Real-time monitoring method of oil well dynamic liquid level combined with dynamometer data

Abstract

The present invention discloses a method for real-time monitoring of the dynamic liquid level of an oil well in combination with dynamometer diagram data, and relates to the field related to oil well data processing. The method comprises: multi-dimensionally collecting the target oil well through an oil well monitoring network, drawing the first dynamometer diagram data of the target oil well based on the multi-dimensional sensor data set to extract features, generating multiple feature parameters to construct a dynamic evaluation model to calculate the dynamic liquid level of the target oil well, obtaining initial dynamic liquid depth data, performing dynamic analysis according to the multi-dimensional sensor data set, generating a correction factor to dynamically feedback the initial dynamic liquid depth data, drawing a dynamic liquid level change trend diagram, and combining the third dynamometer diagram data to monitor the dynamic liquid level of the target oil well in real time. The method solves the technical problem that the measurement accuracy of the dynamic liquid level of a traditional oil well is low, and it is impossible to continuously monitor it in real time, resulting in low accuracy of oil well management, and achieves the technical effect of combining the dynamometer diagram data to detect the dynamic liquid level of the oil well in real time and continuously, thereby improving the accuracy of oil well monitoring.

Description

Technical Field

[0001] The present application relates to the field of oil well data processing, and in particular to a method for real-time monitoring of oil well dynamic liquid level in combination with dynamometer diagram data. Background Art

[0002] Dynamic liquid level monitoring in oil wells is of great significance in the process of oilfield development and can provide key indicators of the operating status of oil wells. Dynamic liquid level refers to the dynamic height of the liquid in the wellbore, reflecting the fluid pressure, liquid supply and production performance of the oil well. Traditional dynamic liquid level measurement mainly relies on acoustic sounding, wellbore pressure monitoring and other means, but these methods are limited by measurement accuracy, time delay and data processing complexity. Acoustic sounding relies on the reflection of sound waves in the wellbore to judge the depth of the liquid surface, but due to complex factors such as the presence of gas-liquid mixtures and changes in wellbore diameter in the wellbore, it often leads to large measurement errors. In addition, acoustic sounding is usually performed regularly or manually, making it difficult to achieve continuous real-time monitoring. On the other hand, although the monitoring method based on wellbore pressure can infer the liquid surface depth to a certain extent, this method is sensitive to the characteristics of the fluid in the well, especially when the pressure fluctuates greatly, it is difficult to maintain the stability of the measurement results. Therefore, the existing technology still faces problems such as low measurement accuracy, inability to continuously monitor in real time, and large interference from the external environment in practical applications, which limits its application potential in intelligent management of oil fields. Summary of the invention

[0003] The present application provides a real-time monitoring method for the dynamic liquid level of oil wells combined with the dynamometer diagram data, and solves the technical problem that the traditional measurement accuracy of the dynamic liquid level of oil wells is low and it is impossible to continuously monitor in real time, resulting in low accuracy of oil well management. The present application achieves the technical effect of real-time and continuous detection of the dynamic liquid level of oil wells combined with the dynamometer diagram data, thereby improving the accuracy of oil well monitoring.

[0004] The present application provides a real-time monitoring method for the dynamic liquid level of an oil well in combination with dynamometer diagram data, comprising: performing multi-dimensional acquisition of a target oil well through an oil well monitoring network to obtain a multi-dimensional sensor data set, and drawing first dynamometer diagram data of the target oil well based on the multi-dimensional sensor data set; traversing the first dynamometer diagram data to perform feature extraction, generate multiple feature parameters, and construct a dynamic evaluation model based on the multiple feature parameters; calculating the dynamic liquid level of the target oil well through the dynamic evaluation model to obtain initial dynamic liquid depth data; performing dynamic analysis based on the multi-dimensional sensor data set to generate a correction factor, dynamically feeding back the initial dynamic liquid depth data based on the correction factor, and drawing a dynamic liquid level change trend graph; and monitoring the dynamic liquid level of the target oil well in real time according to the dynamic liquid level change trend graph in combination with third dynamometer diagram data.

[0005] In a possible implementation, the oil well monitoring network performs the following processing: traverses the target oil well to perform dynamic liquid level analysis and determine multiple deployment points; selects multiple sensor devices to be deployed based on the multiple deployment points, and associates and deploys the multiple sensor devices to be deployed according to the multiple deployment points to generate multiple device association chains, wherein each sensor device to be deployed has an association relationship with at least another sensor device to be deployed that is different from itself; performs multi-dimensional analysis in combination with the target oil well according to the multiple device association chains, and delineates multiple oil well sub-networks according to the analysis results;

[0006] The multiple oil well sub-networks are communicatively connected to a central monitoring platform to construct the oil well monitoring network.

