An oil drilling and production equipment management method and system based on big data analysis

Abstract

The present invention discloses an oil drilling and production equipment management method and system based on big data analysis, which relates to the technical field of equipment management. It includes: installing a variety of sensors to collect the operation parameters and environmental parameters of the equipment in real time; calculating the dynamic baseline value of the equipment operation parameters based on the historical operation data of the equipment, and setting the normal fluctuation range; calculating the correlation degree between equipment and identifying the correlation mode between equipment by analyzing the operation parameters of multiple pieces of equipment in the same operation area; when the equipment operation parameters exceed the dynamic baseline range and the correlation between equipment changes abnormally, determining that the equipment is abnormal and triggering an early warning; evaluating the maintenance priority according to the abnormal detection result, the importance and the operation status of the equipment, and formulating an optimized maintenance plan. By introducing the calculation of the dynamic baseline value and the identification of the correlation mode between equipment, the present invention can issue an early warning in time when the parameters are abnormal and the correlation changes, improving the accuracy of abnormal detection.

Description

Technical Field

[0001] The present invention relates to the technical field of equipment management, and particularly to an oil drilling and production equipment management method and system based on big data analysis. Background Art

[0002] With the continuous growth of global energy demand, the oil drilling and production industry plays a crucial role in ensuring energy supply. Traditional equipment management methods mainly rely on regular manual inspections and experience-based fault diagnosis, which show many limitations when dealing with complex and changing drilling and production environments. In recent years, the introduction of Internet of Things (IoT) technology has made it possible to collect equipment operation parameters in real time through sensors. Further combined with big data analysis technology, a large amount of equipment data can be efficiently processed and deeply mined, so as to achieve accurate monitoring of equipment status and fault prediction.

[0003] However, the existing oil drilling and production equipment management technologies still have many deficiencies. First of all, many existing technologies rely on static parameter thresholds to determine equipment status, lack adaptability to the dynamic changes of equipment operation, and cannot identify deep-level equipment anomalies in a timely manner. Secondly, existing technologies usually take a single device as the management unit, ignoring the mutual correlation between devices, and cannot make full use of the correlation data generated by the collaborative operation of multiple devices, which to a certain extent limits the accuracy of fault diagnosis and the timeliness of early warning. Summary of the Invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides an oil drilling and production equipment management method and system based on big data analysis to solve the problems in the existing oil drilling and production equipment management, such as relying on static parameter thresholds, ignoring the correlation between devices, and being unable to dynamically optimize the maintenance plan.

[0006] To solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, an embodiment of the present invention provides an oil drilling and production equipment management method based on big data analysis, which includes: installing a variety of sensors to collect the operation parameters and environmental parameters of the equipment in real time; calculating the dynamic baseline value of the equipment operation parameters based on the historical operation data of the equipment, and setting the normal fluctuation range; calculating the correlation degree between devices by analyzing the operation parameters of multiple devices in the same operation area, and identifying the correlation mode between devices; when the equipment operation parameters exceed the dynamic baseline range and the correlation between devices changes abnormally, determining that the equipment is abnormal and triggering an early warning; evaluating the maintenance priority according to the abnormal detection result, the importance and the operation status of the equipment, and formulating an optimized maintenance plan.

[0008] As a preferred embodiment of the oil drilling and production equipment management method based on big data analysis according to the present invention, the multiple sensors include a temperature sensor, a pressure sensor, a vibration sensor, a flowmeter, an environmental temperature and humidity sensor, and a barometric pressure sensor; the operating parameters of the equipment include pressure, temperature, vibration, and flow; and the environmental parameters include environmental temperature, humidity, and barometric pressure.

[0009] As a preferred embodiment of the oil drilling and production equipment management method based on big data analysis according to the present invention, based on the historical operating data of the equipment, calculating the dynamic baseline values of the equipment operating parameters and setting the normal fluctuation range includes the following steps:

[0010] Extracting the historical operating data of the equipment from the data storage system;

[0011] Preprocessing the historical data;

[0012] According to the operating characteristics of the equipment, setting the size of the sliding window, setting the data acquisition frequency, and calculating the moving average value of each parameter and the moving standard deviation ;

[0013] Using the calculated moving average value and the moving standard deviation , calculating the lower limit and the upper limit at each time point, and recording the lower limit and the upper limit as the normal fluctuation range of the parameter;

[0014] Whenever a new data point arrives, the sliding window moves forward by one data point, removing the earliest data point , and adding the new data point ;

[0015] Recalculating the moving average value and the moving standard deviation within the current window, and updating the normal fluctuation range;

[0016] Defining a weight for each data point , calculating the weighted moving average value , and using to replace the moving average value ;

[0017] Defining an operating mode detection mechanism to determine whether a mode change has occurred by analyzing the operating parameters of the equipment;

[0018] When the operating mode of the equipment changes, adjusting the baseline value through the weighted moving average, dynamically adjusting the parameter multiple according to the change of the operating mode , and recalculating the lower limit and the upper limit.

[0019] As a preferred solution of the oil drilling and production equipment management method based on big data analysis of the present invention, wherein: the defined operation mode detection mechanism determines whether a mode change occurs by analyzing the operation parameters of the equipment, including the following steps,

[0020] Conduct preliminary data analysis on the operation parameters of the equipment collected in real time, and evaluate the importance of each parameter for operation mode recognition;

[0021] Use correlation analysis to check the correlation between parameters and eliminate redundant parameters;

[0022] Determine the final set of key parameters for operation mode detection based on importance evaluation and redundancy check;

[0023] Based on the set of key parameters, synchronize and align data at different frequencies through interpolation method, standardize the data of each parameter, extract the statistical features of time series data using the sliding window technique, and use principal component analysis to reduce the feature dimension;

[0024] Select the Gaussian mixture model GMM, determine the optimal number of clusters K through the GMM model selection method, use the K-Means algorithm for initial clustering, set the initial mean and covariance of GMM, and use the expectation maximization algorithm to train GMM and optimize the parameter setting;

[0025] Use historical data to train the GMM model, and evaluate the fitting effect of the GMM model through cross-validation and statistical indicators;

[0026] Set the size of the sliding window for real-time detection;

[0027] Input the operation parameter data of the equipment collected in real time into the trained GMM model, and calculate the compliance probability of each sample belonging to each operation mode;

[0028] When the belonging mode of M consecutive data points is different from that before, and the compliance probability exceeds the set threshold, it is determined that a mode change has occurred.

