Predicting maintenance requirements for a valve assembly

• US 10,371,285 B2
• Filed: 10/27/2015
• Issued: 08/06/2019
• Est. Priority Date: 10/27/2015
• Status: Active Grant

• Current Assignee: Dresser, LLC (GE Aerospace)
• Original Assignee: Dresser, LLC (GE Aerospace)
• Inventors: Schoonover, Larry Gene, Shriver, Justin Scott
• Primary Examiner(s): Islam, Mohammad K
• Application Number:
  US14/924,352
• Publication Number:
  US 20170114927A1
• Time in Patent Office: 1,379 Days
• CPC Class Codes: F16K 37/0083 by measuring valve parameters
• CPC Class Codes: G05B 2219/33326 Analyzer, diagnostic for servovalve
• CPC Class Codes: G05B 23/0283 Predictive maintenance, e.g. involving the monitoring of a system and, based on the monitoring results, taking decisions on the maintenance schedule of the monitored system; Estimating remaining useful life [RUL] preventive maintenance, i.e. planning maintenance according to the available resources without monitoring the system G06Q10/06

Abstract

A system and method for predicting failure conditions on a valve assembly. The method can include stages for comparing a maintenance variable to a threshold criteria, each related to an operating feature that uses operating data to quantify operation of a valve assembly, the threshold criteria associating an operating pattern found in the operating data with a failure mode on the valve assembly. The method can also include stages for selecting a maintenance task according to a relationship between the maintenance variable and the threshold criteria, the maintenance task related to the failure mode. The method can further include stages for generating an output conveying the maintenance task.

Claims

1. A method, comprising:

acquiring operating data that describes operation of a valve assembly on a process line;

calculating amplitude for stick-slip cycling and amplitude for setpoint cycling from the operating data;

calculating a maintenance variable, wherein the maintenance variable includes a pattern coefficient that modifies a value of the amplitude for stick-slip cycling and the amplitude for setpoint cycling;

comparing the maintenance variable to a threshold criteria related to the amplitude for stick-slip cycling and the amplitude for setpoint cycling, the threshold criteria delimiting data into two groups including a first group where a combination of stick-slip cycling and setpoint cycling indicates that stick-slip prevails on the valve assembly and a second group where the combination of stick-slip cycling and setpoint cycling indicates cycling of the valve assembly that results from process variations occurring on the process line;

wherein the maintenance variable is calculated in accordance with,

M=C_iX_i+C_i+nX_i+n,

wherein M is the maintenance variable, Ci is a first pattern coefficient, Xi is the amplitude for stick slip cycling, Ci+n is a second pattern coefficient, Xi+n is the amplitude for setpoint cycling, and n is an integer having consecutive values;

selecting a maintenance task according to a relationship between the maintenance variable and the threshold criteria, wherein the maintenance task relates to stick-slip on the valve assembly in response to the combination of stick-slip cycling and setpoint cycling is in the first group that is indicative of cycling that occurs on the valve assembly itself;

using the maintenance task to maintain operation of the process line on which the valve assembly is installed by,

generating an output in the form of an alert that corresponds with corrective action to the valve assembly; and

causing the valve assembly to modulate operation of the valve to change flow through the valve assembly in response to a control signal that relates to the alert so as to avoid a failure condition on the valve assembly that would disrupt operation of the process line.

2. The method of claim 1, wherein the maintenance variable integrates more than one of the pattern coefficient.

3. The method of claim 1, wherein the maintenance variable takes the form of a linear equation.

4. The method of claim 1, further comprising:

updating the relationship between the maintenance variable and the threshold criteria in accordance with the pattern coefficient.

5. The method of claim 1, wherein the maintenance variable includes a pattern coefficient that modifies a value of the amplitude for stick-slip cycling and the amplitude for setpoint cycling.

6. The method of claim 1, wherein the maintenance variable integrates more than one of the pattern coefficient and more than one of the operating feature.

7. The method of claim 1, wherein the maintenance variable corresponds to one or more values for the amplitude for stick-slip cycling and the amplitude for setpoint cycling that defines performance of the valve assembly.

8. The method of claim 1, further comprising:

receiving data that defines the maintenance task that is performed on the valve assembly;

incorporating the data into a training set of data; and

updating the first pattern coefficient and the second pattern coefficient in accordance with the training set of data.