# US 2023/0408476 A1 · Apparatuses, computer-implemented methods, and computer program products for accurate explosion predicting and warning

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

- Number: US 2023/0408476 A1
- Document ID: US20230408476A1
- Status: Published application
- Kind: A1
- Application number: 17/664,802
- Filed: 2022-05-24
- Published: 2023-12-21
- Earliest claimed: 2022-05-24
- Assignees: Honeywell International Inc.
- Applicants: Honeywell International Inc.
- Inventors: Agnel Anto; Surya Lakshmi Subba Rao Pilla; Priyanka Sanjay Vispute
- Web page: https://patentgrep.com/patent/US20230408476A1

## CPC

- G01N 33/0075: {for multiple spatially distributed sensors, e.g. for environmental monitoring}
- G08B 21/16: Combustible gas alarms
- G01N 33/0063: {using a threshold to release an alarm or displaying means}
- G06F 18/2415: based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
- G06F 18/251: {of input or preprocessed data}
- G08B 31/00: Predictive alarm systems characterised by extrapolation or other computation using updated historic data
- G01N 33/0068: {using a computer specifically programmed}
- G08B 17/125: {by using a video camera to detect fire or smoke}
- G01N: INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES (measuring or testing apparatus or processes other than immunoassay, involving enzymes or microorganisms C12M, C12Q)
- G08B: SIGNALLING SYSTEMS, e.g. PERSONAL CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

Embodiments utilize captured data, such as gas data and/or flame/heat data, from sensors in an environment to generate a data-constructed image for use in predicting explosion likelihood within an environment. Some embodiments utilize gas and flame data to generate the data-constructed image that is processable via one or more model(s) to determine whether the environment includes one or more sub-regions at risk of explosion. Some embodiments receive a plurality of gas sensor data and a plurality of flame sensor data, generate a data-constructed image including a plurality of channels based at least in part on such data, and generate explosion prediction data by applying at least a portion of the data-constructed image to a prediction model.

## Claims

### Claim 1

A computer-implemented method comprising:
receiving a plurality of gas sensor data and a plurality of flame sensor data;
generating a data-constructed image comprising a plurality of channels, the plurality of channels comprising at least a first channel assigned based at least in part on the plurality of gas sensor data and at least one additional channel assigned based at least in part on the plurality of flame sensor data; and
generating explosion prediction data by applying at least a portion of the data-constructed image to a prediction model,
wherein the prediction model generates the explosion prediction data based at least in part on explosion feature data determined from at least the plurality of channels corresponding to at least a portion of the data-constructed image.

### Claim 2 (depends on 1)

The computer-implemented method of claim 1, the computer-implemented method further comprising:
applying the data-constructed image to a computer vision model that identifies at least the portion of the data-constructed image determined associated with at least one explosion contribution level indicating presence of at least one explosion contributing factor in an environment; and
extracting the portion of the data-constructed image from the data-constructed image, wherein the prediction model only processes the extracted portion of the data-constructed image.

### Claim 3 (depends on 1)

The computer-implemented method of claim 1, wherein the plurality of gas sensor data comprises at least a first gas sensor data portion and a second gas sensor data portion captured via a first gas sensor, and wherein the plurality of gas sensor data comprises a first gas sensor data portion associated with a first sampling rate, and wherein the plurality of flame sensor data comprises at least a first flame sensor portion captured via a first flame sensor associated with a second sampling rate, wherein the first sampling rate is faster than the second sampling rate, and wherein generating the data-constructed image comprises:
generating an averaged value by averaging the first gas sensor data portion and the second gas sensor data portion; and
assigning at least a first pixel of the first channel based at least in part on the averaged value.

### Claim 4 (depends on 1)

The computer-implemented method of claim 1, wherein the plurality of gas sensor data comprises at least a first gas sensor data portion captured via a first gas sensor, and wherein the plurality of gas sensor data comprises a first gas sensor data portion associated with a first sampling rate, and wherein the plurality of flame sensor data comprises at least a first flame sensor portion and a second flame sensor portion captured via a first flame sensor associated with a second sampling rate, wherein the second sampling rate is faster than the first sampling rate, and wherein generating the data-constructed image comprises:
generating an averaged value by averaging the first flame sensor data portion and the second flame sensor data portion; and
assigning at least a first pixel of an additional channel of the plurality of additional channels based at least in part on the averaged value.

### Claim 5 (depends on 1)

The computer-implemented method of claim 1, wherein the plurality of gas sensor data comprises a first time series of gas sensor data portions captured via at least one gas sensor, and wherein generating the data-constructed image comprises:
assigning a first pixel value of the first channel based at least in part on a first gas sensor data portion of the first time series of gas sensor data portions corresponding to a first timestamp; and
assigning each subsequent pixel value of the first channel based at least in part on a next gas sensor data portion associated with each subsequent timestamp.

### Claim 6 (depends on 1)

The computer-implemented method of claim 1, wherein the plurality of flame sensor data comprises at least first band range data, second band range data, and third band range data, wherein the at least one additional channel comprises a second channel, a third channel, and a fourth channel, and wherein generating the data-constructed image comprises:
assigning the second channel based at least in part on the first band range data;
assigning the third channel based at least in part on the second band range data; and
assigning the fourth channel based at least in part on the third band range data.

### Claim 7 (depends on 1)

The computer-implemented method of claim 1, the computer-implemented method further comprising:
applying the data-constructed image to a computer vision model that at least determines an explosion contribution level; and
determining the explosion contribution level satisfies a threshold,
wherein the generating the explosion likelihood data is initiated in response to determining that the explosion contribution level satisfies the threshold.