[0029] As a preferred solution of the oil drilling and production equipment management method based on big data analysis of the present invention, wherein: by analyzing the operation parameters of multiple equipment in the same operation area, calculate the degree of association between the equipment, and identify the association mode between the equipment, including the following steps,

[0030] Based on the obtained operation parameters of each equipment, use the Pearson correlation coefficient to calculate the correlation coefficient between each set of equipment operation parameters;

[0031] Construct a correlation matrix, where the rows and columns represent the operation parameters of different equipment respectively, and the values in the matrix represent the correlation coefficients of the corresponding parameter pairs;

[0032] Classify the correlation coefficients to identify pairs of highly correlated devices;

[0033] Based on historical data, calculate the correlation coefficients between devices, statistically analyze their distribution, and determine the normal range of the correlation coefficients between each group of devices through statistical analysis;

[0034] Identify devices whose current parameters deviate from the baseline through the set dynamic baseline range, calculate the correlation coefficients between this device and other devices in the same operation area, and determine whether the correlation coefficients are within the normal range. If they exceed the range, record it as an associated anomaly.

[0035] As a preferred solution of the oil drilling and production equipment management method based on big data analysis described in the present invention, wherein: when the device operation parameters exceed the dynamic baseline range and the correlation between devices changes abnormally, determine that the device is abnormal and trigger an early warning, including the following steps,

[0036] Obtain the latest device operation parameter data of the device in real time, obtain the lower and upper limits of the current time point t, and determine whether the device operation parameters exceed the normal fluctuation range;

[0037] If it is determined that the parameter exceeds the standard, record the event;

[0038] Obtain the list of all devices determined to exceed the standard, and for each device with parameter exceeding the standard, calculate its correlation coefficient with other devices in the same operation area;

[0039] For each pair of devices, determine whether their correlation coefficients deviate from the historical average correlation coefficient by more than the set tolerance range, and record all pairs of devices determined to be associated abnormally;

[0040] Obtain all the device pairing information determined to be abnormal from the parameter exceeding the standard and the associated anomaly;

[0041] Determine that the device is abnormal only when the device parameter exceeds the standard and the correlation with other devices is abnormally satisfied at the same time;

[0042] After determining that the device is abnormal, trigger warning information through multiple channels, and maintenance personnel obtain the warning information and respond to the relevant information for fault handling;

[0043] The warning information includes device ID, abnormal parameter, current value, historical baseline value, list of abnormally associated devices, and timestamp.

[0044] As a preferred solution of the oil drilling and production equipment management method based on big data analysis described in the present invention, wherein: according to the anomaly detection results, the importance and operating conditions of the device, evaluate the maintenance priority and formulate an optimized maintenance plan, including the following steps,

[0045] Determine the anomaly severity, equipment criticality, and operating downtime as risk assessment indicators;

[0046] Normalize the data of each indicator;

[0047] Adopt the multi - indicator weighted scoring method, combine different risk factors according to weights, and obtain the comprehensive risk score of the equipment;

[0048] According to the calculated risk score, divide the equipment into three risk levels: high, medium, and low;

[0049] If the equipment is classified as high - risk, immediately arrange maintenance tasks, increase the monitoring frequency, track the equipment operation status in real - time, and shorten the equipment maintenance cycle according to the risk score;

[0050] For medium - risk equipment, perform equipment maintenance regularly according to the pre - determined maintenance plan, dynamically adjust the maintenance cycle according to the equipment operation condition and risk score, and adjust the maintenance strategy based on real - time monitoring and risk score;

[0051] For low - risk equipment, perform maintenance according to the formulated maintenance plan and extend the equipment maintenance cycle according to the risk score.

[0052] In a second aspect, the present invention provides an oil drilling and production equipment management system based on big data analysis, including a data acquisition module responsible for collecting the operation parameters and environmental parameters of the equipment in real - time by installing a variety of sensors; a dynamic baseline setting module responsible for calculating the dynamic baseline value of the equipment operation parameters based on the historical operation data of the equipment and setting the normal fluctuation range; an equipment correlation analysis module responsible for calculating the correlation degree between equipment and identifying the correlation mode between equipment by analyzing the operation parameters of multiple equipment in the same operation area; an anomaly detection and warning module responsible for determining equipment anomalies and triggering warnings when the equipment operation parameters exceed the dynamic baseline range and the correlation between equipment changes abnormally; a maintenance strategy optimization module responsible for evaluating the maintenance priority according to the anomaly detection results, the importance and operation status of the equipment, and formulating an optimized maintenance plan.

[0053] In a third aspect, an embodiment of the present invention provides a computer device, including a memory and a processor, where the memory stores a computer program, and: when the computer program is executed by the processor, any step of the oil drilling and production equipment management method based on big data analysis as described in the first aspect of the present invention is implemented.

[0054] In a fourth aspect, an embodiment of the present invention provides a computer - readable storage medium, on which a computer program is stored, and: when the computer program is executed by the processor, any step of the oil drilling and production equipment management method based on big data analysis as described in the first aspect of the present invention is implemented.

[0055] The beneficial effects of the present invention are as follows: By introducing the calculation of dynamic baseline values and the recognition of the association patterns between devices, it is possible to accurately monitor the fluctuations of the operating parameters of the devices and issue early warnings in a timely manner when the parameters are abnormal and the correlations change. This method not only improves the accuracy of anomaly detection but also significantly enhances the intelligent level and maintenance efficiency of device management by optimizing the maintenance priorities and formulating scientific maintenance plans, providing effective technical support for the device management in the oil drilling and production industry. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention, and those of ordinary skill in the art can obtain other drawings based on these drawings without creative efforts.

