A Computational Design System with Cognitive Features Based on Multi-objective Evolutionary Search with Fuzzy Information Processing

Author(s):  
Michael S. Bittermann
Author(s):  
Kikuo Fujita ◽  
Shinsuke Akagi

Abstract A Framework of computational design method and model is proposed for layout and geometry design of complicated mechanical systems, which is named “configuration network and its viewing control”. In the method, a design object is represented with a set of declarative relationships among various elements of a system, that is, configurations, which is gradually extended from schematic structure to exact layout and geometry through design process. Since a whole of such configurations forms a too complicated network to compute all together, how to view subparts is controlled based on levels of granularity and width of scope range. Such a configuration network is made to grow and refined through embodying geometry and layout corresponding to a focused subpart with a numerical optimization procedure. The framework has also an ability to flexibly integrate with engineering analysis. Moreover, a design system is implemented with an object-oriented programming technique, and it is applied to a design problem of air conditioner units in order to show the validity and effectiveness of the framework.


Author(s):  
Alireza Fathi ◽  
Abdollah Shadaram ◽  
Mohammad Alizadeh

This paper introduces a framework to perform a multi-objective multipoint aerodynamic optimization for an axial compressor blade. This framework considers through-flow design requirements and mechanical and manufacturing constraints. Typically, components of a blade design system include geometry generation tools, optimization algorithms, flow solvers, and objective functions. In particular, optimization algorithms and objective functions are tuned to reduce blade design calculation cost and to match designed blade performance to the through flow design criteria and mechanical and manufacturing constrains. In the present study, geometry parameters of blade are classified to three categories. For each category, a distinct optimization loop is applied. In outer loop, Gradient-based optimization techniques are used to optimize parameters of the second category and a two-dimensional compressible viscous flow code is used to simulate the cascade fluid flow. Surface curvature optimization is carried out in inner loop, and its objective function is defined by integrating the normalized curvature and curvature slope. The genetic algorithm is used to optimize the parameters in the interior loop. To highlight the capabilities of the design method and to develop design know-how, an initial profile is optimized with three different design philosophies. The highest performance improvement in the first case is 15% reduction in loss at design incidence angle. In the second case, 16.5% increase in allowable incidence angle range, improves blade’s performance at off design conditions.


Author(s):  
Z.-Q. Ye

This paper describes a systematic computational design system for two-dimensional turbine cascades. The system includes a sequence of calculations in which airfoil profiles are designed from velocity diagram requirements and specified geometric parameters, followed by an inviscid global streamline curvature analysis, a magnified reanalysis around the leading edge, and a transitional profile boundary layer and wake mixing analysis. A finite area technique and a body-fitted mesh are used for the reanalysis. The boundary layer analysis is performed using the dissipation-integral method of Walz which has been modified in the present application. Several turbine airfoil profile geometry designs are presented. Also two sample cascade design cases and their calculated performance for a range of Mach numbers and incidence angles are given and discussed.


Author(s):  
Keiji Tajiri ◽  
Jinhui Zhao ◽  
William C. Hohlweg ◽  
Haijie Zhang

Automatic optimization techniques have been used in recent years to facilitate more rapid analyses of different design options with multiple performance objectives. Typically, this process has been used during new product development. In this paper, a design system is presented, which enables the multipoint, multi-objective optimization of the centrifugal compressor stage aerodynamic components. Moreover, it is applied to a design modification of a multistage compressor, during the manufacturing cycle, for risk mitigation. The system is based on the application of the Isight code for coupling of one dimensional direct design and analysis with multi-objective genetic algorithms, design of experiment, and response surface method. The design system was applied to a redesign of the diffuser, crossover, and return channel of two stages in a multistage compressor. The geometry parameterization is performed by a one dimensional analysis method where the diffuser width, crossover inlet and exit width and associated inner and outer radii, are used to describe the meridional flow path while holding the return vane geometry unchanged. Centrifugal compressor performance parameters, such as polytropic head and efficiency at the client rated point, head rise to surge, and choke flow capacity are evaluated during the optimization process. The example confirmed the validity of the system to perform the optimization of turbomachine components in a time efficient manner to meet production schedule. The system also allowed for a sensitivity analysis of the impact of geometry parameters on the aerodynamic performance, contributing to the development of guidelines for manufacturers to design new products and mitigate the performance risk on test floor.


2020 ◽  
Vol 10 (3) ◽  
pp. 22
Author(s):  
Andy D. Pimentel

As modern embedded systems are becoming more and more ubiquitous and interconnected, they attract a world-wide attention of attackers and the security aspect is more important than ever during the design of those systems. Moreover, given the ever-increasing complexity of the applications that run on these systems, it becomes increasingly difficult to meet all security criteria. While extra-functional design objectives such as performance and power/energy consumption are typically taken into account already during the very early stages of embedded systems design, system security is still mostly considered as an afterthought. That is, security is usually not regarded in the process of (early) design-space exploration of embedded systems, which is the critical process of multi-objective optimization that aims at optimizing the extra-functional behavior of a design. This position paper argues for the development of techniques for quantifying the ’degree of secureness’ of embedded system design instances such that these can be incorporated in a multi-objective optimization process. Such technology would allow for the optimization of security aspects of embedded systems during the earliest design phases as well as for studying the trade-offs between security and the other design objectives such as performance, power consumption and cost.


Sign in / Sign up

Export Citation Format

Share Document