5GRedcap IoT Control Method and System for Oilfield Equipment Based on Big Data

Abstract

This invention discloses a 5GRedcap IoT control method and system for oilfield equipment based on big data. The technical solution acquires multi-source sensor data generated by the sensor cluster of oilfield equipment through 5GRedcap IoT; constructs a large database based on the attributes of the multi-source sensor data; extracts task event datasets of the oilfield equipment from the large database; predicts the operating status trend of the oilfield equipment based on the task event datasets; identifies abnormal events of the oilfield equipment based on the operating status trend; identifies abnormal equipment links in oilfield production based on the distribution of abnormal events; and isolates abnormal equipment links through IoT and adjusts the working status of these links. By constructing a multi-source sensor database for oilfield equipment and identifying abnormal events, abnormal equipment links in the production scenario can be identified, enabling multi-equipment associated anomaly monitoring and improving the overall operational reliability and fault isolation efficiency of the oilfield production system.

Description

Technical Field

[0001] This invention relates to the field of Internet of Things (IoT) technology, and in particular to a 5GRedcap IoT control method and system for oilfield equipment based on big data. Background Technology

[0002] Oilfield development is a technology-intensive operation that heavily relies on automated equipment and real-time monitoring. To ensure production safety and efficiency, multi-source sensor data is typically collected in real time through sensor clusters deployed on various oilfield equipment, and remote data transmission and centralized processing are achieved using technologies such as 5G RedCap (5G Reduced Capability, 5G Lightweight) IoT. However, existing IoT-based monitoring systems mostly focus on the independent status monitoring and anomaly alarms of single devices. In actual production, oilfield equipment is tightly coupled through process flows, and an anomaly in a single device can propagate along the process chain, triggering a chain of failures. Existing technologies lack effective mechanisms for identifying and collaboratively handling anomalies associated with different devices. Therefore, how to achieve accurate location and collaborative control of abnormal equipment chains based on massive amounts of heterogeneous sensor data has become a challenge for improving the reliability and overall operational efficiency of oilfield IoT monitoring systems. Summary of the Invention

[0003] In view of the above problems, the present invention is proposed to provide a 5GRedcap IoT control method and system for oilfield equipment based on big data that overcomes or at least partially solves the above problems.

[0004] This invention provides a 5GRedcap IoT control method for oilfield equipment based on big data, for use with a server, including:

[0005] Multi-source sensor data generated by the sensor cluster of oilfield equipment is acquired through 5GRedcap IoT; a large database is constructed based on the attributes of the multi-source sensor data;

[0006] Extract task event datasets of oilfield equipment from the large database; predict the operating status trend of the oilfield equipment based on the task event datasets; and determine abnormal events of the oilfield equipment based on the operating status trend.

[0007] Based on the distribution of the abnormal events, the abnormal equipment links in oilfield development are identified; the abnormal equipment links are isolated through the Internet of Things, and the working status of the abnormal equipment links is adjusted.

[0008] In one embodiment, multi-source sensor data generated by a sensor cluster of oilfield equipment is acquired via 5GRedcap IoT; a large database is constructed based on the attributes of the multi-source sensor data, including:

[0009] Monitor the detection logs of the sensor cluster configured in the oilfield equipment to determine the sensor data generation status data of each sensor in the sensor cluster; generate status data based on the sensor data to identify and mark all trusted sensors in the sensor cluster;

[0010] Based on the network access identifiers of all trusted sensors, the multi-source sensor data generated by them is obtained through the 5GRedcap IoT;

[0011] Extract the intrinsic and extrinsic attributes of the multi-source sensor data; wherein, the intrinsic attributes include any one or more of the type, format, and sampling frequency of the multi-source sensor data, and the extrinsic attributes include the device ID where the multi-source sensor data was generated, the location of the device, and the data generation timestamp;

[0012] Based on the intrinsic and extrinsic attributes, sensor data belonging to the same process area and with adjacent time are spatiotemporally aligned and correlated, and the multi-source sensor data are uniformly formatted to construct a large database.

[0013] In one embodiment, extracting the task event dataset of oilfield equipment from the large database includes:

[0014] The target process flow and associated equipment identifiers to be monitored are obtained by parsing the monitoring requests from the user.

[0015] Based on the associated device identifier, extract the task event dataset corresponding to the associated device during the execution of the target process flow from the big data; the task event dataset is a multi-dimensional sensor data sequence arranged in chronological order of process steps.

[0016] In one embodiment, predicting the operating status trend of the oilfield equipment based on the task event dataset, and determining abnormal events of the oilfield equipment based on the operating status trend, includes:

[0017] The task event dataset is divided into time windows to obtain several sub-data sequences arranged in time sequence; the sub-data sequences are input into a pre-trained time-series prediction model to output the operating load change trend and the preset key output flow change trend of the associated equipment in the future preset time period.

[0018] Based on the operating load change trend, determine the predicted load value sequence for the future preset period; based on the preset key output flow change trend, determine the predicted preset key output value sequence for the future preset period.

[0019] If any value in the predicted load value sequence exceeds the preset load safety threshold range, or if any value in the predicted preset key output value sequence exceeds the preset key output flow rate threshold range, then the oilfield equipment is determined to have experienced an abnormal event.

[0020] In one embodiment, predicting the operating status trend of the oilfield equipment based on the task event dataset, and determining abnormal events of the oilfield equipment based on the operating status trend, includes:

[0021] Select the equipment to be analyzed one by one from the associated equipment as the target oilfield equipment, and for each target oilfield equipment:

[0022] Continuous sampling data of the target oilfield equipment is acquired within a time window that traces back a preset duration from the current analysis time. The continuous sampling data includes real-time operating load measurements, real-time preset key output flow measurements, and preset key output planned flow baseline values ​​collected at each continuous sampling time within the time window. Based on the continuous sampling data within the time window, the rated maximum operating load of the target oilfield equipment, and historical failure rate statistics, the equipment operation status index of the target oilfield equipment at each sampling time is determined, forming an equipment operation status index sequence for this analysis.

[0023] Based on the equipment operation status index sequence, determine the standard deviation of the fluctuation degree used to quantify the discrete fluctuation amplitude of the equipment operation status index within the time window, and the first-order autocorrelation coefficient used to quantify the persistence of the evolution pattern of the equipment operation status index over time; determine the system deterioration entropy value of the target oilfield equipment at the current analysis time based on the absolute values ​​of the standard deviation of the fluctuation degree and the first-order autocorrelation coefficient.

[0024] If the system degradation entropy value determined at the current analysis time is greater than the preset judgment threshold, then it is determined that an abnormal event has occurred in the target oilfield equipment within the time window.

