An Algorithm for Cavity Reconstruction in Electrical Impedance Tomography

We consider the inverse problem of finding cavities within some object from electrostatic measurements on the boundary. By a cavity we understand any object with a different electrical conductivity from the background material of the body. We give an algorithm for solving this inverse problem based on the output nonlinear least-square formulation and the regularized Newton-type iteration. In particular, we present a number of numerical results to highlight the potential and the limitations of this method.

The Rapidly Solidified Aging Copper Alloy by BP Neural Network

Rapid solidifiation is a kind of new process for enhancing the hardness and electrical conductivity of Cu-Cr-Zr copper alloy.The use of BP neural network(NN) is presented to model the non-linear relationship between parameters of age hardening processes and the mechanical and electrical properties of rapdily solidified Cu-Cr-Zr alloy.The improved model is developed by the Levenberg-Marquardt training algorithm and the good generalization performance is demonstrated.So,an important foundation has been laid for optimisticaly controlling the rapidly solidified aging processes of Cu-Cr-Zr alloy.

An improved potential field method for mobile robot navigation

In order to overcome the inherent oscillation problem of potential field methods(PFMs) for autonomous mobile robots in the presence of obstacles and in narrow passages,an enhanced potential field method that integrates Levenberg-Marquardt(L-M) algorithm and k-trajectory algorithm into the basic PFMs is proposed and simulated.At first,the mobile robot navigation function based on the basic PFMs is established by choosing Gaussian model.Then,the oscillation problem of the navigation function is investigated when a mobile robot nears obstacles and passes through a long and narrow passage,which can cause large computation cost and system instability.At last,the L-M algorithm is adopted to modify the search direction of the navigation function for alleviating the oscillation,while the k-trajectory algorithm is applied to further smooth trajectories.By a series of comparative experiments,the use of the L-M algorithm and k-trajectory algorithm can greatly improve the system performance with the advantages of reducing task completion time and achieving smooth trajectories.

Simulation of the Ultrasonic Precipitation Process of Nano-hydroxyapatite by an Artificial Neural Network

The ultrasonic precipitation technique for preparing hydroxyapatite nanoparticles is a complex process that was strongly influenced by temperature, reaction time and ultrasonic power. The use of a modified artificial neural network （ANN） was proposed to model the non-linear relationship between ultrasonic precipitation parameters and the hydroxyapatite content. The improved model for processing dataset and selecting its topology was developed using the Levenberg-Marquardt training algorithm and was trained with comprehensive dataset of hydroxyapatite nanoparticles collected from experimental data. A basic repository on the domain knowledge of ultrasonic precipitation process for the preparation of hydroxyapatite is established via sufficient data mining by the network. With the help of the repository stored in the trained network, the influence of preparation temperature, preparation time and ultrasonic sonicating power on the hydroxyapatite content can be analyzed and predicted. The results show that the ANN system is effective and successful in analyzing the influence of ultrasonic precipitation parameters on the preparation of hydroxyapatite nanoparticles.

Structural form selection of the high-rise buildingwith the improved BP neural network

As civil engineering technology development,the structural form selection is more and more critical in design of high-rise buildings.However,structural form selection involves expertise knowledge and changes with the environment which makes the task arduous.An approach utilizing improved back propagation(BP)neural network optimized by the Levenberg-Marquardt(L-M)algorithm is proposed to extract the main controlling factors of structural form selection.Then,an intelligent expert system with artificial neural network is constructed to design high-rise buildings structure effectively.The experiment tests the model in 15 well-known architecture samples and get the prediction accuracy of 93.33%.The results show that the method is feasible and can help designers select the appropriate structural form.

多波长BOTDR系统中布里渊散射谱的特征提取

根据布里渊散射谱的传输特性和高精度特征提取的要求,理论分析了布里渊频移的波长依赖性和多波长探测光传感时的外差检测布里渊散射谱特征,提出了利用Pseudo-Voigt基函数和莱文伯格-马夸尔特(L-M)优化算法对布里渊散射叠加谱进行特征提取。通过与洛伦兹、高斯和5次多项式曲线拟合法进行预测比较,在中心频移为11.122 903GHz的单波长和多波长传感的仿真散射谱模型中,本文所提方法的频移测量误差最小,对应的温度测量误差仅为0.047、0.000和0.112℃,且拟合度最好。在采用多模法布里-珀罗激光器的布里渊散射谱检测系统中,Pseudo-Voigt曲线拟合的综合评价指标优于其他3种拟合方法。仿真分析和实验结果表明,Pseudo-Voigt曲线拟合适用于多波长传感时布里渊散射叠加谱的特征提取,可有效地提高预测精度。

Modeling of CVI process in fabrication of carbon/carbon composites by an artificial neural network

The chemical vapor infiltration(CVI) process in fabrication of carbon-carbon composites is very complex and highly inefficient, which adds considerably to the cost of fabrication and limits the application of the material. This paper tries to use a supervised artificial neural network(ANN) to model the nonlinear relationship between parameters of isothermal CVI(ICVI) processes and physical properties of C/C composites. A model for preprocessing dataset and selecting its topology is developed using the Levenberg-Marquardt training algorithm and trained with comprehensive dataset of tubal C/C components collected from experimental data and abundant simulated data obtained by the finite element method. A basic repository on the domain knowledge of CVI processes is established via sufficient data mining by the network. With the help of the repository stored in the trained network, not only the time-dependent effects of parameters in CVI processes but also their coupling effects can be analyzed and predicted. The results show that the ANN system is effective and successful for optimizing CVI processes in fabrication of C/C composites.