A digital pumping unit monitoring system

Abstract

The present invention discloses a digital pumping unit monitoring system, which relates to the technical field of pumping units. A suspension point initialization unit processes the initial position of the suspension point; a suspension point position monitoring unit monitors the suspension point position to obtain a monitoring result; a dynamometer diagram unit constructs a dynamometer diagram of the suspension point, and based on the monitoring result, performs data cleaning and correction on the constructed dynamometer diagram; a feature extraction unit extracts features from the corrected dynamometer diagram and performs data annotation to obtain an annotation data set; a data processing unit processes vibration frequency data under different fault types to obtain a processing result; and finally, a fault monitoring unit constructs a fault monitoring model, trains the fault monitoring model using the annotation data set and the processing result, and performs real-time monitoring of the pumping unit based on the trained fault monitoring model. The present invention improves the efficiency and accuracy of existing pumping unit monitoring systems.

Description

Technical Field

[0001] The present invention relates to the technical field of oil pumping units, and in particular to a digital oil pumping unit monitoring system. Background Art

[0002] The digital oil pumping system mainly consists of three parts: the oil pump, the oil pumping unit and the oil pumping rod. The main hardware structure of the oil pumping unit includes the donkey head, the walking beam, the connecting rod, the crank mechanism, the reducer, the power equipment and the auxiliary equipment. The oil pumping rod is the rod device of the oil well, which is connected to the polished rod at the top and the oil pump at the bottom, and serves to connect the ground and the underground. The oil pump consists of four parts: the pump barrel, the plunger, the suction valve and the floating valve. The dynamometer (load sensor, displacement sensor), the control box, etc. are responsible for collecting production parameters.

[0003] At present, the monitoring of digital oil pumping system mostly requires staff to rotate around the oil production site for 24 hours, and then use the load displacement sensor to collect the load data and displacement data of the suspension point to form an indicator diagram at the back end (the horizontal axis is the suspension point displacement, and the vertical axis is the suspension point load). Professionals analyze the indicator diagram to determine the type of failure of the oil pumping unit. However, the above method has the following problems: the oil pumping unit converts the rotational motion of the motor into the up and down reciprocating motion of the light rod (suspension point) through the crank-connecting rod mechanism. That is to say, under ideal conditions, when the crank rotates one circle, the output distance of the motor is equal to the distance the suspension point completes an up and down stroke, that is, when the suspension point completes an up and down stroke, it returns to the starting point. However, since the oil pumping unit is continuously operated in a harsh environment full of dust, rain, and saline-alkali air all year round, During operation, the bearings suffer from uneven force and eccentric wear, which leads to problems with the bracket bearings, beam bearings and crank pin bearings, as well as clearance in the crank, connecting rod, rocker and other hinged parts. Long-term operation in harsh environments leads to increased wear and tear, and the "idle stroke" causes displacement loss during the stroke, which ultimately causes the suspension point to be unable to return to the starting point, that is, there is an error between the displacement of a single up and down stroke and the displacement of a single up and down stroke under ideal conditions, and the error increases with time. Finally, the dynamometer diagram formed by collecting displacement and load data cannot accurately reflect the failure of the oil pump, reducing the monitoring accuracy of the oil pump monitoring system. Moreover, the current method of manual overtime and professional personnel to analyze the dynamometer diagram at the back end has a low monitoring efficiency, and cannot detect and handle faults in a timely manner.

[0004] In summary, the monitoring efficiency and accuracy of digital pumping unit systems are currently difficult to guarantee. Summary of the Invention

[0005] In view of this, the present application provides a digital pumping unit monitoring system to improve the monitoring efficiency and accuracy of the digital pumping unit system.

[0006] The first aspect of the present application provides a digital pumping unit monitoring system, comprising:

[0007] A suspension point initialization unit: obtains the initial position of the suspension point and the change value of the initial position at adjacent moments, and determines whether the change value exceeds a first threshold. If so, obtains the real-time position of the suspension point at the adjacent moments as the initial position; if not, takes no action;

[0008] The suspension point position monitoring unit uses the initial position of the suspension point as a reference, obtains the time of one crank movement to divide the time interval, obtains the output distance of the motor and the displacement distance of the suspension point in a single complete time interval, and plots the output distance of the motor and the displacement distance of the suspension point in the same coordinate system. The data in the coordinate system is analyzed and monitored to obtain monitoring results.

[0009] Dynamometer unit: constructs a dynamometer diagram for the suspension point within a single complete time interval, and determines whether to perform data cleaning on the dynamometer diagram of the suspension point based on the monitoring results. If so, cleans the dynamometer diagram of the suspension point; if not, corrects the dynamometer diagram of the suspension point.

[0010] A feature extraction unit is configured to obtain all corrected dynamometer diagrams and place them into a first preset data set. A multi-point random sampling method is used to obtain feature data of different dynamometer diagrams in the first preset data set. Based on the feature data of all dynamometer diagrams in the first preset data set, dynamometer diagrams of different fault types are labeled to obtain a labeled data set.

[0011] Data processing unit: collects vibration frequency data of the digital pumping unit in multiple complete time intervals at multiple time scales, and processes the data using a preset method to obtain multiple processing results;

[0012] Fault monitoring unit: constructs a fault monitoring model, takes all labeled data sets as the first influencing factor and all processing results as the second influencing factor, trains the fault monitoring model together, and performs real-time monitoring of the digital pumping unit based on the trained fault monitoring model.

