on the charge carriers (i.e., holes and electrons), these multilayer films decreased the mobility of the hole carriers in the PPV layers with strong hole transporting characteristics, and therefore increased the recombination probability in the emitting layer with confined geometry. As a result, the device efficiency was significantly improved in comparison with that of a PPV single layer device without PMAA layer and with that of devices with relatively thick PMAA layers of 1.4 or 2.0 nm.     

基于多层前馈神经网络的并联型电能质量控制器

神经网络用于电力系统电能质量分析和控制是一个新研究领域.快速可靠地提取谐波分量决定着并联型电能质量控制器的整体性能,构造了一种和理论分析相一致的基于反向传播算法的三层前馈神经网络,离线训练收敛后可用来在线检测电力系统谐波电流.系统中逆变器补偿电流的产生对系统的补偿性能至关重要,提出了一种基于神经网络的逆变器瞬时电流PWM控制.并联型电能质量控制器投入系统后电流总畸变率由26.29%下降为5.25%.仿真实例表明,所提并联型电能质量控制器动态响应快,可改善电力系统电流波形畸变,提高电能质量.     

基于小波分解与运动补偿的弹迹检测方法

基于小波分解与运动补偿的弹迹检测方法,采用多层离散小波分解图像,以低频图像块匹配估计背景运动矢量.再利用差分算法检测射弹,计算出形心坐标并绘出射弹运动轨迹.其实现步骤包括:图像预处理-靶标定位跟踪、原始图像小波多层分解、低频图像块匹配、背景图像运动补偿及射弹检测与空间坐标推断.并以实验验证.     

Development and performance evaluation of an improved complex valued radar pulse compressor

Pulse compression is an important and burning issue in radar signal processing. In the recent past, many adaptive and neural network based methods have been proposed to achieve effective pulse compression performance for real coded transmitted waveforms. Even though the radar signal is complex, it is mostly processed as real-valued in-phase and quadrature components. Hence it is desirable that for processing complex radar signal for pulse compression both the structure as well as the learning algorithm associated with it need to be complex in nature. Accordingly in this paper a novel adaptive method is proposed by employing a complex valued fully connected cascaded (CFCC) neural network. For training this network, a new complex Levenberg–Marquardt (CLM) algorithm is derived and used for imparting effective training of its weights. The new CLM based CFCC (CFCC-CLM) model offers superior convergence performance with the least residual mean squared error during training phase compared to those provided by the multilayer perceptron (MLP) trained with complex domain backpropagation (CDBP) and CLM based methods. Further the comparison of peak signal-to-sidelobe ratio (PSR) under noisy and Doppler shift conditions of the proposed method exhibits best performance compared to those offered by the MLP-CDBP, MLP-CLM and the matched filter (MF) based methods.     

A versatile strategy for achieving the second-order advantage when applying different artificial neural networks to non-linear second-order data: Unfolded principal component analysis/residual bilinearization

Second-order instrumental signals showing a non-linear behaviour with respect to analyte concentration can still be adequately processed in order to achieve the important second-order advantage. The combination of unfolded principal component analysis with residual bilinearization, followed by application of a variety of neural network models, allows one to obtain the second-order advantage. While principal component analysis models the training data, residual bilinearization models the contribution of the potential interferents which may be present in the test samples. Neural networks such as multilayer perceptron, radial basis functions and support vector machines, are all able to model the non-linear relationship between analyte concentrations and sample principal component scores. Three different experimental systems have been analyzed, all requiring the second-order advantage: 1) pH–UV absorbance matrices for the determination of two active principles in pharmaceutical preparations, 2) fluorescence excitation–emission matrices for the determination of polycyclic aromatic hydrocarbons, and 3) UV-induced fluorescence excitation–emission matrices for the determination of amoxicillin in the presence of salicylate. In all cases, reasonably accurate predictions can be made with the proposed techniques, which cannot be reached using traditional methods for processing second-order data.     

Microstructure, interdiffusion and magnetic properties of ZnO/MnOx multilayers grown by pulse laser deposition

ZnO/MnO_x nano-thin multilayers are fabricated by pulsed laser deposition with a sectioned target or “cake” technique. Structure and magnetic properties are measured, finding a magnetic signal that scales better with the number of interfaces than with the film volume, and an interlayer interdiffusion that happens at remarkably low film growth temperatures.     

A three-dimensional finite element analysis of the temperature field during laser melting of metal powders in additive layer manufacturing

Simulating the transient temperature field in additive layer manufacturing (ALM) processes has presented a challenge to many researchers in the field. The transient temperature history is vital for determining the thermal stress distribution and residual stress states in ALM-processed parts that utilise a moving laser heat source. The modelling of the problem involving multiple layers is equally of great importance because the thermal interactions of successive layers affect the temperature gradients, which govern the heat transfer and thermal stress development mechanisms. This paper uses an innovative simulation technique known as element birth and death, in modelling the three-dimensional temperature field in multiple layers in a powder bed. The results indicate that the heated regions undergo rapid thermal cycles that could be associated with commensurate thermal stress cycles. Deposition of successive layers and subsequent laser scanning produces temperature spikes in previous layers. The resultant effect is a steady temperature build-up in the lower layers as the number of layers increases.