Applied Mechanics and Materials Vols. 148-149

Abstract: Inspired by Bat Algorithm, a novel algorithm, which is called Evolved Bat Algorithm (EBA), for solving the numerical optimization problem is proposed based on the framework of the original bat algorithm. By reanalyzing the behavior of bats and considering the general characteristics of whole species of bat, we redefine the corresponding operation to the bats’ behaviors. EBA is a new method in the branch of swarm intelligence for solving numerical optimization problems. In order to analyze the improvement on the accuracy of finding the near best solution and the reduction in the computational cost, three well-known and commonly used test functions in the field of swarm intelligence for testing the accuracy and the performance of the algorithm, are used in the experiments. The experimental results indicate that our proposed method improves at least 99.42% on the accuracy of finding the near best solution and reduces 6.07% in average, simultaneously, on the computational time than the original bat algorithm.

Abstract: Under the context of deicing in the airport, and in order to solve a kind of Non-identical Parallel Multi-machine Schedule Problem for minimizing the number of passengers delayed, an ant colony algorithm was proposed. Considering the characteristic of scheduling model and its constraints, an improved pheromone update strategy was developed by deploying different parameters. Simulation was made for proving the effectiveness of the algorithm, and the sensitivity analysis was conducted. Besides, the improved ant colony algorithm was compared further with FIFO algorithm used in realistic backgrounds. In all, the result indicated that the improving ant colony algorithm was valid, and it can fit for large-scale non-identical parallel Multi-machine scheduling problem.

Abstract: The performance degeneration of components not only reduces the economy of aircraft engine, but also serious threat to the flight safety. Neural network with learning and reasoning abilities is widely used in fault diagnosis. In this paper, a neural network is established with improved BP method. 9 measured parameters of an aircraft engine are selected as features parameters. 21 kinds of gas path failure modes are coded with 5 bits. The data preprocessing methods and the effect of measurement noise are also discussed.

Abstract: A miniature circularly polarized (CP) microstrip patch antenna with rectangular slots for GPS application is presented. The antenna structure is composed of a square radiating patch and two rectangular slots. The two rectangular slots are incorporated to obtain CP characteristic, with one slot on the lower left corner of the square patch and the other on the upper right corner. By adjusting the dimensions and positions of the slots, the proposed antenna can exhibit CP radiation. Moreover, miniature properties are achieved by the use of very high permittivity substrate material. Simulated results by Ansoft HFSS are given.

Abstract: Knowledge base of milling feature machining based on relational database is proposed, using knowledge representation of production rules, according to the characteristics of feature machining knowledge. Tool selected reasoning mechanism and reasoning processes is presented basing the installed CNC milling tool database , and further reasoning is based on knowledge, milling tool selection method is implemented based on feature machining knowledge using the forward direction inference strategy .In the paper, a complete system of the selecting milling cutter is established. The system connects not only the theoretical knowledge but the expert’s experiences with the computer applications in order to provide a base of realizing the automatic mechanical processing. By the example of machining tool selection for complex surface, the selection process is described, and the system can select the tools to meet the processing requirements within a shot time, and has good versatility.

Abstract: Although average or region energy method are widely used in calculating low-frequency coefficients in multi-sensor fusion, these methods still have shortcomings while fusing infrared and visible remote sensing imagery. Because of the differences of infrared and visible imaging mechanism, the extent of each source image’s contribution to the fused image is uncertain. Much information could be reduced and good result can not be obtained. Fuzzy logic is an efficient way to resolve uncertain problems. Therefore, fuzzy logic and nonsubsampled Contourlet transform (NSCT) are introduced into infrared and visible remote sensing image fusion and a novel multi-scale analysis method based on fuzzy logic is proposed in this paper. The spatial detail information of fused results is improved and the experiment indicates that the proposed method can get better visual quality.

Abstract: FGH95 nickel-based superalloy is produced by powder metallurgy (PM) processing for aerospace applications. Due to lower thermal conductivity, work hardening tendency during machining, and intensive adhesion to the surface of the tooling under operation, machining of FGH95 alloy is a significant challenges. The FGH95 machining process will induce substantial amount of plastic deformation in the surface and subsurface of the workpiece. A theoretical model is developed to predict the plastic deformation in machined surface of FGH95 superalloy. Experimental results are also applied to analyze the influence of cutting speed on plastic deformation in machined surface of FGH95. It is found that cutting speed has significantly effect on the plastic deformation in the machined surface. The increasing the cutting speed creates severer plastic deformation. Surface plastic shear strain increases with the increases of cutting speed, while the depth of plastic deformation decreases contrary.

Abstract: The impedance matching and the optimization of power from a circular piezoelectric energy harvester with a central-attached mass are studied. A finite element model is constructed to analyze the electrical equivalent impedance of the circular piezoelectric energy harvester. Furthermore, the complex conjugate matching load is used to extract the maximum output power of the energy harvester. The power output from complex conjugate matching load is compared with the power output from the resistive matching load and a constant resistance, separately. The results suggest that the complex conjugate matching can result in a significant increase of the output power for all frequencies. The effective bandwidth of the piezoelectric energy harvester is extended significantly.

Abstract: The vibration control problem of repeated eigenvalue systems with random parameters is discussed, which is approximated by a deterministic one. A method for calculating the standard deviations of repeated eigenvalues of the control systems with random parameters is presented by using the random perturbation. It will not require the distribution function of the random parameters of the systems other than their means and variances. The numerical results show that the present method is effective.

Abstract: In this paper, the parameter and optimal model of the sprung parts of unit aluminum plate curtain wall has been proposed, and its analysis software has also been programmed based on the APDL Language of ANSYS. The optimal dimensions of the sprung parts of unit aluminum plate curtain-wall have been obtained under the most dangerous working conditions by using our software. The goals of the automation of design, small size and light weight for the sprung parts are easily realized, and the protection of weight reduction, lower cost and less material reserve has also been gotten in the curtain wall project. It is a real Green Design.