[0007] In a possible implementation, multi-dimensional data collection is performed on the target oil well through the oil well monitoring network to obtain a multi-dimensional sensor data set, and the first indicator diagram data of the target oil well is drawn based on the multi-dimensional sensor data set, and the following processing is performed: multi-dimensional analysis is performed on the target oil well through the multiple oil well sub-networks to determine multiple multi-dimensional sensor data classes; the multiple oil well sub-networks are mapped according to the multiple multi-dimensional sensor data classes to obtain multiple oil well data networks; regression analysis is performed on the multi-dimensional sensor data set based on the multiple oil well data networks to generate an oil well operation data set, real-time analysis is performed based on the oil well operation data set, and the first indicator diagram data of the target oil well is drawn.

[0008] In a possible implementation, the first dynamometer diagram data is traversed to perform feature extraction, generate multiple feature parameters, and perform the following processing: extract a data acquisition time series based on the multidimensional sensor data set; traverse the first dynamometer diagram data according to the data acquisition time series to perform load analysis and determine the working load characteristics; traverse the first dynamometer diagram data according to the data acquisition time series to perform efficiency analysis and determine the working efficiency characteristics; standardize the working load characteristics and the work efficiency characteristics to generate the multiple feature parameters.

[0009] In a possible implementation, a dynamic evaluation model is constructed based on the multiple characteristic parameters, and the following processing is performed: a load difference parameter is calculated based on the workload characteristic in combination with the work efficiency characteristic; a work diagram area parameter is calculated based on the work efficiency characteristic in combination with the first work diagram data; multiple static parameters of the target oil well are retrieved, and a training data set is determined based on the load difference parameter, the work diagram area parameter and the multiple static parameters; multiple operating parameters of the target oil well are introduced, and training and calibration are performed based on the multiple operating parameters in combination with the training data set according to the data acquisition time series to obtain the dynamic evaluation model.

[0010] In a possible implementation, a dynamic analysis is performed on the multidimensional sensing data set to generate a correction factor, and the following processing is performed: the multidimensional sensing data set is dynamically calculated according to the multiple characteristic parameters to obtain a dynamic liquid level change data set, wherein the dynamic liquid level change trend includes dynamic liquid level change rate data and dynamic liquid level change amplitude data; based on the dynamic liquid level change amplitude data, the dynamic liquid level change rate data is iteratively updated in combination with the power diagram area parameter to draw a second power diagram data; the first power diagram data and the second power diagram data are subjected to waveform similarity analysis to generate a waveform similarity coefficient; it is determined whether the waveform similarity coefficient is greater than or equal to an expected similarity threshold; if the waveform similarity coefficient is less than the expected similarity threshold, a tracking instruction is generated, and the dynamic liquid level change data set is dynamically tracked by the tracking instruction to determine abnormal fluctuation data; the abnormal fluctuation data is traced to determine the source of the abnormal fluctuation, and continuous monitoring is performed according to the source of the abnormal fluctuation, and the correction factor is updated based on the monitoring result.

[0011] In a possible implementation, the initial dynamic fluid depth data is dynamically fed back according to the correction factor, a dynamic fluid level change trend graph is drawn, and the following processing is performed: taking the abnormal fluctuation source as an index, traversing the initial dynamic fluid depth data and comparing it with the dynamic fluid level change data set, and extracting the abnormal dynamic fluid depth change data; correcting the abnormal dynamic fluid depth change data according to the correction factor to generate dynamic fluid depth corrected change data; when the waveform similarity coefficient is less than the expected similarity threshold, correcting the first dynamometer diagram data according to the second dynamometer diagram data to generate third dynamometer diagram data; combining the data acquisition time series to construct a dynamic fluid change coordinate system, and based on the third dynamometer diagram data, filling the dynamic fluid depth corrected change data into the dynamic fluid change coordinate system to construct the dynamic fluid level change trend graph.

[0012] In a possible implementation, when the waveform similarity coefficient is less than the expected similarity threshold, the first dynamometer diagram data is corrected according to the second dynamometer diagram data to generate third dynamometer diagram data, and the following processing is performed: when the waveform similarity coefficient is less than the expected similarity threshold, the second dynamometer diagram data is weighted according to the waveform similarity coefficient to generate multiple weight coefficients; the second dynamometer diagram data is processed in descending order according to the multiple weight coefficients to generate a second dynamometer diagram sequence; the first dynamometer diagram data is traversed based on the second dynamometer diagram sequence to perform deviation correction to generate the third dynamometer diagram data.

[0013] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0014] The present application provides a method for real-time monitoring of the dynamic liquid level of an oil well in combination with the dynamometer diagram data, which can achieve the technical effect of improving the accuracy of oil well monitoring by performing real-time and continuous detection of the dynamic liquid level of the oil well in combination with the dynamometer diagram data, while solving the technical problem that the traditional measurement of the dynamic liquid level of an oil well has low accuracy and cannot be continuously monitored in real time, resulting in low accuracy of oil well management. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solution of the embodiment of the present invention, the accompanying drawings of the embodiment of the present invention will be briefly introduced below. A flow chart is used in the present application to illustrate the operations performed by the method according to the embodiment of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps can be processed in reverse order or simultaneously as needed. At the same time, other operations can also be added to these processes, or one or more operations can be removed from these processes.

