Patent Application
20240084681
The present disclosure relates generally to wellbore operations and, more particularly (although not necessarily exclusively), to cloud-based management of a hydraulic fracturing operation in a wellbore.
A wellbore can be formed in a subterranean formation for producing or extracting material from the subterranean formation. The material can include hydrocarbon material such as oil and gas. Various operations can be performed with respect to the wellbore. The various operations can include hydraulic fracturing operations. Hydraulic fracturing operations can enhance the extraction of materials, such as gas or oil, from rock formations in the subterranean formation, thereby increasing production of the wellbore. Hydraulic fracturing operations can be performed by pumping a hydraulic fluid into the wellbore at or above a pressure high enough to create or enhance fractures in the rock formations. Maximizing the efficiency of hydraulic fracturing operations can improve wellbore production and reduce completion costs.
Certain aspects and examples of the present disclosure relate to a cloud-based dashboard that can be supported by cloud architecture and used to control or manage a hydraulic fracturing operation. The cloud architecture can be a combination of elements for providing a cloud service. For example, the cloud architecture can include a front-end platform, a back-end platform or servers, an internet service, a cloud-based delivery service, additional elements, or a combination thereof. The cloud service can be a variety of services delivered to a company, a customer, another suitable entity, or a combination thereof over the internet. For example, the cloud service can provide a cloud data warehouse to integrate and store data for further data analysis. The front-end platform of the cloud architecture can be the cloud-based dashboard, which can display relevant quantities for controlling, adjusting, or managing the hydraulic fracturing operation. The relevant quantities can include estimated hydraulic fracturing operation data, equipment data, hydraulic fracturing treatment data, user feedback, any additional relevant information, or a combination thereof. The cloud-based dashboard may be supported on the back end by the cloud architecture. Raw data can be streamed to the cloud service provided by the cloud architecture and the raw data can be pre-processed into pre-preprocessed data via the cloud service. The pre-processed data can provide useful information, can be used to determine parameters relating to the hydraulic fracturing operation, can be displayed as the relevant quantities on the cloud-based dashboard, or a combination thereof. In some examples, the cloud-based dashboard can be used for controlling, adjusting, or managing the hydraulic fracturing operation in real-time or near real-time. The cloud-based dashboard can also be used for decision making in a future hydraulic fracturing operation. The relevant quantities can be key performance indicators. The cloud-based dashboard can display the key performance indicators and the key performance indicators can be used to increase the productivity or efficiency of the hydraulic fracturing operation.
The relevant quantities can be parameters related to the estimated hydraulic fracturing operation data, the equipment data, or a combination thereof. The parameters can be used to determine, or can be a measurement of, diagnostic issues on pumps, blenders, or other equipment equipped with sensors. In an example, the parameters can include pumping hours over time, number of gear changes per treatment, duration of revolutions per minute (RPM) outside of optimal horsepower range per treatment, duration of cavitation per treatment, historical data, or a combination thereof. Additionally, the relevant quantities can be parameters related to real-time hydraulic fracturing data, such as actual pump rate, proppant concentration, pressure over time, scheduled rate, proppant concentration over time, or a combination thereof. The relevant quantities can further include user feedback and equipment diagnostics from equipment sensors. The user feedback may include feedback from operators, supervisors, crew members, or additional employees related to the hydraulic fracturing operation, or the user feedback can include customer or user feedback.
The cloud-based dashboard can be supported on the back end by the cloud architecture. The raw data from the hydraulic fracturing operation can be streamed in real-time to the cloud service via the cloud architecture. The cloud service can implement on-the-fly computation and visualization of the raw data. In an example, the cloud service may organize, store, or organize and store the raw data into one or more data lakes. The raw data can be pre-processed into a required form for the cloud-based dashboard via the cloud service. Pre-processing of the raw data into pre-processed data for the cloud-based dashboard can be an automated process.
In an example, the automated process can include alerts, defined where raw data enters the cloud architecture, to flag unexpected behavior during the hydraulic fracturing operation. For example, unexpected behavior can be a sudden drop in pressure, deviation from a hydraulic fracturing operation plan, deterioration of equipment conditions, or other suitable unexpected behavior during the hydraulic fracturing operation. The automated process may further include ingesting the raw data into the one or more data lakes. The one or more data lakes can store the raw data in a variety of formats. The variety of formats may include time series data, images, audio, or other suitable data formats. Additionally in the automated process, the raw data ingested to the one or more data lakes can be cleansed, curated, and joined with other available data sources to generate the pre-processed data.
