# US 2024/0176932 A1 · Continuous online validation of process measurements in digital twins

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2024/0176932 A1
- Document ID: US20240176932A1
- Status: Published application
- Kind: A1
- Application number: 18/071,243
- Filed: 2022-11-29
- Published: 2024-05-30
- Earliest claimed: 2022-11-29
- Assignees: Saudi Arabian Oil Company
- Applicants: Saudi Arabian Oil Company
- Inventors: Hassan Mohammadali Al Matouq; Mohamed Ahmed Soliman; Rida Majed Bin Hashim
- Web page: https://patentgrep.com/patent/US20240176932A1

## CPC

- G06F 30/27: using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

Systems and methods include a method for monitoring sensors. Readings of actual field sensors corresponding to virtual sensors of a virtual model are monitored. A deviation between an actual sensor reading (ASR) and a virtual sensor reading (VSR) is determined. When only the VSR is deviating and when sensor readings of sensors proximate to the actual sensor correspond to the VSR, then the ASR is invalid and the actual sensor requires revalidation/recalibration. When only the ASR is deviating and when sensor readings of sensors proximate to the actual sensor correspond to the ASR, then the VSR is invalid and the virtual sensor requires revalidation/recalibration. When both the VSR and ASR are deviating from design ranges and when sensor readings of sensors that are proximate to the actual sensor do not correspond to either the ASR or the VSR, then both the ASR and VSR are invalid, requiring revalidation/recalibration.

## Claims

### Claim 1

A computer-implemented method, comprising:
monitoring readings of actual field sensors corresponding to virtual sensors modeled in a virtual model;
determining a deviation between an actual sensor reading (ASR) and a virtual sensor reading (VSR);
when only the VSR is deviating and when sensor readings of sensors that are proximate to the actual sensor correspond to the VSR, then:
determining that the ASR is invalid and that the actual sensor requires revalidation and recalibration;
when only the ASR is deviating and when sensor readings of sensors that are proximate to the actual sensor correspond to the ASR, then:
determining that the VSR is invalid and that the virtual sensor requires revalidation and recalibration; and
when both of the VSR and the ASR are deviating from their respective design ranges and when sensor readings of sensors that are proximate to the actual sensor do not correspond to either the ASR or the VSR, then:
determining that both the ASR and the VSR are invalid and that both the actual sensor and the virtual sensor require revalidation and recalibration.

### Claim 2 (depends on 1)

The computer-implemented method of claim 1, wherein the actual field sensors include temperature sensors, pressure sensors, and flow sensors.

### Claim 3 (depends on 1)

The computer-implemented method of claim 1, wherein determining the deviation between the ASR and the VSR includes performing calculations using instantaneous sensor readings.

### Claim 4 (depends on 1)

The computer-implemented method of claim 1, wherein determining the deviation between the ASR and the VSR is based on average deviations of historical readings from both types of sensors over a designated time period.

### Claim 5 (depends on 1)

The computer-implemented method of claim 1, wherein determining that the sensor readings of the sensors that are proximate to the actual sensor correspond to the sensor includes using a set of validation rules.

### Claim 6 (depends on 5)

The computer-implemented method of claim 5, wherein the set of validation rules includes mathematical functions based on inequalities between the sensors.

### Claim 7 (depends on 1)

The computer-implemented method of claim 1, wherein determining that the sensor readings of the sensors that are proximate to the actual sensor correspond to the sensor includes using a table correlating the sensors, wherein the table is generated from an offline simulation model based on process design conditions of multiple sensors.

### Claim 8 (depends on 1)

The computer-implemented method of claim 1, wherein determining that the sensor readings of the sensors that are proximate to the actual sensor correspond to the sensor includes using a machine learning model correlating each sensor with its related sensors.

### Claim 9 (depends on 1)

The computer-implemented method of claim 1, wherein determining that only the VSR is deviating includes determining that the deviation D in
D
=
V
⁢
S
⁢
R
-
A
⁢
S
⁢
R
V
⁢
S
⁢
R
*
1
⁢
0
⁢
0
is greater than a threshold positive deviation and that changes in ASR previous values over a specified period of time are below an ASR change threshold.

### Claim 10 (depends on 1)

The computer-implemented method of claim 1, wherein determining that only the ASR is deviating includes determining that the deviation D in
D
=
V
⁢
S
⁢
R
-
A
⁢
S
⁢
R
V
⁢
S
⁢
R
*
1
⁢
0
⁢
0
is less than a threshold negative deviation and that changes in VSR previous values over a specified period of time are below a VSR change threshold.