[0016] Figure 1 A schematic flow chart of a method for real-time monitoring of the dynamic liquid level of an oil well in combination with dynamometer diagram data provided in an embodiment of the present application;

[0017] Figure 2 A schematic diagram of the process of drawing a dynamic liquid level change trend diagram for a real-time monitoring method of oil well dynamic liquid level combined with indicator diagram data provided in an embodiment of the present application. DETAILED DESCRIPTION

[0018] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.

[0019] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.

[0020] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict, and the terms "first\second" involved are merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "including" and "having" and any variations are intended to cover non-exclusive inclusions, for example, a process, method, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by technicians in the technical field of this application. The terms used herein are for the purpose of describing the embodiments of the present application only.

[0021] The present application embodiment provides a method for real-time monitoring of the dynamic liquid level of an oil well in combination with the dynamometer diagram data, such as Figure 1 As shown, the method includes:

[0022] Step A100, performing multi-dimensional acquisition of the target oil well through the oil well monitoring network to obtain a multi-dimensional sensor data set, and drawing the first indicator diagram data of the target oil well based on the multi-dimensional sensor data set; in a possible implementation, step A100 further includes step A110, traversing the target oil well to perform dynamic liquid level analysis and determine multiple deployment points; executing step A120, selecting multiple sensor devices to be deployed based on the multiple deployment points, and associating and deploying the multiple sensor devices to be deployed according to the multiple deployment points to generate multiple device association chains, wherein each sensor device to be deployed has an association relationship with at least one other sensor device to be deployed that is different from itself; executing step A130, performing multi-dimensional analysis in combination with the target oil well according to the multiple device association chains, and delineating multiple oil well sub-networks according to the analysis results; executing step A140, connecting the multiple oil well sub-networks to the central monitoring platform for communication, and constructing the oil well monitoring network.

[0023] Through preliminary monitoring, the dynamic liquid level data of the oil well is obtained, including key parameters such as liquid level depth, fluid pressure and well temperature, to analyze the dynamic changes of the fluid inside the wellbore, and to traverse and analyze the characteristics of the fluid dynamics, well wall structure, pressure distribution and other characteristics in the oil well to determine multiple deployment points. The multiple deployment points can be distributed at key positions of the oil well, such as the wellhead, the middle of the wellbore, the bottom, etc., to ensure that the liquid level changes and wellbore pressure can be fully monitored. According to the multiple deployment points, appropriate sensing equipment is selected, which can include pressure sensors, liquid level sensors, temperature sensors, etc. The sensing equipment will be installed at different positions such as the wellhead, wellbore, and pumping unit to monitor the multi-dimensional parameters in the wellbore, and then multiple sensing devices to be deployed are associated according to the sensing data type according to the multiple deployment points to generate multiple equipment association chains. Each sensor device in the multiple device association chains is associated with at least one other sensor device that is different from itself, ensuring that the data between sensors can complement each other and work together, and then collect multidimensional data in real time during the operation of the oil well according to the multiple device association chains. The multidimensional data can include liquid level dynamics, pressure fluctuations, temperature changes, etc. At the same time, the multidimensional data is comprehensively analyzed to evaluate the operating status and fluid dynamics of the oil well. Through the data integration of the association chain, the system can accurately depict the mutual influence between different points in the wellbore, and finally divide the oil well into multiple sub-networks based on the multidimensional analysis results. Each sub-network contains one or more sensor devices, covering a specific area of ​​the oil well, ensuring the monitoring independence and efficiency of each area. The monitoring range of each sub-network is determined according to the sensor layout of each sub-network, ensuring that all oil well areas are monitored in real time, and multiple oil well sub-networks are connected to the central monitoring platform by wireless or wired communication to ensure that the data in the sub-network can be transmitted to the central monitoring system in real time for unified management. An oil well monitoring network is constructed through the communication connection between multiple oil well sub-networks and the central platform. The oil well monitoring network is used to collect, analyze and feedback the operating status of oil wells in real time, thereby improving the intelligence level of oil well management.

[0024] In one possible implementation, step A100 further includes step A150, performing multidimensional analysis on the target oil well through the multiple oil well sub-networks to determine multiple multidimensional sensor data classes; executing step A160, mapping the multiple oil well sub-networks according to the multiple multidimensional sensor data classes to obtain multiple oil well data networks; executing step A170, performing regression analysis on the multidimensional sensor data set based on the multiple oil well data networks to generate an oil well operation data set, performing real-time analysis based on the oil well operation data set, and drawing the first indicator diagram data of the target oil well.

