Árpád Csík

6701719656

Publications - 15

Accelerating unstructured finite volume computations on field-programmable gate arrays

Csaba Nemes Antal Hiba Zoltán Nagy András Kiss Árpád Csík P. Szolgay Miklõs Ruszinkõ

Publication Name: Concurrency and Computation Practice and Experience

Publication Date: 2014-03-10

Volume: 26

Issue: 3

Page Range: 615-643

Description:

In the paper, an field-programmable gate array (FPGA)-based framework is described to efficiently accelerate unstructured finite volume computations where the same mathematical expression has to be evaluated at every point of the mesh. The irregular memory access patterns caused by the unstructured spatial discretization are eliminated by a novel mesh node reordering technique, and a special architecture is designed to fully utilize the benefits of the predictable memory access patterns. In the proposed architecture, a fixed-size moving window of the input stream of the reordered state variables is cached into the on-chip memory and a pipelined chain of processing elements, which gets input only from the fast on-chip memory, is used to carry out the computations. The arithmetic unit (AU) of the processing elements is generated from the data flow graph extracted from the given mathematical expression. The data flow graph is partitioned with a novel graph partitioning algorithm to break up the AU into smaller locally controlled parts, which can be more efficiently implemented in FPGA than the globally controlled AU. The proposed architecture and algorithms are presented via a case study solving the Euler equations on an unstructured mesh. On the currently available largest FPGA, the generated architecture contains three processing elements working in a pipelined fashion to provide one order of magnitude speedup compared with a high performance microprocessor and three times speedup compared with a high performance graphics processing unit. Copyright © 2013 John Wiley & Sons, Ltd. Copyright © 2013 John Wiley & Sons, Ltd.

Open Access: Yes

DOI: 10.1002/cpe.3022

An expert system for the cost-optimal refurbishment of buildings

Árpád Csík

Publication Name: Advanced Materials Research

Publication Date: 2014-01-01

Volume: 899

Issue: Unknown

Page Range: 599-604

Description:

In the present paper we propose a solid theoretical framework for the automatic parametrization of the cost-optimal refurbishment of arbitrary buildings. The parameters describe the structural parts of the building that can be modified or replaced during the refurbishment process. The methodology also provides the cost-optimal refurbishment cost and the corresponding smallest possible total life-cycle operation cost constrained by the available technological and economical limitations. The theory is implemented into the EnergOpt expert system providing fast and accurate answers to costoptimization problems appearing in common practical applications. The system is built on firm mathematical foundations that is supported by state-of-the-art optimization algorithms capable of finding good optimums in a short amount of CPU time. The modular structure of the IT implementation facilitates effortless localization in different countries. The application of the system may considerably contribute to the decrease of the heating energy consumption and the corresponding environmental load of the building stock. The practical potential of the technology is demonstrated by the energetic analysis of an existing family house. © (2014) Trans Tech Publications, Switzerland.

Open Access: Yes

DOI: 10.4028/www.scientific.net/AMR.899.599

Defining energy- and cost-saving potentials and their application in optimal building refurbishment

Árpád Csík Tamás Csoknyai

Publication Name: Environmental Engineering and Management Journal

Publication Date: 2014-01-01

Volume: 13

Issue: 11

Page Range: 2771-2779

Description:

In this paper, we develop a rigorous theoretical framework and a practical implementation for the evaluation of the energy- and cost-saving potential of buildings. The goal is to promote sustainability in the context of building energetics by achieving the most efficient, optimal exploitation of the financial resources available for refurbishments. The practical realization of this concept requires the evaluation of the optimal refurbishment cost that maximizes the net energetic or financial savings during the life cycle of the building. On the one hand, too small a refurbishment cost might lock-in a substantial amount of energy and cost savings potential. On the other hand, refurbishment costs that are too high due to unnecessarily implemented energy-saving measures are likely to waste financial resources. The key concept behind the theory is the novel definition of the reference value used for the computation of the energy- and cost-saving potentials. From a mathematical point of view, the reference value is obtained by two subsequent optimizations. First, a constrained, single-objective optimization is used to evaluate the best energetic state of the building as a function of the refurbishment cost. Second, a simple unconstrained search must be performed to obtain the minimum value and the minimum place of the one-dimensional cost function. The proposed framework automatically provides personalized solutions corresponding to the actual technical characteristics of the building. These solutions are optimal under the given circumstances of the actual refurbishment, resulting in either the highest possible energy- or cost-saving amounts during the life cycle of the building.