[0057] Figure 1 It is a flowchart of the oil drilling and production equipment management method based on big data analysis in Embodiment 1.

[0058] Figure 2 It is a flowchart of device anomaly determination and early warning triggering in Embodiment 1. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0059] In order to make the above objects, features, and advantages of the present invention more obvious and understandable, the following detailed description of the specific embodiments of the present invention will be made with reference to the accompanying drawings of the specification.

[0060] In the following description, many specific details are set forth in order to fully understand the present invention. However, the present invention can also be implemented in other ways different from those described herein. Those skilled in the art can make similar extensions without departing from the connotation of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0061] Secondly, the so-called "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that can be included in at least one implementation manner of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it an individual or alternative embodiment that is mutually exclusive with other embodiments.

[0062] Embodiment 1, referring to Figure 1 and Figure 2 , is the first embodiment of the present invention. This embodiment provides an oil drilling and production equipment management method based on big data analysis, including the following steps:

[0063] S1. By installing a variety of sensors, the operating parameters and environmental parameters of the devices are collected in real time.

[0064] Specifically, select a high-precision pressure sensor with a pressure range suitable for the equipment requirements, such as the PTP series pressure sensor, which has high sensitivity and linearity. For the temperature sensor, choose an industrial-grade thermocouple or RTD (resistance temperature detector), such as PT100, to ensure the accuracy of temperature measurement. For the vibration sensor, use an accelerometer (such as a triaxial accelerometer) for vibration monitoring to ensure that multi-directional vibration data of the equipment can be captured. Select an electromagnetic flowmeter or a vortex flowmeter to ensure the stability and accuracy of flow measurement. Use a high-precision temperature and humidity sensor, such as DHT22, to ensure the reliability of environmental monitoring data. Select an accurate barometric pressure sensor, such as BMP280, for monitoring atmospheric pressure changes;

[0065] According to the equipment structure and operating characteristics, deploy the sensors at key parts of the equipment. For example, install the pressure sensor at the inlet and outlet of the pressure vessel, and install the vibration sensor at the motor bearing. Deploy temperature and humidity sensors in different areas around the equipment to ensure a comprehensive reflection of the environmental conditions and avoid blind spots;

[0066] Collect equipment operation parameters including pressure, temperature, vibration, and flow, as well as environmental parameters affecting equipment operation including environmental temperature and humidity; ensure that the collected data meets the accuracy requirements of equipment monitoring. For example, the error of the pressure sensor does not exceed ±1%, and the error of the temperature sensor does not exceed ±0.5°C; set the collection frequency of different parameters according to the equipment characteristics. For example, vibration data may require a higher collection frequency (such as 10 times per second), while environmental temperature and humidity can be once per minute;

[0067] Select an edge computing device with high computing power and low power consumption, such as an industrial-grade edge gateway based on the ARM architecture (such as the NVIDIA Jetson series), for near-real-time data processing and preprocessing. Install the edge computing node near the equipment to reduce data transmission latency and improve the response speed; deploy data collection software (such as Node-RED, Ignition SCADA) on the edge computing node, configure the reading interfaces and data formats of each sensor to ensure accurate data reading, and configure the software to support multiple communication protocols (such as Modbus, OPC UA) to ensure compatibility with different types of sensors;

[0068] Use a Moving Average Filter to smooth the data, use the Z-score method to detect and process abnormal data, convert the data collected by different sensors into a unified format (such as JSON, CSV) to ensure the convenience of subsequent data storage and analysis, standardize the timestamps to ensure the synchronization of each source data in time.

[0069] S2. Calculate the dynamic baseline value of the device operation parameters based on the historical operation data of the device, and set the normal fluctuation range, including the following steps:

[0070] S2.1 Extract the historical operation data of the device from the data storage system, determine the time range as the past six months to cover the long-term operation characteristics and working condition changes of the device. The extracted data includes device operation parameters (pressure, temperature, vibration, flow rate, etc.) and timestamps;

[0071] Perform missing value processing, outlier detection and processing, and data standardization processing on the historical data. Specifically, for continuous time series data, identify the positions of missing values in the data. For each missing value, use the two adjacent known data points for linear interpolation; calculate the Z-score of each data point to identify and process extreme outliers. For the data points determined to be outliers, replace them with the mean of adjacent data points; for each device operation parameter, calculate its mean and standard deviation, and perform standardization processing on all data points to ensure that the parameters are analyzed on the same scale.

[0072] S2.2 According to the device operation characteristics, set the size of the sliding window. Generally, select the data of the past 1 hour for calculation, set the data acquisition frequency to 1 time per minute, and ensure that the number of data points in each window is 60, covering the data of the past 1 hour; for each sliding window, calculate the mean of all data points in the window to obtain , when the sliding window moves forward, calculate the new in turn; for each sliding window, calculate the difference between all data points in the window and the moving average , square it, calculate the mean of these differences, and then take the square root to obtain ;

[0073] Use the calculated moving average and moving standard deviation to calculate the lower and upper limits of each time point. The expression is

[0074]

[0075] where and are the lower and upper limits at time respectively, =2 indicates that the normal fluctuation range is within 2 standard deviations above and below the mean; record the lower and upper limits as the normal fluctuation range of the parameter.