[0025] The oil well sub-network collects sensor data of multiple dimensions, including pressure, temperature, liquid level depth, load and displacement, etc., and performs preliminary analysis on the collected sensor data to determine the key monitoring parameters of the oil well, such as pressure distribution, temperature change, liquid level fluctuation, etc. in the wellbore. According to the preliminary analysis results, the multidimensional data in the operation process of the oil well is classified to determine multiple multidimensional sensor data classes. The multiple multidimensional sensor data classes may include: pressure data, temperature data, liquid level data, pump load data, etc., and each type of data reflects different aspects of the oil well. Then, according to the determined multidimensional sensor data class, the sensor data in each sub-network is mapped to generate a corresponding oil well data network, in which each data network corresponds to a multidimensional sensor data class, which is responsible for the transmission and analysis of specific categories of data. Then, by associating the sensor devices in each sub-network, the data flow relationship of the oil well data network is constructed to ensure that the sensor data in each data network can work together and achieve efficient data transmission.

[0026] Further, the regression analysis of the multi-dimensional sensor data set based on the multiple oil well data networks refers to modeling and analyzing the overall operating status of the oil well based on the sensor data of multiple dimensions. Regression analysis is to generate an oil well operation data set for reflecting the operating status of the oil well through the relationship between data of different dimensions, that is, the regression analysis process may include modeling the relationship between key parameters such as pressure, liquid level depth, temperature and pump load, identifying trends and abnormal conditions in the operation of the oil well, and then performing real-time analysis based on the oil well operation data set, and monitoring the current operating status of the target oil well. The real-time analysis is to accurately judge the current working status of the oil well through the continuously updated data set and regression model, and predict the future change trend, and draw the first indicator diagram data of the target oil well according to the real-time analysis results. The first indicator diagram data is based on the real-time pump load and displacement relationship, reflecting the working status of the pumping unit. The first indicator diagram data is also used to display the force and displacement changes of the pump in the lift and lower stroke.

[0027] Execute step A200, traverse the first dynamometer diagram data to perform feature extraction, generate multiple feature parameters, and construct a dynamic evaluation model based on the multiple feature parameters;

[0028] In one possible implementation, step A200 further includes step A210, extracting a data acquisition time series based on the multidimensional sensing data set; executing step A220, traversing the first dynamometer diagram data according to the data acquisition time series to perform load analysis and determine the working load characteristics; executing step A230, traversing the first dynamometer diagram data according to the data acquisition time series to perform efficiency analysis and determine the working efficiency characteristics; executing step A240, standardizing the working load characteristics and the working efficiency characteristics to generate the multiple characteristic parameters.

[0029] Acquire real-time data collected by various sensors from the multidimensional sensor data set, which may include pressure, load, displacement, temperature, etc. When acquiring data, synchronize the timestamp to ensure that all data correspond to the same time axis. And extract the data acquisition time series according to the timestamp to generate a time clue to represent the data changes of each sensor at different time points. Then, based on the extracted time series, traverse each data point in the first dynamometer diagram, analyze the load change trend and extract the load information of the pump at each moment. The load information is used to represent the force change of the pump under different displacements, and on this basis, extract the working load characteristics of the target oil well. Further, traverse the first dynamometer diagram data according to the data acquisition time series, analyze the working efficiency of the pump, which means calculating the ratio of the actual output power of the pump to the input power per unit time, and evaluating the working efficiency of the pump during the lifting and lowering process, which may include the total efficiency of the pump in the entire cycle, the efficiency change trend of the pump in different stages, and identify whether there is a reduction in efficiency or abnormality. On this basis, extract the working efficiency characteristics of the target oil well.

[0030] Finally, the extracted workload characteristics and work efficiency characteristics are standardized, that is, the workload characteristics and work efficiency characteristics are normalized or Z-score processed, so that each characteristic parameter can be compared at the same scale to generate multiple characteristic parameters. The multiple characteristic parameters are used to reflect the load and efficiency status of the pump at different time nodes, which is convenient for the system to model and optimize the overall working status of the oil well.

[0031] In one possible implementation, step A200 further includes step A250, performing calculation based on the workload characteristic in combination with the workload characteristic to obtain a load difference parameter; executing step A260, performing calculation based on the workload characteristic in combination with the first dynamometer diagram data to obtain a dynamometer diagram area parameter; executing step A270, retrieving multiple static parameters of the target oil well, and determining a training data set based on the load difference parameter, the dynamometer diagram area parameter and the multiple static parameters; executing step A280, introducing multiple operating parameters of the target oil well, and performing training and calibration based on the multiple operating parameters in combination with the training data set according to the data acquisition time series to obtain the dynamic evaluation model.