The pre-processed data can provide a comprehensive view of a current hydraulic fracturing operation and can provide historical expectations from similar hydraulic fracturing operations. A subset of pre-processed data can be extracted during the automated process to generate an optimized view of the pre-processed data. The optimized view can be used for quick diagnosis or prognosis of the hydraulic fracturing operation by providing the most significant or relevant pre-processed data. For example, a subset of pre-processed data relating to a pump can be used to quickly determine if the pump is operating at an optimal level during the hydraulic fracturing operation. The optimized view can be further partitioned to enhance extraction and visualization of pre-processed data. For example, the subset of pre-processed data relating to the pump can be partitioned for stages of the hydraulic fracturing operation to enable analysis of the pump's performance during each stage of the hydraulic fracturing operation. The automated process can further include a computational engine such as SQL, Synapse, or Databricks cluster, which can provide the pre-processed data to the cloud-based dashboard with low latency. The cloud-based dashboard can be a business intelligence (BI) dashboard and the pre-processed data can be provided to the BI dashboard with a BI tool, such as PowerBI or web apps.
The cloud-based dashboard can be used to make real-time or near real-time changes to the current hydraulic fracturing operation, or the cloud-based dashboard can be used to make decisions in a future hydraulic fracturing operation. As an example, the cloud-based dashboard can diagnose if a pump is not operating properly. As a result, the cloud-based dashboard can bring the pump offline and replace the pump with another pump, if available. In another example, the cloud-based dashboard can be used to adjust a pumping schedule by altering pump flow rate, proppant concentration, diverter concentration, or a combination thereof. Additionally, the pump can be determined to be operating outside the optimum horsepower range for more than 15 minutes, for example. In this example, the cloud-based dashboard can diagnose the issue and adjust the pump gear to bring the pump back to operating within the optimum horsepower range. As another example, the cloud-based dashboard can suspect cavitation. The cloud-based dashboard can address cavitation by adjusting the pump schedule to stagger individual pump rates to increase boost pressure. Future changes to hydraulic fracturing operations can be based on issues detected for specific crews, wells, basins, locations, or the like. The cloud-based dashboard can be used to take preventative measures in the hydraulic fracturing operation. For example, a pump crew may consistently operate outside an optimal RPM range. The cloud-based dashboard can recognize the issue and investigate the root cause. As a result, the pump crew can be retrained for future hydraulic fracturing operations.
Illustrative examples are given to introduce the reader to the general subject matter discussed herein and are not intended to limit the scope of the disclosed concepts. The following sections describe various additional features and examples with reference to the drawings in which like numerals indicate like elements, and directional descriptions are used to describe the illustrative aspects, but, like the illustrative aspects, should not be used to limit the present disclosure.
The well system 100 can also include a sensing control device 118 and a distributed acoustic sensor system 120. The distributed acoustic sensor system 120 can include one or more fiber optic cables extending along a length of the wellbore 104. The distributed acoustic sensor system 120 can be used to monitor and collect raw data relating to the wellbore 104 and the hydraulic fracturing operation before, during, or after the hydraulic fracturing operation. The raw data collected by the distributed acoustic sensor system 120 can be received and processed by the sensing control device 118 at the surface 106 of the wellbore 104. For example, the sensing control device 118 can convert light signals from the distributed acoustic sensor system 120 to measure wellbore properties such as the size, depth, or location of perforations. In some examples, the distributed acoustic sensor system 120 may be used to detect changes to light signals resulting from acoustic signals, pressure signals, or other disturbance signals within the wellbore 104. The changes to the light signals may be used by the sensing control device 118 to detect the wellbore properties such as the size, depth, or location of perforations, or other properties such as a downhole flow of fracturing fluid, sand out conditions, duration of cavitation, pump rate, pressure over time, proppant concentration, proppant concentration over time, or any other properties relating to the wellbore or the hydraulic fracturing operation. While
The wellbore 104 can further include a work string 112 extending from the surface 106 into the wellbore 104. A pump system 124 can be coupled to the work string 112 for pumping the fracturing fluid 108 into the wellbore 104. The pump system 124 can receive the fracturing fluid 108 from a fluid storage tank (not shown) and combine the fracturing fluid 108 with other components, including proppant from a proppant source, additional fluid from additives, or a combination thereof. The resulting mixture may be pumped downhole into the wellbore 104 under a pressure sufficient to create or enhance one or more pathways or fractures 116 in the subterranean formation 102. The pressure and resulting fractures 116 can stimulate production of fluids, such as oil or gas, from the subterranean formation 102. In some examples, the pump system 124 can provide fracturing fluid into the wellbore 104, proppants into the wellbore 104, or a combination of those components into the wellbore 104.
