Construction method of pumping unit well working condition monitoring knowledge base based on theoretical electric power reconstruction

Abstract

The application provides a pumping unit well working condition monitoring knowledge base construction method based on theoretical electric power reconstruction, belongs to the technical field of oil field informatization construction, acquires measured electric power curves of various working conditions of a pumping unit well and corresponding pumping unit parameters, forms an electric power historical data set, establishes a relationship formula among a suspension point displacement, a suspension point load, electric power and a crank angle, establishes an electric power to suspension point indicator diagram model, calculates a suspension point indicator diagram corresponding to the measured electric power curve, converts the calculated suspension point indicator diagram into a theoretical indicator diagram by using a theoretical indicator diagram generation method, establishes a suspension point indicator diagram to electric power model, calculates a theoretical electric power curve corresponding to the theoretical indicator diagram, extracts shape features of the theoretical electric power curve and the measured electric power curve, and establishes a pumping unit well working condition monitoring knowledge base by using the electric power historical data set. The application can automatically calculate a theoretical electric power curve according to a measured electric power curve, and does not need to involve experts.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of oilfield informatization construction, and particularly relates to a pumping unit well working condition monitoring knowledge base construction method based on theoretical electric power curve reconstruction. BACKGROUND

[0002] The main equipment of the pumping unit well is located in thousands of meters of underground, which is long-term affected by high temperature and high pressure and other factors, and the working condition is complex and changeable. Once the abnormal working condition cannot be found and effectively treated in time, it will not only affect the efficiency of oil production, but also may cause serious damage to the pumping unit well. Therefore, timely and accurate monitoring of the working condition of the pumping unit well is of great significance to the efficient and safe production of the oilfield. The traditional method takes the theoretical indicator diagram as the main basis, and realizes the working condition monitoring of the pumping unit well by drawing the suspension point indicator diagram and analyzing the geometric characteristics of the suspension point indicator diagram. The drawing of the suspension point indicator diagram needs to install load and displacement sensors on the pumping unit, however, the installation steps of the load and displacement sensors are complicated and need to be calibrated regularly, which leads to the problems of high installation and maintenance costs of the monitoring method based on the suspension point indicator diagram. With the continuous advancement of oilfield informatization construction, most motor control cabinets of pumping unit wells have been equipped with terminal monitoring equipment, which can realize real-time acquisition of electric parameters such as electric power without the need to install additional sensors. Electric parameters such as electric power are easy to obtain and contain rich information about the working condition change of the pumping unit well, so they have become an important data source for the current working condition monitoring method of the pumping unit well.

[0003] The Chinese patent "CN106337681A Pumping unit well working condition diagnosis method based on electric power diagram" provides another method for monitoring the working condition of the pumping unit well. This patent first converts the change curve of the electric power with time into an electric power diagram with displacement as the horizontal coordinate and electric power as the vertical coordinate, and then selects multiple typical electric power diagrams to form an electric power diagram for each working condition. Finally, the measured electric power is converted into an electric power diagram and classified into a category corresponding to the electric power diagram similar in characteristics in the electric power diagram, thereby realizing the working condition monitoring of the pumping unit well.

[0004] The Chinese patent "CN111322058A Pumping unit well working condition determination method and device based on electric parameter curve" provides another method for monitoring the working condition of the pumping unit well. This patent introduces the concept of standard electric parameter curve, takes a historical electric power curve of normal working condition as a standard electric parameter curve, and realizes the monitoring of the working condition of the pumping unit well such as liquid supply and rod break by comparing the peak value and the trough value of the measured electric parameter curve with the standard electric parameter curve.

[0005] The above method does not give the selection method of the electrical performance chart or the standard electrical parameter curve. If the electrical performance chart or the standard electrical parameter curve is selected randomly from the electrical performance chart or the standard electrical parameter curve of the normal working condition, the abnormal working condition similar to the chart or the curve will be judged as the normal working condition, and the normal working condition greatly different from the chart or the curve will be judged as the abnormal working condition, so that the misclassification problem occurs, and thus the precision of the pumping unit well working condition monitoring is reduced. SUMMARY

[0006] In view of the deficiencies in the prior art, the present application provides a pumping unit well working condition monitoring knowledge base construction method based on theoretical electrical power reconstruction. The present application establishes an electrical power to suspension point indicator diagram model and a suspension point indicator diagram to electrical power model based on the least square method, and in combination with a theoretical indicator diagram generation method, a theoretical electrical power curve reconstruction method is given, which solves the misclassification problem caused by the selection of the electrical performance chart or the standard electrical parameter curve relying on the expert experience. In addition, the present application extracts the shape features of the theoretical electrical power curve and the measured electrical power curve, analyzes the relationship between the two and the working condition, establishes a pumping unit well working condition monitoring knowledge base, and realizes the monitoring of the multiple working conditions of the pumping unit well based on the measured electrical power curve. The theoretical electrical power curve reconstructed by the present application is calculated from the theoretical indicator diagram, so the reconstructed theoretical electrical power curve and the theoretical indicator diagram have the same important position in the research of the pumping unit well working condition monitoring method.

