Applied Mechanics and Materials Vols. 321-324

Abstract: As all the world looks ahead to the next generation of human space exploration missions (the USA and Russia even plan to develop new program which will take humans back to the Moon by 2020, to Mars, and beyond), the lunar exploration has been one of the hottest field in the world. And some scientists presented lunar impactor missions. This paper proposes an solution for trajectory optimization for lunar impactor with multi constraints by the method of Gauss Psedospectural, and simulation results shows this method is feasible to improve the performance of lunar impactor.

Abstract: Considering discrete demand and time-vary unit production cost under a foreseeable time horizon, this study presents an adaptive genetic algorithm to determine the production policy for one manufacturer supplying single item to multiple warehouses in a supply chain environment. Based on Distribution Requirement Planning (DRP) and Just in Time (JIT) delivery policy, we assume each gene in chromosome represents a period. Standard GA operators are used to generate new populations. These populations are evaluated by a fitness function using the total cost of production scheme. An explicit procedure for obtaining the local optimal solution is provided.

Abstract: A chaos concise differential evolution algorithm (CcDE) is proposed for the embedded controller with limited memory, which introduces chaotic local search based on basic differential evolution algorithm to increase exploring and prevent premature convergence. Using virtual population and Gaussian sampling, the CcDE becomes simple and reduces the memory requirements at run time. Experimental simulation on optimizing parameters of the recurrent fuzzy neural network shows that the proposed CcDE can obtain better performance than other concise algorithm.

Abstract: In this paper the rough set fuzzy neural network is used to monitor the over current fault problem of the mining scraper conveyor motor driving system. The phase current signals are input into the neural network, and then the current signals are processed with fuzzy logic set theory for optimization. Because too many rules may lead to complex computation, the rough set theory is used to reduce the rules after the signal characteristics are extracted. The simulation results show that the precision and reliability of motor driving system of the mining scraper conveyor can be improved by this method.

Abstract: Ship field repair sheduling is a asynchronism and concurrency process. Although it is complex, it is very important in the field repair decision at wartime. Through the optimal field repair scheduling, we can reduce the total repair time and improve repair efficiency. Regretly ship field repair scheduling optimization model has not been researched perfectly. Generally speaking, there are field repair processes of two kinds. Repair processe of one kind is that the repair sequence is fixed and the repair resource need not be scheduled. It is used to schedule repair tasks of the same kind. Repair processe of the other kind is that the repair resoure should be scheduled to reduce the total repair time. It is used to schedule the repair tasks of the different kind. The scheduling method of the first kind has been researched in math work. But the math model turns to be complex when the amount of the object increases. The scheduling method of the second kind is more complex than the first one. It is a typical asynchronism and concurrency process. So the new feasible models should be researched. Based on the repair processes of the two kinds, two ship field repair scheduling models are founded based on timed colored Petri net (TCPN) and depth first search method in this paper. The timed colored Petri net is used to model repair process and its transition rules are used to design the arithmetic. The arithmetics of the two models are explained in detail. In the second arithmetic, the conflict-solution is researched particularly. Conflict-solution method is corresponding to the field repair resource scheduling plan. The conflict-solution problem is solved based on depth first search method of artifical intelligence. Through the examples of the use of the two arithmetics, we find that the arithmetics given in this paper are very feasible and convenient. The field repair scheduling optimization models are the basic of the ship field repair and safety recovery intelligent decision support system. Through the optimization models we can generate the repair scheduling plan intelligently. That is to say the two models are not only optimization models but also intelligent models.

Abstract: The theory and algorithm of BP neural network were introduced, and it was trained by the theoretical amounts of corresponding angle of Cam, then the curve-fitting was obtained. The result proved that the fast computation of the theoretical amounts of any angle in Cam carve-fitting could be achieved by means of BP neural network.

Abstract: Domain reduce methods purpose is to reduce the times of simulation and optimization by decreasing search domain so as to realize the least time-consuming. MLIS (move limit indicator strategy) is one of the common effective methods. After analyzing the drawbacks of the existing optimization problem, propose an MLIS using six indicators with domain reduce method. And next, functional quality, search space, move history, assessment criterion, convergence criterion and constraint conditions of the six indicators are introduced detailedly. Finally, Hesse function example and a design problem with cantilever beam are optimized and simuliated by MLIS. The practices proved that this method has a certain application value and practical meaning in domain reducing.

Abstract: The nearest neighbor search in high-dimensional space is an important operation in many applications, such as data mining and multimedia databases. Evaluating similarity in high-dimensional space requires high computational cost; index-structures are frequently used for reducing computational cost. Most of these index-structures are built by partitioning the data set. However, the partitioning approaches potentially have the problem of failing to find the nearest neighbor that is caused by partitions. In this paper, we propose the Error Minimizing Partitioning (EMP) method with a novel tree structure that minimizes the failures of finding the nearest neighbors. EMP divides the data into subsets with considering the distribution of data sets. For partitioning a data set, the proposed method finds the line that minimizes the summation of distance to data points. The method then finds the median of the data set. Finally, our proposed method determines the partitioning hyper-plane that passes the median and is perpendicular to the line. We also make a comparative study between existing methods and the proposed method to verify the effectiveness of our method.

Abstract: Through research on initial alignment problem of fiber optic gyro SINS in the sway condition, this paper proposed a rapid compass alignment scheme with variable parameters, which can accomplish rapid initial alignment. Firstly, analysted SINS compass alignment principle, and gave a concrete realization method which has the same calculation procedure with full damping navigation algorithm. This method makes the alignment and navigation to use the same set of algorithms, and can effectively reduce algorithm complexity. Simulation and repeatedly sway test results show that the alignment algorithm is effective. The alignment precision and instrument accuracy is consistent, it can meet the requirements of the initial alignment.

Abstract: A method for compensating nonlinear characteristic of thermocouple vacuum gauge is proposed. Least squares support vector machine (LS-SVM) is adopt as compensation model, of which parameters are optimized using particle swarm optimization (PSO) algorithm. Experimental results using data obtained on-site show that the proposed approach effectively compensates the nonlinearity characteristic, and the accuracy of this method is higher than those obtained by SVM model.