The work string 112 can include coiled tubing, jointed pipe, or other suitable structure for enabling fracturing fluid 108 to flow into the wellbore 104. The work string 112 can further include flow control devices, bypass valves, ports, perforations, or other suitable tools or well devices to control a flow of the fracturing fluid 108 through the work string 112. For example, the work string 112 can include perforations corresponding to the perforations formed in the casing 110. The perforations can be formed using shaped charges, a perforating gun, hydro-jetting, or other tools. The perforations of the casing 110 and work string 112 can provide a channel between the subterranean formation 102 and the wellbore 104 for transmitting the fracturing fluid 108 directly into the subterranean formation 102 of interest, enabling produced fluid such as oil or gas to flow to the wellbore, or a combination thereof.
The work string 112 or the wellbore 104 can include one or more sets of packers 114. The packers 114 can seal the annulus between the work string 112 and the wellbore 104 to define an interval of interest 122 of the wellbore 104 into which the fracturing fluid 108 can be pumped.
In the cloud service 202, the raw data files 208 can be received by a processing pipeline 204 in a plurality of formats. For example, the raw data files 208 can be audio files, images, timetables, or other suitable data formats. The processing pipeline 204 can pre-process the raw data files 208 to generate pre-processed data that can be used by the cloud-based dashboard 214. Generating pre-processed data can be an automated process performed by the processing pipeline 204. The processing pipeline 204 may cleanse and curate the raw data files 208. Cleansing the raw data files 208 can be a technique to fix incorrect, incomplete, duplicate, or otherwise erroneous data in the raw data files 208. Curating the raw data files 208 can involve collecting, structuring, indexing, and organizing the raw data in the raw data files 208. The processing pipeline 204 may also combine the raw data files 208 with additional available data sources. Available data sources can be equipment data, user feedback data, historical hydraulic fracturing operation data, or other data relating to the hydraulic fracturing operation.
The processing pipeline 204 can create a variety of pre-processed datasets to provide a complete view of the hydraulic fracturing operation. The processing pipeline 204 can pass the raw data files 208 or the pre-processed data to a multi-layer data lake 206. The multi-layer data lake 206 can be a repository for storing, processing, and securing large amounts of structured, semi-structured, or unstructured data. The processing pipeline 204 may further partition the pre-processed data to create an optimized view of the pre-processed data before transmitting the pre-processed data to the multi-layer data lake 206. Further partitioning of the pre-processed data can be used for quickly diagnosing issues or quickly providing solutions to issues in the hydraulic fracturing operation. The processing pipeline 204 can continuously index or partition the pre-processed data to enhance the ability of the cloud-based dashboard 214 to identify, display, or otherwise use the pre-processed data to improve the hydraulic fracturing operation.
The cloud-based dashboard can further include parameters related to the hydraulic fracturing operation based on the pre-processed data. The parameters can include pumping hours over time, number of gear changes per treatment, duration of revolutions per minute (RPM) outside of optimal horsepower range per treatment, duration of cavitation per treatment, historical data, actual pump rate, proppant concentration, pressure over time, scheduled rate, proppant concentration over time, user feedback, or a combination thereof. The pumping hours over time can be related to a pumping schedule for the hydraulic fracturing operation, in which the pumping hours may differ over a time frame of the hydraulic fracturing operation. The duration of RPM outside optimal horsepower range per treatment can be related to a pump 218 not operating at an optimal pressure throughout the hydraulic fracturing operation. Cavitation can be a formation and collapse of air cavities in a liquid and can cause pump failure during the hydraulic fracturing operation.
