Patent
1995-01-30
1997-01-14
Nguyen, Phu K.
G06F 1500
Patent
active
055948510
DESCRIPTION:
BRIEF SUMMARY
BACKGROUND OF THE INVENTION
The invention relates to a method by which components, such as for example electronic analog components, or mechanical components can be simulated on a computer-based simulator, more rapidly and precisely than hitherto usual.
During recent years, the simulation of electronic components has developed in to an important topic. It is constantly emphasized that the simulation can above all improve the quality of the developed circuits and leads to an increase in productivity (H. Spiro: Simulation of Integrated Circuits by Universal Computer Programs, R. Oldenbourg, Verlag GmbH, Munich, 1985); (H. Wiedemann (Racal Redac Design System GmbH): Superordinate Aspects in the ASIC Design in: Meeting Reports SMT/ASIC/Hybrid '91, 5th International Trade Fair and Congress, VDE Verlag, Berlin, 1991); and (H. Stump (Logic Modeling Systems Incorporated): Trends in Technology and EDA for the 1990s, in: Conference Record Electro/90, Ventura, USA, 1990). However, the extent to which simulation has become widespread is relatively small, above all in the case of the development of circuit board assemblies. The reasons for this are both of an organizational and also of a technical nature.
One of the technical obstacles in simulation--both on the digital and also on the analog side--is the absence of suitable models (H. Stump (Logic Modeling Systems Incorporated): Trends in Technology and EDA for the 1990s, in: Conference Record Electro/90, USA, 1990). While in the case of digital simulation what is involved is the logical and temporal description of a component element using a corresponding programing language, analog simulation comprises the conversion of prescribed measurement curves in to mathematical equations, which must be formulated with the aid of a modeling language. Although at the present time no standard can be identified in the field of analog modeling languages, the majority of simulators do nevertheless offer at least the possibility of the input of mathematical function equations. Accordingly, the problem of the modeling is essentially reduced to obtaining such equations.
The requirements imposed on mathematical equations which are intended to describe the physical behavior of a component are divided in to two groups of demands: on the one hand, the physical behavior is to be described as precisely as possible. In the case of electronic components, the theory frequently also delivers differing description functions for differing working sectors. In most cases, it is not possible to describe a component over the entire working range, even with only one function. On the other hand, the models are to be as efficient as possible, and the simulator must therefore be able to process them rapidly. The more complicated are the functional descriptions, the more slowly the simulation runs. Specifically in the case of the complex circuits which are of particular interest for simulation, the factor "simulation time" plays a decisive role.
SUMMARY OF THE INVENTION
The object underlying the invention consists in specifying a method for simulating at least one component, which operates more rapidly and more precisely than hitherto usual methods.
This object is achieved in accordance with method for simulating at least one component, whose characterizing features are described by a set of pairs of values (A, V) by means of a simulator on a computer. In a first step, a function (s) is found by spline interpolation for all pairs of values (A, V) which function allocates in each instance to the one vlaue (A, V) of each respective pair of values the other value (V, A) of the pair of vlaues. In a second step, the function found is used in the simulator to determine arbitrarily selectable pairs of values (A, V) for the component. A non-differentiable point (D) in the progression of the function is generated in that the set of pairs of values is subdivided in to two subsets, in which the point is included. The spline interpolation is carried out for each subset, the point being used on one occasion as upp
REFERENCES:
patent: 5047971 (1991-09-01), Horwitz
Kahaner, "Numerical Methods and Software", pp. 108-113,1989, Prentice Hall.
"A Table Model For Circuit Simulation", Klaus-Georg Rauh, 12th European Solid State Circuits Conference (1986), pp. 211-213.
"Automatic Tension Adjustment for Interpolatory Splines", Yates Fletcher, IEEE Computer Graphics and Applications, vol. 10, No. 1, Jan. 1990, pp. 10-17.
H. Stump (Logic Modeling Systems Incorporated): Trends in Technology and EDA for the 1990's, in: Conference Record Electro/90, Ventura, USA, 1990, pp. 335-341.
B. Tuck: What Do Digital Designers Need to Master the Art of Analog Design, in: Computer Design, Aug. 1991, pp. 92-106.
Jurgen Haible
Lange Eberhard
Nguyen Phu K.
Siemens Aktiengesellschaft
LandOfFree
Method of simultating at least one component whose characterizin does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method of simultating at least one component whose characterizin, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method of simultating at least one component whose characterizin will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-1395237