US4674028A
Expired / Lapsed
US
⏱️ 0.0 years remaining
Identification method of a process parameter
Assignee: Hitachi Ltd
•
Filing: 1985-01-16
•
Granted: 1987-06-16
•
Expires: 2005-01-16
Abstract
Abstract A calculation unit for calculating a gain matrix from process input and output information and a calculation unit for calculating a parameter estimation have an independent weight setting unit. A weight W i (k+1) is set at a current time point while taking into consideration of an affect of square of residual e 2 (i) in each time step between an initial time step (i=1) to the current time step (i=k+1), and a parameter is estimated in accordance with an evaluation index ##EQU1##
Inventors
- 👤 Makoto Shioya
- 👤 Teruji Sekozawa
- 👤 Motohisa Funabashi
- 👤 Masahiro Ishida
- 👤 Hiroatsu Tokuda
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-10-29 06:23
CPC Classifications
G; G05; G05B; G05B13/00; G05B13/02; G05B13/04
Claims Analysis
Total Claims: 7
Claim 1
Independent Claim
95 words
In a method of identification of a process parameter to minimize a sum of weighted squares of residuals defined in accordance with a relationship between process input information and process output information in a plurality of time steps from an initial time step to a current time step, the improvement comprising the step of optimally identifying the process parameter in a sense of a least square method for each corresponding time step in a recursive form based on said input information and output information when the weight of the squares changes for each time step.
Claim 2
Dependent Claim
95 words
In a method of identification of a process parameter to minimize a sum of weighted squares of residuals defined in accordance with a relationship between process input information and process output information in a plurality of time steps from an initial time step to a current time step, the improvement comprising the step of optimally identifying the process parameter in a sense of a least square method for each corresponding time step in a recursive form based on said input information and output information when the weight of the squares changes for each time step.
Claim 3
Dependent Claim
17 words
An identification method according to claim 1, wherein said weight is expressed by a diagonal weighting matrix.
Claim 4
Dependent Claim
40 words
An identification method according to claim 2, wherein said weighting matrix has a form of an expanded generalized adjustment law in which adjustment coefficients which are two scalar coefficients in a generalized adjustment law are replaced with two matrix coefficients.
Claim 5
Dependent Claim
35 words
An identification method according to claim 3, wherein the two matrix coefficients of said expanded generalized adjustment law are determined from a weighting matrix corresponding to a method of fitting in a moving average period.
Claim 6
Dependent Claim
18 words
An identification method according to claim 3, wherein said expanded generalized adjustment law is in a recursive form.
Claim 7
Dependent Claim
30 words
An identification method according to claim 3, wherein an extended least square method eliminating steady state bias is applied to a least square method of said expanded generalized adjustment law.
References Cited (Backward Citations)
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 1990-06-15 | 1990-12-15 | 1991-06-16 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 1994-06-15 | 1994-12-15 | 1995-06-16 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 1998-06-15 | 1998-12-15 | 1999-06-16 | $7700.00 | Window Closed / Lapsed |