[0025] In one embodiment, determining abnormal equipment links in oilfield development based on the distribution of abnormal events; isolating the abnormal equipment links through the Internet of Things; and adjusting the operating status of the abnormal equipment links include:

[0026] Based on the location distribution of the abnormal events, the distribution of the abnormal start time, and the pre-stored oilfield development process topology map, a set of abnormal oilfield equipment with direct or indirect process flow dependencies is determined, and the set of abnormal oilfield equipment is defined as an abnormal equipment link.

[0027] Based on the communication port address of each abnormal oilfield device in the abnormal device link in the 5GRedcap IoT, a command is sent to the controller of the corresponding device to isolate the original control command transmission channel of the abnormal device link.

[0028] Obtain the real-time working status parameters of each abnormal oilfield device within the abnormal device link; generate a device reset sequence based on the reverse order of device dependencies in the process topology diagram or the order of real-time load rate of each device from low to high, and send status reset commands to each device controller in sequence to adjust each device to a preset standby state.

[0029] Corresponding to the aforementioned method, this embodiment of the invention also provides a 5GRedcap IoT control system for oilfield equipment based on big data. This system is used as a server and includes:

[0030] The database construction module is used to acquire multi-source sensor data generated by the sensor cluster of oilfield equipment through 5GRedcap IoT; and to construct a large database based on the attributes of the multi-source sensor data.

[0031] An abnormal event determination module is used to extract a task event dataset of oilfield equipment from the large database; predict the operating status trend of the oilfield equipment based on the task event dataset; and determine the abnormal events of the oilfield equipment based on the operating status trend.

[0032] An abnormal link processing module is used to determine abnormal equipment links in oilfield development based on the distribution of the abnormal events; to isolate the abnormal equipment links through the Internet of Things; and to adjust the working status of the abnormal equipment links.

[0033] In one embodiment, the database construction module includes:

[0034] The sensor data processing unit is used to monitor the detection logs of the sensor cluster configured in the oilfield equipment, determine the sensor data generation status data of each sensor under the sensor cluster, generate status data based on the sensor data, and identify and mark all trusted sensors under the sensor cluster.

[0035] The sensor data acquisition unit is used to acquire the multi-source sensor data generated by all the trusted sensors through the 5GRedcap IoT based on the network access identifiers of all the trusted sensors.

[0036] An attribute extraction unit is used to extract the intrinsic and extrinsic attributes of the multi-source sensor data; wherein, the intrinsic attributes include any one or more of the type, format, and sampling frequency of the multi-source sensor data, and the extrinsic attributes include the device ID where the multi-source sensor data was generated, the location of the device, and the data generation timestamp;

[0037] The database construction unit is used to perform spatiotemporal alignment and association of sensor data belonging to the same process area and with adjacent time based on the intrinsic attributes and the extrinsic attributes, and to perform unified format conversion on the multi-source sensor data to construct a large database.

[0038] In one embodiment, the abnormal event determination module includes:

[0039] The task event data extraction unit is used to parse the target process flow and associated equipment identifiers to be monitored from the monitoring request from the user terminal; based on the associated equipment identifiers, it extracts the task event dataset corresponding to the associated equipment during the execution of the target process flow from the big data; the task event dataset is a multi-dimensional sensor data sequence arranged in the time sequence of process steps.

[0040] In one embodiment, the abnormal event determination module further includes a first processing unit or a second processing unit:

[0041] The first processing unit is used to divide the task event dataset into time windows to obtain several sub-data sequences arranged in time sequence; input the sub-data sequences into a pre-trained time-series prediction model to output the operating load change trend and preset key output flow change trend of the associated equipment in a future preset time period; determine the predicted load value sequence in the future preset time period based on the operating load change trend; determine the predicted preset key output value sequence in the future preset time period based on the preset key output flow change trend; if any value in the predicted load value sequence exceeds a preset load safety threshold range, or any value in the predicted preset key output value sequence exceeds a preset preset key output flow threshold range, then it is determined that an abnormal event has occurred in the oilfield equipment;

[0042] The second processing unit is used to select the equipment to be analyzed from the associated equipment one by one as the target oilfield equipment. For each target oilfield equipment, it acquires continuous sampling data of the target oilfield equipment within a time window that is tracing back a preset duration from the current analysis time. The continuous sampling data includes real-time operating load measurement values, real-time preset key output flow measurement values, and preset key output planned flow baseline values ​​collected at each continuous sampling time within the time window. Based on the continuous sampling data within the time window, the rated maximum operating load of the target oilfield equipment, and historical failure rate statistics, it determines the target oilfield equipment at each sampling time. The equipment operation status index is used to form an equipment operation status index sequence for this analysis. Based on the equipment operation status index sequence, the standard deviation of the volatility of the equipment operation status index within the time window and the first-order autocorrelation coefficient of the evolution pattern of the equipment operation status index over time are determined. The absolute values ​​of the standard deviation of the volatility and the first-order autocorrelation coefficient are used to determine the system degradation entropy value of the target oilfield equipment at the current analysis time. If the system degradation entropy value determined at the current analysis time is greater than a preset judgment threshold, it is determined that an abnormal event has occurred in the target oilfield equipment within the time window.

[0043] The beneficial effects of the above-mentioned technical solutions provided in the embodiments of the present invention include at least the following:

[0044] The technical solution provided in this invention acquires multi-source sensor data generated by a sensor cluster of oilfield equipment through 5GRedcap IoT; constructs a large database based on the attributes of the multi-source sensor data; extracts task event datasets of oilfield equipment from the large database; predicts the operating status trend of oilfield equipment based on the task event datasets; identifies abnormal events of oilfield equipment based on the operating status trend; identifies abnormal equipment links in oilfield production based on the distribution of abnormal events; and isolates abnormal equipment links through IoT and adjusts the working status of abnormal equipment links. By constructing a multi-source sensor database for oilfield equipment and identifying abnormal events of equipment, abnormal equipment links in the production scenario can be identified, enabling multi-equipment associated anomaly monitoring and improving the overall operational reliability and fault isolation efficiency of the oilfield production system.

[0045] Other features and advantages of the invention will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the written description and the accompanying drawings.

[0046] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0047] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0048] Figure 1 This is a flowchart illustrating the 5GRedcap IoT control method for oilfield equipment based on big data provided in this embodiment of the invention.