[0032] Extracting key data based on the working load characteristics can be the maximum and minimum loads of the pump, which are used to analyze the load fluctuation of the pump. On this basis, according to the working efficiency characteristics, the energy efficiency level of the pump under different working conditions is calculated, which means that by combining the load change and efficiency characteristics, the load difference parameter is calculated, and the load difference parameter is used to reflect the relationship between the load fluctuation and efficiency of the pump during the working cycle. Then, by integrating the dynamometer data and combining the working efficiency characteristics, the dynamometer area parameter is obtained, which reflects the actual working efficiency and output power of the pump during the entire working cycle.

[0033] Further, multiple static parameters of the target oil well are retrieved, including well depth, fluid density, wellhead pressure, wellbore temperature, etc. The multiple static parameters are used to characterize the geological and physical characteristics of the oil well. According to the load difference parameter and the work diagram area parameter, combined with the static parameters of the oil well, the data is classified and marked, and the training data sets under different working conditions are divided for subsequent model training and calibration. At the same time, multiple operating parameters of the target oil well are introduced. The multiple operating parameters may include the operating frequency of the pump, the up and down stroke time of the pumping unit, the liquid supply speed, etc. The multiple operating parameters are used to reflect the operating status of the oil well at different production stages.

[0034] Based on the operating parameters and training data sets, the dynamic evaluation model is trained according to the data acquisition time series. This means that the dynamic evaluation model is calibrated iteratively multiple times according to different operating parameters during the training process. The model can be trained using regression analysis, machine learning algorithms, etc. to ensure that the dynamic evaluation model can accurately predict the working status of the oil well under various operating conditions, thereby obtaining a dynamic evaluation model to provide support for the intelligent management of the oil well.

[0035] Execute step A300, calculate the dynamic liquid level of the target oil well through the dynamic evaluation model, and obtain the initial dynamic liquid depth data; first load the trained and calibrated dynamic evaluation model, and initialize it according to the real-time data and parameters of the oil well, fit the liquid level change trend, pressure distribution and load dynamics in the dynamic evaluation model according to the sensor data, ensure that the calculation results of the dynamic evaluation model reflect the actual situation of the current oil well, and then use the algorithm in the dynamic evaluation model to calculate the dynamic liquid level, which means that the dynamic evaluation model is based on information such as pressure, pump load and fluid density. The dynamic depth change of the liquid in the wellbore is calculated by combining the pressure gradient in the sensor data and the working state of the pump to output the initial dynamic liquid depth data. The initial dynamic liquid depth data is used to represent the liquid level in the wellbore of the current oil well, which is a key parameter in oil well management and provides accurate liquid level monitoring data for the production and management of the oil well.

[0036] Execute step A400, perform dynamic analysis based on the multidimensional sensing data set, generate a correction factor, dynamically feedback the initial dynamic liquid depth data based on the correction factor, and draw a dynamic liquid level change trend graph; in a possible implementation, step A400 further includes step A410, dynamically calculate the multidimensional sensing data set according to the multiple characteristic parameters to obtain a dynamic liquid level change data set, and the dynamic liquid level change trend includes dynamic liquid level change rate data and dynamic liquid level change amplitude data; execute step A420, based on the dynamic liquid level change amplitude data, iteratively update the dynamic liquid level change rate data in combination with the work diagram area parameter, and draw a dynamic liquid level change trend graph. Prepare second dynamometer diagram data; execute step A430, perform waveform similarity analysis on the first dynamometer diagram data and the second dynamometer diagram data, and generate a waveform similarity coefficient; execute step A440, determine whether the waveform similarity coefficient is greater than or equal to an expected similarity threshold; execute step A450, if the waveform similarity coefficient is less than the expected similarity threshold, generate a tracking instruction, dynamically track the dynamic liquid level change data set through the tracking instruction, and determine abnormal fluctuation data; execute step A460, trace the source according to the abnormal fluctuation data, determine the source of the abnormal fluctuation, perform continuous monitoring according to the source of the abnormal fluctuation, and update the correction factor based on the monitoring result.

[0037] Based on the multiple characteristic parameters extracted previously, namely the workload characteristics, efficiency characteristics, etc., dynamic calculation of the multidimensional sensing data set refers to respectively calculating the rate of liquid level rise or fall and the amplitude of liquid level fluctuation to obtain dynamic liquid level change rate data and dynamic liquid level change amplitude data. The dynamic liquid level change rate data is used to characterize the fluctuation speed of the liquid level in different time periods, and the dynamic liquid level change amplitude data is used to characterize the maximum rise and fall height of the liquid level in one cycle. The dynamic liquid level change rate data and the dynamic liquid level change amplitude data are then integrated to generate a dynamic liquid level change data set, which is used to reflect the real-time dynamic changes of the liquid level in the oil well.