### Claim 11

A non-transitory, computer-readable medium storing one or more instructions executable by a computer system to perform operations comprising:
monitoring readings of actual field sensors corresponding to virtual sensors modeled in a virtual model;
determining a deviation between an actual sensor reading (ASR) and a virtual sensor reading (VSR);
when only the VSR is deviating and when sensor readings of sensors that are proximate to the actual sensor correspond to the VSR, then:
determining that the ASR is invalid and that the actual sensor requires revalidation and recalibration;
when only the ASR is deviating and when sensor readings of sensors that are proximate to the actual sensor correspond to the ASR, then:
determining that the VSR is invalid and that the virtual sensor requires revalidation and recalibration; and
when both of the VSR and the ASR are deviating from their respective design ranges and when sensor readings of sensors that are proximate to the actual sensor do not correspond to either the ASR or the VSR, then:
determining that both the ASR and the VSR are invalid and that both the actual sensor and the virtual sensor require revalidation and recalibration.

### Claim 12 (depends on 11)

The non-transitory, computer-readable medium of claim 11, wherein determining that the sensor readings of the sensors that are proximate to the actual sensor correspond to the sensor includes using a table correlating the sensors, wherein the table is generated from an offline simulation model based on process design conditions of multiple sensors.

### Claim 13 (depends on 11)

The non-transitory, computer-readable medium of claim 11, wherein determining that the sensor readings of the sensors that are proximate to the actual sensor correspond to the sensor includes using a machine learning model correlating each sensor with its related sensors.

### Claim 14 (depends on 11)

The non-transitory, computer-readable medium of claim 11, wherein determining that only the VSR is deviating includes determining that the deviation D in
D
=
V
⁢
S
⁢
R
-
A
⁢
S
⁢
R
V
⁢
S
⁢
R
*
1
⁢
0
⁢
0
is greater than a threshold positive deviation and that changes in ASR previous values over a specified period of time are below an ASR change threshold.

### Claim 15 (depends on 11)

The non-transitory, computer-readable medium of claim 11, wherein determining that only the ASR is deviating includes determining that the deviation D in
D
=
V
⁢
S
⁢
R
-
A
⁢
S
⁢
R
V
⁢
S
⁢
R
*
1
⁢
0
⁢
0
is less than a threshold negative deviation and that changes in VSR previous values over a specified period of time are below a VSR change threshold.

### Claim 16

A computer-implemented system, comprising:
one or more processors; and
a non-transitory computer-readable storage medium coupled to the one or more processors and storing programming instructions for execution by the one or more processors, the programming instructions instructing the one or more processors to perform operations comprising:
monitoring readings of actual field sensors corresponding to virtual sensors modeled in a virtual model;
determining a deviation between an actual sensor reading (ASR) and a virtual sensor reading (VSR);
when only the VSR is deviating and when sensor readings of sensors that are proximate to the actual sensor correspond to the VSR, then:
determining that the ASR is invalid and that the actual sensor requires revalidation and recalibration;
when only the ASR is deviating and when sensor readings of sensors that are proximate to the actual sensor correspond to the ASR, then:
determining that the VSR is invalid and that the virtual sensor requires revalidation and recalibration; and
when both of the VSR and the ASR are deviating from their respective design ranges and when sensor readings of sensors that are proximate to the actual sensor do not correspond to either the ASR or the VSR, then:
determining that both the ASR and the VSR are invalid and that both the actual sensor and the virtual sensor require revalidation and recalibration.

### Claim 17 (depends on 16)

The computer-implemented system of claim 16, wherein determining that the sensor readings of the sensors that are proximate to the actual sensor correspond to the sensor includes using a table correlating the sensors, wherein the table is generated from an offline simulation model based on process design conditions of multiple sensors.

### Claim 18 (depends on 16)

The computer-implemented system of claim 16, wherein determining that the sensor readings of the sensors that are proximate to the actual sensor correspond to the sensor includes using a machine learning model correlating each sensor with its related sensors.

### Claim 19 (depends on 16)

The computer-implemented system of claim 16, wherein determining that only the VSR is deviating includes determining that the deviation D in
D
=
V
⁢
S
⁢
R
-
A
⁢
S
⁢
R
V
⁢
S
⁢
R
*
1
⁢
0
⁢
0
is greater than a threshold positive deviation and that changes in ASR previous values over a specified period of time are below an ASR change threshold.

### Claim 20 (depends on 16)

The computer-implemented system of claim 16, wherein determining that only the ASR is deviating includes determining that the deviation D in
D
=
V
⁢
S
⁢
R
-
A
⁢
S
⁢
R
V
⁢
S
⁢
R
*
1
⁢
0
⁢
0
is less than a threshold negative deviation and that changes in VSR previous values over a specified period of time are below a VSR change threshold.