[0076] S2.3 Whenever a new data point arrives, the sliding window moves forward by one data point, removes the earliest data point , and adds the new data point ; Recalculate the moving average within the current window and the moving standard deviation , and update the normal fluctuation range;

[0077] Define weights for each data point such that the most recent data points have the highest weights, and calculate the weighted moving average , with the expression being

[0078]

[0079] where is the weight coefficient, usually giving higher weights to more recent data, is the original data value at time point , is the current time point, represents the sum of the products of each data point and its weight, represents the sum of all weights, ensuring that the calculation result is on the correct scale;

[0080] is defined as follows:

[0081]

[0082] represents the most recent data point with the highest weight, and the weights of the previous data points decrease sequentially; where is the sum of the integer sequence from 1 to , which is used to normalize the weight coefficient to ensure that the sum of the weights is 1, is the index variable, representing the relative position of the data points within the sliding window, is an index variable used to represent each term in the summation operation, ranges from 1 to , i.e., k = 1, 2, 3, …, ;

[0083] Use to replace the moving average to enhance the response speed to new data;

[0084] Define a running mode detection mechanism to determine whether a mode change has occurred by analyzing the operating parameters of the device; when the operating mode of the device changes (such as an increase in load, a large change in ambient temperature), adjust the baseline value through weighted moving average to ensure that the baseline can promptly reflect the new operating state, and dynamically adjust the parameter multiples according to the change in the operating mode To adapt to the fluctuating characteristics of the device and recalculate the lower and upper limits, the expression is

[0085]

[0086] where Dynamically adjust according to the current operating state of the device, and the specific strategy is as follows:

[0087] When the device is operating stably, = 2, that is, the normal fluctuation range is within 2 standard deviations above and below the mean. When the device is operating at high load, increase value (such as = 2.5), widen the fluctuation range to adapt to larger fluctuations. When the device is operating at low load, decrease value (such as = 1.5) to improve the sensitivity of anomaly detection.

[0088] S3. Define an operating mode detection mechanism to determine whether a mode change has occurred by analyzing the operating parameters of the device, including the following steps

[0089] S3.1. Determine the device operating state parameters for operating mode detection, including pressure, temperature, vibration, and flow rate. These parameters can comprehensively reflect the operating state and working condition changes of the device. Conduct preliminary data analysis on the real-time collected device operating parameters, evaluate the importance of each device operating parameter for operating mode recognition, and preferentially select the device operating parameters with greater influence;

[0090] Specifically, use the preprocessed device operating parameter data to train a random forest classifier, with the target variable being the operating mode of the device, and extract the feature importance score, expressed as

[0091]

[0092] where is the number of decision trees in the random forest, is the feature importance of the th tree;

[0093] Calculate the mutual information value between each feature and the target variable, which reflects the information contribution of each device operating parameter in operating mode recognition; normalize the random forest feature importance and mutual information scores, and combine them with weighted average to obtain the comprehensive feature importance score;

[0094] According to the comprehensive feature importance score, sort all device operating parameters in descending order to determine the importance order of each device operating parameter;

[0095] Set a threshold value for screening out device operation parameters that have a greater impact on the running mode recognition. The threshold value can be determined based on the cumulative importance contribution ratio. For example, select the parameters with a cumulative importance contribution reaching 80% as the core device operation parameters;

[0096] According to the cumulative contribution ratio, select the top k parameters as the core device operation parameters that have a greater impact on the running mode recognition.

[0097] Use correlation analysis to check the correlation between device operation parameters, eliminate redundant device operation parameters, and retain device operation parameters with independent information; determine the set of device operation parameters finally used for running mode detection according to importance evaluation and redundancy check.

[0098] S3.2. Based on the set of device operation parameters, synchronize and align the data at different frequencies through interpolation method to ensure that the data of all device operation parameters have the same timestamp. Standardize the data of each device operation parameter to eliminate the influence of dimension, so that different device operation parameters can be analyzed on the same scale. Use the sliding window technique to extract the statistical features of time series data, such as mean, standard deviation, maximum value, minimum value, kurtosis, and skewness, etc. Use principal component analysis to reduce the feature dimension, retain the main information, and reduce the computational complexity.

[0099] Select the Gaussian mixture model GMM, determine the optimal number of clusters K through the GMM model selection method (such as Akaike information criterion AIC, Bayesian information criterion BIC), use the K-Means algorithm for initial clustering, and set the initial mean and covariance of GMM; use the expectation maximization algorithm (EM) to train GMM and optimize the parameter setting;

[0100] Use historical data to train the GMM model, evaluate the fitting effect of the GMM model through cross-validation and statistical indicators (such as log-likelihood, AIC, BIC), select the appropriate value of K to make the GMM model avoid overfitting while ensuring the fitting degree; set the size of the sliding window for real-time detection, which is usually the same as the window size in the baseline setting (such as the past 1 hour);

[0101] Input the real-time collected device operation parameter data into the trained GMM model, and calculate the compliance probability of each sample belonging to each running mode; when the belonging modes of consecutive M data points are different from those before and the compliance probability exceeds the set threshold (such as 80%), it is determined that the mode has changed.

[0102] S4. By analyzing the operation parameters of multiple devices in the same operation area, calculate the degree of association between devices and identify the association mode between devices, including the following steps,

[0103] S4.1. Based on the obtained operating parameters of each device, use the Pearson correlation coefficient to calculate the correlation coefficient between each set of device operating parameters. For example, calculate the pressure correlation coefficient between device A and device B, the temperature correlation coefficient between device A and device C, and so on. The calculation formula is

[0104]

[0105] where is the Pearson correlation coefficient between parameter and , and are the parameter values of device and device in the th time window respectively, and are the average values of the parameters of device and device respectively, is the number of data points in the time window (such as 60 data points);

[0106] Construct a correlation matrix, where the rows and columns represent the operating parameters of different devices, and each value in the matrix represents the correlation coefficient between the corresponding parameter pairs. Specifically,

[0107] According to the calculated Pearson correlation coefficient, construct a correlation matrix. The rows and columns of the matrix represent the operating parameters of different devices, and each value in the matrix represents the correlation coefficient between the corresponding parameter pairs;

[0108] The following is a specific example of a correlation matrix, showing the Pearson correlation coefficients between the operating parameters of four devices (device A, device B, device C, device D):

[0109] Table 1 Correlation Matrix

[0110] Parameter device Device A Pressure (Pa) Device A_Temperature C9 Device B_Vibration (m / s) Device C_Flow rate (L / min) Device D Ambient Temperature (G Device D Humidity (%) Device D_Air pressure (hPa) Device A Pressure (Pa) 1.00 0.60 0.40 0.25 0.15 0.10 0.05 Device A Temperature (G 0.60 1.00 0.55 0.35 0.20 0.12 0.08 Device B Vibration (m / s²) 0.40 0.55 1.00 0.45 0.10 0.60 0.15 Device C Flow rate (L / min) 0.25 0.35 0.45 1.00 0.05 0.20 0.10 Device D Ambient Temperature (G) 0.15 0.20 0.10 0.05 1.00 0.25 0.30 Device D Humidity (%) 0.10 0.12 0.60 0.20 0.25 1.00 0.40 Device D Air pressure (hPa) 0.05 0.08 0.15 0.10 0.30 0.40 1.00

[0111] Matrix Explanation:

[0112] Autocorrelation: The values on the main diagonal of the matrix are all 1.00, indicating the perfect correlation of each parameter with itself.

