Applied Mechanics and Materials Vols. 239-240

Abstract: Hybrid air-cushion vehicles (ACVs) provide a solution to transportation on soft terrain, whereas they also bring a new problem of excessive energy consumption. In order to minimize energy consumption in a computational manner, a Genetic Algorithm (GA) - Neural Network (NN) joint optimization algorithm is proposed to online calculate the corresponding optimal vehicle running parameters for given soil conditions. This is realized in three steps. (1) The energy consumption index is firstly simplified as a constrained function with respect to only two independent vehicle running parameters. (2) The optimal solutions are figured out offline with respect to some specific soil conditions by the designed GA optimizer. (3) The optimal solutions are figured out online with respect to general soil conditions by the designed NN optimizer, which is trained using the above offline-obtained data. The feasibility of the joint algorithm is supported by experiments, whose results show an effective integration of the GA’s advantage in complex function optimization and the NN’s advantage in generalization ability and computing speed.

Abstract: With respect to the voltage distortion caused by harmonics and unbalanced loads, traditional digital phase lock algorithm doesn't work properly. And this paper presents a novel estimating algorithm based on BP neutral network model. The voltage signal is the input of the network and then by training the network with the prepared samples, the network can calculate the grid phase correctly. This algorithm has no relationship with the harmonics so it can work in a wide range. A network program is finished and the algorithm is validated. The results show that the algorithm works effectively and the result is correct.

Abstract: The study on convergence of GA is always one of the most important theoretical issues. This paper analyses the sufficient condition which guarantees the convergence of GA. Via analyzing the convergence rate of GA, the average computational complexity can be implied and the optimization efficiency of GA can be judged. This paper proposes the approach to calculating the first expected hitting time and analyzes the bounds of the first hitting time of concrete GA using the proposed approach.

Abstract: A soft sensor modeling method is presented in this paper，it selects optimal fuzzy rules by tuning the radius of a subtractive cluster center to generate a T-S fuzzy model. The radius of a cluster center is adjusted to select optimal number of fuzzy rules, to acquire a fuzzy model with perfect generalization capability. Then, the parameter is fine-tuned by means of a hybrid gradient descent (GD) and least-squares estimation (LSE) approach. Finally, the method is used to model a PDU Naphtha’s Dry Point, simulation results show that it can determine the optimal model quickly and achieve satisfactory prediction precision.

Abstract: The bottleneck assignment (BA) and the generalized assignment (GA) problems and their exact solutions are explored in this paper. Firstly, a determinant elimination (DE) method is proposed based on the discussion of the time and space complexity of the enumeration method for both BA and GA problems. The optimization algorithm to the pre-assignment problem is then discussed and the adjusting and transformation to the cost matrix is adopted to reduce the computational complexity of the DE method. Finally, a synthesis method for both BA and GA problems is presented. The numerical experiments are carried out and the results indicate that the proposed method is feasible and of high efficiency.

Abstract: To improve the performance of Quantum-inspired Evolutionary algorithm based on P Systems (QEPS), this paper presents an improved QEPS with a Dynamic Membrane Structure (QEPS-DMS) to solve knapsack problems. QEPS-DMS combines quantum-inspired evolutionary algorithms (QIEAs) with a P system with a dynamic membrane structure. In QEPS-DMS, a QIEA is considered as a subalgorithm to put inside each elementary membrane of a one-level membrane structure, which is dynamically adjusted in the process of evolution by applying a criterion for measuring population diversity. The dynamic adjustment includes the processes of membrane dissolution and creation. Knapsack problems are applied to test the effectiveness of QEPS-DMS. Experimental results show that QEPS-DMS outperforms QEPS and three variants of QIEAs recently reported in the literature.

Abstract: Linear discriminant analysis (LDA) is a popular method among pattern recognition algorithms of machine olfaction. However, “Small Sample Size” (SSS) problem would occur while using LDA algorithm with traditional Fisher criterion if the within-class scatter matrix is singular. In this paper, maximum scatter difference (MSD) criterion and LDA were combined to solve SSS problem, so that three kinds of Chinese herbal medicines from different growing areas were accurately classified. At the same time, the classification result was enhanced. It works out that only a few samples of Anhui Atractylodes are classified incorrectly, however, the classification rate reaches 97.8%.

Abstract: Space index is one of the key technologies of spatial database, and also one of the biggest problems with puzzling GIS workers. So , how to build a more effective spatial index structure ,which has been the most realistic, most urgent, also to the forefront of research subject to GIS domain and graphics processing. This paper introduces several typical spatial index methods in the e GIS domain and graphics processing fields, and puts forward a binary tree index structure generated space grid step by step. The average time complexity of the index structure is the same with the quad tree index structure, but the maximum time complexity is reduced a third than the quadtree index structure, that is, the search efficiency improving a third.

Abstract: This paper, with the analysis of BP neural network learning and execution algorithm on single computer unit, a parallel neural network on many computer units is constructed, a system on programmable chip based on FPGA and uClinux is provided. Because it’s flexibility and reliability, this parallel neural network can be widely used where there is a large quantity of data to be processed. In addition to it, the system based on the SOPC has good versatility and easy to transplant because the reconfiguration of the hardware logic and software system.

Abstract: Application of machine vision method for MEMS dynamic parameters were measured, the testing image have a certain degree of ambiguity.This paper presents a sub-pixel algorithm based on fractal and wavelet transform: Firstly, using self-similar characteristics of fractal interpolation to overcome the problem ,that can not be accurate interpolation and the edge of the image reconstruction. Then because of abilities of high resolution and anti-noise,using wavelet transform modulus maxima,the image edge detection.The experimental results show that the algorithm can reach 0.02 pixel accuracy.