• Journal of Computing Science and Engineering
• /
• 제6권4호
• /
• pp.294-309
• /
• 2012

In the age of multi-core and specialized accelerators in high performance computing (HPC) systems, it is critical to understand application characteristics and apply suitable optimizations in order to fully utilize advanced computing system. Often time, the process involves multiple stages of application performance diagnosis and a trial-and-error type of approach for optimization. In this study, a general guideline of performance optimization has been demonstrated with two class-representing applications. The main focuses are on node-level optimization and inter-node scalability improvement. While the number of optimization case studies is somewhat limited in this paper, the result provides insights into the systematic approach in HPC applications performance engineering.

• International Journal of Aeronautical and Space Sciences
• /
• 제18권4호
• /
• pp.780-787
• /
• 2017

The propellant grain configuration is a design variable that determines the shape and performance of a solid rocket motor. Grain configuration variables have complicated effects on the motor performance; so the global optimization problem has to be solved in order to design the configuration variables. The grain performance has been analyzed by means of the grain burn-back and internal ballistic analysis, and the optimization technique searches for the configuration variables that satisfy the requirements. The deterministic and stochastic optimization techniques have been applied for the grain optimization, but the results are imperfect. In this study, the optimization design of the configuration variables has been performed using the hybrid optimization technique, which combines those two techniques. As a result, the hybrid optimization technique has proved to be efficient for the grain optimization design.

• IEIE Transactions on Smart Processing and Computing
• /
• 제4권6호
• /
• pp.450-454
• /
• 2015

Short execution time is the major performance factor for computer systems. This performance factor is directly determined by code quality, which is influenced by the compiler's optimizations. However, a compiler has limitations when optimizing source code due to insufficient information. Thus, if programmers can learn the reasons why a compiler fails to apply optimizations, they can rewrite code that is more easily understood by the compiler, and thus improve performance. In this paper, we propose a compiler that provides a programmer with reasons for failed optimization and recognizes programmer's additional information to obtain better optimization. As a result, we obtain performance improvement, i.e., reducing execution time and code size, by taking advantage of additional optimization opportunities.

• Nuclear Engineering and Technology
• /
• 제42권4호
• /
• pp.460-467
• /
• 2010

In this study, the parameters of the feedwater control system (FWCS) of the OPR1000 type nuclear power plant (NPP) are optimized by response surface methodology (RSM) in order to acquire better level control performance from the FWCS. The objective of the optimization is to minimize the steam generator (SG) water level deviation from the reference level during transients. The objective functions for this optimization are relationships between the SG level deviation and the parameters of the FWCS. However, in this case of FWCS parameter optimization, the objective functions are not available in the form of analytic equations and the responses (the SG level at plant transients) to inputs (FWCS parameters) can be evaluated by computer simulations only. Classical optimization methods cannot be used because the objective function value cannot be calculated directely. Therefore, the simulation optimization methodology is used and the RSM is adopted as the simulation optimization algorithm. Objective functions are evaluated with several typical transients in NPPs using a system simulation computer code that has been utilized for the system performance analysis of actual NPPs. The results show that the optimized parameters have better SG level control performance. The degree of the SG level deviation from the reference level during transients is minimized and consequently the control performance of the FWCS is remarkably improved.

• Nuclear Engineering and Technology
• /
• 제55권6호
• /
• pp.2125-2138
• /
• 2023

The printed circuit heat exchanger (PCHE) with airfoil fins has the benefits of high compactness, high efficiency and superior heat transfer performance. A novel multi-objective optimization approach is presented to design the airfoil fin PCHE in this paper. Three optimization design variables (the vertical number, the horizontal number and the staggered number) are obtained by means of dimensionless airfoil fin arrangement parameters. And the optimization objective is to maximize the Nusselt number (Nu) and minimize the Fanning friction factor (f). Firstly, in order to investigate the impact of design variables on the thermal-hydraulic performance, a parametric study via the design of experiments is proposed. Subsequently, the relationships between three optimization design variables and two objective functions (Nu and f) are characterized by an improved particle swarm optimization-backpropagation artificial neural network. Finally, a multi-objective optimization is used to construct the Pareto optimal front, in which the non-dominated sorting genetic algorithm II is used. The comprehensive performance is found to be the best when the airfoil fins are completely staggered arrangement. And the best compromise solution based on the TOPSIS method is identified as the optimal solution, which can achieve the requirement of high heat transfer performance and low flow resistance.