[0049] Figure 2 This is a schematic diagram of the structure of the 5GRedcap IoT control system for oilfield equipment based on big data, provided in an embodiment of the present invention. Detailed Implementation

[0050] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0051] In the description of this invention, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," "outer," "far," "near," "front," and "rear," etc., indicating the orientation or positional relationship, are based on the orientation or positional relationship shown in the accompanying drawings and are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0052] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0053] Please see Figure 1 As shown, this embodiment of the invention provides a 5GRedcap IoT control method for oilfield equipment based on big data, used in a server, comprising:

[0054] The system acquires multi-source sensor data generated by the sensor cluster of oilfield equipment through 5GRedcap IoT; and constructs a large database based on the attributes of the multi-source sensor data.

[0055] Extract task event datasets of oilfield equipment from large databases, predict the operating status trend of oilfield equipment based on task event datasets, and identify abnormal events of oilfield equipment based on operating status trends.

[0056] Based on the distribution of abnormal events, identify abnormal equipment links in oilfield development; isolate abnormal equipment links through the Internet of Things and adjust the working status of abnormal equipment links.

[0057] The beneficial effects of the above embodiments are that the 5GRedcap IoT control system for oilfield equipment based on big data constructs a multi-source sensor database for oilfield equipment, identifies abnormal events of the equipment, thereby determining abnormal equipment links in the mining scenario, realizing multi-equipment associated abnormal monitoring, and improving the overall operational reliability and fault isolation efficiency of the oilfield production system.

[0058] In another embodiment, multi-source sensor data generated by a sensor cluster of oilfield equipment is acquired via 5GRedcap IoT; based on the attributes of the multi-source sensor data, a large database is constructed, including:

[0059] Monitor the detection logs of the sensor cluster configured in the oilfield equipment to determine the sensor data generation status data of each sensor in the sensor cluster; generate status data based on the sensor data to identify and mark all trusted sensors in the sensor cluster;

[0060] Based on the network access identifiers of all trusted sensors, the multi-source sensor data generated by them is obtained through the 5GRedcap IoT;

[0061] Extract the intrinsic and extrinsic attributes of the multi-source sensor data; wherein, the intrinsic attributes refer to any one or more of the type, format, and sampling frequency of the multi-source sensor data, and the extrinsic attributes refer to the device ID, device location, and data generation timestamp of the multi-source sensor data;

[0062] Based on the intrinsic and extrinsic attributes, sensor data belonging to the same process area and with adjacent time are spatiotemporally aligned and correlated, and the multi-source sensor data are converted into a unified format to construct a large database for spatiotemporally correlated storage.

[0063] The beneficial effects of the above embodiments are that oilfield development includes different types of oilfield equipment such as pumping equipment and water injection equipment. The operating mode and corresponding output results (such as the production of media such as oil or water and the impact on the surrounding environment) of each oilfield equipment are different. In order to comprehensively monitor the operating status of each oilfield equipment, it is necessary to configure a corresponding type and number of sensors for each oilfield equipment. The sensors mentioned above can be, but are not limited to, those for the operating status of the oilfield equipment itself and the environmental conditions in which the oilfield equipment is located. In order to comprehensively detect the operating status and environmental conditions of each oilfield equipment, a sensor cluster is configured for all oilfield equipment. It can be understood that the sensor cluster can include, but is not limited to, several current sensors, several voltage sensors, several liquid level sensors, several temperature and humidity sensors, several hydraulic sensors, and other different types of sensors. Furthermore, all sensors in the sensor cluster are connected to the 5GRedcap IoT. The multi-source sensor data generated by the sensor cluster is uploaded to the server through the 5GRedcap IoT, realizing the server's collection of multi-source sensor data. Specifically, considering that sensor malfunctions can affect the accuracy of sensing data during operation, it is necessary to identify and calibrate each sensor in the sensor cluster to obtain sensing data generated only by sensors without malfunctions. To this end, the detection logs of the sensor cluster are monitored to obtain the sensing data generation status data of each sensor. This data may include, but is not limited to, the sensing data generation flow rate, the sensing data bit error rate, the data packet transmission success rate, and the signal strength. If the generation flow rate is greater than a preset flow rate threshold and / or the bit error rate is less than a bit error rate threshold, the corresponding sensor is marked as a trusted sensor; and / or, if the data packet transmission success rate is higher than a preset success rate threshold and the signal strength index is higher than a preset strength threshold, the corresponding sensor is marked as a trusted sensor. Then, based on the network access identifiers (e.g., access port address labels) of all trusted sensors in the Internet of Things (IoT), multi-source sensing data generated by all trusted sensors is obtained through 5GRedcap IoT.

[0064] Furthermore, since multi-source sensor data is generated from different types of sensors, it varies in type, format, corresponding generating device, and generation time. To perform correlation calibration and standardization on the multi-source sensor data, the intrinsic and extrinsic attributes of each sub-data item under the multi-source sensor data are first extracted. Then, based on these intrinsic and extrinsic attributes, spatiotemporal correlation and unified format transformation are performed on the multi-source sensor data to construct a large database. Spatiotemporal correlation can include, but is not limited to, correlation calibration of sensor data generated by sensors corresponding to the same location or region, and correlation calibration of sensor data generated at the same time. The unified format transformation can include, but is not limited to, converting each sub-data item under the multi-source sensor data into the same format. Finally, the multi-source sensor data that has undergone spatiotemporal correlation and unified format transformation is assembled into a large database to achieve multi-dimensional state characterization of oilfield equipment.

[0065] In one embodiment, extracting a task event dataset of oilfield equipment from the large database includes:

[0066] The target process flow and associated equipment identifiers to be monitored are obtained by parsing the monitoring requests from the user.

[0067] Based on the associated device identifier, extract the task event dataset corresponding to the associated device during the execution of the target process flow from the big data; the task event dataset is a multi-dimensional sensor data sequence arranged in chronological order of process steps.

[0068] In one embodiment, the above-mentioned extraction of task event datasets of oilfield equipment from a large database, prediction of the operating status trend of the oilfield equipment based on the task event datasets, and determination of abnormal events of the oilfield equipment based on the operating status trend can be implemented as follows: Embodiment 1 or 2:

[0069] Example 1

[0070] Extracting task event datasets from large databases for oilfield equipment, predicting operational status trends based on these datasets, and identifying abnormal events based on these trends can be implemented as follows:

[0071] The task event dataset is divided into time windows to obtain several sub-data sequences arranged in time sequence; the sub-data sequences are input into a pre-trained time-series prediction model to output the operating load change trend and the preset key output flow change trend of the associated equipment in the future preset time period.

[0072] Based on the operating load change trend, determine the predicted load value sequence for the future preset period; based on the preset key output flow change trend, determine the predicted preset key output value sequence for the future preset period.

