http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Prediction of Power Generation in China Using Process Neural Network
Jianghua Ge,Jiwei Wen,Chuntao Zhang,Yaping Wang,Gang Ding 보안공학연구지원센터 2015 International Journal of u- and e- Service, Scienc Vol.8 No.5
The power generation prediction problem can be seen as a time series prediction problem in nature. The traditional time series prediction methods based on regression analysis do not take account into the time accumulation effect existed in the time series due to their discrete input values. This limitation causes the low prediction accuracy of the time series prediction methods based on regression analysis. To solve this problem, a power generation time series prediction model based on the process neural network is proposed. The inputs of the proposed prediction model can be continuous time-varying functions. The time accumulation effect existed in the power generation time series can be expressed and computed by the integration operator of the process neural network. The proposed prediction model is trained and the efficiency of the proposed prediction model is tested by the month power generation data form January 2001 to April 2012, and the comparison experiment results indicate that the process neural network performs better than the auto regression analysis.
Research on the Error Detection Technology of Complex Parts based on Improved Bee Colony Algorithm
Jianghua Ge,Huihui Zhang,Yaping Wang,Xiulin Sui,Yongguo Sun 보안공학연구지원센터 2015 International Journal of Security and Its Applicat Vol.9 No.12
For Parts of complex structures with multiple characteristics, traditional methods are often difficult and inefficient, so the three-dimensional detection techniques have been developed. But the detection used for the complex structural components still get some problems, such as : it’s difficult to collect data of complex structural components in the data collection process, because of size is always limited ; features of parts retained difficultly in data processing; there are many iterations and calculations in the optimization algorithm of geometric error evaluation, and the search direction is blind. In this paper, the improved colony algorithm is used to get the cylindrical error by establishing the mathematical model with the point cloud data that has been processed. First, collect data acquisition of more structural parts with hand-held laser scanners; Secondly, denoise and streamline the raw point cloud collected before, and then make registration and alignment with the actual CAD model of part; Finally, according to the resultant point cloud data above, in accordance with the principle of minimum condition assessment, establish evaluation model of cylinder error based on improved bee colony algorithm. Experimental results show that for various structural parts, the precision of this detection method is high enough, and detecting speed is also fast.
The Numerical Simulation Analysis of Flow Field in Level Control Valve of Water Storage Tank
Yaping Wang,Lei Gao,Jianghua Ge,Han Gao,Jing Zhang 보안공학연구지원센터 2014 International Journal of Control and Automation Vol.7 No.10
Level control valve of water storage tank is installed in the outlet of tank, which control the water level of water storage tank by adjusting the water emissions, which make great significance for the safe operation of the super (super) criticality thermal power unit boiler . Complete the simulation analysis of internal flow field by the method of numerical simulation ,in order to achieve structure improvements of level control valve of water storage tank, that make level control valve of water storage tank better meet the work requirements. Establish flow channel model and mesh models of level control valve of water storage tank in various stages of the open process. Make three dimensional viscous numerical simulation of the internal complex flow field by the computational fluid dynamics software of ANSYS CFX. The error between the simulation values of the flow coefficient and the theoretical values , trial values is less than 5% ,which verifies the correctness of the simulation results . Get pressure contour , velocity streamline diagrams and outlet mass flow of level control valve of water storage tank in various stages of the open process. According to the pressure contours and velocity streamline diagrams, make detailed analysis of the flow field of level control valve of water storage tank. According to the dates of outlet mass flow, fit the flow characteristic curve of level control valve of water storage tank. Provide a theoretical basis for the structural design of level control valve of water storage tank.
Bing Dai,J. P Shao,Xuemei Wu,Guangbin Yu,Ye Song,Ge Jianghua 보안공학연구지원센터 2015 International Journal of u- and e- Service, Scienc Vol.8 No.10
Aiming at high nonlinear problem of special transmission gear invalidation analysis, provided a method which based on Meshless local Petrov-Galerkin method coupled with finite element method and method to solve fracture problem of special transmission gear. Simulation calculation has been done to non-involute beveloid gear developed by project team. Calculation result verified efficiency of the simulation method. The method has important meaning to novel gear development and research.