[0113] Pressure of device A and temperature of device A (0.60): It indicates a moderately positive correlation between pressure and temperature, meaning that when the pressure of device A increases, the temperature has an upward trend.

[0114] Pressure of Equipment A and Vibration of Equipment B (0.40): It indicates a weak positive correlation between the pressure of Equipment A and the vibration of Equipment B, suggesting a certain association between the two.

[0115] Vibration of Equipment B and Flow Rate of Equipment C (0.35): It indicates a moderately positive correlation between the vibration of Equipment B and the flow rate of Equipment C.

[0116] Ambient Temperature of Equipment D and Other Parameters (0.05): It indicates that there is almost no linear correlation between the ambient temperature of Equipment D and other equipment parameters;

[0117] Classify the correlation coefficients to identify pairs of equipment with a high degree of correlation. Specifically, ∣ = 1 indicates a perfect positive or negative correlation, ∣ ∣ > 0.8 indicates a high degree of correlation, 0.5 < ∣ ∣ ≤ 0.8 indicates a moderate degree of correlation, ∣ ∣ ≤ 0.5 indicates a low degree of correlation.

[0118] S4.2. Calculate the correlation coefficients between each pair of equipment based on the historical data of the past six months, calculate the mean and standard deviation of the correlation coefficients, and statistically analyze their distribution. Determine the normal range of the correlation coefficients between each group of equipment. For example, a correlation coefficient between 0.8 and 1 is considered normal;

[0119] Identify the equipment whose current parameters deviate from the baseline through the set dynamic baseline. Calculate the correlation coefficient between this equipment and other equipment in the same operation area, and determine whether the correlation coefficient is within the normal range. If it exceeds the range, record it as an abnormal association. The determination condition for abnormal association is

[0120]

[0121] where, is the real-time correlation coefficient at time , is the historical average correlation coefficient, is the set tolerance range, such as 0.1.

[0122] S5. When the equipment operation parameters exceed the dynamic baseline range and the association between equipment changes abnormally, determine that the equipment is abnormal and trigger an alarm, including the following steps

[0123] S5.1. Real-time obtain the latest equipment operation parameter data of the equipment, obtain the lower and upper limits at the current time point t, and determine whether the equipment operation parameters exceed the normal fluctuation range;

[0124] If it is determined that the parameter exceeds the standard, record the event;

[0125] Obtain the list of all devices determined to be out of standard. For each device with out-of-standard parameters, calculate its correlation coefficient with other devices in the same operation area;

[0126] For each pair of devices , determine whether its correlation coefficient deviates from the historical average correlation coefficient by more than the set tolerance range , and record all pairs of devices determined to be associated abnormally;

[0127] Obtain all the paired device information determined to be abnormal from out-of-standard parameters and associated abnormalities;

[0128] Determine that a device is abnormal only when both the device parameters are out of standard and the association with other devices is abnormal;

[0129] S5.2. After determining that a device is abnormal, trigger warning information through multiple channels. The maintenance personnel obtain the warning information and respond to the relevant information to handle the fault;

[0130] Specifically, the content of the warning information includes: the device ID is the identifier of the abnormal device, the abnormal parameter is the specific parameter that exceeds the standard (such as pressure), the current value is the actual measured value of the current parameter , the historical baseline value is that of the corresponding time point and , the list of abnormally associated devices is other devices that are abnormally associated with this device, and the timestamp is the specific time when the abnormality occurred;

[0131] Furthermore, send an emergency warning message to the mobile phone of the maintenance personnel through the SMS gateway, send an email containing detailed warning information to the work email of the maintenance personnel, push real-time warning notifications through the enterprise internal mobile application, support instant viewing and operation, highlight the abnormal devices on the visualization interface of the monitoring system, and provide detailed information and operation guidelines;

[0132] Integrate the above transmission methods to ensure that the warning information is sent through all configured channels.

[0133] S6. Evaluate the maintenance priority according to the abnormal detection results, the importance and operating conditions of the devices, and formulate an optimized maintenance plan, including the following steps

[0134] S6.1. Determine the abnormal severity, device criticality, and operating downtime as risk assessment indicators;

[0135] Specifically, the anomaly severity: The severity of the anomaly is evaluated based on the magnitude and duration of the deviation of the device operating parameters from the baseline (e.g., the greater the deviation and the longer the duration, the higher the score). The device criticality: The device is scored according to its importance in the production process (e.g., critical devices have a high score, and non-critical devices have a low score). The operating downtime: The device is scored according to the operating downtime caused by the device failure (e.g., the longer the downtime, the higher the score).

[0136] Normalize the data of each index to ensure that the scores of different indexes are within the same range;

[0137] Adopt a multi-index weighted scoring method, combine different risk factors according to their weights, and obtain the comprehensive risk score of the device.