A combination of the parameters can be displayed on the cloud-based dashboard 214. Additional parameters can be determined from real-time hydraulic fracturing data, estimated hydraulic fracturing data, equipment data, or a combination thereof. In some examples, further information displayed on the cloud-based dashboard can be historical trends from past hydraulic fracturing operations, crew or customer feedback, or other information relating to the hydraulic fracturing operation. The cloud-based dashboard 214 can be a business intelligence (BI) dashboard. The cloud-based dashboard 214 can be used to make real-time, near real-time, or future changes to the hydraulic fracturing operation. Real-time changes to the hydraulic fracturing operation can be characterized by low latency. In an example, the latency for real-time changes to the hydraulic fracturing operation can be one to sixty seconds. Near real-time changes to the hydraulic fracturing operation can be characterized by a longer latency. For example, the near real-time latency can be one to five minutes.
At block 220, changes to the hydraulic fracturing operation can be identified and implemented based on the cloud-based dashboard 214. The change can be implemented automatically, or changes can be made by an operator, crew member, or other suitable user based on the cloud-based dashboard. At block 222, the cloud-based dashboard identifies bugs and makes future changes to the hydraulic operation. The bugs in the hydraulic fracturing operation may be due to equipment issues or non-optimized parameters. As an example, the cloud-based dashboard 214 can flag cavitation and the future pump schedule can be changed to stagger individual pump rates to prevent the cavitation during future operations. At block 224, the cloud-based dashboard can identify a training opportunity for crews operating the hydraulic fracturing operation. For example, a training opportunity can be identified by the cloud-based dashboard if a pump crew is consistently operating outside optimal ranges of the hydraulic fracturing operation. Additionally, at block 226 real-time changes to the hydraulic fracturing operations can be made based on the cloud-based dashboard. For example, one or more pumps 218 can be operating outside an optimum horsepower range. The cloud-based dashboard 214 can include a parameter for a difference between a current horsepower value and the optimum horsepower range. The pump can be adjusted in real-time to bring the pump back to the optimum house power range based on the difference identified by the cloud-based dashboard 214.
The computing device 302 can include the processor 304, the memory 307, and a bus 306. The processor 304 can execute one or more operations for controlling or managing the hydraulic fracturing operation using one or more optimization models subject to one or more constraints. The processor 304 can execute instructions 312 stored in the memory 307 to perform the operations. The processor 304 can include one processing device or multiple processing devices or cores. Non-limiting examples of the processor 304 include a Field-Programmable Gate Array (“FPGA”), an application-specific integrated circuit (“ASIC”), a microprocessor, etc.
The processor 304 can be communicatively coupled to the memory 307 via the bus 306. Non-volatile memory 307 may include any type of memory device that retains stored information when powered off. Non-limiting examples of the memory 307 may include EEPROM, flash memory, or any other type of non-volatile memory. In some examples, at least part of the memory 307 can include a medium from which the processor 304 can read instructions 312. A computer-readable medium can include electronic, optical, magnetic, or other storage devices capable of providing the processor 304 with computer-readable instructions or other program code. Non-limiting examples of a computer-readable medium include (but are not limited to) magnetic disk(s), memory chip(s), ROM, RAM, an ASIC, a configured processor, optical storage, or any other medium from which a computer processor can read instructions 312. The instructions 312 can include processor-specific instructions generated by a compiler or an interpreter from code written in any suitable computer-programming language, including, for example, C, C++, C #, Perl, Java, Python, etc.
In some examples, the memory 307 can be a non-transitory computer readable medium and can include computer program instructions 312. For example, the computer program instructions 312 can be executed by the processor 304 for causing the processor 304 to perform various operations. For example, the processor 304 can receive raw data 316 relating to a hydraulic fracturing operation. The raw data 316 can be streamed from a data acquisition and control system 216 via a streaming hub 212 to a cloud service 314. The raw data 316 can be pre-processed into pre-processed data 318 by the cloud service 314. The pre-processed data 318 can be used or further organized and partitioned to identify parameters 320 relating to the hydraulic fracturing operation. The parameters 320 can include pumping hours over time, number of gear changes per treatment, duration of RPM outside optimal horsepower range per treatment, duration of cavitation per treatment, actual pumped rate, proppant concentration, pressure over time, scheduled rate, proppant concentration over time, or a combination thereof. The parameters 320 can be displayed on the cloud-based dashboard 310 as graphs, tables, charts, or other suitable displays.