[0073] If any value in the predicted load value sequence exceeds the preset load safety threshold range, or if any value in the predicted preset key output value sequence exceeds the preset key output flow rate threshold range, then the oilfield equipment is determined to have experienced an abnormal event.

[0074] The abnormal events include overload anomalies indicated by predicted load values ​​exceeding limits and efficiency anomalies indicated by predicted output values ​​exceeding limits.

[0075] The pre-trained time-series prediction model refers to a machine learning or deep learning model that can learn evolutionary patterns from historical time-series data and predict future sequences accordingly. Based on multi-dimensional sensor data sequences (task event datasets) of historical oilfield equipment operation, it can model the dynamic patterns of key equipment parameters (such as operating load and key output flow rates) over time, thereby enabling the prediction of future operational trends.

[0076] The model can be built and trained by following these steps. In actual implementation, the steps can be selected and adjusted according to the specific application scenario:

[0077] The time-series prediction model can be a Long Short-Term Memory (LSTM) network, a gated recurrent unit (GRU), a temporal convolutional network, or a Transformer-based encoder-decoder architecture, or other model structures suitable for sequence prediction. A large number of timestamped, multi-dimensional sensor data sequences generated by oilfield equipment during its historical normal operation and execution of various processes are extracted from the large database as a training sample set. Each sample contains sequence data for a historical observation window and its subsequent real value sequence over a period of time. Using the historical observation window data as model input and the future real value sequence as the prediction target, the model is trained using supervised learning. During training, optimization algorithms (such as Adam and SGD) are used to minimize the loss function (such as mean squared error and mean absolute error) between the predicted and actual values, enabling the model to learn to capture reliable temporal dependencies and trends from historical data. Finally, a validation dataset is used to evaluate the model's prediction accuracy and generalization ability. The trained model can then be deployed on the server to predict the task event dataset, outputting the operating load change trend and the preset key output flow change trend within a future preset time period.

[0078] The above embodiment is a predictive monitoring system based on the process flow. First, based on user requests, the specific production process and associated equipment are identified, and historical sensor data sequences of that equipment during the execution of that process are extracted from the database, forming a dataset with production task context. Then, a pre-trained time-series prediction model is used to analyze the evolution of the data and predict the changing trends of key operating parameters (such as load and flow rate) of the equipment over a future period. Finally, by determining whether these predicted values ​​will exceed safety thresholds, an early warning is issued before the equipment actually fails.

[0079] This technology shifts from reactive alarms to proactive warnings, providing a buffer time for failures, supporting preventative maintenance, and avoiding unplanned downtime. Furthermore, because the analysis process is integrated with specific technological flows, it can distinguish between normal operating condition fluctuations and true anomalies, reducing false alarm rates and improving the accuracy and foresight of oilfield production monitoring.

[0080] Example 2

[0081] Based on the task event dataset, predict the operating status trend of oilfield equipment; based on the operating status trend, identify abnormal events of oilfield equipment, which can be implemented as follows:

[0082] Select the device to be analyzed from the associated devices one by one as the target oilfield device, and perform the following steps C1-C3 for each target oilfield device:

[0083] Step C1: Obtain the target oilfield equipment at the current analysis time. Preset duration for the endpoint and backward tracking The continuous sampling data within a time window, wherein the continuous sampling data includes each continuous sampling moment within the time window. The collected real-time operating load measurement value, real-time preset key output flow measurement value, and preset key output planned flow baseline value; based on the continuous sampling data within the time window, the rated maximum operating load of the target oilfield equipment, and the historical failure rate statistics, the equipment operation status index of the target oilfield equipment at each sampling time is determined, forming an equipment operation status index sequence for this analysis;

[0084] Preferably, the sampling time of the target oilfield equipment within the time window can be determined according to the following formula. Equipment operating status index:

[0085] ;

[0086] ;

[0087] ;

[0088] in, This item calculates the real-time health status, which is a health assessment based on current real-time data, reflecting the device's current operating status. This is a historical reliability correction factor, reflecting the level of confidence in real-time health. Devices with a high history of failures will have their real-time health discounted (multiplied by a number less than 1), even if the current real-time health appears healthy.

[0089] in, For load health factor; It is a real-time operating load measurement value, indicating the target oilfield equipment at the sampling time. The actual load, such as motor current or power, is obtained by directly querying the target oilfield equipment from the large database. The data is obtained from the load sensor records corresponding to the given time. The rated load of the target oilfield equipment; The preset optimal load rate for the target oilfield equipment (a preset value, such as 0.7); For the maximum value function, when When the calculation result is less than 0, Set it to 0 to ensure its value is non-negative. This is the determination. The formula is used to calculate the current load factor. With optimal load factor The absolute deviation; if the deviation is 0, then... (Optimal); the greater the deviation, The smaller the value; when the deviation is ≥ 1, (Abnormal) This reflects that the closer the load is to the optimal value, the healthier it is, and the further it deviates, the less healthy it is.

[0090] For output conformity factor; for Real-time preset key output flow measurement values ​​at any given moment; To pre-determine key outputs in The planned flow baseline value at a given time represents the flow rate at that moment. At any given time, the normal output flow rate that the target oilfield equipment is expected to achieve according to the production plan is a time variable whose value is greater than zero. The method for obtaining this value is to query or read it in real time from the production planning management system or production instruction database that is bound to the target process flow. The theoretical production flow rate value that is scheduled to be issued to the equipment at any given time; This is a minimum value function, ensuring that the value inside the parentheses does not exceed 1, thus guaranteeing... This needs to be determined. The formula calculates the relative deviation of actual output from planned output. If the output perfectly matches the plan (deviation is 0), then... The deviation increased. Linear decrease; when the deviation ≥ 100%, .

[0091] These are the first and second weighting coefficients, respectively, both positive real numbers, whose values ​​directly reflect the workload health factor. Consistency factor with output The relative importance of the overall score.

[0092] This is a historical failure rate statistic of the target oilfield equipment, used to reflect the inherent reliability level of the target oilfield equipment. Its value ranges from zero to positive infinity, and the unit is the number of failures per thousand operating hours. This value is calculated by counting the total number of failures and downtimes of the target oilfield equipment in the past statistical period from the historical records of the equipment maintenance management system, and then dividing it by the total number of operating hours in that period. As a reference failure rate, used to... Normalized to a dimensionless relative value, its value can be determined by one of the following methods: using the industry average failure rate of the type of equipment in the target oilfield; or using the theoretical failure rate specified in the design documents of the equipment in the target oilfield; or using the historical average failure rate of all similar equipment in this oilfield. and They have the same dimensions. The preset baseline confidence coefficient is a real number greater than 0, representing the historical failure rate of the equipment. hour, The preferred value range is [4, 99], which corresponds to a trust level of 80% to 99%. The historical sensitivity index is a real number greater than 0. The larger the value, the more sensitive the historical failure rate is to the current implementation health confidence level. The preferred value range is [0.5, 2.0].