[0138] S6.2. According to the calculated risk score, divide the devices into three risk levels: high, medium, and low, and formulate corresponding maintenance strategies to ensure the reasonable allocation and efficient utilization of maintenance resources;

[0139] Specifically, according to the risk score, allocate the devices to high, medium, and low risk levels,

[0140]

[0141] If the device is classified as high risk, immediately arrange maintenance tasks to reduce the risk of device failure, increase the monitoring frequency, track the device operation status in real time, promptly discover potential problems, and appropriately shorten the maintenance cycle of the device according to the risk score to ensure that the device is in the best operating state;

[0142] For medium-risk devices, perform device maintenance regularly according to the predetermined maintenance plan to prevent potential failures, dynamically adjust the maintenance cycle according to the device operation status and risk score, improve the utilization efficiency of maintenance resources, and timely adjust the maintenance strategy based on the real-time monitoring and risk assessment results to ensure the healthy operation of the device;

[0143] For low-risk devices, perform maintenance according to the established maintenance plan to keep the device in good operating condition, appropriately extend the maintenance cycle of the device according to the risk score, reduce the maintenance frequency, optimize the allocation of maintenance resources, and perform maintenance only when necessary to reduce unnecessary maintenance operations and costs;

[0144] Generate corresponding maintenance plans according to the risk levels of the devices, clarify the maintenance tasks, time, and responsible persons, and assign the maintenance tasks to relevant maintenance personnel to ensure the timely execution and tracking of the tasks.

[0145] This embodiment also provides an oil drilling and production equipment management system based on big data analysis, including: a data acquisition module, which is responsible for collecting the operation parameters and environmental parameters of the equipment in real time by installing a variety of sensors; a dynamic baseline setting module, which is responsible for calculating the dynamic baseline value of the equipment operation parameters based on the historical operation data of the equipment and setting the normal fluctuation range; an equipment correlation analysis module, which is responsible for calculating the correlation degree between equipment and identifying the correlation mode between equipment by analyzing the operation parameters of multiple equipment in the same operation area; an abnormal detection and warning module, which is responsible for determining equipment abnormalities and triggering warnings when the equipment operation parameters exceed the dynamic baseline range and the correlation between equipment changes abnormally; a maintenance strategy optimization module, which is responsible for evaluating the maintenance priority according to the abnormal detection results, the importance and operation status of the equipment, and formulating an optimized maintenance plan.

[0146] This embodiment also provides a computer device applicable to the situation of the oil drilling and production equipment management method based on big data analysis, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the oil drilling and production equipment management method based on big data analysis as proposed in the above embodiment.

[0147] This computer device can be a terminal. The computer device includes a processor, a memory, a communication interface, a display screen, and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner. The wireless manner can be implemented through WIFI, a carrier network, NFC (Near Field Communication), or other technologies. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer covered on the display screen, or a button, a trackball, or a touchpad set on the outer shell of the computer device, or an external keyboard, touchpad, or mouse, etc.

[0148] This embodiment also provides a storage medium, on which a computer program is stored. When the program is executed by a processor, it implements the method for managing oil drilling and production equipment based on big data analysis as proposed in the above embodiment; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM for short), electrically erasable programmable read-only memory (EEPROM for short), erasable programmable read-only memory (EPROM for short), programmable read-only memory (PROM for short), read-only memory (ROM for short), magnetic memory, flash memory, magnetic disk or optical disc.

[0149] In summary, the present invention: By installing a variety of sensors to collect the operating parameters and environmental parameters of the equipment in real time, multi-dimensional and real-time equipment data is obtained, achieving comprehensive coverage and high-precision monitoring of data in equipment management, and significantly improving the timeliness and accuracy of equipment management; By extracting the historical operating data of the equipment, calculating the dynamic baseline values of the equipment operating parameters, and setting the normal fluctuation range, the real-time update and adaptability of the baseline values are ensured, and abnormal fluctuations during equipment operation can be more accurately identified, reducing false alarms and missed alarms; Using the Pearson correlation coefficient to construct a correlation matrix, and performing clustering analysis through the Gaussian mixture model (GMM) and K-Means algorithm to identify the association patterns between equipment, realizing dynamic monitoring and collaborative management of the mutual influence between equipment, enhancing the accuracy and comprehensiveness of fault diagnosis, and helping to early warn potential linkage faults; By comprehensively considering the abnormal deviation of the dynamic baseline of the equipment operating parameters and the abnormal changes in the correlation between equipment, the equipment abnormality can be accurately determined and early warning information can be sent out in a timely manner. This not only improves the accuracy and timeliness of abnormality identification, but also ensures that maintenance personnel can respond quickly and perform fault handling through multi-channel transmission of early warning information, thus guaranteeing the continuity of the production process and the safe operation of the equipment, and reducing the risk of production stagnation and maintenance costs.

[0150] Embodiment 2, referring to Table 1, is the second embodiment of the present invention. To further verify the technical solution of the present invention, experimental simulation data of the method for managing oil drilling and production equipment based on big data analysis is given.

[0151] To verify the effectiveness and advantages of the present invention, the following specific embodiments are designed. Four key devices (Device A, Device B, Device C, and Device D) at a certain oil drilling and production site are selected as the test objects. These devices are respectively equipped with temperature sensors, pressure sensors, vibration sensors, flow meters, as well as environmental temperature and humidity sensors and air pressure sensors to collect various device operation parameters and environmental parameters in real time. During the test, the device operation data is monitored and analyzed simultaneously by the existing static parameter threshold management method and the method of the present invention based on big data analysis to ensure the integrity and comparability of the data.

[0152] First, by installing a variety of sensors, the operation parameters of the device (including pressure, temperature, vibration, flow) and environmental parameters (including environmental temperature, humidity, air pressure) are collected in real time. The data collection frequency is set to once per minute to ensure the high timeliness and accuracy of the data. Subsequently, based on the historical operation data of the device in the past six months, the method of the present invention is used to calculate the dynamic baseline value of the device operation parameters and set the normal fluctuation range. During the data processing process, missing value processing, outlier detection and processing, and data normalization processing are carried out to eliminate data noise and interference and improve the data quality.

[0153] After the dynamic baseline value is calculated, by analyzing the operation parameters of multiple devices in the same operation area, the correlation degree between the devices is calculated to identify the correlation mode between the devices. The specific steps include using the Pearson correlation coefficient to construct a correlation matrix and applying the Gaussian mixture model (GMM) and K-Means algorithm for clustering analysis to identify the mutual influence relationship and collaborative operation mode between the devices. This step helps to discover potential linkage failures and improve the accuracy and timeliness of fault diagnosis.