The computing device 302 can additionally include an input/output 308. The input/output 308 can connect to a keyboard, a pointing device, a display, other computer input/output devices or any combination thereof. A user may provide input using the input/output 308. Data relating to a wellbore 104, the hydraulic fracturing operation, or a combination thereof can be displayed to the user related to the hydraulic fracturing operation via the cloud-based dashboard 310 that can be connected to, part of, or displayed on the input/output 308. The displayed values can be observed by an operator, a crew member, a customer, a supervisor, or other user related to the hydraulic fracturing operation, who can adjust the hydraulic fracturing operation based on the cloud-based dashboard 310. Alternatively, the computing device 302 can automatically control or adjust the hydraulic fracturing operation based on the cloud-based dashboard 310.
In some examples, the user interface 400 can be used to provide parameters that have been optimized by the computing device 302. For example, raw data 316 can be collected by a data acquisition and control system 216 and streamed in real-time or near real-time to a cloud service 314. The raw data 316 can be stored in a data lake in the cloud service 314. The cloud service 314 can pre-process the raw data 316 to generate pre-processed data 318. A subset of the pre-processed data 318 can be extracted to generate an optimized version the pre-processed data 318 that can be displayed on the user interface 400 as a parameter. A display of pre-processed data 318, subsets of pre-processed data 318, or parameters on the user interface 400 can include various types of graphs, tables, or other suitable data display methods.
For example, the user interface 400 can include one or more key performance indicator (KPI) plots 402a-c. In some examples, KPI plots 402a-c can summarize parameters of the hydraulic fracturing operation. As illustrated, KPI plot 402a-b can provide information on the production of the hydraulic fracturing operation over stages or a time frame of the hydraulic fracturing operation. The parameters of the hydraulic fracturing operation used for KPI plots 402a-c can be based on equipment data or estimated hydraulic fracturing operation data. The estimated hydraulic fracturing operation data can be derived from previous hydraulic fracturing operations or theoretical data. The parameters of the hydraulic fracturing operation can include pumping hours over time, number of gear changes per treatment, duration of RPM outside of optimal horsepower range per treatment, duration of cavitation per treatment, or a combination thereof. In other examples, KPI plot 402c can provide information on a difference between estimated hydraulic fracturing operation data and real-time hydraulic operation data. For example, KPI plot 402c provides information on pump performance by comparing a set rate of the pump to an actual rate the pump functions at during the hydraulic fracturing operation. In other examples, the KPI plot 402a-c can provide information on crew performance, wellbore production, proppant concentration, pressure over time, scheduled rate, proppant concentration over time, any additional aspects of the hydraulic fracturing operation, or a combination thereof.
The user interface 400 can further include one or more overview plots 404. In some examples, the overview plot 404 can provide various key performance indicators for comparison on one plot. The KPI plots 402a-c and the overview plot 404 can further include historical trends by comparing data from past hydraulic fracturing operations to current hydraulic fracturing operations. In additional examples, KPI plots 402a-c can provide information from user feedback. The user feedback can be from a crew working at the hydraulic fracturing operation site, an operator in a real-time operating center, a customer, or any other user related to the cloud-based dashboard 310 or the hydraulic fracturing operation.
