A VLSI analog computer/math co-processor for a digital computer

2005 ◽  
Author(s):  
G.E.R. Cowan ◽  
R.C. Melville ◽  
Y.P. Tsividis
Keyword(s):  
Digital Computer ◽  
Analog Computer ◽  
Computer Math

The design of an automatic patching system

SIMULATION ◽  
1968 ◽  
Vol 10 (6) ◽  
pp. 281-288 ◽  
Author(s):  
David A. Starr ◽  
Jens J. Jonsson
Keyword(s):  
Control System ◽  
Digital Computer ◽  
Analog Computer ◽  
Master Of Science ◽  
The Third ◽  
Specialized Software ◽  
Work Done ◽  
Switching Devices

An automatic patching system for analog computers is proposed. The system patches any set of component ter minals together, yet size and cost are practical. The design is accomplished in three steps: First, a "model" analog computer is defined and its components divided into modules. Components within a module connect to one another through an "intraface" which allows any terminal to connect to any other ter minal. Problem sections are patched within the modules and then joined together by an "intermodular" trunking system. Secondly, the principle of module chaining permits in dividual modules to access a limited number of other modules, rather than all other modules. This reduces the number of patching switches required by the intermodular trunking system and reduces the overall number of switches required. The third step, called concentration, brings the required number of patching switches down to a practical value by assuming that all analog computer terminals will not be in use at once. Switching devices called "concentrators" exploit this fact to reduce switch requirements. The automatic patching system is driven by a digital computer under control of specialized software, and to gether with the automatic patching control system it can patch large problems in times of the order of fifteen sec onds. The information in this paper is a result of work done for a Master of Science thesis.1

Impedance loading servo pump system for excised canine ventricle

1982 ◽  
Vol 243 (2) ◽  
pp. H346-H350 ◽  
Author(s):  
K. Sunagawa ◽  
D. Burkhoff ◽  
K. O. Lim ◽  
K. Sagawa
Keyword(s):  
Digital Computer ◽  
Computer Control ◽  
Complex Impedance ◽  
Impedance Spectrum ◽  
Ventricular Volume ◽  
Analog Computer ◽  
Arterial System ◽  
Windkessel Model ◽  
Pump System ◽  
Hydraulic Impedance

To investigate ventricular-arterial system interaction, we have developed a hybrid-computer-controlled impedance loading servo pump system that enables us to impose a simulated arterial hydraulic impedance on an excised canine ventricle. An analog computer programmed to simulate a three-element Windkessel model of the arterial system computes instantaneous aortic flow from the instantaneous ventricular pressure. The time integral of this flow is used to command a volume servo pump system that controls the instantaneous ventricular volume. All parameter values in the loading system are controlled by a digital computer. The actual impedance spectrum generated by the system was reasonably close to that expected from the arterial model. The unique features of this system are the following. 1) The instantaneous volume of the ventricle, which is crucial information, can be measured. 2) If needed, the arterial impedance model can easily be reprogrammed to generate more complex impedance spectra. 3) The vascular parameters can be made nonlinear or time varying through the digital computer control.

The DES-1A New Digital Computer for Solving Differential Equations

SIMULATION ◽  
1965 ◽  
Vol 4 (4) ◽  
pp. 264-276 ◽  
Author(s):  
Leon Levine
Keyword(s):  
Differential Equations ◽  
Data Storage ◽  
Digital Computer ◽  
High Speed ◽  
Initial Conditions ◽  
Analog Computer ◽  
Rapid Communication ◽  
Floating Point Arithmetic ◽  
Point Arithmetic ◽  
Computer Operator

The DES-1 is a new digital computer system, espe cially designed for the solution of ordinary differ ential equations. The major components are (1) the SDS 9300, a high-speed digital computer, ( 2) a mathe matical operator language, and (3) a console. The mathematical operator language is analogous to the computing elements of the analog computer. Thus both the DES-1 and analog computer can be programmed the same way (although the DES-1 has several more operations). Consequently an analog computer operator can learn to program the DES-1 easily and can use much of the programming expe rience acquired on the analog computer. Further more, the use of floating-point arithmetic eliminates all need for amplitude scaling. The console provides rapid communication between the operator and the DES-1 and allows problem mode control and changes even as the com putations are being made. Some of the modes are RESET, OPERATE, and HOLD, while typical changes include parameters, initial conditions, and even equa tion structure. The DES-1, therefore, combines the advantages of the digital computer 1. Accuracy 2. Reliability 3. Ease of program and data storage 4. Ability to perform algebraic operations with those of the analog computer 1. Convenience of programming through use of the mathematical operator language 2. Rapid communication between operator and computer to produce a computer which has advantages over both conventional analog and digital computers in solving differential equations.

scholarly journals Response of a Parametrically Excited System to a Nonstationary Excitation

1995 ◽  
Vol 1 (1) ◽  
pp. 57-73 ◽  
Author(s):  
Harold L. Neal ◽  
Ali H. Nayfeh
Keyword(s):  
Computer Simulations ◽  
Governing Equation ◽  
Digital Computer ◽  
Multiple Scales ◽  
Method Of Multiple Scales ◽  
Saddle Points ◽  
Period Doubling ◽  
Excitation Amplitude ◽  
Analog Computer ◽  
Modulation Equations

