Method for evaluating the operating conditions of a machine or an installation

• US 7,194,381 B2
• Filed: 08/27/2003
• Issued: 03/20/2007
• Est. Priority Date: 08/30/2002
• Status: Expired due to Fees

• Current Assignee: Siemens AG
• Original Assignee: Siemens AG
• Inventors: Adam, Gottfried
• Primary Examiner(s): Barlow; John
• Assistant Examiner(s): KHUU, HIEN DIEU THI
• Application Number:
  US10/648,930
• Publication Number:
  US 20040073416A1
• Time in Patent Office: 1,301 Days
• Field of Search: 702/182
• US Class Current: 702/182
• CPC Class Codes: G01M 15/00 Testing of engines

Abstract

A method for evaluating and modifying the operating conditions of a machine or an installation includes measuring at least one parameter a number of times to create a database, of the parameter, determining an extrapolation range, performing an extrapolation in the range, and determining a measure of quality, K, of the extrapolation calculated on the basis of the database. K is a function of at least two variables. The parameter value is controlled so that the value will be retained within a limit.

Claims

1. A method for the evaluating and modifying the operating conditions of a machine or an installation, comprising the steps of:

measuring at least one parameter a number of times to create a database, which comprises values (x₁,y₁) to (x_n,y_n) of the parameter;

determining an extrapolation range, x_s;

performing an extrapolation in the range x_s;

determining a measure of quality, K, of the extrapolation calculated on the basis of the database wherein K is a function of at least two variables taken from the group consisting of V, Δ

I, S, C, with V being a ratio of the value range of the database to the extrapolation range, x_s which is determined by x_s>

x₁, x_n with Δ

I being the x uncertainty of the adjustment curve in the x direction, with S being continuity as a measure of the change in the y values in the database and with C being the time constancy of the extrapolation; and

controlling the parameter value so that the value will be retained within a limit,

wherein the evaluation of the operating conditions is used to optimize the operation of the machine or the installation.

2. A method according to claim 1, wherein the value of the parameter is controlled based on the measure of quality K.

3. A method according to claim 2, wherein the evaluation of the operating conditions increases the operational dependability of the machine or the installation, by influencing the parameter accordingly based on the measure of quality K.

4. A method according to claim 2, wherein evaluation of the operating conditions and controlling the parameter value optimizes operation of the machine or the installation.

5. A method according to claim 1, wherein the optimization is effective to increase the operational dependability of the machine or the installation, by influencing the parameter accordingly based on the measure of quality K.

6. A method according to claim 5, wherein a limit value is predetermined for the parameter and a period is determined in which the limit value of the parameter is not to be exceeded.

7. A method according to claim 1, wherein the variables are selected so that the measure of quality K does not depend on the gradient of an adjustment curve in respect of the database.

8. A method according to claim 1, wherein the measure of quality K is standardized, in particular by 1−

e<sup>−

K</sup>.

9. A method according to claim 8, wherein the measure of quality K is standardized to a value range of 0 to 100%.

10. A method according to claim 1, wherein the measure of quality K is defined by:

11. A method according to claim 9, wherein the ratio V of the value range of the database is defined by (x_l−

X₁)/(x_s−

x₁).

12. A method according to claim 1, wherein the ratio V of the value range of the database is defined by (x_n−

x₁)/(x_s−

x₁).

13. A method according to claim 1, wherein:

the database is divided into at least three segments;

a mean value g₁, g₂, g₃ and a linear adjustment function y₁, y₂, y₃ with gradients c₁, c₂ and c₃ are each calculated for each segment from the database;

a numerical curvature measure p, whereby p=g₁−

2*g₂+g₃ is calculated, which reflects the current direction of curvature of the gradient pattern;

from a curve repertoire of curve types at least of the group;

Linear function->

f(x)=y=a₀+a₁*x

Potency function->

f(x)=ln y=ln a₀+a₁*ln x

Logarithmic function->

f(x)=y=a₀+a₁*ln x

Exponential function->

f(x)=ln y=ln a₀+a₁x

 

that curve type of the adjustment function is selected iteratively and adjusted in respect of the value range of the entire current database, with the curve type selected from the curve repertoire having to satisfy the following conditions;

the direction of curvature of the curve, which is determined from the extrapolation, must correspond to that of p and the quotient Q_k of numerator and denominator must be minimal;

14. A method according to claim 13, wherein the value range of continuity S is in the range 0 and +∞

.

15. A method according to claim 1, wherein determination of the x uncertainty includes:

selection of an extrapolation function, which can be transferred to linear structures from the group consisting of

Linear function->

y=a₀+a₁*x

Potency function->

ln y=ln a₀+a₁*ln x

Logarithmic function->

y=a₀+a₁*ln x

Exponential function->

ln y=ln a₀+a₁*x,

 

determination of a database,

 

with the database comprising n correlated x and y values

 

Calculation of x and y of the database and the variable Σ

x_iy_i,

 

Calculation of

16. A method according to claim 1, wherein the measure of quality K changes over time and an evaluation of the time variance of the extent of extrapolation C is calculated as follows: