Shen W.,Sichuan Engineering Technical College
Advanced Materials Research | Year: 2011
Data mining is a variety of analysis tools using the data found in the mass relationship between model and data technology, and relationships of these models can be used to predict. Organically with the combination of expert system, can solve the problem of automatic acquisition of knowledge. It from a large number, incomplete, noisy, fuzzy, random data mining the hidden decision-making has important reference value to generate the information. Data mining technology used in machinery fault diagnosis system, the scene of large quantities of raw data into valuable knowledge for specialists to explore the information they are interested in, but also describes the evolution trend of mechanical malfunction, the expert decision-making support information. And, after years of development, data mining technology has matured and widely used. The purpose of this study is that data mining technology used in machinery fault diagnosis and to study its feasibility. If applicable, must be able to open up a mechanical fault diagnosis, a new diagnostic method, which greatly promote the development of mechanical fault diagnosis technology. © (2011) Trans Tech Publications, Switzerland.
Yin J.,Sichuan Engineering Technical College
Journal of Convergence Information Technology | Year: 2012
PID controller is one of the earliest-developed control strategies, and genetic algorithm is an iterated self-fitness probability searching algorithm taking example by the natural selection of the living beings and natural genetic mechanism. The paper puts forward a kind of optimization design for PID controller parameters based on genetic algorithm. Genetic algorithm simulates the evolution steps of living beings, introduces the selection operator, crossover operator and mutation operator into the optimization procedures. The selection is to optimize the population by choosing the individuals of higher adaptability from the paternal population; the crossover is to randomly select some to cross to gain the new samples by randomly choosing a pair of individuals from the population, which can amplify the hunting space; the mutation is to avoid of the important genetic information missed by the selection and crossover, which can develop the searching efficiency and the global searching ability of GA by operating the individual according to its location. Compared with some conventional adjusting methods, GA is easier and has more accurate adjustment to confirm the optimization direction by the fitness function. The paper uses the genetic algorithm to optimize the PID parameter of the speed control system of the crude oil engine, and has a simulation research on the system. The experimental result shows that genetic algorithm is a simple and high effective optimization algorithm, and it obviously improves the dynamic performance of the system comparing with the traditional methods.
Shen W.,Sichuan Engineering Technical College
Advances in Intelligent Systems and Computing | Year: 2013
According to the teaching objectives of vocational CNC machining process design background, performance analysis to determine the problem and determine the design criteria of effective teaching objectives by clarifying teaching objectives, the goal of teaching is also constant feedback to refine and improve. © 2013 Springer-Verlag.
Han X.,Sichuan Engineering Technical College
Tezhong Zhuzao Ji Youse Hejin/Special Casting and Nonferrous Alloys | Year: 2015
Effects of back pressure on ECAE Ti alloy were investigated. The results show that with the increase of the back pressure, the deformation uniformity of the samples is improved as a result of decrease of outer-round angle, meanwhile, the loading is increased, which increases the requirement of mould and extruding equipment. The temperature field at the whole sample exhibits slightly change, and it becomes more uniform at the bottom compared with that with absence of back pressure. The maximum equivalent strain at the bottom sample is increased with the backpressure increases gradually, however, excessive back pressure can cause strain uneven distribution. The maximum principal stress is decreased with the pressure increase, which reduces effectively the crack tendency of specimen surface. ©, 2015, Journal Office of Special Casting and Nonferrous Alloys. All right reserved.
Chen J.S.,Tsinghua University |
Su G.D.,Tsinghua University |
Xiang S.B.,Sichuan Engineering Technical College
Science and Technology of Welding and Joining | Year: 2012
Extracting single pixel wide paths is the core image processing problem in many vision based welding seam tracking applications. Traditional techniques such as edge detection and image thinning may not be effective for solving this problem on low quality welding images polluted by process noises. Based on the understanding of the prior knowledge about the welding images, we present in this paper a welding seam tracking method using the image seam extraction technique. Owing to its optimisation nature, the proposed method is robust to strong image noises caused by environmental factors such as uneven illumination, light interferences and welding spatters. In addition, the proposed method is of low time complexity and is highly adaptive to different types of welding images. Experimental results on a variety of welding images validate the effectiveness and efficiency of the proposed method. © 2012 Institute of Materials, Minerals and Mining.