[0013] In a possible implementation of the first aspect, obtaining the time required for one crank movement to divide the time intervals includes:

[0014] Obtain the starting position of the crank, and record the moment when the crank starts to move as time t0, and the moment when the crank returns to the starting position as time t1;

[0015] The time interval between the time t0 and the time t1 is a time interval.

[0016] In a possible implementation of the first aspect, obtaining the output distance of the motor and the displacement distance of the suspension point within a single complete time interval includes:

[0017] In a single complete time interval, the displacement distance of the suspension point is obtained by the displacement sensor, and the output distance of the motor is obtained by obtaining the motor speed and the circumference of the motor shaft.

[0018] In a possible implementation of the first aspect, plotting the output distance of the motor and the displacement distance of the suspension point in the same coordinate system includes:

[0019] Taking time as the horizontal axis, and dividing the horizontal axis into multiple time intervals;

[0020] The vertical axis is dimensionless, and points A and B are determined on the vertical axis;

[0021] Get the output distance of the motor in all time intervals, take point A as the starting point, and plot the output distance of the motor in the corresponding time interval in equal proportion;

[0022] Obtain the displacement distance of the suspension point in all time intervals, take point B as the starting point, and draw the displacement distance of the suspension point in equal proportion within the corresponding time interval.

[0023] In a possible implementation of the first aspect, analyzing and monitoring the data in the coordinate system to obtain a monitoring result includes:

[0024] In the coordinate system, the output distance of the motor in any time interval is obtained and recorded as L1, and the displacement distance of the suspension point is recorded as L2;

[0025] If L2 is greater than L1, data cleaning is performed on the indicator diagram of the hanging point in the corresponding time interval;

[0026] If L2 is less than or equal to L1, calculate the difference between L1 and L2, and determine whether the difference is greater than a second threshold. If so, modify the indicator diagram of the hanging point in the corresponding time interval; if not, do nothing.

[0027] In a possible implementation of the first aspect, correcting the dynamometer diagram of the suspension point includes:

[0028] Obtain any indicator diagram of the suspension point as a first indicator diagram, and obtain a difference between an output distance of the motor and a displacement distance of the suspension point within a time interval corresponding to the first indicator diagram as a first difference;

[0029] Obtain a continuous longitudinal line feature between the upper right vertex and the lower right vertex in the first dynamometer diagram, record it as a first line feature, obtain a continuous transverse line feature extending from the upper right vertex to the longitudinal axis, record it as a second line feature, and obtain a continuous transverse line feature extending from the lower right vertex to the longitudinal axis, record it as a third line feature;

[0030] determining an upper right incremental vertex and a lower right incremental vertex in the first dynamometer diagram based on the first difference, the second line feature, and the third line feature;

[0031] Based on the second line feature, a line is connected between the upper right vertex and the upper right incremental vertex; based on the third line feature, a line is connected between the lower right vertex and the lower right incremental vertex; based on the first line feature, a line is connected between the upper right incremental vertex and the lower right incremental vertex;

[0032] The correction of the first indicator diagram is completed.

[0033] In a possible implementation of the first aspect, the preset method includes:

[0034] Based on the vibration frequency data collected in multiple complete time intervals at multiple time scales, multiple coordinate curves of the vibration frequency data changing with time are constructed;

[0035] Obtain any fault type and record it as a first fault type, and obtain multiple coordinate curves under the first fault type and store them in a second preset data set;

[0036] Segmenting each coordinate curve in the second preset data set by the time interval to obtain M1 time intervals, and collecting the slopes of N1 data points in each time interval to obtain a collection result;

[0037] Based on the acquisition results, calculating the average slopes of corresponding data points in corresponding time intervals of a plurality of coordinate curves in the second preset data set;

[0038] Establishing a database to store the average slopes of corresponding data points of corresponding time intervals of a plurality of coordinate curves in the second preset data set;

[0039] When there is a single coordinate curve recorded as the first coordinate curve, and the slopes of N2 data points in a single time interval of the first coordinate curve meet a preset condition, it is determined that the time interval corresponding to the first coordinate curve matches the time interval corresponding to the first fault type;

[0040] When there are M2 matching time intervals in the first coordinate curve and the ratio of M2 / M1 exceeds a fourth threshold, it is determined that the first coordinate curve matches the coordinate curve corresponding to the first fault type.

[0041] In a possible implementation of the first aspect, the preset condition includes:

[0042] The difference between the slopes of the N2 data points and the average slope of the corresponding data points in the database is less than a set value, and the ratio N2 / N1 exceeds a third threshold.

[0043] In a possible implementation of the first aspect, performing real-time monitoring of the digital oil pumping unit based on the trained fault monitoring model includes:

[0044] Obtaining a corrected indicator diagram of the digital oil pumping unit and recording it as a second indicator diagram, and recording a processing result of corresponding vibration frequency data within a preset time period as a first processing result;

[0045] The second dynamometer diagram and the first processing result are input into the trained fault monitoring model, and the fault type of the digital pumping unit is output.

[0046] In a possible implementation of the first aspect, constructing a dynamometer diagram about the suspension point within a single complete time interval includes:

[0047] In a single complete time interval, the displacement data of the suspension point is collected by a displacement sensor, and the load data of the suspension point is collected by a load sensor; and an indicator diagram about the suspension point is constructed with the displacement data of the suspension point as the horizontal coordinate and the load data of the suspension point as the vertical coordinate.

