Apparatus and method for system error monitoring

Patent metadata

Field	Value
Publication number	US-9798699-B2
Application number	US-201414526891-A
Country	US
Kind code	B2
Filing date	Oct 29, 2014
Priority date	Dec 20, 2013
Publication date	Oct 24, 2017
Grant date	Oct 24, 2017

Abstract

An information processing method for system identification includes: generating a fitting curve represented by a sum of exponential functions for each of a set of digital inputs and a set of digital outputs for a physical system that is represented by one or plural equations including m-order differential operators (m is an integer equal to or greater than 1); and calculating coefficients of the differential operators, which are included in first coefficients, so that a first coefficient of each exponential function included in an expression obtained by a product of the differential operators and the fitting curve for the set of the digital inputs is equal to a second coefficient of the same exponential function, which is included in the fitting curve for the set of the digital outputs.

First claim

What is claimed is:

1. An information processing apparatus for prediction of system outputs, comprising:
a memory; and
a processor coupled to the memory and configured to:
generate a first fitting curve represented by a sum of exponential functions for a set of first digital inputs measured by a first sensor and a second fitting curve represented by a sum of exponential functions for a set of first digital outputs measured by a second sensor, for a physical system that is represented by one or a plurality of equations including m-order differential operators, wherein m is an integer equal to or greater than 1;
calculate coefficients of the differential operators, which are included in first coefficients, so that a first coefficient of each exponential function included in an expression obtained by a product of the differential operators and the first fitting curve for the set of the first digital inputs is equal to a second coefficient of a same exponential function, which is included in the second fitting curve for the set of the first digital outputs; and
predict second digital outputs of the physical system based on an output function obtained from the second fitting curve and coefficients of the output function, which are calculated from the coefficients of the differential operators and second digital inputs of the physical system.

2. The information processing apparatus as set forth in claim 1 , wherein the calculating comprises:
generating a plurality of combinations of a third coefficient and a difference between the second coefficient and a constant term, for which a coefficient of the differential operators is not multiplied, among the first coefficients, wherein the third coefficient is a coefficient, for which a coefficient of the differential operators is multiplied, among the first coefficients; and
performing linear regression calculation for the plurality of combinations to calculate the coefficients of the differential operators.

3. The information processing apparatus as set forth in claim 1 , wherein the first coefficient is a product of a value of a differential operator for a coefficient in an exponent of a first exponential function corresponding to the first coefficient and a coefficient of a same exponential function as the first exponential function, which is included in the first fitting curve for the set of the first digital inputs.

4. The information processing apparatus as set forth in claim 1 , wherein the predicting comprises:
generating a third fitting curve represented by a sum of exponential functions for a set of the second digital inputs of the physical system;
calculating, for each exponential function of the third fitting curve, a third coefficient of an exponential function in the output function, by using a coefficient of the exponential function and a value of a differential operator, which includes a coefficient of the differential operator for a coefficient in an exponent of the exponential function; and
calculating digital output values from the output function including calculated third coefficients of the output function.

5. The information processing apparatus as set forth in claim 1 , wherein the generating comprises:
generating the first fitting curve for the set of the first digital inputs by calculating n+1 frequencies by performing discrete Fourier transformation for 2n+1 digital inputs and generating n-order Fourier series from the n+1 frequencies, wherein n is an integer; and
generating the second fitting curve for the set of the first digital outputs by calculating n+1 frequencies by performing discrete Fourier transformation for 2n+1 digital outputs and generating n-order Fourier series from the n+1 frequencies.

6. The information processing apparatus as set forth in claim 5 , wherein the generating comprises:
calculating l+1 frequencies by performing discrete Fourier transformation for 2l+1 third digital inputs, and generate l-order Fourier series from the l+1 frequencies to generate a fourth fitting curve represented by a sum of exponential functions for the third digital inputs; and
calculating the 2n+1 digital inputs from the generated fourth fitting curve.

7. The information processing apparatus as set forth in claim 1 , wherein the generating comprises:
generating a first curve by performing differential or integral for the first fitting curve for the set of the first digital inputs; and
generating fourth digital inputs by using the generated first curve.

8. A non-transitory computer-readable storage medium storing a program that causes a computer to execute a process for prediction of system outputs, the process for the prediction of the system outputs comprising:
generating a first fitting curve represented by a sum of exponential functions for a set of first digital inputs measured by a first sensor and a second fitting curve represented by a sum of exponential functions for a set of first digital outputs measured by a second sensor, for a physical system that is represented by one or a plurality of equations including m-order differential operators, wherein m is an integer equal to or greater than 1;
calculating coefficients of the differential operators, which are included in first coefficients, so that a first coefficient of each exponential function included in an expression obtained by a product of the differential operators and the first fitting curve for the set of the first digital inputs is equal to a second coefficient of a same exponential function, which is included in the second fitting curve for the set of the first digital outputs; and
predicting second digital outputs of the physical system based on an output function obtained from the second fitting curve and coefficients of the output function, which are calculated from the coefficients of the differential operators and second digital inputs of the physical system.

9. An information processing method for prediction of system outputs, comprising:
generating, by using a computer, a first fitting curve represented by a sum of exponential functions for a set of first digital inputs measured by a first sensor and -a second fitting curve represented by a sum of exponential functions for a set of first digital outputs measured by a second sensor, for a physical system that is represented by one or a plurality of equations including m-order differential operators, wherein in is an integer equal to or greater than 1;
calculating, by using the computer, coefficients of the differential operators, which are included in first coefficients, so that a first coefficient of each exponential function included in an expression obtained by a product of the differential operators and the first fitting curve for the set of the first digital inputs is equal to a second coefficient of a same exponential function, which is included in the second fitting curve for the set of the first digital outputs; and
predicting, by using the computer, second digital outputs of the physical system based on an output function obtained from the second fitting curve and coefficients of the output function, which are calculated from the coefficients of the differential operators and second digital inputs of the physical system.

Assignees

• Fujitsu Ltd

Inventors

• Ito Toshio

Classifications

• G06F2218/02

  Preprocessing · CPC title

• G06F17/13 Primary

  Differential equations (using digital differential analysers G06F7/64) · CPC title

• G07C5/0808

  Diagnosing performance data (testing of vehicles G01M17/00; testing of electrical installation on vehicles G01R31/005) · CPC title

• G06K9/00503

  Physics · mapped topic

• G06F17/12 Primary

  Simultaneous equations {, e.g. systems of linear equations} · CPC title