### Claim 8 (depends on 1)

The computer-implemented method of claim 1, wherein the plurality of gas sensor data is collected via a plurality of gas sensors.

### Claim 9 (depends on 1)

The computer-implemented method of claim 1, wherein the plurality of flame sensor data is collected via a plurality of flame sensors.

### Claim 10 (depends on 1)

The computer-implemented method of claim 1, wherein the plurality of gas sensor data comprises at least a first gas data portion associated with a first gas sensor corresponding to a first environment region and a second gas data portion associated with a second gas sensor corresponding to a second environment region, and wherein the plurality of flame sensor data comprises a first flame data portion associated with a first flame sensor corresponding to the first environment region and a second flame data portion associated with a second flame data portion corresponding to the second environment region, wherein generating the data-constructed image comprises:
generating a first sub-image corresponding to the first environment region based at least in part on the first gas data portion and the first flame data portion;
generating a second sub-image corresponding to the second environment region based at least in part on the second gas data portion and the second flame data portion; and
generating the data-constructed image by assigning a first portion of the data-constructed image to the first sub-image and assigning a second portion of the data-constructed image to the second sub-image.

### Claim 11 (depends on 1)

The computer-implemented method of claim 1, wherein the data-constructed image comprises a plurality of sub-image, each sub-image corresponding to an assigned pixel sub-region of the data-constructed image.

### Claim 12 (depends on 1)

The computer-implemented method of claim 1, wherein the explosion prediction data comprises a data value indicating a probability of an explosion.

### Claim 13 (depends on 1)

The computer-implemented method of claim 1, the computer-implemented method further comprising:
determining the explosion prediction data satisfies a threshold by at least comparing the explosion prediction data to the threshold; and
in response to determining the explosion prediction data satisfies the threshold, generating a warning signal.

### Claim 14 (depends on 1)

The computer-implemented method of claim 1, wherein the prediction model comprises a specially trained machine learning model.

### Claim 15

A computing apparatus comprising:
at least one processor; and
at least one memory storing instructions that, when executed by the at least one processor, configure the apparatus to:
receive a plurality of gas sensor data and a plurality of flame sensor data;
generate a data-constructed image comprising a plurality of channels, the plurality of channels comprising at least a first channel assigned based at least in part on the plurality of gas sensor data and at least one additional channel assigned based at least in part on the plurality of flame sensor data; and
generate explosion prediction data by applying at least a portion of the data-constructed image to a prediction model,
wherein the prediction model generates the explosion prediction data based at least in part on explosion feature data determined from at least the plurality of channels corresponding to at least a portion of the data-constructed image.

### Claim 16 (depends on 15)

The apparatus of claim 15, wherein the plurality of flame sensor data comprises at least first band range data, second band range data, and third band range data, wherein the at least one additional channel comprises a second channel, a third channel, and a fourth channel, and wherein to generate the data-constructed image the apparatus is configured to:
assign the second channel based at least in part on the first band range data;
assign the third channel based at least in part on the second band range data; and
assign the fourth channel based at least in part on the third band range data.

### Claim 17 (depends on 15)

The apparatus of claim 15, the instructions further configure the apparatus to:
apply the data-constructed image to a computer vision model that at least determines an explosion contribution level; and
determine the explosion contribution level satisfies a threshold,
wherein the generating the explosion likelihood data is initiated in response to determining that the explosion contribution level satisfies the threshold.

### Claim 18 (depends on 15)

The computing apparatus of claim 15, wherein the plurality of gas sensor data comprises at least a first gas data portion associated with a first gas sensor corresponding to a first environment region and a second gas data portion associated with a second gas sensor corresponding to a second environment region, and wherein the plurality of flame sensor data comprises a first flame data portion associated with a first flame sensor corresponding to the first environment region and a second flame data portion associated with a second flame data portion corresponding to the second environment region, wherein generating the data-constructed image comprises:
generate a first sub-image corresponding to the first environment region based at least in part on the first gas data portion and the first flame data portion;
generate a second sub-image corresponding to the second environment region based at least in part on the second gas data portion and the second flame data portion; and
generate the data-constructed image by assigning a first portion of the data-constructed image to the first sub-image and assigning a second portion of the data-constructed image to the second sub-image.

### Claim 19

A non-transitory computer-readable storage medium, the computer-readable storage medium including instructions that when executed by at least one processor, cause the at least one processor to:
receive a plurality of gas sensor data and a plurality of flame sensor data;
generate a data-constructed image comprising a plurality of channels, the plurality of channels comprising at least a first channel assigned based at least in part on the plurality of gas sensor data and at least one additional channel assigned based at least in part on the plurality of flame sensor data; and
generate explosion prediction data by applying at least a portion of the data-constructed image to a prediction model,
wherein the prediction model generates the explosion prediction data based at least in part on explosion feature data determined from at least the plurality of channels corresponding to at least a portion of the data-constructed image.

### Claim 20 (depends on 19)

The computer-readable storage medium of claim 19, wherein the plurality of gas sensor data comprises at least a first gas data portion associated with a first gas sensor corresponding to a first environment region and a second gas data portion associated with a second gas sensor corresponding to a second environment region, and wherein the plurality of flame sensor data comprises a first flame data portion associated with a first flame sensor corresponding to the first environment region and a second flame data portion associated with a second flame data portion corresponding to the second environment region, wherein generating the data-constructed image comprises:
generate a first sub-image corresponding to the first environment region based at least in part on the first gas data portion and the first flame data portion;
generate a second sub-image corresponding to the second environment region based at least in part on the second gas data portion and the second flame data portion; and
generate the data-constructed image by assigning a first portion of the data-constructed image to the first sub-image and assigning a second portion of the data-constructed image to the second sub-image.