[0048] Compared with the existing technology, the present application provides a digital pumping unit monitoring system, including:

[0049] The suspension point initialization unit obtains the initial position of the suspension point and the change in the initial position at adjacent moments. If the change exceeds a first threshold, the real-time position of the suspension point at the adjacent moment is obtained as the new initial position, completing the initialization of the suspension point position. If the initial position of the suspension point is not initialized, a deviation will occur between the suspension point displacement distance measured by the displacement sensor and the actual suspension point displacement distance. This deviation will persist, directly affecting the generation of the indicator diagram and thus affecting the judgment of the operating condition of the pumping unit.

[0050] The suspension point position monitoring unit, because the oil pumping unit is working continuously in a harsh environment full of dust, rain, and saline-alkali all year round, the "idle stroke" will cause displacement loss during the stroke, resulting in the suspension point not being able to return to the starting point and the error will continue to increase with time. Finally, the indicator diagram formed by collecting displacement load data cannot accurately reflect the failure of the oil pumping unit. Therefore, the suspension point position monitoring unit uses a complete cycle of crank movement as a benchmark to obtain the output distance of the motor and the displacement distance of the suspension point in a single complete time interval. In theory, the output distance of the motor is equal to the displacement distance of the suspension point. However, due to the long-term operation of the oil pumping unit in a harsh environment, the wear is aggravated, resulting in the output distance of the motor being greater than the displacement distance of the suspension point. This module establishes a unified coordinate system to more intuitively monitor the output distance of the motor and the displacement distance of the suspension point to obtain the monitoring results;

[0051] The working principle of the dynamometer unit is to collect the displacement distance and load of the suspension point within a single complete stroke (time interval), construct a coordinate system of the suspension point load with respect to the suspension point displacement distance, obtain the dynamometer diagram of the oil pumping unit, and judge the working condition of the oil pumping unit by analyzing the dynamometer diagram. However, as mentioned earlier, due to the fact that the displacement distance of the suspension point may not be equal to the output distance of the motor during actual operation (firstly, the initial position of the suspension point may change, and secondly, the suspension point cannot return to the initial position in a single stroke), the dynamometer diagram generated based on the collected data is not accurate, which of course directly affects the judgment of the working condition of the oil pumping unit. Therefore, based on the monitoring results, this module first cleans the data of part of the dynamometer diagram (if the suspension point displacement distance is greater than the motor output distance, the data is judged to be abnormal), and then corrects the remaining dynamometer diagrams. Its correction logic is to supplement the part of the dynamometer diagram corresponding to the missing stroke.

[0052] The feature extraction unit obtains all corrected dynamometer diagrams, and then extracts feature data from different dynamometer diagrams using a multi-point random sampling method. Based on the extracted feature data, the dynamometer diagrams of different fault types are standardized to obtain a labeled data set.

[0053] The data processing unit, in order to further improve the accuracy of the monitoring system's judgment of the pumping unit's working condition, introduces other data that affects the pumping unit's working condition, namely vibration frequency data, based on the indicator diagram. Under different fault conditions, the vibration frequency data of the pumping unit also varies. This module analyzes and processes the collected vibration frequency data to obtain the corresponding processing results;

[0054] The fault monitoring unit builds a fault monitoring model, uses the labeled data set and the corresponding processing results as the training set, trains the fault monitoring model, and then performs real-time monitoring of the digital oil pumping unit based on the trained fault monitoring model. Both the monitoring efficiency and accuracy are greatly improved compared with the existing monitoring system.

[0055] Its beneficial effects are as follows: the present invention discloses a digital oil pumping unit monitoring system, which initializes the initial position of the suspension point by setting a suspension point initialization unit; monitors the suspension point position by a suspension point position monitoring unit, and its monitoring logic is to compare the motor output distance and the suspension point displacement distance to obtain a monitoring result; then the dynamometer diagram unit constructs the dynamometer diagram of the suspension point, and based on the monitoring result, performs data cleaning and correction on the constructed dynamometer diagram, and its correction logic is to supplement the part of the dynamometer diagram with missing stroke; utilizes the feature extraction unit to extract features from the corrected dynamometer diagram and performs data annotation to obtain an annotation data set; based on the data processing unit, processes the vibration frequency data under different fault types to obtain a processing result; finally, the fault monitoring unit constructs a fault monitoring model, uses the annotation data set and the processing result to jointly train the fault monitoring model, and performs real-time monitoring of the oil pumping unit based on the trained fault monitoring model. The present invention achieves the effect of data cleaning and correction of the dynamometer diagram by initializing the suspension point and monitoring the suspension point position, then performs feature extraction and data annotation on the corrected dynamometer diagram to obtain an annotated data set, and simultaneously analyzes and processes the vibration frequency data of the pumping unit under different fault types to obtain processing results. A neural network model learning method is adopted, and training is performed using the annotated data set of the dynamometer diagram and the processing results of the vibration frequency data, which greatly improves the monitoring efficiency and accuracy of the monitoring system. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are merely embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without any creative work.

