Papers by Keyword: Multi-Variable

Abstract: The development of turbine blade root style is one of the key problems in the blade design technology as the root carries most of the loads of the whole blade. The optimal design problem and the corresponding numerical method were established for a straddle root structure with the minimum equivalent stress of the root and rim as the optimal objective. A multi-variable parametric model of the blade and rim, which took eight critical geometrical variables of the root and rim as design variables, was built by APDL (ANSYS parametric design language) and the optimal problem was numerically solved by combining pattern search algorithm with finite element method. The results indicate that the optimized structure has better strength performance, whose maximum equivalent stress of the root sharply decreased by 25.18% comparing with the original design. It eventually confirms the feasibility and validity of the proposed optimal design method.

Abstract: The paper analyzes the characteristics and the situation of the control of multivariable coupling system, combining improved PID neural networks with PSO algorithm, and finally designs a suitable controller model. To achieve a controller model, an initial structure of PID neural networks is established in the first place. Then weights in the network is initialized by PSO algorithm and optimized by improved gradient descent algorithm. A simulation, MIMO coupling system, prove this controller has such characteristics with short time and high precision. The research of the paper provides a new idea and approach for control of complex coupling system.

Abstract: A multivariable predictive functional control (M-PFC) algorithm based on a two-inputs/two-outputs system with the transfer function model is presented in this paper. A simple and explicit solution of manipulated variables of the control system can be obtained by optimizing the objective function. Simulations of the system applying M-PFC are also provided in here, showing that the presented algorithm has good performance of tracking set-point without steady-state error, disturbance rejection and robustness. Finally, the application of temperature control for the methylamine synthesizing tower is addressed, demonstrating the effectiveness of the proposed PFC algorithm.

Abstract: In this paper, a multivariable analysis of grey prediction model MGM (1,n) is proposed ,which is based on single-variable gray prediction model GM (1,1). For multiple variables of a system, they influence each other and interrelate, MGM (1,n) model is description of each of the major relevant variables in the system from system point of view.Through comparison of an example of forecasting relay failure, can be obtained: multivariable grey prediction is more accurate and more close to the actual value than the single variable prediction.

Abstract: This paper presents a general numerical model to solve non-linear inverse heat conduction problems with multi-variables which include thermal parameters and boundary conditions, and can be identified singly or simultaneously. The direct problems are numerically modeled via FEM, facilitating to sensitivity analysis that is required in solving inverse problems via a least-square based CGM (Conjugate Gradient Method). Inhomogeneous distribution of parameters is considered, and a number of numerical examples are given to illustrate the work proposed.

Abstract: At present, with the popularization of higher education and acceleration of college students every year, colleges and universities are increasing scale. Expansion of the scale for colleges and universities on the one hand, broadens the scope for development, on the other hand it also brings many problems, including the issue of the quality of teaching which is particularly prominent. Owing to the problems existing in the previous system of teaching quality, based on the teaching characteristics, a new multivariate teaching quality evaluation model is set up by means of GM(1,N) theory. And the application processes of the model are illuminated in detail. The model is applied into the evaluation of teaching quality. By analyzing a lot of practical examples, the experiment result indicates that this mathematical model has better appraisal effect than other appraisal model, which can overcome the complexity of traditional evaluation model. Compared with other methods, this method is scientific, simple and operable. And its structure and method will have a bright future.

Abstract: The present paper studies the process of electro discharge machining, estimating the relative importance of main influence factors. The first step is to evaluate the possibilities of a design of experiments based on an orthogonal array, an appropriated method to systems with multiple factors changing in numerous levels After of the realization of the experiences the theory of grey systems is used to know which combination of factors optimize all of the variables. Finally the analysis of the variance is applied to investigate which of the factors significantly affects the global answer of the system.