فصلنامه مدل سازی در مهندسی
پیاپی 28 (بهار 1391)

In this paper, Group Method of Data Handling (GMDH)-type neural networks based on Genetic Algorithm, as an effective means to model the complex and nonlinear system, used for the modeling of turbofan engine. Fuel consumption is an important factor in turbofan engines, particularly in commercial and passenger aircraft engines. Regarding to its importance, fuel/air ratio influence on two main parameters, namely, specific fuel consumption and thrust investigated in this study at different altitudes for lower Mach numbers. Therefore, GMDH-type neural networks based on Genetic Algorithm developed for modeling. The impact of fuel / air ratio and flight altitude on thrust and specific fuel consumption of a turbofan engine that mathematical functions obtained for thrust and specific fuel consumption will be applied to optimization of the engine and estimation of best operating points. Finally, the GMDH model`s value compared with experimental data that show the effectiveness and accuracy of the network.

The rotary regenerators are the type of compact heat exchangers that thermal energy of hot gas is absorbed by the matrix and then released to a cold gas stream. Mathematical analysis of this type of heat exchangers and the effect of various parameters has a considerable effect in improving their performance. In this paper, the mathematical model for fluid and thermal analysis of matrix is used. In the model, the effects of heat conduction in the direction of the fluid flow, rotating speed of the wheel and time are considered. The model is solved in the visual basic program. The effect of various parameters such as rotating speed of the wheel, the fluid flow of hot and cold stream and mass of matrix on dynamic responses are analyzed. Results showed that with increase of length and rotating speed of the wheel, the yield increased. With the increase of diameter, the yield increased initially, reached to a maximum and then decreased. The numerical results are compared with experimental measurements. Results showed that predicted results were matching satisfactorily with the experimental results.

Wind tunnel، as an experimental laboratory، has many applications nowadays. Installation of wind tunnel is very expensive، so designers try to reduce wind tunnel’s installation and test costs by optimizing the design of various components of the wind tunnel. In this paper، costs are reduced and efficiency is increased due to a new design and optimization of a wind tunnel diffuser. In many supersonic wind tunnels، the convergent parts of the diffuser have walls with fixed slope. So، a new model of the supersonic wind tunnel diffuser is presented. This model of diffuser includes a flexible wall and three jacks in its direction. Then the two-dimensional and compressible flow was analyzed using k-w SST turbulence model and AUSM+ method in a supersonic wind tunnel. Finally، in Mach 4 and by the use of genetic algorithm، a new model of supersonic wind tunnel diffuser was optimized with the lowest overall pressure drop. For practical investigation the sample of supersonic wind tunnel diffuser of Ghadr Aerodynamic Research Center has been used. The results indicate that the new optimized wind tunnel has higher efficiency with 83 percent in comparison to Ghadr Center’s sample of supersonic wind tunnel.

To analyze the heat transfer and fluid flow of mixed convection through nanofluid in an inclined square cavity having two hot obstacles, a numerical method based on the finite volume approach is applied. The effect of the Rayleigh number, solid volume fraction of nanoparticle, height of heated obstacle and cavity inclination angles are studied and obtained results are presented in the form of streamline and isotherm plots and Nusselt diagram. In this investigation Al2O3-water at T=300 and diameter of nanoparticle equal to 40, have been used as the working fluid. The study has been executed for the Rayleigh number between 104 and 105, solid volume fraction of nanoparticles altered from 0 to 0.06 while the height of the obstacle changed from 0.1 to 0.2 length of the cavity. Results show that increasing solid volume fraction and Ra number lead to heat transfer increases. Also with increasing of height of the obstacle, the average Nusselt number decreases.

In this paper, forward extrusion process of Bimetal sheets through taper dies is analyzed by upper bound technique and simulated by FEM. Deformation region is divided into two deformation zones and based on a proposed velocity field for each deformation zone, total internal, shear and friction powers are calculated. The extrusion force is estimated by equating the external power with the total power. Analytical results are compared with the results given by finite element method (ABAQUS Software). These comparisons show a good agreement. Finally the effect of the process parameters including reduction in the area and the friction factor on the extrusion force are estimated.

This paper presents a new approach based on Benders decomposition (BD) to solve short-term hydrothermal coordination problem with AC power flow, voltage stability and security constraints. The proposed method decomposes the problem into a master problem and a sub-problem. The master problem applies the Mixed Integer Programming (MIP) method and finds the on/off states of the thermal and hydro units and the amount of power produced by each hydro unit. The sub-problem apply the nonlinear Programming (NLP) method and find the amount of power produced by each thermal unit while modeling the transmission network using AC power flow considering security and voltage stability constraints. Index evaluates voltage stability of the power system. Due to AC power flow and security constraints, the sub-problem may become infeasible. To cope with the infeasibility problem, positive penalty terms are added to the sub-problem formulation.The methodology is tested on the 9-bus and IEEE 118-bus test systems. The obtained results confirm the validity of the developed approach.

In this article, a reverse engineering procedure has been presented to achieve an upgraded design applicable to legacy mechanical and structural components of aeronautical systems. In this way, a methodology comprising re-engineering and rapid prototyping methods has been proposed. Assuming no available useful original design data, this procedure focuses on re-producing engineering design data to develop an engineering platform for innovative design. In this procedure, first cloud point data are extracted by laser scanning to create 3D solid parametric modeling. Then, reasonable changes are applied to the original design to achieve a new design compatible with requirements and available rapid prototyping techniques and material resources. A special attention has been paid to stress and fracture analysis to ensure reliability and damage tolerance capability of the new design. The method is benchmarked using an actual aircraft component.