[0057] Figure 1 This is a schematic diagram of the composition of a digital pumping unit monitoring system provided in an embodiment of the present application;

[0058] Figure 2 Schematic diagram of the output distance of the motor and the displacement distance of the suspension point in the same coordinate system in the embodiment of the present application;

[0059] Figure 3 It is a schematic diagram of correcting the indicator diagram in the embodiment of the present application. DETAILED DESCRIPTION

[0060] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0061] In this application, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus comprising the element.

[0062] In order to better understand this application, the following are corresponding explanations of the technical names involved in this application:

[0063] Dynamometer diagram: A closed curve showing the relationship between the load on the sucker rod string at the suspension point and the displacement of the suspension point. By analyzing the shape, area and other characteristics of the suspension point dynamometer diagram, we can understand the operating conditions of the pumping unit and determine whether it is working properly, such as whether there are any faults such as fixed valve stuck, valve wax accumulation, or wax accumulation on the oil pipe and oil rod.

[0064] Example

[0065] As can be seen from the above background, a digital pumping unit system primarily consists of three components: the pumping unit, the unit, and the sucker rod. The unit's primary hardware structure includes the pumping head, walking beam, connecting rod, crank mechanism, reducer, power unit, and auxiliary equipment. The sucker rod is the rod assembly in the oil well, connecting the polished rod at the top and the pump at the bottom, serving as the connection between the surface and the wellbore. The pump consists of the pump barrel, plunger, suction valve, and floating valve. The dynamometer (load sensor, displacement sensor) and control box are responsible for collecting production parameters. Efficient and accurate real-time monitoring of the digital pumping unit not only ensures efficient production and reduces risks, but also reduces operation and maintenance costs and improves management efficiency.

[0066] In the existing technology, the monitoring of digital oil pumping system mostly requires staff to rotate around the oil production site for 24 hours, and then use the load displacement sensor to collect the load data and displacement data of the suspension point to form an indicator diagram at the back end (the horizontal axis is the suspension point displacement, and the vertical axis is the suspension point load). Professionals analyze the indicator diagram to determine the type of oil pump failure. However, the above method has the following problems: Since the oil pump is constantly working in a harsh environment full of dust, rain, and saline-alkali air, the bearings are subject to uneven force and eccentric wear, which leads to problems with the bracket bearings, beam bearings and crank pin bearings, as well as the crank There is a fit clearance in the hinged parts such as the connecting rod and the walking beam. After long-term operation in a harsh environment, the wear is aggravated. The "idle stroke" will cause displacement loss during the stroke, and eventually the suspension point cannot return to the starting point, that is, the displacement of a single up and down stroke is different from the displacement of a single up and down stroke under the ideal state. There is an error, and the error will continue to increase with time. Finally, the dynamometer diagram formed by collecting displacement and load data cannot accurately reflect the failure of the oil pump, reducing the monitoring accuracy of the oil pump monitoring system. Moreover, the current method of manual overtime and professional personnel to analyze the dynamometer diagram at the back end has a low monitoring efficiency and cannot detect and handle faults in time.

[0067] Therefore, this application provides a digital pumping unit monitoring system, such as Figure 1 As shown, including:

[0068] A suspension point initialization unit: obtains the initial position of the suspension point and the change value of the initial position at adjacent moments, and determines whether the change value exceeds a first threshold. If so, obtains the real-time position of the suspension point at the adjacent moments as the initial position; if not, takes no action;

[0069] The suspension point position monitoring unit uses the initial position of the suspension point as a reference, obtains the time of one crank movement to divide the time interval, obtains the output distance of the motor and the displacement distance of the suspension point in a single complete time interval, and plots the output distance of the motor and the displacement distance of the suspension point in the same coordinate system. The data in the coordinate system is analyzed and monitored to obtain monitoring results.

[0070] Dynamometer unit: constructs a dynamometer diagram for the suspension point within a single complete time interval, and determines whether to perform data cleaning on the dynamometer diagram of the suspension point based on the monitoring results. If so, cleans the dynamometer diagram of the suspension point; if not, corrects the dynamometer diagram of the suspension point.

[0071] A feature extraction unit is configured to obtain all corrected dynamometer diagrams and place them into a first preset data set. A multi-point random sampling method is used to obtain feature data of different dynamometer diagrams in the first preset data set. Based on the feature data of all dynamometer diagrams in the first preset data set, dynamometer diagrams of different fault types are labeled to obtain a labeled data set.

[0072] Data processing unit: collects vibration frequency data of the digital pumping unit in multiple complete time intervals at multiple time scales, and processes the data using a preset method to obtain multiple processing results;

[0073] Fault monitoring unit: constructs a fault monitoring model, takes all labeled data sets as the first influencing factor and all processing results as the second influencing factor, trains the fault monitoring model together, and performs real-time monitoring of the digital pumping unit based on the trained fault monitoring model.

[0074] Among them, the suspension point initialization unit is used to initialize the initial position of the suspension point. During the actual operation of the oil pump, the initial point of the suspension point may also change. If the initial point of the suspension point is not corrected, there will be a deviation between the suspension point displacement distance measured by the displacement sensor and the actual displacement distance of the suspension point, which directly affects the generation of the indicator diagram. Therefore, by obtaining the initial position of the suspension point and the change value of the initial position at adjacent moments, the adjacent moments can be adjacent every second, and can also be adjusted according to actual conditions. If the change value of the initial position of the suspension point at adjacent moments exceeds the set first threshold, if the indicator diagram is still drawn based on the original initial position, then there will be a large error. Therefore, this embodiment obtains the real-time position of the suspension point and uses it as the new initial position; if the change value of the initial position of the suspension point at adjacent moments does not exceed the set first threshold, no action is taken.