At block 502, the processor 304 can collect raw data 316 relating to the hydraulic fracturing operation, the raw data can be streamed to the cloud service from the hydraulic fracturing operation. The raw data 316 can be collected during the hydraulic fracturing operation by a sensing control device 118 in a data acquisition and control system 216. The raw data 316 can be collected from one or more pumps 218 during the hydraulic fracturing operation. In some examples, the raw data 316 can be collected in the form of audio, images, timetables, or other suitable data formats. The raw data 316 can be streamed to the cloud service 314 via a streaming hub that receives the raw data from the data acquisition and control system. The cloud service 314 can provide computation, visualization, ingestion, or a combination thereof of the raw data 316. In some examples, a large amount of the raw data 316 can be streamed in real-time or near real-time to the cloud service 314. In other examples, a subset of the raw data 316 can be prioritized. Thus, the subset of raw data 316 can be streamed to the cloud service 314 before additional subsets of the raw data 316 to optimize availability of data on the cloud-based dashboard 310. Additionally, one or more alerts can be defined at a gateway of raw data 316 entry to the cloud service 314 to flag a problem with the hydraulic fracturing operation. The one or more alerts can increase the efficiency of addressing a problem and minimize the impact of the problem on the hydraulic fracturing operation. The one or more alerts may occur due to a sudden decrease or increase in pressure or rate of the hydraulic fracturing operation, a significant deviation from a plan for the hydraulic fracturing operation, a significant deterioration in equipment conditions, or a combination thereof.
At block 504, the processor 304 can pre-process, via the cloud service 314, the raw data 316 to generate pre-processed data 318. Pre-processing of the raw data 316 can be an automated process. The raw data 316 can be sorted and ingested into one more data lakes during pre-processing. Generating pre-processed data 318 can include cleansing and curating the raw data 316. In an example, raw data 316 can be cleansed and curated by identifying incomplete or inaccurate data, deleting repetitive, unnecessary, or inaccurate data, organizing the remaining raw data 316, or through any additional operations to improve the raw data 316. In some examples, pre-processing the raw data 316 further includes combining the raw data 316 with other available data sources. Available data sources can include historical data related to a wellbore 104, a crew, equipment, or the hydraulic fracturing operation or any other data that relates to or affects the hydraulic fracturing operation. Pre-processing data may further include partitioning the pre-processed data 318 one or more times to create an optimized version of the pre-processed data 318. Machine learning models can be applied to pre-processing the raw data 316 to predict data with a highest impact on the hydraulic fracturing operation. Therefore, machine learning can improve the efficiency of pre-processing. A sub-set of the pre-processed data 318 can be extracted as a quick diagnostic indicator. The quick diagnostic indicator can be used to identify a problem with the hydraulic fracturing operation, to identify an opportunity for optimization in the hydraulic fracturing operation, or to identify a combination thereof. In some examples, the quick diagnostic indicator can be used to adjust or control the hydraulic fracturing operation.
At block 506, the processor 304 can identify at least one parameter relating to the hydraulic fracturing operation based on the pre-processed data 318. The at least one parameter can be updated based on the raw data 316 being streamed and pre-processed. Machine learning models can also be implemented in identifying parameters 320 to predict sets of data that can significantly impact parameters 320 or machine learning models can be implemented to predict parameters 320 with a significant impact on the efficiency of the hydraulic fracturing operation. Parameters 320 relating to the hydraulic fracturing operation can be based on equipment data, hydraulic fracturing operation data, or a combination thereof. Examples of parameters 320 include pumping hours over time, number of gear changes per treatment, duration of RPM outside optimal horsepower range per treatment, duration of cavitation per treatment, actual pumped rate, proppant concentration, pressure over time, scheduled rate, proppant concentration over time, or a combination thereof. Historical hydraulic fracturing operation data or crew or customer feedback may also be used as parameters 320.
At block 508, the processor 304 can determine a difference between the at least one parameter and at least one optimized parameter. For example, a parameter can be identified from the pre-processed data for the number of gear changes per treatment, the actual pumped rate, the proppant concentration, the pressure over time, or other suitable parameters of the hydraulic fracturing operation. Additionally, the cloud service may determine optimal vales for the parameters based on historical data, theoretical data, a theoretical model, etc. The difference between the parameter and the optimal value for the parameter can be displayed, for example, at a user interface that includes the cloud-based dashboard,
At block 510 the processor 304 can adjust the hydraulic fracturing operation based on the difference between the at least one parameter and at least one optimized parameter. For example, the actual pumped rate, proppant concentration, pressure over time, scheduled rate, proppant concentration over time, additional parameters 320, or a combination thereof can be adjusted via the cloud-based dashboard 310. The cloud-based dashboard 310 can be displayed at user interface 400. The cloud-based dashboard 310 can be used to manage a current hydraulic fracturing operation or a future hydraulic fracturing operation. The hydraulic fracturing operation can be controlled autonomously by the cloud-based dashboard 310 based on the differences between parameters 320 and optimal values for parameters 320. In additional examples, an operator or other user viewing user interface 400 can determine an adjustment to the hydraulic fracturing operation based on KPI plots 402a-c, overview plot 404, or a combination thereof.