[0154] When the device operation parameters exceed the dynamic baseline range and the correlation between the devices changes abnormally, the method of the present invention can accurately determine the device abnormality and trigger an alarm to ensure that maintenance personnel can quickly obtain key information and make a response. In addition, according to the abnormal detection results, the importance and operation status of the device, the maintenance priority is evaluated and an optimized maintenance plan is formulated. The formulation of the maintenance plan is based on the multi-index weighted scoring method, comprehensively considering the abnormal severity of the device, the importance score in the production process, and the historical operation downtime data, so as to realize the reasonable allocation and efficient utilization of maintenance resources.

[0155] During the test, by comparing the existing static parameter threshold management method and the big data analysis management method of the present invention, data on the abnormal detection rate, false alarm rate, maintenance response time, and maintenance cost of both are collected. The test results show that the method of the present invention is superior to the existing method in all indicators, showing significant advantages and innovation.

[0156] Specifically, as shown in Table 2 below:

[0157] Table 2 Experimental data table

[0158] Parameter Existing method of Device A Method of the present invention for Device A Existing method of Device B Method of the present invention for Device B Existing method of Device C Method of the present invention for Device C Existing method of Device D Method of the present invention for Device D Abnormal detection rate (%) 75 95 70 90 80 93 65 88 False alarm rate (%) 15 5 20 4 18 6 25 3 Maintenance response time (hours) 48 24 50 22 45 20 60 18 Maintenance cost (ten thousand yuan) 10 7 12 8 11 7.5 15 9 Device downtime (hours / month) 30 10 35 8 25 7 40 6

[0159] It can be clearly seen from the above experimental data table that the present invention is significantly superior to the traditional existing static parameter threshold management method in multiple key performance indicators. First of all, in terms of the anomaly detection rate, the detection rate of device A has increased from 75% to 95%, that of device B has increased from 70% to 90%, that of device C has increased from 80% to 93%, and that of device D has increased from 65% to 88%. This shows that the present invention has higher accuracy and sensitivity in identifying device anomalies, can detect potential problems earlier, and reduce the impact of device failures on production.

[0160] Secondly, in terms of the false alarm rate, the false alarm rate of device A has decreased from 15% to 5%, that of device B has decreased from 20% to 4%, that of device C has decreased from 18% to 6%, and that of device D has decreased from 25% to 3%. The reduction of the false alarm rate not only reduces unnecessary maintenance operations, saves maintenance resources, but also improves the trust of maintenance personnel in early warning information.

[0161] Under the existing method, the maintenance response time of device A is 48 hours, that of device B is 50 hours, that of device C is 45 hours, and that of device D is 60 hours. After adopting the method of the present invention, the response time is significantly shortened to 24 hours, 22 hours, 20 hours and 18 hours. This means that when a device anomaly occurs, maintenance personnel can take measures more quickly, reduce the risk of fault expansion, and ensure the continuity of the production process and the normal operation of the device.

[0162] Regarding the maintenance cost, the maintenance cost of device A has decreased from 100,000 yuan to 70,000 yuan, that of device B has decreased from 120,000 yuan to 80,000 yuan, that of device C has decreased from 110,000 yuan to 75,000 yuan, and that of device D has decreased from 150,000 yuan to 90,000 yuan. This is mainly due to the reduction of the false alarm rate and the shortening of the maintenance response time, which reduces unnecessary maintenance operations and device downtime.

[0163] Finally, from the data of device downtime, the downtime of device A under the existing method is 30 hours / month, that of device B is 35 hours / month, that of device C is 25 hours / month, and that of device D is 40 hours / month. After adopting the method of the present invention, the downtime is reduced to 10 hours / month, 8 hours / month, 7 hours / month and 6 hours / month respectively. This result shows that the method of the present invention can effectively prevent the occurrence of major device failures and improve the availability and production efficiency of the device.

[0164] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered within the scope of the claims of the present invention.

Claims

1. A method for managing oil drilling equipment based on big data analysis, characterized in that: include, By installing a variety of sensors, the operating parameters and environmental parameters of the equipment can be collected in real time; Based on the historical operation data of the equipment, calculate the dynamic baseline value of the equipment operation parameters and set the normal fluctuation range; By analyzing the operating parameters of multiple devices in the same operating area, the degree of correlation between devices is calculated and the correlation pattern between devices is identified; When the equipment operating parameters exceed the dynamic baseline range and the correlation between equipment changes abnormally, the equipment is judged to be abnormal and an early warning is triggered; Evaluate maintenance priorities and develop optimized maintenance plans based on abnormal detection results, equipment importance, and operating conditions; By analyzing the operating parameters of multiple devices in the same operating area, calculating the degree of association between devices, and identifying the association pattern between devices, the following steps are included: Based on the acquired operating parameters of each device, the Pearson correlation coefficient is used to calculate the correlation coefficient between each set of device operating parameters; Construct a correlation matrix, where the rows and columns represent the operating parameters of different devices, and the values ​​in the matrix represent the correlation coefficients of the corresponding parameter pairs; Classify the correlation coefficients and identify highly correlated device pairs; Based on historical data, calculate the correlation coefficients between devices, count their distribution, and determine the normal range of the correlation coefficients between each group of devices through statistical analysis; The device whose current parameters deviate from the baseline is identified through the set dynamic baseline range, the correlation coefficient between the device and other devices in the same operating area is calculated, and it is determined whether the correlation coefficient is within the normal range. If it exceeds the normal range, it is recorded as an associated abnormality.

2. The oil drilling equipment management method based on big data analysis according to claim 1, characterized in that: The multiple sensors include temperature sensors, pressure sensors, vibration sensors, flow meters, ambient temperature and humidity sensors, and air pressure sensors; The operating parameters of the equipment, including pressure, temperature, vibration and flow; The environmental parameters include ambient temperature, humidity and air pressure.

