Evaluation of the computation times for direct and iterative resolution methods of MTJ library matrices applied in a thermal simulation system
Abstract:
Thermal simulation systems in buildings, contribute to the design of energy-efficient structures; however, a significant amount of computational time is required in order to obtain the results of the simulation process. The main focus of this paper is examining the possibility of reducing time required for thermal simulation systems calculations. More specifically, the paper focuses on one of the major processes that requires computing resources at solving the system of equations obtained as a result of the thermal modeling of a building. This research was undertaken in order to determine the performance of Java library methods: Matrix Tool for Java (MTJ) for solving systems of linear equations, and identifying which of these was the optimum in terms of computation time. For this purpose, tests for the iterative methods combined with pre-conditioners were conducted. The tests of the direct method were done through the development of a software implemented in two case studies of buildings, and that was modeled with the parameters of the thermal simulation software called JEner, from the Thermal Engineering Research Group from the University of Cadiz in Spain.
Año de publicación:
2018
Keywords:
- MTJ
- thermal simulation
- iterative methods
Fuente:
Tipo de documento:
Conference Object
Estado:
Acceso restringido
Áreas de conocimiento:
- Simulación por computadora
- Simulación por computadora
Áreas temáticas:
- Ciencias de la computación