In some aspects, systems, computer-implemented methods, or non-transitory computer-readable mediums for managing a hydraulic fracturing operation are provided according to one or more of the following examples:
As used below, any reference to a series of examples is to be understood as a reference to each of those examples disjunctively (e.g., “Examples 1-4” is to be understood as “Examples 1, 2, 3, or 4”).
Example 1 is a system comprising: a processor; and a memory device that includes instructions executable by the processor for causing the processor to perform operations comprising: receiving, at a cloud service, raw data streamed to the cloud service from a hydraulic fracturing operation; pre-processing, via the cloud service, the raw data to generate pre-processed data ingestible by a cloud-based dashboard of the cloud service; identifying, via the cloud service, at least one parameter relating to the hydraulic fracturing operation using the pre-processed data; determining, via the cloud service, a difference between the at least one parameter and at least one optimized parameter, the at least one optimized parameter determined by the cloud service based on historical data or theoretical data; and adjusting the hydraulic fracturing operation based on the difference between the at least one parameter and the at least one optimized parameter.
Example 2 is the system of example 1, further comprising displaying the at least one parameter on the cloud-based dashboard.
Example 3 is the system of example 2, wherein the at least one parameter comprises pumping hours over time, number of gear changes per treatment, pumped rate over time, proppant concentration over time, pressure over time, duration of cavitation per treatment, other diagnostic parameters of hydraulic fracturing equipment, or a combination thereof.
Example 4 is the system of examples 1-3, wherein the operation of pre-processing, via the cloud service, the raw data to generate the pre-processed data ingestible by the cloud-based dashboard further comprises: receiving the raw data at a data lake in a plurality of formats; performing at least one operation on the raw data and combining the raw data with preexisting data to generate the pre-processed data; and extracting a subset of the pre-processed data to generate a diagnostic indicator for adjusting the hydraulic fracturing operation.
Example 5 is the system of examples 1-4, wherein the operation of receiving, by the cloud service, the raw data related to the hydraulic fracturing operation further comprises: identifying a subset of raw data, the subset of raw data related to an unexpected behavior of the hydraulic fracturing operation; and generating an alert for display at the cloud-based dashboard for the unexpected behavior of the hydraulic fracturing operation based on the subset of raw data.
Example 6 is the system of examples 1-5, wherein the operation of adjusting the hydraulic fracturing operation based on the difference between the at least one parameter and the at least one optimized parameter further comprises: adjusting the hydraulic fracturing operation in real-time or near real-time.
Example 7 is the system of examples 1-6, wherein the operation of receiving, by the cloud service, the raw data relating to the hydraulic fracturing operation further comprises: assigning a first pre-processing priority level to a first subset of raw data from the raw data; assigning a second pre-processing priority level to a second subset of raw data from the raw data; and pre-processing, by the cloud service, the first subset of raw data prior to pre-processing the second subset of raw data based on the first pre-processing priority level exceeding the second pre-processing priority level.
Example 8 is the system of examples 1-7, wherein the operation of pre-processing the raw data or the operation of identifying the at least one parameter further comprises executing a machine learning model that predicts data with a highest impact on the hydraulic fracturing operation.
Example 9 is a computer-implemented method comprising: receiving, at a cloud service, raw data streamed to the cloud service from a hydraulic fracturing operation; pre-processing, via the cloud service, the raw data to generate pre-processed data ingestible by a cloud-based dashboard of the cloud service; identifying, via the cloud service, at least one parameter relating to the hydraulic fracturing operation using the pre-processed data; determining, via the cloud service, a difference between the at least one parameter and at least one optimized parameter, the at least one optimized parameter determined by the cloud service based on historical data or theoretical data; and adjusting the hydraulic fracturing operation based on the difference between the at least one parameter and the at least one optimized parameter.