The response of a single-degree-of-freedom system to a nonstationary excitation is investigated by using the method of multiple scales as well as analog- and digital-computer simulations. The unexcited system has one focus and two saddle points. The system can be used to model rolling of ships in head or follower seas. The method of multiple scales is used to derive equations governing the modulation of the amplitude and phase of the response. The modulation equations are used to find the stationary solutions and their stability. The response to nonstationary excitations is found by integrating the original governing equation as well as the modulation equations. There is good agreement between the results of both approaches. For some frequency and amplitude sweeps, the nonstationary response found from integrating the original governing equation exhibits behaviors that are analogous to symmetry-breaking bifurcations, period-doubling bi furcations, chaos, and unboundedness present in the stationary case. The maximum response amplitude and the excitation amplitude or frequency at which the response becomes unbounded are found as functions of the sweep rate. The results of the digital-computer simulations are verified with an analog computer.

Analog simulation of satellite times-in-view

SIMULATION ◽  
1967 ◽  
Vol 8 (2) ◽  
pp. 105-110
Author(s):  
Roger N. Bryan
Keyword(s):  
Real Time ◽  
Digital Computer ◽  
Satellite Orbit ◽  
Analog Computer ◽  
Paper Strip ◽  
Analysis Tool ◽  
Analog Simulation ◽  
Satellite Geometry ◽  
The One ◽  
Orbit Geometry

During the planning stages of any communication or com mand/control system utilizing satellites, it is necessary to examine the ground-terminal/satellite geometry for many proposed configurations in order to find the one most suitable. By "most suitable," I mean the configuration exhibiting the greatest ground-terminal-to-satellite acces sibility as demonstrated by a maximum time-in-view. The procedures I describe in this paper provide a means for rapidly generating ground terminal/satellite visibility records via an analog computer simulation of the Earth's surface/satellite orbit geometry. The analog simulation has several significant advantages relative to similar digital computer approaches-(1) it is significantly faster (easily 15,000 times real-time) and less expensive to operate; (2) the output, ink-on-paper strip-chart recordings, is directly interpretable without further data reduction; (3) system parameter changes are readily introduced while a simula tion is in progress; (4) the Systems Analyst can maintain a decidedly greater rapport with his configuration analysis. The disadvantages are: (1) diminished accuracy, and (2) exclusion of Earth/orbit non-linearities and anomalies—in a real-time tracking or scheduling simulation, these dis advantages are serious; in a systems analysis tool such as this time-in-view simulation, they are not of great con sequence. I have devoted the first five sections of this paper to a description of the development of the vector-matrix model underlying the simulation. The sixth section describes the mechanization of the model using an electronic analog computer. The equations from which the computer potentiometer settings are derived are quite tedious to manipulate man ually—even aided by a desk calculator. In an appendix, I describe the simple digital computer (in my case, a RE- COMP II) program used to obtain the desired potentiom eter settings. It strikes me that, while lack of simultaneity in the usage of the analog and digital machines precludes the use of the word hybrid, we may validly coin the expression "syner getic computing" to describe this simulation.

Optimization using a remote control console for a digital computer

SIMULATION ◽  
1966 ◽  
Vol 7 (1) ◽  
pp. 36-41 ◽  
Author(s):  
D.B. Kirby
Keyword(s):  
Digital Computer ◽  
Analog Computer ◽  
General Purpose ◽  
Control Console ◽  
Current Interaction ◽  
Equipment Configuration ◽  
Tion System ◽  
Work Done ◽  
Machine Interaction ◽  
Param Eters

A general-purpose optimization program, using the meth ods of steepest descent and multidimensional Newton- Raphson, has been developed for the IBM 7094. The program can be used independently or as part of a man- machine optimization system in which the engineer can quickly appraise the effect of changes in problem param eters. Man-machine interaction at a remote location is provided through an experimental console which con sists of a Packard-Bell PB-250 computer and associated input-output equipment. The program and equipment configuration used for interaction will be described and an example of network optimization using this system will be presented. An optimization scheme which is designed for a digital computer offers a more sophisticated strategy and better accuracy than a scheme designed for an analog computer. The current interaction system is of limited form; never theless, work done with it has aided in defining the ob jectives for a more flexible man-digital machine interac tion system.

Digital Computer Control of Analog Computer Potentiometers

SIMULATION ◽  
1965 ◽  
Vol 4 (1) ◽  
pp. 14-18 ◽  
Author(s):  
W.J. Chandler ◽  
R.B. McGhee
Keyword(s):  
Digital Computer ◽  
Computer Control ◽  
Analog Computer

A VLSI Analog Computer/Digital Computer Accelerator

2006 ◽  
Vol 41 (1) ◽  
pp. 42-53 ◽  
Author(s):  
G.E.R. Cowan ◽  
R.C. Melville ◽  
Y.P. Tsividis
Keyword(s):  
Digital Computer ◽  
Analog Computer

Development of Memory Core

SIMULATION ◽  
1963 ◽  
Vol 1 (1) ◽  
pp. R-21-R-25
Author(s):  
Stanley H. Jury
Keyword(s):  
Digital Computer ◽  
Analog Computer ◽  
Memory Unit

Using this memory unit, an analog computer can operate in the same way as a digital computer. Computer size no longer limits the size of the problem which can be handled

Sign in / Sign up

Export Citation Format

Share Document