[0093] The equipment operation status index sequence used in this analysis can be represented as follows: ;in, ,and , It is the total number of sampling points, i.e., the time window. The number of sampling times included.

[0094] Step C2: Based on the equipment operation status index sequence, determine the standard deviation of the volatility of the equipment operation status index within the time window, and the first-order autocorrelation coefficient, which quantifies the persistence of the evolution pattern of the equipment operation status index over time; determine the target oilfield equipment at the current analysis time based on the absolute values ​​of the standard deviation of the volatility and the first-order autocorrelation coefficient. The system deterioration entropy value.

[0095] Preferably, the standard deviation of the fluctuation level is calculated. The formula is:

[0096] ;

[0097] Calculate the first-order autocorrelation coefficient The formula is:

[0098] ;

[0099] in, , is a sequence The arithmetic mean;

[0100] Preferably, the target oilfield equipment at the current analysis time is determined according to the following formula. System deterioration entropy value :

[0101]

[0102] in, The target oilfield equipment at the current analysis time The system degradation entropy value represents the target oilfield equipment at the current analysis time. The value indicates the degree of instability and chaos in the operation state. The larger the value, the more chaotic the state and the higher the risk of anomalies.

[0103] The index characterizes the dispersion of the equipment's operating status around its mean, i.e., the intensity of instantaneous fluctuations; the larger the value, the more drastic the fluctuations in key parameters such as equipment load and efficiency, and the more unstable the operation.

[0104] It reflects the linear dependency between state values ​​at adjacent sampling times and is used to characterize the memory and predictability of states. (Take the absolute value.) The reason is that, regardless of whether the first-order autocorrelation coefficient is positive (continuous gradual change of state) or negative (regular oscillation of state), the closer its absolute value is to 1, the clearer the time series pattern, the stronger the persistence of the evolution pattern, and the more predictable the system behavior; the closer it is to 0, the closer the state evolution is to random noise, and the more unstable the system tends to be. Therefore, It is used to characterize temporal instability and measure the degree to which a system deviates from an ordered evolutionary pattern.

[0105] It is a volatility weighting coefficient used for adjustment. The contribution ratio of the coefficient to the system degradation entropy value is preferably taken in the range of [0.5, 2.0], and this coefficient is a configurable value.

[0106] It is an instability weighting coefficient used to adjust the timing instability term. The contribution ratio of the system degradation entropy value, its value range is a positive real number, the preferred value range is [0.5, 2.0], and this coefficient is a configurable value.

[0107] The standard deviation of the fluctuation degree and the first-order autocorrelation coefficient are both statistical quantities calculated based on the dimensionless equipment operation status index, and both are dimensionless parameters.

[0108] Step C3, if the current analysis time If the determined system degradation entropy value is greater than a preset judgment threshold, then it is determined that an abnormal event has occurred in the target oilfield equipment within the time window.

[0109] The preset judgment threshold can be a fixed preset threshold, or it can be a dynamic judgment threshold set for the target oilfield equipment at the current analysis time t. , The method for determining it is as follows:

[0110] Based on multiple system degradation entropy value samples calculated during the historical normal operation of the target oilfield equipment, an initial threshold is determined; and based on the target oilfield equipment at the current analysis time... The operating load and the rated maximum load of the target oilfield equipment are used to determine the dynamic judgment threshold set for the current analysis time t of the target oilfield equipment.

[0111] Specifically, the dynamic judgment threshold of the current analysis time t set by the target oilfield equipment can be determined according to the following formula. :

[0112]

[0113] in, This indicates that the target oilfield equipment is at the current analysis time. Operating load (such as current, power). This indicates the rated maximum load of the target oilfield equipment. The reference load rate (e.g., 0.7, indicating that 70% of the rated load is the typical reference operating condition) is a preset value. λ is the load sensitivity coefficient, ranging from [0.1, 0.5]. λ represents the proportional increment by which the normal operating condition of the target oilfield equipment is allowed to become disordered when the operating load deviates from the reference condition. This absolute value ensures that the threshold increases when the load deviates from the reference operating condition, thus enhancing robustness.

[0114] This is the initial statistical threshold, obtained by calculating a specified high quantile (e.g., the 95th percentile) of multiple system degradation entropy value samples from the target oilfield equipment during its historical normal operation. The specified high quantile is used to determine the initial threshold. A statistical method is used. Specifically, multiple system degradation entropy value samples generated by the target oilfield equipment during its historical normal operation are collected, and these sample values ​​are arranged in ascending order; the specified high quantile refers to the sample value located at a specified percentage position in this sorted sequence. For example, when the specified high quantile is the 95th percentile, it means that 95% of the historical normal sample values ​​are less than or equal to this threshold, and only 5% of the normal sample values ​​are higher than it. This value is used as... This ensures that the threshold can cover the fluctuation range of most normal equipment operation, while also remaining sensitive to abnormal states that exceed normal levels, thereby effectively achieving early warning while reducing false alarms.

[0115] The above steps C1-C3 have the following beneficial effects:

[0116] By constructing an equipment operation status index, which integrates real-time operating load, preset key output flow compliance, and historical equipment failure rate, a unified quantitative indicator of the equipment's current health level and inherent reliability is formed, providing a comprehensive basis for anomaly judgment and realizing a quantitative assessment of the equipment's overall operating status. By calculating the standard deviation of the fluctuation degree and the first-order autocorrelation coefficient of the equipment operation status index sequence, and based on this, the system degradation entropy value is calculated. This entropy value objectively quantifies the temporal fluctuation amplitude and pattern persistence of the equipment's operating status; its increase indicates the evolution of equipment operation from an orderly and predictable state to an unstable and chaotic state, realizing a quantitative diagnosis of the orderliness of the equipment's operating status. Abnormal events are judged by comparing the system degradation entropy value with a preset judgment threshold (or a dynamic judgment threshold). This threshold is determined based on the statistical analysis of entropy value samples during the equipment's historical normal operation and can be dynamically adjusted according to the equipment's current operating load, thereby reducing false alarms and achieving more accurate early warning of early performance degradation, realizing threshold-based adaptive anomaly judgment.

