Applied Mechanics and Materials Vol. 422

Abstract: BP network model has become one of the important neural network model which is used in many fields, but it has some defects. From a mathematical perspective, it is a nonlinear optimization problem, which inevitably has the local minima problem; BP neural network learning algorithm has slow convergence rate, and the convergence speed and the initial weights of choice; network structure, namely the hidden layer nodes selection still has no theory, but according to the experience. Based on the BP algorithm local extreme values, considering the genetic algorithm, combining with BP algorithm, the BP neural network optimization is achieved. Neural network using genetic algorithm optimization mainly includes three aspects: the connection weights of evolution, evolutionary network structure, learning the rules of evolution.

Abstract: For joint robot system contains inevitable model error in the modeling process, an effective method is proposed for self-adaptive stability control in this paper. After building the robot dynamics model, error factors are analyzed in the model. Based on robust control theory, an improved self-adaptive PID controller is designed and its Lyapunov stability is verified. Finally, by simulation for a two-link manipulator, the result which shows the control method has well efficiency and practicality for robust stability control. The results will be significant for the precise control of the robot system.

Abstract: Through analyzing and comparing existing wireless communication protocols, this paper selects Bluetooth protocol suitable for CNC manufacturing and constructs communication system base on Bluetooth protocol.this system can solve some shortcomings of the wired network process, and make CNC machining faster, more flexible, more efficient.

Abstract: From view of design of automatic control system, wastewater treatment is a complex industrial process that is difficult to control. Certain improvements have been obtained on intelligent control, water quality prediction, fault diagnosis, operating instructions by artificial intelligence technology in field of wastewater treatment such as ANN control, fuzzy control, ES control, GA control, RS control, but there are still some shortcomings and problems especially on architecture of PCS. Distributed intelligent system based on multi-agent architecture is hopeful to meet synthetic system performance requirement of wastewater treatment plant in the future.

Abstract: Using MATLAB for development tools,we has developed a set of end mills CNC grinding programming system by analysis the end mills grinding craft ,research on cutter location trajectory calculation and post-processing technology. Meanwhile, build a five-axis tool grinder grinding simulation processing by VERICUT(a kind of CNC machining simulation software), through the edge grinding simulation , we confirm the cutter location trajectory and the NC code is errorless.

Abstract: In this paper, BOF steelmaking cost data from a steelmaking plant is analyzed, results show that production fluctuations greatly impact BOF cost, the cost of the best heat is much lower than average. Its presented that if every heat can be operated as the best heat, the average cost will reduced dramatically. The software pro-vides a method which is based on PLC network; and assisted with k-means clustering and feedback approach. After compiling software features, covering data acquisition, data storage, data filter and analysis, and data feedback control, the software is finished and implemented in production. With optimization of software, BOF smelting consumption declined and smelting fluctuation reduced significantly.

Abstract: Due importance has not yet been attached to the application of computer-aided translation (CAT) in the teaching activities in the institutes of higher learning in China. With the differentiation of machine translation and computer-aided translation, this paper tries to clarify the process involved in CAT and illustrate what is expected to be improved within the field of the research of computer-aided translation. It is pointed out in this paper that much effort is required to be made to combine the teaching of translation with computer-aided translation in a scientific way, with the training of computer-aided translation technology taking up its due position in the curriculum of the institutes of higher learning in an effort to facilitate the development of market-oriented and vocation-oriented talents of foreign language and translation.

Abstract: In this day and age, there is a disruptive shift in technology concerning the powerful ability of cloud storage that back up and store large data in the cloud, as well as syncing it across multiple devices automatically, and its drawbacks in data security. This paper mainly presents an overview of cloud storage, highlights many of security issues and vulnerabilities in cloud storage, lists various solutions, and then gives a summary.

Abstract: With the rapid development of the information and communication technology (ICT) and the considerable increasing of the network business, the energy consumption of the network equipments improves continually. Thus building the technology of the next generation electric power communication network based on energy efficiency has become the research focus of the current electric power communication field. This paper researches the problem of the QoS graded optimization in the next generation electric power communication network. First of all, we discuss the network model of the networks QoS graded optimization, and analyze the delay and the packet loss ratio in time-variant networks. Secondly, we introduce the mathematical description of the throughput capacity and the energy consumption in time-variant networks and build the optimizing model of the energy efficiency which describes the networks QoS classification through considering the QoS constraints such as the networks delay and packet loss ratio etc. Thirdly, we propose using Genetic Algorithm to solve the model and find the QoS classes which make the networks reach the maximum energy efficiency through iterative optimization. Finally, the simulation results indicate the method proposed in this paper is available.

Abstract: Mobile IPv6 is the key application in 3G network which is putting into commercial use. In this paper, we introduce the basic principle and key technology of Mobile IPv6 as well as its main implementing details and build a Mobile IPv6 test bed with Linux and MIPL. The mechanism of Mobile IPv6 is also validated through our experiments.