Open Access: Yes

DOI: 10.30638/eemj.2014.309

Multi-objective optimization of building envelopes by bacterial memetic algorithms

J. Botzheim Árpád Csík

Publication Name: 2013 World Congress on Nature and Biologically Inspired Computing Nabic 2013

Publication Date: 2013-11-22

Volume: Unknown

Issue: Unknown

Page Range: 245-252

Description:

In the present paper, we apply bacterial memetic algorithms in the context of multi-objective optimization. The goal of the development is to optimize the energetic quality of residential building envelopes in a sustainable manner. Our interest is to determine a good compromise between the quality of the building envelope and the total cost of the installed energy saving components. We provide a description of the applied bacterial operators and the particular ordering of the population that is a non unique procedure. In order to refine the ordering we implement a variant of the distance metric for comparing the non-comparable individuals. The numerical computations are performed by the EnergOpt computational framework of the authors. The performance of the algorithm is demonstrated on the solution of a benchmark test case. Finally, the optimization of an existing building is presented to demonstrate the potential of the methodology. © 2013 IEEE.

Open Access: Yes

DOI: 10.1109/NaBIC.2013.6617870

Examining the accuracy and the precision of PDEs for FPGA computations

Zoltán Nagy András Kiss Árpád Csík P. Szolgay

Publication Name: International Workshop on Cellular Nanoscale Networks and their Applications

Publication Date: 2012-12-13

Volume: Unknown

Issue: Unknown

Page Range: Unknown

Description:

There are a large number of problems which can be accelerated by using architectures on Field Programmable Gate Arrays (FPGA). However sometimes the complexity of a problem does not allow to map it onto a specific FPGA. In that case analysis of precision of the arithmetic unit which may solve the computational problem can be a good attempt to fit the architecture and to accelerate its computation. Numerical algorithm can be implemented using fixed-point or floating point arithmetic (or mixed (both)) with different precision. The aim of the article is not to optimize the numerical algorithm but to find a smaller arithmetic unit precision, which results enough accuracy and fits to smaller FPGA-s. In the paper, one particular problem type is investigated, namely the accuracy of the solution of a simple Partial Differential Equation (PDE). The accuracy measurement is done on an FPGA with different bit width. The solution of the advection equation is analyzed using first and second order discretization methods. As a result we managed to find an optimal bit width for the solution on a specific FPGA. © 2012 IEEE.

Open Access: Yes

DOI: 10.1109/CNNA.2012.6331439

FPGA based acceleration of computational fluid flow simulation on unstructured mesh geometry

Csaba Nemes Antal Hiba Zoltán Nagy András Kiss Árpád Csík P. Szolgay

Publication Name: Proceedings 22nd International Conference on Field Programmable Logic and Applications Fpl 2012

Publication Date: 2012-12-12

Volume: Unknown

Issue: Unknown

Page Range: 128-135

Description:

Numerical simulation of complex computational fluid dynamics problems evolving in time plays an important role in scientific and engineering applications. Accurate behavior of dynamical systems can be understood using large scale simulations which traditionally requires expensive super-computing facilities. In the paper a Field Programmable Gate Array (FPGA) based framework is described to accelerate simulation of complex physical spatio-temporal phenomena. Simulating complicated geometries requires unstructured spatial discretization which results in irregular memory access patterns severely limiting computing performance. Data locality is improved by mesh node renumbering technique which results in a sequential memory access pattern. Additionally storing a small window of cell-centered state values in the on-chip memory of the FPGA can increase data reuse and decrease memory bandwidth requirements. Generation of the floating-point data path and control structure of the arithmetic unit containing dozens of operators is a very challenging task when the goal is high operating frequency. Efficiency and use of the framework is described by a case study solving the Euler equations on an unstructured mesh using finite volume technique. On the currently available largest FPGA the generated architecture contains three processing elements working in parallel providing 75 times speedup compared to a high performance microprocessor. © 2012 IEEE.

Open Access: Yes

DOI: 10.1109/FPL.2012.6339276

Energy and cost optimal design for the reconstruction of residential building envelopes by bacterial memetic algorithms

J. Botzheim József Levente Hontvári János Balázs Árpád Csík Tamás Csoknyai

Publication Name: 6th International Conference on Soft Computing and Intelligent Systems and 13th International Symposium on Advanced Intelligence Systems Scis Isis 2012

Publication Date: 2012-12-01

Volume: Unknown

Issue: Unknown

Page Range: 1226-1231

Description:

In this paper we apply bacterial memetic algorithms for the energy and cost optimal renovation of residential buildings. Following economical demands, there are two types of optimizations considered. First, the total cost of the renovation is prescribed and the best energy quality of the building envelope is determined with a total construction cost not exceeding the given limit. Second, the targeted energy quality of the renovated building is prescribed, and the algorithm determines the optimal renovation plan requiring the smallest costs. Only the optimization of the building envelope is performed, the optimization of the heating-ventilation-air-condition system is ignored. The chromosome of the bacteria contains genes taking only integer values and genes taking real values as well. The value of the genes taking only integer numbers are improved by a simple local search algorithm. The value of the genes taking real numbers are improved locally by the Levenberg-Marquardt approach. Results of actual building optimizations reveal the potential of the proposed algorithm. © 2012 IEEE.