[0038] Furthermore, based on the dynamic liquid level change amplitude data, combined with the previously calculated power diagram area parameters, the dynamic liquid level change rate data is iteratively updated, which means that by analyzing the relationship between the liquid level change amplitude and the power diagram area, the liquid level change rate is adjusted to ensure the accuracy of the liquid level change rate. After completing the iterative update of the dynamic liquid level change rate, the second power diagram data is drawn based on the corrected rate data. The second power diagram data is used to reflect the impact of the liquid level change on the pump load and displacement, and the waveform similarity analysis of the first power diagram data and the second power diagram data is performed through cosine similarity or dynamic time warping, and the waveform similarity coefficient between the two is calculated. The waveform similarity coefficient is used to measure the matching degree of the two sets of data. Then the calculated waveform similarity coefficient is compared with the expected similarity threshold to determine whether the similarity meets the requirements. If the waveform similarity coefficient is greater than or equal to the expected similarity threshold, it means that the first dynamometer data and the second dynamometer data are sufficiently similar and no further adjustment is required. If the waveform similarity coefficient is less than the expected similarity threshold, a tracking instruction is generated to dynamically track the dynamic liquid level change data set to monitor the changes in the liquid level in real time. At the same time, by tracking the fluctuations in the liquid level changes, possible abnormal fluctuation data can be identified, including sudden and drastic rise or fall in the liquid level, or abnormal matching with load changes.

[0039] Furthermore, the detected abnormal fluctuation data is combined with multi-dimensional sensor data, power diagram data, etc. to trace the source, analyze the possible causes of abnormal oil well operating conditions, equipment failures, etc. to determine the source of abnormal fluctuations, and after determining the source of abnormal fluctuations, the area is continuously monitored to monitor the liquid level and pump working status in real time. Based on the monitoring results, the correction factor is updated to ensure that the subsequent calculation of liquid level changes and power diagram data is more accurate.

[0040] In one possible implementation, Figure 2 Said step A400 further includes step A470, taking the abnormal fluctuation source as an index, traversing the initial dynamic liquid depth data and the dynamic liquid level change data set for comparison, and extracting the abnormal change data of the dynamic liquid depth; executing step A480, correcting the abnormal change data of the dynamic liquid depth according to the correction factor, and generating corrected change data of the dynamic liquid depth; taking the detected abnormal fluctuation source as an index, traversing the initial dynamic liquid depth data and the dynamic liquid level change data set, the index points to the time period or position of the abnormal change of the liquid level, which means that by traversing the initial dynamic liquid depth data and the dynamic liquid level change data set, the abnormal change data of the dynamic liquid depth is identified and extracted, and the abnormal change data of the dynamic liquid depth is used to characterize the depth change of the liquid level during the abnormal fluctuation period. Then, according to the correction factor generated as above, the extracted abnormal change data of the dynamic liquid depth is matched, and the corresponding matched data is corrected or replaced according to the correction factor, wherein the correction factor is used to correct the abnormal fluctuation of the liquid level depth to ensure that the corrected data is more in line with the actual working conditions.

[0041] Execute step A490. When the waveform similarity coefficient is less than the expected similarity threshold, the first dynamometer diagram data is corrected according to the second dynamometer diagram data to generate third dynamometer diagram data. In one possible implementation, step A490 further includes: first, when the waveform similarity coefficient is less than the expected similarity threshold, the second dynamometer diagram data is weighted according to the waveform similarity coefficient to generate a plurality of weight coefficients. Secondly, the second dynamometer diagram data is processed in descending order according to the plurality of weight coefficients to generate a second dynamometer diagram sequence. Finally, the first dynamometer diagram data is traversed based on the second dynamometer diagram sequence to perform deviation correction to generate the third dynamometer diagram data.

[0042] When it is detected that the waveform similarity coefficient is lower than the preset expected similarity threshold, the second dynamometer data will be weighted according to the waveform similarity coefficient obtained above, which means that a weight coefficient is assigned to each data point in the second dynamometer data according to the similarity degree of the waveform similarity coefficient. The weight assignment method can be realized by cosine similarity or weighted average algorithm. The data point with higher similarity has a larger weight assigned, and vice versa, the weight is smaller. Each data point corresponds to a weight coefficient, and the system generates multiple weight coefficients, which are used to characterize the similarity and importance of different data points, and then the second dynamometer data is sorted according to the assigned weight coefficients. Data points with larger weights are arranged in front, and data points with smaller weights are arranged in the back, generating a second dynamometer sequence arranged according to the weight size. The second dynamometer sequence is used to ensure that more similar data participates in the subsequent deviation correction process first, reducing the correction error.

[0043] Furthermore, based on the generated second dynamometer diagram sequence, the first dynamometer diagram data is traversed point by point, the two sets of data are compared and analyzed, and the deviation between each data point is calculated. The deviation is calculated based on the difference between displacement and load. Combined with the weight coefficient, the system quantifies the deviation of each data point, and corrects the data in the first dynamometer diagram according to the calculated deviation value to generate corrected third dynamometer diagram data, which better reflects the actual working state of the pump and ensures the accuracy of liquid level depth monitoring and analysis.