3. The oil drilling equipment management method based on big data analysis as claimed in claim 2, characterized in that: Based on the historical operation data of the equipment, the dynamic baseline value of the equipment operation parameters is calculated and the normal fluctuation range is set, including the following steps: Extract historical operation data of the equipment from the data storage system; Preprocess historical data; According to the equipment operation characteristics, set the size of the sliding window, set the data collection frequency, and calculate the moving average of each parameter and moving standard deviation ; Use the calculated moving average and moving standard deviation , calculate the lower limit and upper limit of each time point, and record the lower limit and upper limit as the normal fluctuation range of the parameter; Every time a new data point Arrived, the sliding window moves forward one data point and removes the earliest data point , and add new data points ; Recalculate the moving average within the current window and moving standard deviation , update the normal fluctuation range; Define weights for each data point , calculate the weighted moving average ,use Alternative moving average ; Define the operation mode detection mechanism to determine whether the mode changes occur by analyzing the operation parameters of the device; When the operating mode of the device changes, the baseline value is adjusted through weighted moving average, and the parameter multiples are dynamically adjusted according to the change of the operating mode. , and recalculate the lower and upper bounds.

4. The oil drilling equipment management method based on big data analysis as claimed in claim 3 is characterized by: The operation mode detection mechanism is defined to determine whether a mode change occurs by analyzing the operation parameters of the device, including the following steps: Conduct preliminary data analysis on the equipment operating parameters collected in real time to evaluate the importance of each parameter to the identification of the operating mode; Use correlation analysis to check the correlation between parameters and eliminate redundant parameters; Based on importance assessment and redundancy check, determine the key parameter set that will be used for operation mode detection; Based on the key parameter set, the data of different frequencies are synchronized and aligned by interpolation method, the data of each parameter is standardized, the statistical features of time series data are extracted by sliding window technology, and the feature dimension is reduced by principal component analysis; Select Gaussian mixture model GMM, determine the optimal number of clusters K through GMM model selection method, use K-Means algorithm for initial clustering, set the initial mean and covariance of GMM, use expectation maximization algorithm to train GMM, and optimize parameter settings; Use historical data to train the GMM model, and evaluate the fitting effect of the GMM model through cross-validation and statistical indicators; Set the sliding window size for real-time detection; Input the real-time collected equipment operation parameter data into the trained GMM model to calculate the probability of each sample belonging to each operation mode; When the mode of M consecutive data points is different from the previous one and the matching probability exceeds the set threshold, it is judged as a mode change.

5. The oil drilling equipment management method based on big data analysis according to claim 4 is characterized in that: When the equipment operating parameters exceed the dynamic baseline range and the correlation between equipment changes abnormally, the equipment is judged to be abnormal and an early warning is triggered, including the following steps: Obtain the latest equipment operating parameter data of the equipment in real time, obtain the lower limit and upper limit of the current time point t, and determine whether the equipment operating parameters exceed the normal fluctuation range; If it is determined that the parameter exceeds the limit, the event is recorded; Obtain a list of all equipment that is judged to be exceeding the standard, and for each equipment whose parameters exceed the standard, calculate its correlation coefficient with other equipment in the same operating area; For each pair of devices, determine whether the correlation coefficient deviates from the historical average correlation coefficient beyond the set tolerance range, and record all device pairs that are determined to be abnormally correlated; Obtain all device pairing information that is determined to be abnormal from parameter exceeding limits and associated anomalies; Only when the device parameters exceed the standard and the associated abnormalities of other devices are met at the same time, the device is judged to be abnormal; After determining that the equipment is abnormal, early warning information is triggered through multiple channels. Maintenance personnel obtain the early warning information and respond to relevant information to handle the fault; The warning information includes device ID, abnormal parameters, current values, historical baseline values, a list of abnormal associated devices and a timestamp.

6. The oil drilling equipment management method based on big data analysis according to claim 5, characterized in that: According to the abnormal detection results, the importance and operating status of the equipment, the maintenance priority is evaluated and an optimized maintenance plan is formulated, including the following steps: Identify abnormality severity, equipment criticality, and operational downtime as risk assessment indicators; Normalize the data of each indicator; The multi-index weighted scoring method is used to combine different risk factors according to weights to obtain a comprehensive risk score for the equipment; Based on the calculated comprehensive risk score, the equipment is divided into three risk levels: high, medium, and low; If the equipment is classified as high risk, immediately arrange maintenance tasks, increase monitoring frequency, track the equipment operating status in real time, and shorten the equipment maintenance cycle based on the risk score; For medium-risk equipment, regular equipment maintenance is carried out according to the scheduled maintenance plan. The maintenance cycle is dynamically adjusted according to the equipment operating status and risk score. The maintenance strategy is adjusted based on real-time monitoring and risk score. For low-risk equipment, maintenance is carried out according to the established maintenance plan, and the maintenance cycle of the equipment is extended based on the risk score.

7. A petroleum drilling equipment management system based on big data analysis, based on the petroleum drilling equipment management method based on big data analysis according to any one of claims 1 to 6, characterized in that: include, The data acquisition module is responsible for collecting the operating parameters and environmental parameters of the equipment in real time by installing a variety of sensors; The dynamic baseline setting module is responsible for calculating the dynamic baseline value of the equipment operating parameters based on the historical operating data of the equipment and setting the normal fluctuation range; The equipment association analysis module is responsible for calculating the degree of association between equipment and identifying the association pattern between equipment by analyzing the operating parameters of multiple equipment in the same operation area; The anomaly detection and warning module is responsible for determining that the equipment is abnormal and triggering an alarm when the equipment operating parameters exceed the dynamic baseline range and the correlation between the equipment changes abnormally; The maintenance strategy optimization module is responsible for evaluating maintenance priorities and formulating optimized maintenance plans based on anomaly detection results, equipment importance, and operating conditions.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the oil drilling equipment management method based on big data analysis described in any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the oil drilling equipment management method based on big data analysis described in any one of claims 1 to 6 are implemented.

Images