[0117] In another embodiment, based on the distribution of abnormal events, abnormal equipment links in oilfield development are identified; the abnormal equipment links are isolated via the Internet of Things (IoT), and their operating status is adjusted. This can be implemented through the following steps A1-A3:

[0118] Step A1: Based on the location distribution of the abnormal events, the distribution of abnormal start times, and the pre-stored oilfield development process topology map, determine the set of abnormal oilfield equipment with direct or indirect process flow dependencies, and define the set of abnormal oilfield equipment as an abnormal equipment link.

[0119] The oilfield extraction process topology diagram is pre-generated digitally based on the oilfield extraction process flow design drawings and stored in the server database in the form of a graph structure or relational table, recording the material flow direction, control signal transmission and other dependencies between equipment.

[0120] Step A2: Based on the communication port address of each abnormal oilfield device in the abnormal device link in the 5GRedcap IoT, send instructions to the controller of the corresponding device to isolate the original control instruction transmission channel of the abnormal device link.

[0121] Step A3: Obtain the real-time working status parameters of each abnormal oilfield device in the abnormal device link; generate a device reset sequence according to the reverse order of the device dependency relationship in the process topology diagram or according to the order of the real-time load rate of each device from low to high, and send status reset instructions to each device controller in sequence to adjust each device to a preset standby state.

[0122] The implementation methods of the above technical solutions will be explained in detail below.

[0123] Step A1 is implemented as follows:

[0124] Extract the device identifier and the start time of each current abnormal event.

[0125] Based on the device identifier, query the pre-stored oilfield extraction process topology map. The process topology map defines the material transfer and control signal dependencies between devices, and obtains all abnormal device pairs with direct or indirect connection paths.

[0126] For each abnormal device pair, verify whether the abnormal start time of the upstream device in the process flow is earlier than that of the downstream device, and calculate the time difference between the two; only when the abnormal device pair simultaneously satisfies the correct process flow sequence and the time difference is less than a preset propagation time threshold, it is determined that the device pair has a fault propagation association.

[0127] Construct an abnormal device association network graph by treating all device pairs that pass the verification and have fault propagation associations as edges.

[0128] In the associated network graph, all connected subgraphs formed by devices that are directly or indirectly connected by edges are identified.

[0129] Each set of device identifiers contained in the connected subgraph is defined as an independent abnormal device link.

[0130] Step A2 is implemented as follows:

[0131] Based on the pre-stored device communication address mapping table, obtain the network address and communication port of each device controller in the abnormal device link in the 5G RedCap IoT.

[0132] Based on the network address and communication port, a channel isolation command is sent to the controller of each device.

[0133] Upon receiving the channel isolation command, the device controller performs a network communication strategy modification operation to block preset standard control commands from the central control system (i.e., the system that controls the device to operate normally) while maintaining communication with the server.

[0134] Step A3 is implemented as follows:

[0135] Through 5G RedCap IoT, a status query request is sent to each device controller within the abnormal device link.

[0136] The system receives response data returned by each device controller and extracts the real-time operating status parameters of each device from it; the real-time operating status parameters include at least the device operating status code and the real-time load rate.

[0137] Based on the process topology diagram, determine the process flow direction within the abnormal equipment link, and sort all equipment within the link in the opposite order to the material or signal flow direction as the equipment reset sequence; when the process flow direction within the abnormal equipment link cannot be determined, sort all equipment within the link in ascending order according to the real-time load rate in the real-time working status parameters as the equipment reset sequence.

[0138] According to the generated device reset sequence, perform the following state reset operation on each device in the sequence in sequence: send a preset standard state reset instruction set predefined for its device type to the controller of the device; after confirming that the response signal returned by the controller of the device indicating that it has entered the preset standby state is received, perform the state reset operation on the next device in the device reset sequence, until all devices in the abnormal device link have completed the state reset operation.

[0139] Furthermore, once it is confirmed that all devices within the abnormal device link have entered a preset standby state, the collaborative handling process is completed. Further, the status of the abnormal device link can be marked as securely isolated and in standby mode, and a handling completion notification containing the identifiers of each device within the abnormal device link can be generated and sent to the monitoring center. The monitoring center performs subsequent processing according to preset processing rules, which may include: sending a maintenance notification containing the identifiers of each device within the abnormal device link to the maintenance client to guide on-site maintenance; or switching the production task of the abnormal device link to a preset backup link.

[0140] The beneficial effects of the above technical solution are as follows:

[0141] Step A1, by performing causal correlation analysis between the spatiotemporal information of the abnormal event and the process topology diagram, accurately identifies the set of devices with fault propagation relationships (i.e., abnormal device links) from discrete abnormal points, achieving the localization of associated faults and overcoming the limitations of traditional monitoring in isolating single-point anomalies. Step A2 utilizes 5G RedCap IoT to quickly issue isolation commands to the identified abnormal device links, achieving logical isolation of the links at the control network level, cutting off the fault propagation path within the control system, preventing the expansion of the abnormal impact, and creating conditions for safe recovery. Step A3, based on the acquired real-time status and process topology, generates a safe operation sequence using a process reversal or load progressive strategy, and executes a coordinated reset sequentially. This ensures that in complex systems with multiple coupled devices, the state recovery operation follows physical dependencies, smoothly and orderly adjusting all devices within the link to the preset standby state—a safe state—avoiding secondary faults caused by improper operation.

[0142] Please see Figure 2 As shown, another embodiment of the present invention provides a 5GRedcap IoT control system for oilfield equipment based on big data. This system is used as a server and includes:

[0143] The database construction module is used to acquire multi-source sensor data generated by the sensor cluster of oilfield equipment through 5GRedcap IoT; and to construct a large database based on the attributes of the multi-source sensor data.

[0144] An abnormal event determination module is used to extract a task event dataset of oilfield equipment from the large database; predict the operating status trend of the oilfield equipment based on the task event dataset; and determine the abnormal events of the oilfield equipment based on the operating status trend.

[0145] An abnormal link processing module is used to determine abnormal equipment links in oilfield development based on the distribution of the abnormal events; to isolate the abnormal equipment links through the Internet of Things; and to adjust the working status of the abnormal equipment links.

[0146] In one embodiment, the database construction module includes:

[0147] The sensor data processing unit is used to monitor the detection logs of the sensor cluster configured in the oilfield equipment, determine the sensor data generation status data of each sensor under the sensor cluster, generate status data based on the sensor data, and identify and mark all trusted sensors under the sensor cluster.

[0148] The sensor data acquisition unit is used to acquire the multi-source sensor data generated by all the trusted sensors through the 5GRedcap IoT based on the network access identifiers of all the trusted sensors.