[0075] Among them, the suspension point position monitoring unit is used to monitor the position of the suspension point. Taking the initial position of the suspension point as the benchmark, the time for the crank to move one circle is obtained to divide the time interval. Specifically, the starting position of the crank is obtained, and the moment when the crank starts to move is recorded as moment t0, and the moment when the crank returns to the starting position is recorded as t1. The time interval between moments t0 and t1 constitutes a complete time interval.

[0076] Among them, the crank moves one circle, that is, the suspension point completes a complete up and down stroke. In a single complete time interval, the output distance of the motor and the displacement distance of the suspension point are obtained respectively. The output distance of the motor is obtained by obtaining the motor speed and the circumference of the motor shaft, that is, in a single complete time interval, the rotation distance of the motor shaft is obtained as the output distance of the motor; the displacement distance of the suspension point is obtained by the displacement sensor.

[0077] Among them, the output distance of the motor and the displacement distance of the suspension point are plotted in the same coordinate system, and the data in the coordinate system are analyzed and monitored to obtain the monitoring results; please refer to Figure 2 , Figure 2 It is a schematic diagram of the output distance of the motor and the displacement distance of the suspension point in the embodiment of the present application in the same coordinate system, with the time interval as the horizontal axis and the vertical axis dimensionless, which is only used to distinguish the starting position A of the motor and the starting position B of the suspension point. Then, the output distance of the motor and the displacement distance of the suspension point are drawn in equal proportion in each time interval. For example, in the t0-t1 time interval, it can be seen intuitively that the output distance of the motor is equal to the displacement distance of the suspension point; in the t1-t2 time interval, the output distance of the motor is less than the displacement distance of the suspension point; in the t2-t3 and t3-t4 time intervals, the output distance of the motor is greater than the displacement distance of the suspension point. It should be noted that this coordinate system is only for more intuitive display of the difference between the output distance of the motor and the displacement distance of the suspension point in different time intervals, so it is not necessary to strictly follow the relevant standards of the coordinate system. In addition, Figure 2 It only lists the possible situations of motor output distance and suspension point displacement distance in a single complete time interval, and does not represent actual data.

[0078] The dynamometer unit is used to construct a dynamometer diagram about the suspension point within a single complete time interval, such as Figure 3 The closed graph formed by vertices D1, D2, D3 and D4 is shown in the figure. Based on the monitoring results, it is determined whether to perform data cleaning on the indicator diagram of the suspension point. When the displacement distance of the suspension point exceeds the output distance of the motor, it is determined that the displacement data collected by the displacement sensor is abnormal. Figure 2 In the time interval t1-t2 in the figure, it is necessary to clean the data of the indicator diagram corresponding to the time interval t1-t2; if the displacement distance of the suspension point does not exceed the output distance of the motor, the difference between the motor output distance and the suspension point displacement distance is calculated. If the difference is greater than the second threshold, the indicator diagram of the suspension point in the corresponding time interval needs to be corrected. Figure 2 If the difference between the time intervals t2-t3 and t3-t4 exceeds the second threshold, the indicator diagram in these two time intervals needs to be corrected; if the difference is not greater than the second threshold, no action is taken. Figure 2 The dynamometer diagram in the time interval t0-t1 does not need to be corrected.

[0079] Among them, when the indicator diagram of the hanging point is corrected, such as Figure 3The dynamometer diagram of the middle suspension point is corrected. First, the upper right vertex and the lower right vertex in the dynamometer diagram are obtained. It should be noted that in the dynamometer diagram, the coordinate feature of the upper right vertex is the point with the largest displacement and the largest load. Its physical meaning is that after the upstroke is completed, the suspension point bears the maximum load; the coordinate feature of the lower right vertex is the only point with the largest displacement and the smallest load. Its physical meaning is the starting point of the downstroke, and the suspension point load drops suddenly; the determination of the four vertices (upper left, lower left, upper right, and lower right) in the dynamometer diagram can be based on the recognition of characteristic points with physical meaning, or it can be identified by data processing and algorithm, which will not be elaborated in this embodiment. Obtain a first difference between the motor output distance and the suspension point displacement distance within a time interval corresponding to the dynamometer diagram, obtain a continuous line feature between points D3 and D4 as a first line feature, a continuous line feature between points D3 and D2 as a second line feature, and a continuous line feature between points D4 and D1 as a third line feature. Determine the upper right incremental vertex D3' based on the first difference and the second line feature, and determine the lower right incremental vertex D4' based on the first difference and the third line feature. Then, based on the second line feature, connect D3 and D3', based on the third line feature, connect D4 and D4', and based on the first line feature, connect D3' and D4', thereby obtaining a closed figure consisting of vertices D1, D2, D3', and D4' as the corrected dynamometer diagram.

[0080] Among them, the feature extraction unit obtains the corrected dynamometer diagram and adopts a multi-sampling point random sampling method to obtain the feature data of different dynamometer diagrams. Based on the feature data of the dynamometer diagram, the dynamometer diagrams of different fault types are labeled to obtain a labeled data set.