[0044] Execute step A4100, construct a dynamic fluid change coordinate system in combination with the data acquisition time series, fill the dynamic fluid depth correction change data into the dynamic fluid change coordinate system based on the third dynamometer data, and construct the dynamic fluid level change trend diagram.

[0045] According to the data acquisition time series, combined with the relationship between the liquid level depth change, load and displacement, a multi-dimensional dynamic liquid change coordinate system is constructed. The dynamic liquid change coordinate system is used to represent the changing trend of the liquid level depth over time. The generated dynamic liquid depth correction change data is then filled in according to the time and position coordinates in the dynamic liquid change coordinate system to ensure that the changing trend of the liquid level depth is accurately expressed in the coordinate system. Based on the third indicator diagram data and the dynamic liquid depth correction change data, a dynamic liquid level change trend graph is drawn. The dynamic liquid level change trend graph is used to characterize the fluctuation of the liquid level in different time periods and the trend of the liquid level change, including the rate and amplitude of rise and fall, and can accurately reflect the real-time trend of the liquid level change in the oil well.

[0046] Next, step A500 is executed to monitor the dynamic liquid level of the target oil well in real time according to the dynamic liquid level variation trend diagram combined with the third indicator diagram data.

[0047] Start the real-time monitoring function, and load the dynamic liquid level change trend graph and the third dynamometer diagram data. The dynamic liquid level change trend graph is used to reflect the historical changes in the liquid level depth, and the third dynamometer diagram data is used to reflect the actual working state of the pump, so that the dynamic liquid level change trend graph can be used to monitor the depth changes of the liquid level in real time. The dynamic liquid level change trend graph is used to display the fluctuation of the liquid level at different time points, including the rising rate and the falling rate. If the liquid level depth changes abnormally, the dynamic liquid level change trend graph can be used to capture the rapid rise or fall of the liquid level depth, and make a real-time warning. At the same time, the third dynamometer diagram data is used to analyze the real-time working state of the pump, and the currently collected pump working data is compared with the third dynamometer diagram data to ensure that the working state of the pump is consistent with expectations. By comparing the third dynamometer diagram data with the actual working state of the current pump, abnormalities in the pump operation can be identified, such as reduced pump efficiency, abnormal load fluctuations, and other problems.

[0048] Finally, through the combination of the dynamic liquid level change trend chart and the third indicator diagram data, the operating status of the oil well is fed back in real time. If an abnormal change in the liquid level depth or an abnormal pump working state is detected, an early warning message is generated. The early warning message may include rapid rise or fall of the liquid level, abnormal operation of the pump, etc., prompting the operator to adjust the production parameters of the oil well in time to ensure the safe and efficient management of the oil well operation.

[0049] The embodiment of the present application solves the technical problem that the measurement accuracy of the dynamic liquid level of traditional oil wells is low and cannot be continuously monitored in real time, resulting in low accuracy of oil well management. It achieves the technical effect of combining the indicator diagram data to continuously detect the dynamic liquid level of the oil well in real time, thereby improving the accuracy of oil well monitoring.

[0050] The above specific implementation manner does not constitute a limitation to the protection scope of the present application. It should be understood by those skilled in the art that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application should be included in the protection scope of the present application. In some cases, the actions or steps recorded in the present application can be performed in an order different from that in the embodiment and can still achieve the desired results. In addition, the process depicted in the accompanying drawings does not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

