Physical quantity analyzing method and apparatus thereof

Boots – shoes – and leggings

Patent

Rate now

  [ 0.00 ] – not rated yet Voters 0   Comments 0

Details

364554, 379382, G01N 2700

Patent

active

056468687

DESCRIPTION:

BRIEF SUMMARY
TECHNICAL FIELD

The present invention relates to physical quantity (measurable or calculated physical characteristic, quantity, aspect, atribute, property, trait or the like) analyzing methods and apparatus therefore, more particularly, relates to novel methods and apparatus for analyzing unknown physical quantities u.sub.i (i=1, 2, . . . , m) in a physical system which has a relationship of ##EQU1## (wherein, i=1, 2, . . . , m; j=1, 2, . . . , n; .alpha..sub.ij is a proportional constant) between physical quantities u.sub.i of individual physical sources and physical quantities O.sub.j which occur at arbitrary points caused by the individual physical sources.


BACKGROUND ART

In the past, methods for analyzing magnetic field sources using a super computer were based upon output from plural flux meters and Monte Carlo analysis; methods applied to adaptive noise cancelers; methods which incorporate an impulse response estimating apparatus; and methods which determine threshold values, coupling coefficients, and the like, of neuron devices which constitute a neural network based upon physical formulae of a system so as to omit learning time required for these neuron devices (refer to Japanese Patent Laid Open Tokukaihei 5-94543), are proposed as physical quantity analyzing methods. Further, a finite element method is also proposed as physical quantity analyzing method.
The method for analyzing magnetic field sources typically involves the following processing: space which is searched by plural flux meters; estimated error calculating process; corresponding current element and the total estimated error; using random numbers; process; error calculated in f), and restore the information which are evacuated at d) when the evacuated total estimated error is smaller than the total estimated error calculated in f).
An adaptive noise canceler incorporating the method is illustrated in FIG. 27 and operates as follows. An input S.sub.j which is made by mixing a noise from a noise source 72 to information from a signal source 71 is supplied to a non-inversed input terminal of an error computing element 73. Only the noise from the noise source 72 is supplied to an inversed input terminal of the error computing element 73 through an FIR filter 74. An output from the error computing element 73 is feedbacked to the FIR filter 74. The FIR filter 74 employs a LMS (Least Mean Square) algorithm.
The adaptive noise canceler can remove only noise in information by determining an estimation gain to be a proper value. Therefore, noise of a duct of an air conditioner can be eliminated, and the passenger area of an automobile can be made to be quiet and so on. That is, noise which should be removed are accurately estimated.
An impulse response estimating apparatus incorporating the method performs analysis of frequency components using a fast Fourier transform (hereinafter referred to as FFT) and estimates an impulse response based upon the analysis result.
A method for solving threshold values, coupling coefficients, and the like, for neural systems has already been filed in a patent application by the applicant of this patent application, and is illustrated in FIG. 28. An input pattern is supplied to plural physical formula operating units 811, 812, . . . , 81m so as to operate based upon a known physical formula. Outputs from all physical formula operating units 811, 812, . . . , 81m are supplied to a sigma unit 82 so as to obtain a sum. The sum and an actual measurement value are supplied to an error operating device 83 so as to obtain an error (a difference between the estimated and actual values). The obtained error is fed back to correction sections 811a, 812a, . . . , 81ma of the physical formula operating units 811, 812, . . . , 81m. Values of variables which are estimated in the physical formula operating units 811, 812, . . . , 81m are collected and outputted as analysis results by an information collection unit 84.
Therefore, no physical formula operating units are required learning of a physical formula. The analysis

REFERENCES:
Fujio Kobayashi, Shoichiro Yamaguchi "Successive Approximation of Hypercomplex Simultaneous equations using Measurements" IEE Journal thesis, No. 92-vol. A, 9th edit. pp. 431-439 1972.
Kenro Murata, Chikara Oguni "Super Computer Application to Scientific technological calculations"Maruzen pp. 136-153 Mar. 15, 1985.
Fujio Kobayashi, Shoichiro Yamaguchi "Solution of Simultaneous linear equations in bad conditions with restrictions", Information Processing, vol. 16, 9th edit. pp. 789-794 1975.

LandOfFree

Say what you really think

Search LandOfFree.com for the USA inventors and patents. Rate them and share your experience with other people.

Rating

Physical quantity analyzing method and apparatus thereof does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Physical quantity analyzing method and apparatus thereof, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Physical quantity analyzing method and apparatus thereof will most certainly appreciate the feedback.

Rate now

     

Profile ID: LFUS-PAI-O-2412495

  Search
All data on this website is collected from public sources. Our data reflects the most accurate information available at the time of publication.