Open Access: Yes

DOI: 10.1109/SCIS-ISIS.2012.6505181

A bullwhip type of instability induced by time varying target inventory in production chains

P. Földesi Árpád Csík

Publication Name: International Journal of Innovative Computing Information and Control

Publication Date: 2012-08-01

Volume: 8

Issue: 8

Page Range: 5885-5897

Description:

We present an analytic investigation of the bullwhip effect developing in production-distribution chains. All common considered causes of the effect are excluded and the only mean to induce a bullwhip type of instability is the adoption of an inventory replenishment policy involving a variable target level. The policy is designed to maintain a safety stock that is proportional to the actual demand. In order to achieve our goals we develop a particular discrete model of supply chains by introducing some fresh concepts into the field of supply chain stability. The basic idea is to derive an update scheme describing the status of the whole chain over the entire time-space (period-stage) domain of interest. We prove that the strategy of demand driven target inventory inherently leads to an instability developing in the chain, which is precisely a manifestation of the bullwhip effect. Following the identification of the source and the nature of the instability, we propose a new production plan, which is stable and does not exhibit the bullwhip effect at all. Thus, the amplitude of the variation of the production rate never exceeds the amplitude of the oscillation of the market demand within the entire supply chain. © 2012 ICIC International.

Open Access: Yes

DOI: DOI not available

Application of evolutionary algorithms for energy efficient building design

J. Botzheim Attila Talamon János Balázs Mihály Retek Árpád Csík Tamás Csoknyai

Publication Name: Iwaciii 2011 International Workshop on Advanced Computational Intelligence and Intelligent Informatics Proceedings

Publication Date: 2011-12-01

Volume: Unknown

Issue: Unknown

Page Range: 6

Description:

The topic of energy efficient building design is an attractive philosophy in the era of diminishing fossil energy sources drained by the ever increasing thirst of the Earth's population for energy. In this paper we propose to apply evolutionary algorithms aiding the design of energy efficient buildings. We employ a simple mathematical model involving a large number of parameters. Energy efficiency is defined in terms of these parameters through algebraic evaluations. The quasi-optimal values of the fitness function representing the level of energy efficiency are obtained by Bacterial Evolutionary Algorithms. Results indicate that in the case of a prescribed total construction cost close to 80% improvement can be achieved in terms of energy efficiency. On the other hand, if one targets a prescribed energy efficiency, more than 30% of the total construction costs can be saved by proper optimization.

Open Access: Yes

DOI: DOI not available

Computational fluid flow simulation on body fitted mesh geometry with IBM cell broadband engine architecture

Sándor Kocsárdi Zoltán Nagy András Kiss Árpád Csík

Publication Name: Ecctd 2009 European Conference on Circuit Theory and Design Conference Program

Publication Date: 2009-12-10

Volume: Unknown

Issue: Unknown

Page Range: 827-830

Description:

The solutions of partial differential equations (PDEs) play a key role in today's real world simulations. Computational Fluid Dynamics (CFD) is an important part of this area, which involves the problem of gas or fluid flow over different obstacles, e.g., air flow around vehicles, buildings, or the flow of water in the oceans. In engineering applications the temporal evolution of non-ideal, compressible fluids is quite often modeled by the system of Navier-Stokes equations. They are a coupled set of nonlinear hyperbolic partial differential equations and form a relatively simple, yet efficient model of compressible fluid dynamics. In the paper the implementation of a CFD on Body Fitted Mesh geometry on the Cell Broadband Engine is described. An arbitrary surface can be more easily simulated on body fitted mesh than on rectangular computation domain. ©2009 IEEE.

Open Access: Yes

DOI: 10.1109/ECCTD.2009.5275111

Supersonic flow simulation on IBM cell processor based emulated digital cellular neural networks

Sándor Kocsárdi Zoltán Nagy András Kiss Árpád Csík

Publication Name: Proceedings IEEE International Symposium on Circuits and Systems

Publication Date: 2009-10-26

Volume: Unknown

Issue: Unknown

Page Range: 1225-1228

Description:

In the area of mechanical, aerospace, chemical and civil engineering the solution of partial differential equations (PDEs) has been one of the most important problems of mathematics for a long time. In this field, one of the most exciting areas is the simulation of fluid flow, which involves for example problems of air, sea and land vehicle motion. In engineering applications the temporal evolution of non-ideal, compressible fluids is quite often modeled by the system of Navier-Stokes equations. They are a coupled set of nonlinear hyperbolic partial differential equations and form a relatively simple, yet efficient model of compressible fluid dynamics. Unfortunately the necessity of the coupled multi-layered computational structure with nonlinear, space-variant templates does not make it possible to utilize the huge computing power of the analog Cellular Neural Network Universal Machine (CNN-UM) chips. To improve the performance of our solution emulated digital CNN-UM implemented on IBM Cell Broadband Engine has been used. The goal is to perform the operations with the highest possible parallelism. ©2009 IEEE.