Example 10 is the computer-implemented method of example 9, further comprising displaying the at least one parameter on the cloud-based dashboard.
Example 11 is the computer-implemented method of example 10, wherein the at least one parameter comprises pumping hours over time, number of gear changes per treatment, duration of cavitation per treatment, pumped rate, proppant concentration, pressure over time, proppant concentration over time, or any combination thereof.
Example 12 is the computer-implemented method of examples 9-11, wherein pre-processing, via the cloud service, the raw data to generate the pre-processed data ingestible by the cloud-based dashboard further comprises: receiving the raw data at a data lake in a plurality of formats; performing at least one operation on the raw data and combining the raw data with preexisting data to generate the pre-processed data; and extracting a subset of the pre-processed data to generate a diagnostic indicator for adjusting the hydraulic fracturing operation.
Example 13 is the computer-implemented method of examples 9-12, wherein receiving, by the cloud service, the raw data related to the hydraulic fracturing operation further comprises: identifying a subset of raw data, the subset of raw data related to an unexpected behavior of the hydraulic fracturing operation; and generating an alert for display at the cloud-based dashboard for the unexpected behavior of the hydraulic fracturing operation based on the subset of raw data.
Example 14 is the computer-implemented method of examples 9-13, wherein adjusting the hydraulic fracturing operation based on the difference between the at least one parameter and the at least one optimized parameter further comprises: adjusting the hydraulic fracturing operation in real-time or near real-time.
Example 15 is the computer-implemented method of examples 9-14, wherein receiving, by the cloud service, the raw data related to the hydraulic fracturing operation further comprises: assigning a first pre-processing priority level to a first subset of raw data from the raw data; assigning a second pre-processing priority level to a second subset of raw data from the raw data; and pre-processing, by the cloud service, the first subset of raw data prior to pre-processing the second subset of raw data based on the first pre-processing priority level exceeding the second pre-processing priority level.
Example 16 is a non-transitory computer-readable medium comprising instructions that are executable by a processing device for causing the processing device to perform operations comprising: receiving, at a cloud service, raw data streamed to the cloud service from a hydraulic fracturing operation; pre-processing, via the cloud service, the raw data to generate pre-processed data ingestible by a cloud-based dashboard of the cloud service; identifying, via the cloud service, at least one parameter relating to the hydraulic fracturing operation using the pre-processed data; determining, via the cloud service, a difference between the at least one parameter and at least one optimized parameter, the at least one optimized parameter determined by the cloud service based on historical data or theoretical data; and adjusting the hydraulic fracturing operation based on the difference between the at least one parameter and the at least one optimized parameter.
Example 17 is the non-transitory computer-readable medium of example 16, further comprising displaying the at least one parameter on the cloud-based dashboard.
Example 18 is the non-transitory computer-readable medium of example 17, wherein the at least one parameter comprises pumping hours over time, number of gear changes per treatment, duration of cavitation per treatment, pumped rate, proppant concentration, pressure over time, proppant concentration over time, or any combination thereof.
Example 19 is the non-transitory computer-readable medium of examples 16-18, wherein the operation of pre-processing, via the cloud service, the raw data to generate the pre-processed data ingestible by the cloud-based dashboard further comprises: receiving the raw data at a data lake in a plurality of formats; performing at least one operation on the raw data and combining the raw data with preexisting data to generate the pre-processed data; and extracting a subset of the pre-processed data to generate a diagnostic indicator used to adjust the hydraulic fracturing operation.
Example 20 is the non-transitory computer-readable medium of examples 16-19, wherein the operation of receiving, by the cloud service, the raw data related to the hydraulic fracturing operation further comprises: identifying a subset of raw data, the subset of raw data related to an unexpected behavior of the hydraulic fracturing operation; and generating an alert for display at the cloud-based dashboard for the unexpected behavior of the hydraulic fracturing operation based on the sub-set of raw data.
The foregoing description of certain examples, including illustrated examples, has been presented only for the purpose of illustration and description and is not intended to be exhaustive or to limit the disclosure to the precise forms disclosed. Numerous modifications, adaptations, and uses thereof will be apparent to those skilled in the art without departing from the scope of the disclosure.