[0149] An attribute extraction unit is used to extract the intrinsic and extrinsic attributes of the multi-source sensor data; wherein, the intrinsic attributes include any one or more of the type, format, and sampling frequency of the multi-source sensor data, and the extrinsic attributes include the device ID where the multi-source sensor data was generated, the location of the device, and the data generation timestamp;

[0150] The database construction unit is used to perform spatiotemporal alignment and association of sensor data belonging to the same process area and with adjacent time based on the intrinsic attributes and the extrinsic attributes, and to perform unified format conversion on the multi-source sensor data to construct a large database.

[0151] In one embodiment, the abnormal event determination module includes:

[0152] The task event data extraction unit is used to parse the target process flow and associated equipment identifiers to be monitored from the monitoring request from the user terminal; based on the associated equipment identifiers, it extracts the task event dataset corresponding to the associated equipment during the execution of the target process flow from the big data; the task event dataset is a multi-dimensional sensor data sequence arranged in the time sequence of process steps.

[0153] In one embodiment, the abnormal event determination module further includes a first processing unit or a second processing unit:

[0154] The first processing unit is used to divide the task event dataset into time windows to obtain several sub-data sequences arranged in time sequence; input the sub-data sequences into a pre-trained time-series prediction model to output the operating load change trend and preset key output flow change trend of the associated equipment in a future preset time period; determine the predicted load value sequence in the future preset time period based on the operating load change trend; determine the predicted preset key output value sequence in the future preset time period based on the preset key output flow change trend; if any value in the predicted load value sequence exceeds a preset load safety threshold range, or any value in the predicted preset key output value sequence exceeds a preset preset key output flow threshold range, then it is determined that an abnormal event has occurred in the oilfield equipment;

[0155] The second processing unit is used to select the equipment to be analyzed from the associated equipment one by one as the target oilfield equipment. For each target oilfield equipment, it acquires continuous sampling data of the target oilfield equipment within a time window that is tracing back a preset duration from the current analysis time. The continuous sampling data includes real-time operating load measurement values, real-time preset key output flow measurement values, and preset key output planned flow baseline values ​​collected at each continuous sampling time within the time window. Based on the continuous sampling data within the time window, the rated maximum operating load of the target oilfield equipment, and historical failure rate statistics, it determines the target oilfield equipment at each sampling time. The equipment operation status index is used to form an equipment operation status index sequence for this analysis. Based on the equipment operation status index sequence, the standard deviation of the volatility of the equipment operation status index within the time window and the first-order autocorrelation coefficient of the evolution pattern of the equipment operation status index over time are determined. The absolute values ​​of the standard deviation of the volatility and the first-order autocorrelation coefficient are used to determine the system degradation entropy value of the target oilfield equipment at the current analysis time. If the system degradation entropy value determined at the current analysis time is greater than a preset judgment threshold, it is determined that an abnormal event has occurred in the target oilfield equipment within the time window.

[0156] The 5GRedcap IoT control system for oilfield equipment provided in this invention constructs a multi-source sensor database for oilfield equipment, identifies abnormal events of the equipment, thereby determining abnormal equipment links in the mining scenario, realizing multi-equipment correlation anomaly monitoring, and improving the overall operational reliability and fault isolation efficiency of the oilfield production system.

[0157] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. This disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this disclosure is limited only by the appended claims. Thus, if these modifications and variations of the invention fall within the scope of the claims of the invention and their equivalents, the invention is also intended to include these modifications and variations.

Claims

1. A 5GRedcap IoT control method for oilfield equipment based on big data, used in a server, characterized in that... include: 5GRedcap IoT acquires multi-source sensor data generated by sensor clusters of oilfield equipment; A large database is constructed based on the attributes of the multi-source sensor data; Extract task event datasets of oilfield equipment from the large database, and predict the operating status trend of the oilfield equipment based on the task event datasets; Based on the operational status trend, determine the abnormal events of the oilfield equipment; Based on the distribution of the abnormal events, identify the abnormal equipment links in oilfield development; The abnormal device link is isolated through the Internet of Things, and the working status of the abnormal device link is adjusted.

2. The 5GRedcap IoT control method for oilfield equipment based on big data as described in claim 1, characterized in that: 5GRedcap IoT acquires multi-source sensor data generated by sensor clusters of oilfield equipment; Based on the attributes of the multi-source sensor data, a large database is constructed, including: Monitor the detection logs of the sensor cluster configured in the oilfield equipment to determine the sensor data generation status data of each sensor in the sensor cluster; generate status data based on the sensor data to identify and mark all trusted sensors in the sensor cluster; Based on the network access identifiers of all trusted sensors, the multi-source sensor data generated by them is obtained through the 5GRedcap IoT; Extract the intrinsic and extrinsic attributes of the multi-source sensor data; wherein, the intrinsic attributes include any one or more of the type, format, and sampling frequency of the multi-source sensor data, and the extrinsic attributes include the device ID where the multi-source sensor data was generated, the location of the device, and the data generation timestamp; Based on the intrinsic and extrinsic attributes, sensor data belonging to the same process area and with adjacent time are spatiotemporally aligned and correlated, and the multi-source sensor data are uniformly formatted to construct a large database.

3. The 5GRedcap IoT control method for oilfield equipment based on big data as described in claim 1, characterized in that: The extraction of the task event dataset for oilfield equipment from the large database includes: The target process flow and associated equipment identifiers to be monitored are obtained by parsing the monitoring requests from the user. Based on the associated device identifier, extract the task event dataset corresponding to the associated device during the execution of the target process flow from the big data; the task event dataset is a multi-dimensional sensor data sequence arranged in chronological order of process steps.

4. The 5GRedcap IoT control method for oilfield equipment based on big data as described in claim 3, characterized in that: The operation status trend of the oilfield equipment is predicted based on the task event dataset; Based on the operational status trend, abnormal events of the oilfield equipment are determined, including: The task event dataset is divided into time windows to obtain several sub-data sequences arranged in time sequence; the sub-data sequences are input into a pre-trained time-series prediction model to output the operating load change trend and the preset key output flow change trend of the associated equipment in the future preset time period. Based on the operating load change trend, determine the predicted load value sequence for the future preset period; based on the preset key output flow change trend, determine the predicted preset key output value sequence for the future preset period. If any value in the predicted load value sequence exceeds the preset load safety threshold range, or if any value in the predicted preset key output value sequence exceeds the preset key output flow rate threshold range, then the oilfield equipment is determined to have experienced an abnormal event.

