Unlit flare detection using satellite images
A method can include receiving data that include satellite data of a region of interest where the region of interest includes multiple hydrocarbon production sites that include gas flaring equipment, identifying one or more unlit gas flares at one or more of the multiple hydrocarbon production sites by using a trained machine learning model and at least a portion of the data and, for an unlit gas flare, issuing an instruction related to operation of the gas flaring equipment in the region of interest.
1 . A method comprising:
receiving data that comprise satellite data of a region of interest that includes multiple hydrocarbon production sites that comprise gas flaring equipment, wherein the data includes weather data for the region of interest;
processing at least a portion of the satellite data using the weather data for the region of interest, wherein the processing includes filtering based on atmospheric conditions indicated by the weather data for the region of interest for reducing false identification of an unlit gas flare state at one or more of the multiple hydrocarbon production sites;
identifying one or more unlit gas flares at the one or more of the multiple hydrocarbon production sites using a trained machine learning model and at least a portion of the data, wherein the trained machine learning model is trained using at least spatial and temporal satellite data to classify gas flares in the region of interest; and
based at least in part on the identifying, issuing at least one instruction related to operation of the gas flaring equipment in the region of interest.
2 . The method of claim 1 , further comprising:
identifying a lit gas flare; and
classifying the lit gas flare as an intermittent gas flare or a continuous gas flare using the trained machine learning model.
3 . The method of claim 1 , further comprising comparing the one or more unlit gas flares to historic data for the one or more unlit gas flares to determine one or more occurrences of a change from a lit state to an unlit state.
4 . The method of claim 1 , wherein the receiving, the identifying, and the issuing are performed automatically without manual intervention.
5 . The method of claim 1 , wherein the trained machine learning model comprises a convolutional neural network.
6 . The method of claim 1 , wherein the identifying one or more unlit gas flares comprises utilizing location data of previously lit flares.
7 . The method of claim 1 , wherein the data comprise data from a plurality of different satellites.
8 . The method of claim 1 , further comprising, based at least in part on the identifying, estimating methane emissions at the one or more of the multiple hydrocarbon production sites.
9 . The method of claim 1 , wherein the data comprise night time data and day time data.
10 . The method of claim 1 , further comprising determining a reason for an unlit flare being unlit at one of the multiple hydrocarbon production sites.
11 . The method of claim 1 , wherein the data comprise transportation data and detecting a change in gas transportation.
12 . The method of claim 1 , wherein the data comprise one or more of gas production chain data, gas supply chain data, and gas utilization data.
13 . The method of claim 1 , further comprising rendering a graphical user interface to a display that comprises an indicator for at least one of the identified one or more unlit flares.
14 . The method of claim 1 , wherein the data comprise data indicative of smoke generated by one or more gas flaring operations.
15 . The method of claim 1 , further comprising:
forming a composite image using a series of images of the region of interest; and
utilizing the composite image in training a machine learning model to generate the trained machine learning model with increased sensitivity for flare identification in the region of interest.
16 . A system comprising:
a processor;
a memory accessible by the processor; and
processor-executable instructions stored in the memory that are executable to instruct the system to:
receive data that comprise satellite data of a region of interest that includes multiple hydrocarbon production sites that comprise gas flaring equipment, wherein the data includes weather data for the region of interest;
process at least a portion of the satellite data using the weather data for the region of interest, wherein the processing includes filtering based on atmospheric conditions indicated by the weather data for the region of interest for reducing false identification of an unlit gas flare state at one or more of the multiple hydrocarbon production sites;
identify one or more unlit gas flares at the one or more of the multiple hydrocarbon production sites using a trained machine learning model and at least a portion of the data, wherein the trained machine learning model is trained using at least spatial and temporal satellite data to classify gas flares in the region of interest; and
based at least in part on identification of the one or more unlit gas flares, issue at least one instruction related to operation of the gas flaring equipment in the region of interest.
17 . One or more non-transitory computer-readable storage media comprising computer-executable instructions executable to instruct a computer to:
receive data that comprise satellite data of a region of interest that includes multiple hydrocarbon production sites that comprise gas flaring equipment, wherein the data includes weather data for the region of interest;
process at least a portion of the satellite data using the weather data for the region of interest, wherein the processing includes filtering based on atmospheric conditions indicated by the weather data for the region of interest for reducing false identification of an unlit gas flare state at one or more of the multiple hydrocarbon production sites;
identify one or more unlit gas flares at the one or more of the multiple hydrocarbon production sites using a trained machine learning model and at least a portion of the data, wherein the trained machine learning model is trained using at least spatial and temporal satellite data to classify gas flares in the region of interest; and
based at least in part on identification of the one or more unlit gas flares, issue at least one instruction related to operation of the gas flaring equipment in the region of interest.