The present invention relates to monitoring oil well site development and operations.
As described in commonly assigned U.S. Pat. No. 8,717,434, which is incorporated herein by reference, liquid energy commodities, such as crude oil, comprise a multi-billion dollar economic market. These commodities are bought and sold by many parties, and as with any traded market, information about the traded commodities is very valuable to market participants. Specifically, the operations of facilities associated with the production, transportation, storage, and distribution of these commodities can have significant impacts on the price and availability of these commodities, making information about said operations valuable. Furthermore, such information generally is not disclosed publicly by the various component owners or operators, and access to said information is therefore limited.
One type of information that is of interest to those involved in the trading of liquid energy commodities is the status of development of oil wells, including oil wells in a particular region or oil wells of a particular owner/operator. As mentioned above, such information generally is not disclosed publicly by the owners/operators. Furthermore, in many oil-producing developing countries, to the extent any information about new oil well drilling activity is available at all, it is usually inadequate, error-prone, and/or incomplete. This is particularly true in “conflict countries,” i.e., developing countries with unstable governments and some degree of internal violent conflict.
The present invention is a method and system for monitoring oil well development and operations.
In development of an oil well, there are a number of steps or milestones. Various of these steps or milestones can be identified and monitored via an analysis of satellite (or similar) imagery of the oil well site.
In an exemplary implementation of the present invention, the method commences by defining an area of interest. This area of interest may be defined by geographic coordinates or other criteria. Once the area of interest has been defined, the next step is to identify (or generate) a list of known or potential oil well locations in the area of interest. The list of such identified oil well locations is preferably stored in a database for subsequent monitoring.
For each oil well location that is identified, satellite (or similar) imagery is received and analyzed to determine if construction has commenced at the site of the oil well, i.e., to identify and confirm the presence of a cleared area of land. Various image processing techniques can be employed to identify this change (i.e., ground clearing and construction of an access road) across a series of received images in order to confirm the commencement of the construction of the oil well. For example, in one preferred image processing technique, a baseline image is first received and stored. Then, as new images are received, they are aligned and compared with the baseline image, and differences between the respective pixels in the image can be used to identify and confirm the presence of a cleared area of land.
Subsequent satellite (or similar) imagery is then analyzed to identify and confirm the arrival of a drilling rig at the site, and then the departure of the drilling rig from the site. Again, various image processing techniques can be employed to identify this change (i.e., arrival or departure of the drilling rig) across a series of received images. For example, with respect to the preferred image processing technique described above for confirmation that construction of an oil well has commenced, the same techniques can be applied to identify a drilling rig and associated equipment. For another example, objects on the drilling pad can be detected by using a machine learning model, such as a trained convolutional neural network, using satellite imagery from other similar sites.
After departure of the drilling rig from the site, subsequent satellite (or similar) imagery may be analyzed to determine if and when construction of the oil well has been completed. For example, with respect to determining if and when construction of the oil well has been completed, the presence of certain equipment (such as frac pumping trucks) can provide such confirmation. For another example, with respect to such a determination as to if and when construction of the oil well has been completed, short-wave infrared imagery collected at night may be used to identify natural gas flaring that is indicative of an oil well that is producing oil.
Subsequent satellite (or similar) imagery is then analyzed to determine if and when production facilities have been built at the site of the oil well. Again, various image processing techniques can be employed to identify this change (i.e., construction of production facilities at the site of the oil well) across a series of received images. For example, a machine learning model can be trained and applied as described above, but with images in the training set to be taken from other sites at times before and after installation of relevant production facilities.
The final result of the above-described analysis is an identification of a completed oil well (or wells). That information is then communicated to third-party market participants and other interested parties, e.g., third parties who would not ordinarily have ready access to such information. Of course, as information about the status of the development of each oil well is determined at each analysis step, that information can also be communicated to third-party market participants and other interested parties.
The present invention is a method and system for monitoring oil well development and operations.
As illustrated in
As also illustrated in
Referring now to
Referring still to
Referring still to
For example, in one preferred image processing technique, a baseline image is first received and stored. Then, as new images are received, the baseline image is aligned to each new image via geo-coordinates that are included in such images via metadata, including the minimum and maximum latitude and longitude boundaries. Once the images are aligned, a difference between the two images is calculated, where each aligned pixel is converted from having red, green, and blue channels into a color space which is comprised of a pixel intensity channel and two color channels. In this regard, there are a number of color spaces which can be used, but a preferred color space is the Lab color space, which mathematically describes all perceivable colors in the three dimensions: L for lightness, and a and b for the color opponents green-red and blue-yellow. Then, the Euclidean distance between each pixel in the color space is calculated. A threshold is determined beforehand based on the area being searched, and the image is thresholded for all of the pixels in the Euclidean distance, such that any pixels that are above the threshold are considered one, and any pixels that are below the threshold are considered a difference. This result can be noisy, so in order to find a large area change, blob detection is performed, such that all pixels which are one and are adjacent to another pixel which has a value of one are grouped together and considered “blobs.” An example of the result of such a blob detection, which is a binary image, is shown in
Furthermore, in some cases, oil wells may not have been previously identified if and until an analysis of images of certain geographic locations results in the identification of ground clearing and construction of an access road at one or more sites. However, after such an identification of ground clearing and construction of an access road, the site can be subsequently monitored, e.g., by storing that site in the database 160 for subsequent monitoring.
Referring again to
For example, with respect to the preferred image processing technique described above for confirmation that construction of an oil well has commenced, the same techniques can be applied to identify a drilling rig and associated equipment. For another example, objects on the drilling pad can be detected by using a machine learning model, such as a trained convolutional neural network, using satellite imagery from other similar sites. These images are chosen using publicly available data during the appropriate time window, e.g., times after pad construction but before drilling to train the model for no drilling activity, and times after drilling has begun but before the rig leaves the site to train the machine learning model for drilling activity. Widely available software libraries can be used to build and train the models such as: TensorFlow™, which is an open source machine learning framework for high performance numerical computation available from Google LLC of Mountain View, Calif.; and/or Keras, which is a high-level neural networks application program interface (API) that runs on top of TensorFlow™ or other machine learning framework.
Referring again to
Referring again to
Referring again to
Referring again to
As an additional refinement, information derived from the above-described analysis can be combined with other sources of data to generate production forecasts for wells in a particular region or wells of a particular owner/operator, and then communicated to third-party market participants and other interested parties.
The above-described operational and analysis steps of this method are preferably achieved through the use of a digital computer program (i.e., computer-readable instructions executed by a processor of a computer) that includes appropriate modules for executing the requisite instructions (which are stored in a memory component of the computer). Thus, an exemplary system for monitoring oil well development in accordance with the present invention includes: (a) an imagery-receiving module for receiving imagery for an area of interest, which is then stored in a database; (b) an analysis module for analyzing the imagery to (i) identify an oil well location in the area of interest and/or (ii) identify one or more milestones associated with development of an oil well; and (c) a communications module for communicating information about the development of the oil well to an interested party.
One of ordinary skill in the art will recognize that additional embodiments and implementations are also possible without departing from the teachings of the present invention. This detailed description, and particularly the specific details of the exemplary embodiments and implementations disclosed therein, is given primarily for clarity of understanding, and no unnecessary limitations are to be understood therefrom, for modifications will become obvious to those skilled in the art upon reading this disclosure and may be made without departing from the spirit or scope of the invention.
The present application claims priority to U.S. Patent Application Ser. No. 62/565,493 filed on Sep. 29, 2017, which is incorporated herein by reference.
Number | Date | Country | |
---|---|---|---|
62565493 | Sep 2017 | US |