Open Access: Yes

DOI: 10.1109/ISCAS.2009.5117983

Simulation of 2D inviscid, adiabatic, compressible flows on emulated digital CNN-UM

Sándor Kocsárdi Zoltán Nagy Árpád Csík P. Szolgay

Publication Name: International Journal of Circuit Theory and Applications

Publication Date: 2009-05-01

Volume: 37

Issue: 4

Page Range: 569-585

Description:

In the area of mechanical, aerospace, chemical and civil engineering the solution of partial differential equations has been one of the most important problems in mathematics for a long time. In this field, one of the most exciting areas is the simulation of fluid flow, which involves for example, problems of air, sea and land vehicle motion. In this paper a CNN-UM based solver of 2D inviscid, adiabatic, compressible fluids will be presented. The governing equations are solved by using first- and second-order numerical methods. Unfortunately the necessity of the coupled multi-layered computational structure with nonlinear, space-variant templates does not make it possible to utilize the huge computing power of the analog CNN-UM chips. To improve the performance of our solution emulated digital CNN-UM implemented on FPGA has been used. Properties of the implemented specialized architecture is examined in terms of area, speed and accuracy. Copyright © 2008 John Wiley and Sons, Ltd.

Open Access: Yes

DOI: 10.1002/cta.565

Simulation of Two-Dimensional supersonic flows on emulated-digital CNN-UM

Sándor Kocsárdi Zoltán Nagy Árpád Csík P. Szolgay

Publication Name: Eurasip Journal on Advances in Signal Processing

Publication Date: 2009-04-09

Volume: 2009

Issue: Unknown

Page Range: Unknown

Description:

Computational fluid dynamics (CFD) is the scientific modeling of the temporal evolution of gas and fluid flows by exploiting the enormous processing power of computer technology. Simulation of fluid flow over complex-shaped objects currently requires several weeks of computing time on high-performance supercomputers. A CNN-UM-based solver of 2D inviscid, adiabatic, and compressible fluids will be presented. The governing partial differential equations (PDEs) are solved by using first- and second-order numerical methods. Unfortunately, the necessity of the coupled multilayered computational structure with nonlinear, space-variant templates does not make it possible to utilize the huge computing power of the analog CNN-UM chips. To improve the performance of our solution, emulated digital CNN-UM implemented on FPGA has been used. Properties of the implemented specialized architecture is examined in terms of area, speed, and accuracy.

Open Access: Yes

DOI: 10.1155/2009/923404

Experimental result on supersonic flow simulation on emulated digital CNN-UM

Sándor Kocsárdi Zoltán Nagy Árpád Csík P. Szolgay

Publication Name: Proceedings of the IEEE International Workshop on Cellular Neural Networks and their Applications

Publication Date: 2008-09-23

Volume: Unknown

Issue: Unknown

Page Range: 5

Description:

In the area of mechanical, aerospace, chemical and civil engineering the solution of partial differential equations (PDEs) has been one of the most important problems of mathematics for a long time. In this field, one of the most exciting areas is the simulation of fluid flow, which involves for example problems of air, sea and land vehicle motion. ©2008 IEEE.

Open Access: Yes

DOI: 10.1109/CNNA.2008.4588636

Two-dimensional compressible flow simulation on emulated digital CNN-UM

Sándor Kocsárdi Zoltán Nagy Árpád Csík P. Szolgay

Publication Name: Proceedings of the IEEE International Workshop on Cellular Neural Networks and their Applications

Publication Date: 2008-09-23

Volume: Unknown

Issue: Unknown

Page Range: 169-174

Description:

In the area of mechanical, aerospace, chemical and civil engineering the solution of partial differential equations (PDEs) has been one of the most important problems of mathematics for a long time. In this field, one of the most exciting areas is the simulation of fluid flow, which involves for example problems of air, sea and land vehicle motion. In this paper a CNNUM based solver of 2D inviscid, adiabatic, compressible fluids will be presented. The governing equations are solved by using firstand second-order numerical methods. Unfortunately the necessity of the coupled multi-layered computational structure with nonlinear, space-variant templates does not make it possible to utilize the huge computing power of the analog CNN-UM chips. To improve the performance of our solution emulated digital CNN-UM implemented on FPGA has been used. Properties of the implemented specialized architecture is examined in terms of area, speed and accuracy. ©2008 IEEE.

Open Access: Yes

DOI: 10.1109/CNNA.2008.4588672