[0081] Among them, the data processing unit is used to collect vibration frequency data of the digital oil pumping unit in multiple complete time intervals at multiple time scales, and process it using a preset method, specifically: construct multiple coordinate curves of the vibration frequency data changing with time; then obtain any fault type and record it as the first fault type, and obtain multiple coordinate curves under the first fault type (such as 5, namely C1, C2, C3, C4 and C5), and store them in a second preset data set, and segment each coordinate curve in the second preset data set by time interval to obtain M1 time intervals, and collect the slopes of N1 data points in each time interval, and calculate the average slopes of corresponding data points in the corresponding time intervals in the C1, C2, C3, C4 and C5 coordinate curves. It should be noted that the data points collected in each coordinate curve are one-to-one corresponding, and then the average slope of each data point is stored by constructing a database. When a single coordinate curve exists, designated as the first coordinate curve, and the slopes of N2 data points within a single time interval of the first coordinate curve meet a preset condition (i.e., the difference between the slopes of these N2 data points and the average slope of the data points corresponding to the first fault type in the database is less than a set value, and the ratio N2 / N1 exceeds a third threshold (e.g., if N1 is 10, N2 is 8, and the third threshold is 0.7), the single time interval is determined to match the time interval corresponding to the first fault type. The number of matching time intervals in the first coordinate curve is then calculated, e.g., M2. If the ratio M2 / M1 exceeds a fourth threshold, the fault type corresponding to the first coordinate curve is determined to be consistent with the first fault type. The processing logic of this embodiment involves performing multiple differential operations on the vibration frequency data and then employing a quadratic matching approach to extract features from the vibration frequency data for different fault types. It should be noted that the above data is for illustrative purposes only and can be adjusted based on actual conditions. This embodiment does not impose specific limitations.

[0082] Among them, a fault monitoring model is constructed, and a neural network learning method is adopted. The labeled data set corresponding to the indicator diagram and the processing results corresponding to the vibration frequency data are used as training sets to train the fault monitoring model. Its training architecture can be optimized by matrix operation or cross structure, that is, the fault monitoring model outputs the fault type of the oil pumping unit through the labeled data set, and at the same time uses all the processing results to output the fault type of the oil pumping unit. The matrix E is used to store the fault type output by the labeled data set, and the matrix F is used to store the fault type output by the processing result. When the fault type output in the matrix E is consistent with the fault type output in the matrix F, the corresponding fault type is output through the fault monitoring model. The final output is the type; when the fault type output in matrix E is inconsistent with the fault type output in matrix F, the corresponding coefficients are set by analyzing the performance of matrices E and F for different fault types. For example, for fault type G, the prediction coefficient of matrix E is 0.9, and the prediction coefficient of matrix F is 0.8. For fault type H, the prediction coefficient of matrix E is 0.95, and the prediction coefficient of matrix F is 0.85. When the fault monitoring model outputs the fault type of the pumping unit as fault type G based on the labeled data set and stores it in matrix E, and outputs the fault type of the pumping unit as fault type H based on all processing results and stores it in matrix F, then the fault monitoring model ultimately outputs fault type G. Simply put, the fault monitoring model is trained using two influencing factors. When the output results of the two factors are consistent, the corresponding fault type is output; when the output results of the two factors are inconsistent, the monitoring result with better performance for the specific fault type is output. It should be noted that this embodiment only provides a method for handling inconsistent output results. When the output results of the two factors are inconsistent, manual verification can also be used to determine the final fault type. This embodiment does not specifically limit this.

[0083] Furthermore, in addition to adding vibration frequency data to the fault monitoring model, other data that may affect the operating conditions of the oil pump, such as noise data and temperature data, can also be added. This is because such data have the same or similar characteristic data under the same operating conditions. By analyzing and processing the characteristic data of such data under different operating conditions and using a neural network model for learning, the effect of inferring the operating conditions of the oil pump can be achieved.

[0084] In some embodiments, obtaining the time for one crank movement to divide the time intervals includes:

[0085] Obtain the starting position of the crank, and record the moment when the crank starts to move as time t0, and the moment when the crank returns to the starting position as time t1;

[0086] The time interval between the time t0 and the time t1 is a time interval.

[0087] In some embodiments, obtaining the output distance of the motor and the displacement distance of the suspension point within a single complete time interval includes:

[0088] In a single complete time interval, the displacement distance of the suspension point is obtained by the displacement sensor, and the output distance of the motor is obtained by obtaining the motor speed and the circumference of the motor shaft.

[0089] In some embodiments, plotting the output distance of the motor and the displacement distance of the suspension point in the same coordinate system includes:

[0090] Taking time as the horizontal axis, and dividing the horizontal axis into multiple time intervals;

[0091] The vertical axis is dimensionless, and points A and B are determined on the vertical axis;

[0092] Get the output distance of the motor in all time intervals, take point A as the starting point, and plot the output distance of the motor in the corresponding time interval in equal proportion;

[0093] Obtain the displacement distance of the suspension point in all time intervals, take point B as the starting point, and draw the displacement distance of the suspension point in equal proportion within the corresponding time interval.

[0094] In some embodiments, analyzing and monitoring the data in the coordinate system to obtain monitoring results includes:

[0095] In the coordinate system, the output distance of the motor in any time interval is obtained and recorded as L1, and the displacement distance of the suspension point is recorded as L2;

[0096] If L2 is greater than L1, data cleaning is performed on the indicator diagram of the hanging point in the corresponding time interval;

[0097] If L2 is less than or equal to L1, calculate the difference between L1 and L2, and determine whether the difference is greater than a second threshold. If so, modify the indicator diagram of the hanging point in the corresponding time interval; if not, do nothing.