Claims

1. A real-time monitoring method for the dynamic liquid level of an oil well in combination with the dynamometer diagram data, characterized in that: The method comprises: Perform multi-dimensional data collection on the target oil well through the oil well monitoring network to obtain a multi-dimensional sensor data set, and draw first indicator diagram data of the target oil well based on the multi-dimensional sensor data set; Traversing the first dynamometer diagram data to perform feature extraction, generate a plurality of feature parameters, and construct a dynamic evaluation model based on the plurality of feature parameters; The dynamic liquid level of the target oil well is calculated by the dynamic evaluation model to obtain initial dynamic liquid depth data; Performing dynamic analysis on the multi-dimensional sensor data set to generate a correction factor, dynamically feeding back the initial dynamic liquid depth data according to the correction factor, and drawing a dynamic liquid level change trend diagram; According to the dynamic liquid level change trend diagram and the third indicator diagram data, the dynamic liquid level of the target oil well is monitored in real time; Dynamically calculating the multidimensional sensing data set according to the multiple characteristic parameters to obtain a dynamic liquid level change data set, wherein the dynamic liquid level change trend includes dynamic liquid level change rate data and dynamic liquid level change amplitude data; Based on the dynamic liquid level change amplitude data, the dynamic liquid level change rate data is iteratively updated in combination with the work diagram area parameter to draw the second work diagram data; Performing waveform similarity analysis on the first dynamometer data and the second dynamometer data to generate a waveform similarity coefficient; determining whether the waveform similarity coefficient is greater than or equal to an expected similarity threshold, and if the waveform similarity coefficient is less than the expected similarity threshold, generating a tracking instruction, and dynamically tracking the dynamic liquid level change data set through the tracking instruction to determine abnormal fluctuation data; Tracing the abnormal fluctuation data to determine the source of the abnormal fluctuation, continuously monitoring the source of the abnormal fluctuation, and updating the correction factor based on the monitoring result; The step of dynamically feeding back the initial dynamic liquid depth data according to the correction factor and drawing a dynamic liquid level change trend diagram comprises: Taking the abnormal fluctuation source as an index, traversing the initial dynamic liquid depth data and comparing with the dynamic liquid level change data set, and extracting the abnormal dynamic liquid depth change data; Correcting the abnormal change data of the dynamic fluid depth according to the correction factor to generate corrected change data of the dynamic fluid depth; When the waveform similarity coefficient is less than the expected similarity threshold, the first dynamometer diagram data is corrected according to the second dynamometer diagram data to generate third dynamometer diagram data; In combination with the data acquisition time series, a dynamic fluid change coordinate system is constructed, and based on the third dynamometer diagram data, the dynamic fluid depth correction change data is filled into the dynamic fluid change coordinate system to construct the dynamic fluid level change trend diagram; When the waveform similarity coefficient is less than the expected similarity threshold, the first dynamometer diagram data is corrected according to the second dynamometer diagram data to generate third dynamometer diagram data, including: When the waveform similarity coefficient is less than the expected similarity threshold, weighting the second dynamometer diagram data is performed according to the waveform similarity coefficient to generate a plurality of weight coefficients; Processing the second dynamometer diagram data in descending order according to the plurality of weight coefficients to generate a second dynamometer diagram sequence; The first dynamometer diagram data is traversed based on the second dynamometer diagram sequence to perform deviation correction to generate the third dynamometer diagram data.

2. The method for real-time monitoring of oil well dynamic liquid level in combination with dynamometer diagram data according to claim 1, characterized in that: The construction method of the oil well monitoring network includes: Traverse the target oil wells to conduct dynamic liquid level analysis and determine multiple deployment points; Selecting a plurality of sensor devices to be deployed based on the plurality of deployment points, and deploying the plurality of sensor devices to be deployed in association with the plurality of deployment points to generate a plurality of device association chains, wherein each sensor device to be deployed has an association relationship with at least one other sensor device to be deployed that is different from itself; Perform multi-dimensional analysis in combination with the target oil wells according to the plurality of equipment association chains, and define a plurality of oil well sub-networks according to the analysis results; The multiple oil well sub-networks are communicatively connected to a central monitoring platform to construct the oil well monitoring network.

3. The method for real-time monitoring of oil well dynamic liquid level in combination with dynamometer diagram data as claimed in claim 2, characterized in that: Performing multi-dimensional acquisition of a target oil well through an oil well monitoring network to obtain a multi-dimensional sensor data set, and drawing first indicator diagram data of the target oil well based on the multi-dimensional sensor data set, the method comprising: Performing multi-dimensional analysis on the target oil well through the multiple oil well sub-networks to determine multiple multi-dimensional sensor data classes; Mapping the plurality of oil well sub-networks according to the plurality of multi-dimensional sensor data classes to obtain a plurality of oil well data networks; Based on the multiple oil well data networks, regression analysis is performed on the multidimensional sensor data set to generate an oil well operation data set, and real-time analysis is performed based on the oil well operation data set to draw the first indicator diagram data of the target oil well.

4. The method for real-time monitoring of oil well dynamic liquid level in combination with dynamometer diagram data according to claim 1, characterized in that: Traversing the first dynamometer diagram data to extract features and generate multiple feature parameters, the method includes: extracting a data acquisition time series based on the multidimensional sensor data set; Traversing the first dynamometer diagram data according to the data acquisition time sequence to perform load analysis and determine working load characteristics; Traversing the first dynamometer diagram data according to the data acquisition time series to perform efficiency analysis and determine work efficiency characteristics; The workload characteristics and the work efficiency characteristics are standardized to generate the multiple characteristic parameters.

5. The method for real-time monitoring of oil well dynamic liquid level in combination with dynamometer diagram data as claimed in claim 4, characterized in that: The dynamic evaluation model is constructed based on the multiple characteristic parameters, and the method includes: Calculating based on the workload characteristics in combination with the workload characteristics to obtain a load difference parameter; Calculating based on the working efficiency characteristics and the first indicator diagram data to obtain a dynamometer area parameter; Retrieving multiple static parameters of the target oil well, and determining a training data set according to the load difference parameter, the work diagram area parameter and the multiple static parameters; A plurality of operating parameters of the target oil well are introduced, and training and calibration are performed based on the plurality of operating parameters in combination with the training data set according to the data acquisition time series to obtain the dynamic evaluation model.

Images