• 한국CDE학회논문집
• /
• 제16권6호
• /
• pp.397-406
• /
• 2011

Since tolerance allocation in a mobile phone camera manufacturing process greatly affects production cost and reliability of optical performance, a systematic design methodology for allocating optimal tolerances is required. In this study, we proposed the tolerance optimization procedure for determining tolerances that minimize production cost while satisfying the reliability constraints on important optical performance indices. We employed Latin hypercube sampling for evaluating the reliabilities of optical performance and a function-based sequential approximate optimization technique that can reduce computational burden and well handle numerical noise in the tolerance optimization process. Using the suggested tolerance optimization approach, the optimal production cost was decreased by 30.3 % compared to the initial cost while satisfying the two constraints on the reliabilities of optical performance.

• Steel and Composite Structures
• /
• 제29권1호
• /
• pp.67-75
• /
• 2018

High performance concrete (HPC) depends on various parameters such as the type of cement, aggregate and water reducer amount. Generally, the ready concrete company in various regions according to the requirements and costs, mix design of concrete as well as type of cement, aggregates, and, amount of other components will vary as a result of moment decisions or dynamic optimization, though the ideal conditions will be more applicable for the design of mix proportion of concrete. This study aimed to apply dynamic optimization for mix design of HPC; consequently, the objective function, decision variables, input and output variables and constraints are defined and also the proposed dynamic optimization model is validated by experimental results. Results indicate that dynamic optimization objective function can be defined in such a way that the compressive strength or performance of all constraints is simultaneously examined, so changing any of the variables at each step of the process input and output data changes the dynamic of the process which makes concrete mix design formidable.

• 한국정보전자통신기술학회논문지
• /
• 제15권2호
• /
• pp.124-133
• /
• 2022

하드웨어 기술이 발달하고 수치 모델 방식이 고도화됨에 따라 더욱 정밀한 기상예보를 진행할 수 있게 되었다. 본 논문에서는 CESM의 간소화 버전인 SCAM에 포함된 적운모수화 코드 (Unicon, Fortran)를 최적화하고 유지보수성을 증가시키기 위해 Loop Vectorization, Dependency Vectorization, Code Modernization 3가지가 결합한 Unicon Optimization 기법을 제안하고 이를 테스트 하기 위하여 SCAM 전체 실행 구조도를 제시하였다. 본 논문에서는 구축한 SCAM 실행 환경에서 논문에서 제안한 Unicon Optimization 기법을 테스트 하였고 기존 소스 코드 대비 Loop Vectorization은 3.086% Dependency Vectorization은 0.4572% 성능 향상을 이끌어 냈다. 그리고 이를 모두 적용한 Unicon Optimization의 경우 기존 소스 코드 대비 3.457%의 성능 향상을 이끌어 냈다. 이는 본 논문에서 제안한 Unicon Optimization 기법이 우수한 성능을 제공하고 있음을 입증한다.

• Structural Engineering and Mechanics
• /
• 제42권3호
• /
• pp.335-352
• /
• 2012

Particle Swarm Optimization (PSO) is a stochastic population based optimization algorithm which has attracted attentions of many researchers. This method has great potentials to be applied to many optimization problems. Despite its robustness the standard version of PSO has some drawbacks that may reduce its performance in optimization of complex structures such as laminated composites. In this paper by suggesting a new variation scheme for acceleration parameters and inertial weight factors of PSO a novel optimization algorithm is developed to enhance the basic version's performance in optimization of laminated composite structures. To verify the performance of the new proposed method, it is applied in two multi-objective design optimization problems of laminated cylindrical. The numerical results from the proposed method are compared with those from two other conventional versions of PSO-based algorithms. The convergancy of the new algorithms is also compared with the other two versions. The results reveal that the new modifications inthe basic forms of particle swarm optimization method can increase its convergence speed and evade it from local optima traps. It is shown that the parameter variation scheme as presented in this paper is successful and can evenfind more preferable optimum results in design of laminated composite structures.

• 한국시뮬레이션학회논문지
• /
• 제27권2호
• /
• pp.83-90
• /
• 2018

본 연구에서는 선형의 기본 파라메타가 주어졌을 때, 선형 최적화를 통하여 고속으로 주행하는 활주선의 저항성능을 개선하는 기법을 제안하였다. 먼저 선행연구 된 활주선형을 기준 선형으로 채택한 뒤, 선형 변경지점을 정의해 설계변수로 하여 최적화 문제를 수립하였다. 계산 효율을 위하여 탐색공간을 이산화하고, 최적화 문제를 풀기위하여 DPSO(Discrete binary version of Particle Swarm Optimization) 알고리즘을 사용하였다. 최적화 수행 후 기준 선형과 수정 선형의 목적함수 출력의 비교를 수행하였고, 이를 통해 고속영역에서의 저항성능의 개선을 확인하였다.