US5825645A
Expired / Lapsed
US
⏱️ 0.0 years remaining
Two-level system identifier apparatus with optimization
Assignee: Honeywell Inc
•
Filing: 1995-05-15
•
Granted: 1998-10-20
•
Expires: 2015-05-15
Abstract
Abstract An apparatus for identifying the structure and estimating the parameters in the structure of a controlled system is described. The apparatus uses a structure identifier, in a preferred embodiment, employing a neural network, to match the relationship of the system inputs and outputs to a mathematical relation. This identified structure is communicated to a parameter estimator. The parameter estimator may employ a neural network to give an initial parameter vector which can be used to generate, iteratively, better parameter estimates. This increases the efficacy of the parameter estimator. The apparatus can be inserted into controllers for various systems to improve performance. Using a neural network for parameter estimation as shown can be generalized to any nonlinear optimization problem.
Inventors
- 👤 Ahmet Ferit Konar
- 👤 Tariq Samad
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-11-01 04:24
CPC Classifications
G; G05; G05B; G05B17/00; G05B17/02
Claims Analysis
Total Claims: 18
Claim 1
Independent Claim
98 words
An two level system identifier apparatus for identifying both the structure and estimating uncontrollable and unmeasured parameters of a controlled system comprising: structure identifier means for determining the structure of a function which describes the mathematical relationship of the inputs to a system and the outputs from said system, communication means for transferring the determined function structure and generating the function, the relationship and parameters, including uncontrollable and unmeasured parameters, of which describes said mathematical relationship in terms of said parameters, parameter estimator means for receiving said function from said communication means and for estimating said parameters' value.
Claim 2
Dependent Claim
98 words
An two level system identifier apparatus for identifying both the structure and estimating uncontrollable and unmeasured parameters of a controlled system comprising: structure identifier means for determining the structure of a function which describes the mathematical relationship of the inputs to a system and the outputs from said system, communication means for transferring the determined function structure and generating the function, the relationship and parameters, including uncontrollable and unmeasured parameters, of which describes said mathematical relationship in terms of said parameters, parameter estimator means for receiving said function from said communication means and for estimating said parameters' value.
Claim 3
Dependent Claim
33 words
An apparatus as set forth in claim 1 wherein the structure identifier means is a neural network trained on a training data set not derived using the process instrumentation or the control system.
Claim 4
Dependent Claim
133 words
An apparatus as set forth in claim 2 wherein said training data set comprises the values of inputs to said system and outputs from said system and structural descriptions, wherein said values of inputs and outputs in said training data set are produced by any of the following methods: A) computation of outputs based on randomly generated inputs employed in said function, B) computation of outputs based on randomly generated inputs employed in randomly generated functions, C) computation of outputs based on thoughtfully generated inputs employed in said function, D) computation of outputs based on thoughtfully generated inputs employed in randomly generated functions, E) recording real world generated inputs and outputs to and from said system, and F) recording real world generated inputs and outputs to and from a test data generating system.
Claim 5
Dependent Claim
122 words
An apparatus as set forth in claim 1 wherein the function is comprised of subfunctions, each subfunction determining the relationship of a set of zero or more parameters to said function, and wherein said communication means employs a set of switches having two logical positions, wherein each switch in one position turns off and in the other position turns on the subfunction of one of said parameter sets within said function, and wherein the position of the switches is determined by output received from said structure identifier so that the position of the set of logical switches operates as a communication of the identified system structure by a connection of an output of said set of switches to said parameter estimator means.
Claim 6
Dependent Claim
19 words
An apparatus as set forth in claim 1 wherein the parameter estimator receives as input an initial parameter value.
Claim 7
Dependent Claim
18 words
An apparatus as set forth in claim 5 wherein the initial parameter value is set by the user.
Claim 8
Dependent Claim
37 words
An apparatus as set forth in claim 5 wherein the initial parameter value is set by a neural network trained on a training data set comprising inputs to said system, outputs from said system and parameter values.
Claim 9
Dependent Claim
27 words
An apparatus as set forth in claim 7 wherein said neural network is also used to generate the structure identification function provided by the structure identifier means.
Claim 10
Dependent Claim
132 words
An apparatus as set forth in claim 1 wherein the parameter estimator means comprises: a simulator element and an optimizer element, wherein said simulator element receives input values from said system and parameter estimate values output from said optimizer element and wherein said simulator element produces an output to said optimizer element that models the system output values based on the system input values and said parameter estimates by applying these two inputs to a simulation based on the structure identified by the structure identifier means, and wherein said optimizer element receives as input, said system output values and compares this system output input to the model of the system output, and based on the error and the gradient performs an optimization of the parameter estimate to produce a new parameter estimate.
Claim 11
Dependent Claim
35 words
An Apparatus as set forth in claim 9 wherein the model output and input pairs together with the parameter estimate outputs are used as part of the training data set for the system identifier means.
Claim 12
Dependent Claim
30 words
An apparatus as set forth in claim 10 wherein the training data set is continually expanded during on-line operations so as to produce initial parameter estimates of continually increasing accuracy.
Claim 13
Dependent Claim
30 words
An apparatus as set forth in claim 10, wherein the input values are a constant step function and the input values are not provided as specific input to the network.
Claim 14
Dependent Claim
29 words
An apparatus as set forth in claim 10, wherein the input is a single fixed function and the input values are not provided as specific input to the network.
Claim 15
Dependent Claim
30 words
An apparatus as set forth in claim 3, wherein the input values are a constant step function and the input values are not provided as specific input to the network.
Claim 16
Dependent Claim
29 words
An apparatus as set forth in claim 3, wherein the input is a single fixed function and the input values are not provided as specific input to the network.
Claim 17
Independent Claim
202 words
An apparatus for identifying the structure and estimating the parameters of a controlled system comprising: structure identifier means for determining the structure of a function which describes the mathematical relationship of the inputs to a system and the outputs from said system; communication means for transferring the determined function structure and generating the function, the relationship and the parameters of which describes said mathematical relationship in terms of said parameters; parameter estimator means for receiving said function from said communication means and for estimating said parameters' value; and wherein said function is comprised of subfunctions, each subfunction determining the relationship of a set of zero or more parameters to said function, and wherein said communication means employs a set of switches having two logical positions, wherein each switch in one position turns off and in the other position turns on the subfunction of one of said parameter sets within said function, and wherein the position of the switches is determined by output received from said structure identifier so that the position of the set of logical switches operates as a communication of the identified system structure by a connection of an output of said set of switches to said parameter estimator means.
Claim 18
Independent Claim
202 words
An apparatus for identifying the structure and estimating the parameters of a controlled system comprising: structure identifier means for determining the structure of a function which describes the mathematical relationship of the inputs to a system and the outputs from said system; communication means for transferring the determined function structure and generating the function, the relationship and the parameters of which describes said mathematical relationship in terms of said parameters; parameter estimator means for receiving said function from said communication means and for estimating said parameters' value; and wherein said function is comprised of subfunctions, each subfunction determining the relationship of a set of zero or more parameters to said function, and wherein said communication means employs a set of switches having two logical positions, wherein each switch in one position turns off and in the other position turns on the subfunction of one of said parameter sets within said function, and wherein the position of the switches is determined by output received from said structure identifier so that the position of the set of logical switches operates as a communication of the identified system structure by a connection of an output of said set of switches to said parameter estimator means.
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) | 2001-10-19 | 2002-04-20 | 2002-10-20 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 2005-10-19 | 2006-04-20 | 2006-10-20 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2009-10-19 | 2010-04-20 | 2010-10-20 | $7700.00 | Window Closed / Lapsed |