Automated modeling of physical systems
Automated modeling of physical systems
Irrelevance reasoning in knowledge-based systems
Irrelevance reasoning in knowledge-based systems
Operationalizing engineering models of steady-state equations into efficient simulation programs
Operationalizing engineering models of steady-state equations into efficient simulation programs
Graph Algorithms
Scaling up self-explanatory simulators polynomial time compilation
IJCAI'95 Proceedings of the 14th international joint conference on Artificial intelligence - Volume 2
| Hi-index | 0.00 |
In recent years, compositional modeling and self-explanatory simulation techniques have simplified the process of building dynamic simulators of physical systems. Building steady-state simulators is, conceptually, a simpler task consisting in solving a set algebraic equations. This simplicity hides delicate technical issues of convergence and search-space size due to the potentially large number of unknown parameters. We present an automated technique for reducing the dimensionality of the problem by 1) automatically identifying feedback loops (a generally NP-complete problem), 2) hierarchically decomposing the set of equations in terms of feedback loops, and 3) structuring a simulator where equations are solved either serially without search or in isolation within a feedback loop. This paper describes the key algorithms and the results of their implementation on building simulators for a two-phase evaporator loop system across multiple combinations of causal and non-causal approximations.