[0098] In some embodiments, based on the monitoring result, modifying the dynamometer diagram of the suspension point includes:

[0099] Obtain any indicator diagram of the suspension point as a first indicator diagram, and obtain a difference between an output distance of the motor and a displacement distance of the suspension point within a time interval corresponding to the first indicator diagram as a first difference;

[0100] Obtain a continuous longitudinal line feature between the upper right vertex and the lower right vertex in the first dynamometer diagram, record it as a first line feature, obtain a continuous transverse line feature extending from the upper right vertex to the longitudinal axis, record it as a second line feature, and obtain a continuous transverse line feature extending from the lower right vertex to the longitudinal axis, record it as a third line feature;

[0101] determining an upper right incremental vertex and a lower right incremental vertex in the first dynamometer diagram based on the first difference, the second line feature, and the third line feature;

[0102] Based on the second line feature, a line is connected between the upper right vertex and the upper right incremental vertex; based on the third line feature, a line is connected between the lower right vertex and the lower right incremental vertex; based on the first line feature, a line is connected between the upper right incremental vertex and the lower right incremental vertex;

[0103] The correction of the first indicator diagram is completed.

[0104] In some embodiments, the preset method includes:

[0105] Based on the vibration frequency data collected in multiple complete time intervals at multiple time scales, multiple coordinate curves of the vibration frequency data changing with time are constructed;

[0106] Obtain any fault type and record it as a first fault type, and obtain multiple coordinate curves under the first fault type and store them in a second preset data set;

[0107] Segmenting each coordinate curve in the second preset data set by the time interval to obtain M1 time intervals, and collecting the slopes of N1 data points in each time interval to obtain a collection result;

[0108] Based on the acquisition results, calculating the average slopes of corresponding data points in corresponding time intervals of a plurality of coordinate curves in the second preset data set;

[0109] Establishing a database to store the average slopes of corresponding data points of corresponding time intervals of a plurality of coordinate curves in the second preset data set;

[0110] When there is a single coordinate curve recorded as the first coordinate curve, and the slopes of N2 data points in a single time interval of the first coordinate curve meet a preset condition, it is determined that the time interval corresponding to the first coordinate curve matches the time interval corresponding to the first fault type;

[0111] When there are M2 matching time intervals in the first coordinate curve and the ratio of M2 / M1 exceeds a fourth threshold, it is determined that the first coordinate curve matches the coordinate curve corresponding to the first fault type.

[0112] In some embodiments, the preset conditions include:

[0113] The difference between the slopes of the N2 data points and the average slope of the corresponding data points in the database is less than a set value, and the ratio N2 / N1 exceeds a third threshold.

[0114] In some embodiments, real-time monitoring of the digital oil pumping unit based on the trained fault monitoring model includes:

[0115] Obtaining a corrected indicator diagram of the digital oil pumping unit and recording it as a second indicator diagram, and recording a processing result of corresponding vibration frequency data within a preset time period as a first processing result;

[0116] The second dynamometer diagram and the first processing result are input into the trained fault monitoring model, and the fault type of the digital pumping unit is output.

[0117] In some embodiments, constructing a dynamometer diagram about the suspension point within a single complete time interval includes:

[0118] In a single complete time interval, the displacement data of the suspension point is collected by a displacement sensor, and the load data of the suspension point is collected by a load sensor; and an indicator diagram about the suspension point is constructed with the displacement data of the suspension point as the horizontal coordinate and the load data of the suspension point as the vertical coordinate.

[0119] Professionals may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computing software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the above description has generally described the components and steps of each example according to their functions. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians may use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0120] Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0121] Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if such changes and modifications fall within the scope of the claims and their equivalents, the present invention is intended to include such changes and modifications.