5. The 5GRedcap IoT control method for oilfield equipment based on big data as described in claim 3, characterized in that: The operation status trend of the oilfield equipment is predicted based on the task event dataset; Based on the operational status trend, abnormal events of the oilfield equipment are determined, including: Select the equipment to be analyzed one by one from the associated equipment as the target oilfield equipment, and for each target oilfield equipment: Continuous sampling data of the target oilfield equipment is acquired within a time window that traces back a preset duration from the current analysis time. The continuous sampling data includes real-time operating load measurements, real-time preset key output flow measurements, and preset key output planned flow baseline values ​​collected at each continuous sampling time within the time window. Based on the continuous sampling data within the time window, the rated maximum operating load of the target oilfield equipment, and historical failure rate statistics, the equipment operation status index of the target oilfield equipment at each sampling time is determined, forming an equipment operation status index sequence for this analysis. Based on the equipment operation status index sequence, determine the standard deviation of the fluctuation degree used to quantify the discrete fluctuation amplitude of the equipment operation status index within the time window, and the first-order autocorrelation coefficient used to quantify the persistence of the evolution pattern of the equipment operation status index over time; determine the system deterioration entropy value of the target oilfield equipment at the current analysis time based on the absolute values ​​of the standard deviation of the fluctuation degree and the first-order autocorrelation coefficient. If the system degradation entropy value determined at the current analysis time is greater than the preset judgment threshold, then it is determined that an abnormal event has occurred in the target oilfield equipment within the time window.

6. The 5GRedcap IoT control method for oilfield equipment based on big data as described in claim 1, characterized in that: The abnormal equipment links in oilfield development are determined based on the distribution of the abnormal events. Isolating the abnormal device link through the Internet of Things and adjusting the working status of the abnormal device link includes: Based on the location distribution of the abnormal events, the distribution of the abnormal start time, and the pre-stored oilfield development process topology map, a set of abnormal oilfield equipment with direct or indirect process flow dependencies is determined, and the set of abnormal oilfield equipment is defined as an abnormal equipment link. Based on the communication port address of each abnormal oilfield device in the abnormal device link in the 5GRedcap IoT, a command is sent to the controller of the corresponding device to isolate the original control command transmission channel of the abnormal device link. Obtain the real-time working status parameters of each abnormal oilfield device within the abnormal device link; generate a device reset sequence based on the reverse order of device dependencies in the process topology diagram or the order of real-time load rate of each device from low to high, and send status reset commands to each device controller in sequence to adjust each device to a preset standby state.

7. A 5GRedcap IoT control system for oilfield equipment based on big data, used as a server, characterized in that: include: The database construction module is used to acquire multi-source sensor data generated by the sensor cluster of oilfield equipment through 5GRedcap IoT; A large database is constructed based on the attributes of the multi-source sensor data; An abnormal event determination module is used to extract task event datasets of oilfield equipment from the large database; and predict the operating status trend of the oilfield equipment based on the task event datasets. Based on the operational status trend, determine the abnormal events of the oilfield equipment; An abnormal link processing module is used to determine the abnormal equipment links in oilfield development based on the distribution of the abnormal events. The abnormal device link is isolated through the Internet of Things, and the working status of the abnormal device link is adjusted.

8. The 5GRedcap IoT control system for oilfield equipment based on big data as described in claim 7, characterized in that, The database construction module includes: The sensor data processing unit is used to monitor the detection logs of the sensor cluster configured in the oilfield equipment, determine the sensor data generation status data of each sensor under the sensor cluster, generate status data based on the sensor data, and identify and mark all trusted sensors under the sensor cluster. The sensor data acquisition unit is used to acquire the multi-source sensor data generated by all the trusted sensors through the 5GRedcap IoT based on the network access identifiers of all the trusted sensors. An attribute extraction unit is used to extract the intrinsic and extrinsic attributes of the multi-source sensor data; wherein, the intrinsic attributes include any one or more of the type, format, and sampling frequency of the multi-source sensor data, and the extrinsic attributes include the device ID where the multi-source sensor data was generated, the location of the device, and the data generation timestamp; The database construction unit is used to perform spatiotemporal alignment and association of sensor data belonging to the same process area and with adjacent time based on the intrinsic attributes and the extrinsic attributes, and to perform unified format conversion on the multi-source sensor data to construct a large database.

9. The 5GRedcap IoT control system for oilfield equipment based on big data as described in claim 7, characterized in that, The abnormal event determination module includes: The task event data extraction unit is used to parse the target process flow and associated equipment identifiers to be monitored from the monitoring request from the user terminal; based on the associated equipment identifiers, it extracts the task event dataset corresponding to the associated equipment during the execution of the target process flow from the big data; the task event dataset is a multi-dimensional sensor data sequence arranged in the time sequence of process steps.

10. The 5GRedcap IoT control system for oilfield equipment based on big data as described in claim 9, characterized in that, The abnormal event determination module further includes a first processing unit or a second processing unit: The first processing unit is used to divide the task event dataset into time windows to obtain several sub-data sequences arranged in time sequence; input the sub-data sequences into a pre-trained time-series prediction model, and output the operating load change trend and preset key output flow change trend of the associated equipment in a future preset time period. Based on the operating load change trend, determine the predicted load value sequence for the future preset time period; Based on the trend of the preset key output flow rate, a sequence of predicted preset key output values ​​is determined within a preset future time period; if any value in the predicted load value sequence exceeds a preset load safety threshold range, or any value in the predicted preset key output value sequence exceeds a preset key output flow rate threshold range, then an abnormal event is determined to have occurred in the oilfield equipment. The second processing unit is used to select the equipment to be analyzed from the associated equipment one by one as the target oilfield equipment. For each target oilfield equipment, it acquires the continuous sampling data of the target oilfield equipment within a time window that traces back a preset time period with the current analysis time as the endpoint. The continuous sampling data includes the real-time operating load measurement value, the real-time preset key output flow measurement value, and the preset key output planned flow benchmark value collected at each continuous sampling time within the time window. Based on the continuous sampling data within the time window, the rated maximum operating load of the target oilfield equipment, and the historical failure rate statistics, the equipment operation status index of the target oilfield equipment at each sampling time is determined, forming an equipment operation status index sequence for this analysis; based on the equipment operation status index sequence, the standard deviation of the fluctuation degree used to quantify the discrete fluctuation amplitude of the equipment operation status index within the time window, and the first-order autocorrelation coefficient used to quantify the persistence of the evolution pattern of the equipment operation status index in the time series are determined; The system degradation entropy value of the target oilfield equipment at the current analysis time is determined based on the absolute value of the standard deviation of the fluctuation degree and the first-order autocorrelation coefficient; if the system degradation entropy value determined at the current analysis time is greater than the preset judgment threshold, it is determined that an abnormal event has occurred in the target oilfield equipment within the time window.

Images