Applied Mechanics and Materials Vols. 333-335

Abstract: Mastermind is code-breaking board game whose principle is to find as fast as possible the guess pattern of code pegs in color and position based on feedback realized by key pegs. The game is targeted at development of logical thinking. The process of the decoding can be realized also using computer. Mathematical analysis, strategy solution, algorithm development of solution and computer simulation program in MS Excel of decoding of the color regardless of the position ordering are shown in the paper.

Abstract: Mastermind is code-breaking board game whose principle is to find as fast as possible the guess pattern of code pegs in color and position based on feedback realized by key pegs. The game is targeted at development of logical thinking. The process of the decoding can be realized also using computer. Mathematical analysis, strategy solution, algorithm development of solution and computer simulation program in MS Excel of decoding of the position order of already found and disordered colored pegs are shown in the paper.

Abstract: In this paper, author presents a new computational model where both the results of emotional interactions from the intrinsic emotion and recognition and the extrinsic environmental stimulation (another agent), the two parts play an important role in everyday life. We take a kind of six basic emotion states (happiness, surprise, anger, fear, sadness, disgust), and updates its state depending on its current emotional and cognitive state and meeting with another one in random environment. At last, we also design some experiments to verify the effects of affective cognitive algorithm. Those experimental results are accordance with the emotion principle of human being.

Abstract: In order to overcome the difficult of large amount of calculation and to satisfy multiple design indicators in the design of control laws, an improved multi-objective particle swarm optimization (PSO) algorithm was used to design control laws of aircraft. Firstly, the hybrid concepts of genetic algorithm were introduced to particle swarm optimization (PSO) algorithm to improve the algorithm. Then based on aircraft flying quality the reference models were built, and then the tracking error, settling time and overshoot were used as the optimization goal of the control laws design. Based on this multi-objective optimize problem the attitude hold control laws were designed. The simulation results show the effectiveness of the algorithm.

Abstract: The traditional particle PHD filter only uses kinematic characters and parameters to measure,However,when the kinematic characters of targets are similar with each other (in distance,location,speed,acceleration,etc.) the weighted measurement will become more and more similar,so the tracks of targets will be closer and closer to each other.Eventually the tracks will mingle with each other and can not be distinguished,thus causing misjudgment and failing in achieving accurate tarcking.To solve such kind of target tracking problem,this paper proposes the particle PHD filter tracking algorithm based on multi-parameter assistance.That is,the target property information and parameters will be imported into the algorithm and the measurement difference is increased by calculating the combined likelihood value of the target property parameter and the kinetic character parameter,and then the combined likelihood value is used to update the particle PHD to filter miscellaneous waves and update multi-target State collection.The simulative analysis will enable the proposed algorithm to realize the short-distance multi-traget tracking and filtering in complex environment.

Abstract: Inspired by thermodynamics molecular system, the particle swarm optimization algorithm based on the molecular force model (MFMPSO) was proposed. Two parameters were introduced in the MFMPSO algorithm. In this paper, the orthogonal test design method is applied to optimize the parameter combinations of three levels and four factors, which include d_l and d_h, the population size and the iteration number. The experimental results prove that the population size and the iteration number have litter influence on the MFMPSO algorithm, however d_l and d_h play a key role and thus the MFMPSO algorithm has good search performance when d_l and d_h take the values respectively in a certain range which is related with the length of the longest diagonal in the search space.

Abstract: According to the defect of falling into a local optimum when dealing with multimodal problems with basic particle swarm optimization, a dynamic neighborhood particle swarm optimization with external archive (EA-DPSO) is proposed. The Ring topology, All topology and Von Neumann topology are adopted, and dynamically refining particle history optimal position, and then store them on the external archive. In terms of particles characteristics in the external archive, a kind of effective extract mechanism method is designed to choose learning sample. Three peak problems as simulation function are chosen and the results show that EA-DPSO can effectively jump out of local optimal solution. Therefore, it can be seen as an effective algorithm for solving multimodal problems.

Abstract: In dynamic environments, it is difficult to track a changing optimal solution over time. Over the years, many approaches have been proposed to solve the problem with genetic algorithms. In this paper a new space-based immigrant scheme for genetic algorithms is proposed to solve dynamic optimization problems. In this scheme, the search space is divided into two subspaces using the elite of the previous generation and the range of variables. Then the immigrants are generated from both the subspaces and inserted into current population. The main idea of the approach is to increase the diversity more evenly and dispersed. Finally an experimental study on dynamic sphere function was carried out to compare the performance of several genetic algorithms. The experimental results show that the proposed algorithm is effective for the function with moving optimum and can adapt the dynamic environments rapidly.

Abstract: Target localization technology has been intensively studied and broadly applied in many fields. This paper presents one improved particle swarm optimization technique in training a back-propagation neural network for position estimation in target localization. The proposed scheme combines particle swarm optimization (PSO), back-propagation neural network (BP), adaptive inertia weight and hybrid mutation, called IPSO-BP. To verify the proposed IPSO-BP approach, comparisons between the PSO-based BP approach (PSO-BP) and general back-propagation neural network (BP) are made. The computational results show that the proposed IPSO-BP approach exhibits much better performance in the training process and better prediction ability in the validation process than those using the other two base line approaches.

Abstract: This paper proposes a combined local best particle swarm optimization algorithm (CLBPSO) which combined with local optimum particle information. And it gives three ways of combination local information. Experimental results indicate that the CLBPSO algorithm improves the search performance on the benchmark functions significantly. On the basis of experimental results, we will also compare these three methods with each other to find the best one.