Claims

1. A digital pumping unit monitoring system, characterized in that: include: A suspension point initialization unit: obtains the initial position of the suspension point and the change value of the initial position at adjacent moments, and determines whether the change value exceeds a first threshold. If so, obtains the real-time position of the suspension point at the adjacent moments as the initial position; if not, takes no action; The suspension point position monitoring unit uses the initial position of the suspension point as a reference, obtains the time of one crank movement to divide the time interval, obtains the output distance of the motor and the displacement distance of the suspension point in a single complete time interval, and plots the output distance of the motor and the displacement distance of the suspension point in the same coordinate system. The data in the coordinate system is analyzed and monitored to obtain monitoring results. Drawing the output distance of the motor and the displacement distance of the suspension point in the same coordinate system includes: taking time as the horizontal axis and dividing the horizontal axis into multiple time intervals; the vertical axis is dimensionless, and point A and point B are determined on the vertical axis; obtaining the output distance of the motor in all time intervals, taking point A as the starting point, and drawing the output distance of the motor in proportion within the corresponding time interval; obtaining the displacement distance of the suspension point in all time intervals, taking point B as the starting point, and drawing the displacement distance of the suspension point in proportion within the corresponding time interval; Analyzing and monitoring the data in the coordinate system to obtain monitoring results includes: obtaining, in the coordinate system, an output distance of the motor in any time interval, recorded as L1, and a displacement distance of the suspension point, recorded as L2; if L2 is greater than L1, performing data cleaning on the dynamometer diagram of the suspension point in the corresponding time interval; if L2 is less than or equal to L1, calculating a difference between L1 and L2, and determining whether the difference is greater than a second threshold; if so, correcting the dynamometer diagram of the suspension point in the corresponding time interval; if not, performing no action; Dynamometer unit: constructs a dynamometer diagram for the suspension point within a single complete time interval, and determines whether to perform data cleaning on the dynamometer diagram of the suspension point based on the monitoring results. If so, cleans the dynamometer diagram of the suspension point; if not, corrects the dynamometer diagram of the suspension point. A feature extraction unit is configured to obtain all corrected dynamometer diagrams and place them into a first preset data set. A multi-point random sampling method is used to obtain feature data of different dynamometer diagrams in the first preset data set. Based on the feature data of all dynamometer diagrams in the first preset data set, dynamometer diagrams of different fault types are labeled to obtain a labeled data set. Data processing unit: collects vibration frequency data of the digital pumping unit in multiple complete time intervals at multiple time scales, and processes the data using a preset method to obtain multiple processing results; The preset method includes: Based on the vibration frequency data collected in multiple complete time intervals at multiple time scales, multiple coordinate curves of the vibration frequency data changing with time are constructed; Obtain any fault type and record it as a first fault type, and obtain multiple coordinate curves under the first fault type and store them in a second preset data set; Segmenting each coordinate curve in the second preset data set by the time interval to obtain M1 time intervals, and collecting the slopes of N1 data points in each time interval to obtain a collection result; Based on the acquisition results, calculating the average slopes of corresponding data points in corresponding time intervals of a plurality of coordinate curves in the second preset data set; Establishing a database to store the average slopes of corresponding data points of corresponding time intervals of a plurality of coordinate curves in the second preset data set; When there is a single coordinate curve recorded as the first coordinate curve, and the slopes of N2 data points in a single time interval of the first coordinate curve meet a preset condition, it is determined that the time interval corresponding to the first coordinate curve matches the time interval corresponding to the first fault type; When there are M2 matching time intervals in the first coordinate curve and the ratio of M2 / M1 exceeds a fourth threshold, it is determined that the first coordinate curve matches the coordinate curve corresponding to the first fault type; The preset conditions include: a difference between the slopes of the N2 data points and the average slope of corresponding data points in the database is less than a set value, and a ratio of N2 / N1 exceeds a third threshold; Fault monitoring unit: constructs a fault monitoring model, takes all labeled data sets as the first influencing factor and all processing results as the second influencing factor, trains the fault monitoring model together, and performs real-time monitoring of the digital pumping unit based on the trained fault monitoring model.

2. A digital pumping unit monitoring system according to claim 1, characterized in that: Obtaining the time for one crank movement to divide the time intervals includes: Obtain the starting position of the crank, and record the moment when the crank starts to move as time t0, and the moment when the crank returns to the starting position as time t1; The time interval between the time t0 and the time t1 is a time interval.

3. A digital pumping unit monitoring system according to claim 1, characterized in that: In a single complete time interval, the output distance of the motor and the displacement distance of the suspension point are obtained respectively, including: In a single complete time interval, the displacement distance of the suspension point is obtained by the displacement sensor, and the output distance of the motor is obtained by obtaining the motor speed and the circumference of the motor shaft.

4. A digital pumping unit monitoring system according to claim 1, characterized in that: Based on the monitoring result, correcting the indicator diagram of the suspension point includes: Obtain any indicator diagram of the suspension point as a first indicator diagram, and obtain a difference between an output distance of the motor and a displacement distance of the suspension point within a time interval corresponding to the first indicator diagram as a first difference; Obtain a continuous longitudinal line feature between the upper right vertex and the lower right vertex in the first dynamometer diagram, record it as a first line feature, obtain a continuous transverse line feature extending from the upper right vertex to the longitudinal axis, record it as a second line feature, and obtain a continuous transverse line feature extending from the lower right vertex to the longitudinal axis, record it as a third line feature; determining an upper right incremental vertex and a lower right incremental vertex in the first dynamometer diagram based on the first difference, the second line feature, and the third line feature; Based on the second line feature, a line is connected between the upper right vertex and the upper right incremental vertex; based on the third line feature, a line is connected between the lower right vertex and the lower right incremental vertex; based on the first line feature, a line is connected between the upper right incremental vertex and the lower right incremental vertex; The correction of the first indicator diagram is completed.

5. A digital pumping unit monitoring system according to claim 1, characterized in that: Based on the trained fault monitoring model, real-time monitoring of the digital pumping unit includes: Obtaining a corrected indicator diagram of the digital oil pumping unit and recording it as a second indicator diagram, and recording a processing result of corresponding vibration frequency data within a preset time period as a first processing result; The second dynamometer diagram and the first processing result are input into the trained fault monitoring model, and the fault type of the digital pumping unit is output.

6. A digital pumping unit monitoring system according to claim 1, characterized in that: Constructing a dynamometer diagram about the suspension point in a single complete time interval includes: In a single complete time interval, the displacement data of the suspension point is collected by a displacement sensor, and the load data of the suspension point is collected by a load sensor; and an indicator diagram about the suspension point is constructed with the displacement data of the suspension point as the horizontal coordinate and the load data of the suspension point as the vertical coordinate.

Images