Key Engineering Materials Vols. 467-469

Abstract: Complete coverage path planning (CCPP) is an essential issue for Autonomous Underwater Vehicles’ (AUV) tasks, such as submarine search operations and complete coverage ocean explorations. A CCPP approach based on biologically inspired neural network is proposed for AUVs in the context of completely unknown environment. The AUV path is autonomously planned without any prior knowledge of the time-varying workspace, without explicitly optimizing any global cost functions, and without any learning procedures. The simulation studies show that the proposed approaches are capable of planning more reasonable collision-free complete coverage paths in unknown underwater environment.

Abstract: Association rules mining is attracting much attention in research community due to its broad applications. Existing web data mining methods suffer the problems that 1) the large number of candidate itemsets, which are hard to be pruned, should be pruned in advance. 2) the time of scanning the database, which are needed to scan transactional database repeatedly, should be reduced. In this paper, a new association rules mining model is introduced for overcoming above two problems. We develop an efficient algorithm-WARDM(Weighted Association Rules Data Mining) for mining the candidate itemsets. The algorithm discusses the generation of candidate-1 itemset, candidate-2 itemset and candidate-k itemset(k>2),which can avoid missing weighted frequent itemsets. And the transactional database are scanned only once and candidate itemsets are pruned twice, which can reduce the amount of candidate itemsets. Theoretical analysis and experimental results show the space and time complexity is relatively good, Meanwhile the algorithm decreases the number of candidate itemsets, enhances the execution efficiency.

Abstract: Gene Expression Programming(GEP) is a novel and accurate approach for classification. With the shortcoming of GEP, it often falls into the local optimums. In this paper, we introduce the virus evolutionary mechanism into GEP, with the infection operation of virus population, the diversity of the host population is increased, and the system is much easier to jump out of the local optimums, and much faster to obtain better results. Experiments on several benchmark data sets show that our approach can get close average accuracy and much better best accuracy compared with available results. What’s more, the average execution time is largely decreased due to smaller population size and maximum generation.

Abstract: U.S. dollar index, oil prices, silver prices, DOW index, OECD leading index and the CRB index are selected and varying-coefficient regression model which has dynamic response to the various variables influence is applied to predict the gold price and improve the prediction accuracy in this paper. In addition, the weighted least squares is adopted as an estimation of the parameters, corrects the traditional least squares method defect which assumes the sample data weights equal points to the prediction, making sample weights larger closer with prediction points. In the choice of weighting function, the paper uses cross validation to gain smoothing parameter. In the last, we predicted the 12 months gold prices from January 2010 December 2010 applies varying-coefficient regression model.

Abstract: A comprehensive introduction of current aluminum refining technology is reviewed in this work especially the effects of chlorine on refining process is discussed. The mechanism of chlorine on improving hydrogen diffusion has been studied by ab initio molecular dynamics calculations and we obtain the diffusivity of hydrogen in liquid aluminum which is in agreement with the experimental data. It can be concluded that the diffusion of hydrogen in aluminum melts can be enhanced on the presence of chlorine.

Abstract: In order to achieve the real-time monitoring internal state of composite specimen embedded shape memory alloy wires, a new wireless monitoring method in which PS021 is used to measure the capacitance change of composite structure is proposed. The micro-capacitance measuring principle of PS021 and the design of hardware and software of system are introduced in detail. The capacitance signal between some two wires of composite structure embedded SMA is measured by the PS021 module, and the measurement data are processed by C8051F330 micro-controller and wirelessly sent to PC by PTR4000. Compared the tested results with that of the high precision LCR meter, experimental results show that the relative error of system is less than 4.5%.

Abstract: This paper deals with the condition for a fuzzy neural network to realize memory, gives the concept of lattice point and proves that if sample set is of lattice point distribution the sample set can be memorized by a fuzzy neural network, otherwise the fuzzy neural network will lost its memory. A theorem shows that Choice-memory method not only reduces computation but also ensures memory of fuzzy neural network.

Abstract: This paper deals with the under-actuated characteristic of a quad-rotor unmanned aerial vehicle (UAV). By designing the double loop configuration, the autonomous trajectory tracking is realized. The model uncertainty, external disturbance and the senor noise are also taken into consideration. Then the controller is put forward in the inner loop. An optimal stability augmentation control (SAC) method is used to stabilize the horizon position and keep it away from oscillation. By calculating the nonlinear decouple map, control quantity is converted to the speeds of the four rotors. At last some simulation results and the prototype implementation prove that the control method is effective.

Abstract: For complicated nonlinear systems, the data inevitably have noise, random disturbance, Traditional kernel principal component analysis (KPCA) methods are very difficult to calculate the kernel matrix K for fault detection with large sample sets. So an improved KPCA method based on wavelet denoising is proposed. First, wavelet denoising method is used for data processing, then the improved KPCA method can reduce calculational complexity of fault detection. The proposed method is applied to the benchmark of Tennessee Eastman (TE) processes. The simulation results show that the proposed method can effectively improve the speed of fault detection.

Abstract: In recent year, the rise of economic growth and technology advance leads to improve the quality of service of traditional transport system. Intelligent Transportation System (ITS) has become more and more popular. At present, the collection of real-time traffic information is executed in two ways: (1) Stationary Vehicle Detectors (VD) and (2) Global Position System (GPS)-based probe cars reporting. However, VD devices need a large sum of money to build and maintain. Therefore, we propose the linear regression model to infer the equation between vehicle speed and traffic flow. The traffic flow can be estimated from the speed which is obtained from GPS-based probe cars. In experiments, the Speed Error Ratio (SER) and Flow Error Ratio (FER) of linear regression model are 4.60% and 24.63% respectively. The estimated speed and traffic flow by using linear regression model is better than by using linear model, power law model, exponential model, and normal distribution model. Therefore, the linear regression model can be used to estimate traffic information for ITS.