Applied Mechanics and Materials Vols. 333-335

Abstract: For the lack of crossover operation, from three aspects of crossover operation , systemically proposed one kind of improved Crossover operation of Genetic Algorithms, namely used a kind of new consistent Crossover Operator and determined which two individuals to be paired for crossover based on relevance index, which can enhance the algorithms global searching ability; Based on the concentrating degree of fitness, a kind of adaptive crossover probability can guarantee the population will not fall into a local optimal result. Simulation results show that: Compared with the traditional cross-adaptive genetic Algorithms and other adaptive genetic algorithm, the new algorithms convergence velocity and global searching ability are improved greatly, the average optimal results and the rate of converging to the optimal results are better.

Abstract: This paper proposes the PI Algorithm and also presents its principle and application. Compared with ordinary algorithms, it could compute multi-byte data more effectively. Experiments demonstrate that this algorithm is feasible and accurate.

Abstract: The evaluation for comprehensive performance of electric power listed companies generally exist some subjectivity and one-sidedness unavoidably. On the basis of the characteristics of complexity and the uncertainty of optimal evaluation on comprehensive performance of electric power listed companies, grey relational analysis model in evaluation on comprehensive performance of electric power listed companies has been put forward in this paper. Application model and procedures of the grey relational analysis method are introduced. According to the degree how close it is with the ideal dot, the optimal evaluation result is easy to be obtained. This method is convenient, quantitative, and strict. It provided a new method and a possible new way for evaluation of comprehensive performance of electric power listed companies.

Abstract: this paper proposes a hierarchical division method that divides a data set into two subsets along each dimension, and merges them into a division of the data set. Then the initial cluster centers are located in dense and separate subsets of the data set, and the means of data point in these subsets are selected as the initial cluster centers. Thus a new cluster center initialization method is developed. Experiments on real data sets show that the proposed cluster center initialization method is desirable.

Abstract: A class of the balanced orthogonal multiwavelets was constructed by defining a specific matrix filter structure, in which the multifilter banks of multiwavelets have had the desired symmetry. The multifilter banks have possessed symmetric/antisymmetric, which resembled the filters of scalar wavelet and have in favor of application, notwithstanding the multiwavelets constructed in this paper lose the linear phase, they have formed a new type of multiwavelets undoubtedly.

Abstract: With the rapid development of air transportation industry, many air traffic controllers are facing overload, which brings potential danger to the flight safety. In order to reduce the workload of air traffic controllers, this paper studies sector optimum partition of terminal airspace. Firstly, airspace topology structure is built to describe the relation of key way points, legs and air routes. Power polygon divides the terminal airspace into several units, the workload of each way point can be calculated. Then the sector optimum partition mathematical model is built. A kind of permutation and combination algorithm is used to recombine the units, the objective function value of each unit is regarded as optimization function, and the optimal solution can be worked out in combination with the Particle Swarm Optimization. At last, an example of Beijing Terminal Airspace sector optimum is given, which shows Particle Swarm Optimization is an effective method to be applied into sector optimum partition based on way points workload.

Abstract: According to fire protection safety system, such as students' dormitories, put forward a fire prediction method based on SVM to deduce the possibility and risk level, in which peoples behavior state is combined with static fire factors .Then, the simulation results are shown the method is effective.

Abstract: Aiming at the complexity of campus fire, the factors in connection with fire is classified as static and dynamic factors. Firstly, points out that the dynamic factors play an important role in the fire. Then, taking the student dormitory area as the research object, and considering the factors of fire comprehensively, a fire prediction method is proposed based on generalized RBF neural network, Finally, by the simulation, the results are shown the proposed method is effective

Abstract: Through compares three methods of Standard Deviation mathematical expectation and variance, a classification algorithm based on the Standard Deviation which in the training set is proposed in this paper. The algorithm first mapped the discrete attribute values to the corresponding values, and calculates Standard Deviation, mathematical expectation and Variance of each attribute in each category. The Standard Deviation, mathematical expectation and Variance of each attribute in each category used as coordinates. When there are new datas need to determine the category, we just need to use the attributes of the new data as coordinates, and calculate its distance to each category, and then the data type is the shortest distance category. Comparison of three methods, the Standard Deviation is the most stable and most accurate. This algorithm has advantages in dealing with the noisy date.

Abstract: In order to improve the precision of oilfield single well production prediction, a single well production prediction model based on improved extreme learning machine (RWELM) is proposed. Substituting wavelet function for common activation function, structural risk minimization principle is integrated into the model in order to avoid the local minimum and over-fitting problem commonly faced by traditional extreme learning machine (ELM) in single well production forecasting. Dynamic data of an oil well production is simulated of Lun Nan oilfield. Experimental results show that the forecasting model is better than ELM, LM-BP neural networks, BP network with delay time sequence in both generalization performance and predictive accuracy.