[0007] The present application provides a pumping unit well working condition monitoring knowledge base construction method based on theoretical electrical power reconstruction, which comprises the following steps:

[0008] Obtaining the measured electrical power curve of each working condition of the pumping unit well and the corresponding pumping unit parameters to form an electrical power historical data set;

[0009] Establishing a relationship between the suspension point displacement, the suspension point load, the electrical power and the crank angle;

[0010] According to the relationship between the suspension point displacement, the suspension point load, the electrical power and the crank angle, an electrical power to suspension point indicator diagram model is established, and the suspension point indicator diagram corresponding to the measured electrical power curve is calculated;

[0011] Using a theoretical indicator diagram generation method, the calculated suspension point indicator diagram is converted into a theoretical indicator diagram;

[0012] According to the relationship between the suspension point displacement, the suspension point load, the electrical power and the crank angle, a suspension point indicator diagram to electrical power model is established, and the theoretical electrical power curve corresponding to the theoretical indicator diagram is calculated;

[0013] Extracting the shape features of the theoretical electrical power curve and the measured electrical power curve, and using the electrical power historical data set to establish a pumping unit well working condition monitoring knowledge base.

[0014] The measured electric power curve of each working condition of the pumping unit well and the corresponding pumping unit parameter are acquired to form an electric power historical data set, including:

[0015] The electric power curve and the corresponding pumping unit parameter under normal, insufficient liquid supply, gas influence, gas lock, rod break, fixed valve leakage and traveling valve leakage working conditions are collected to form an electric power historical data set.

[0016] The relationship between the suspension point displacement, the suspension point load, the electric power and the crank angle is established, including:

[0017] The relationship between the suspension point displacement and the crank angle is established;

[0018] The relationship between the suspension point load, the crank angle and the suspension point load torque is established;

[0019] The relationship between the suspension point load torque and the electric power is established.

[0020] The electric power to suspension point indicator diagram model is established according to the relationship between the suspension point displacement, the suspension point load, the electric power and the crank angle, and the suspension point indicator diagram corresponding to the measured electric power curve is calculated, including:

[0021] The electric power to suspension point load torque model is established according to the relationship between the suspension point load torque and the electric power, and the suspension point load torque corresponding to the measured electric power is calculated;

[0022] The crank angle to torque factor model is established according to the relationship between the crank angle and the suspension point load torque, and the torque factor corresponding to the crank angle is calculated;

[0023] The suspension point load torque to suspension point load model is established according to the relationship between the suspension point load and the suspension point load torque, and the suspension point load corresponding to the suspension point load torque is calculated;

[0024] The crank angle to suspension point displacement model is established according to the relationship between the suspension point displacement and the crank angle, and the suspension point displacement corresponding to the crank angle is calculated.

[0025] The generating method of the theoretical indicator diagram is adopted to convert the calculated suspension point indicator diagram into a theoretical indicator diagram, including:

[0026] The loading section and unloading section slopes of the theoretical indicator diagram are calculated;

[0027] The upper load line of the theoretical indicator diagram is calculated;

[0028] The lower load line of the theoretical indicator diagram is calculated;

[0029] The theoretical indicator diagram corresponding to the calculated suspension point indicator diagram is generated according to the loading section and unloading section slopes, the upper load line and the lower load line of the theoretical indicator diagram.

[0030] The establishing of the suspension point indicator diagram to electric power model, the calculation of the theoretical electric power curve corresponding to the theoretical indicator diagram, comprises:

[0031] According to the relationship between the suspension point displacement and the crank angle, a suspension point displacement to crank angle model is established to calculate the theoretical crank angle corresponding to the theoretical suspension point displacement.

[0032] According to the crank angle to torque factor model, the pumping unit torque factor corresponding to the theoretical crank angle is calculated.

[0033] According to the relationship between the suspension point load and the suspension point load torque, a suspension point load to suspension point load torque model is established to calculate the theoretical suspension point load torque corresponding to the theoretical suspension point load.

[0034] According to the relationship between the suspension point load torque and the electric power, a suspension point load to electric power model is established to calculate the theoretical electric power corresponding to the theoretical suspension point load torque.

[0035] The extraction of the shape features of the theoretical electric power curve and the measured electric power curve, the establishment of the pumping unit well working condition monitoring knowledge base by using the electric power historical data set, comprises:

[0036] The determination of the key features of the theoretical electric power curve;

[0037] The determination of the key features of the measured electric power curve;

[0038] The determination of the relationship between the features of the theoretical electric power curve, the features of the measured electric power curve and the working condition;

[0039] The establishment of the pumping unit well working condition monitoring knowledge base by using the electric power historical data set according to the key features of the theoretical electric power curve, the key features of the measured electric power curve, the relationship between the features of the theoretical electric power curve and the features of the measured electric power curve and the working condition.

[0040] Beneficial effects:

[0041] 1. The pumping unit well working condition monitoring knowledge base construction method based on the theoretical electric power reconstruction is proposed, the theoretical electric power curve is automatically calculated according to the measured electric power curve, no expert is needed to participate, and the problem of improper selection of the electric diagram chart or the standard electric parameter curve caused by the subjective factors such as the insufficient experience of the experts is avoided.

[0042] 2. The pumping unit well working condition monitoring knowledge base construction method based on the theoretical electric power reconstruction is proposed, the least square method is used to calculate the suspension point indicator diagram according to the discrete sequence of the measured electric power data, and the problem of the non-closure of the upper and lower dead points in the suspension point indicator diagram is solved.

[0043] 3. The application provides a pumping unit well working condition monitoring knowledge base construction method based on theoretical electric power reconstruction, which fully utilizes information of a theoretical electric power curve and a measured electric power curve, and improves the accuracy of pumping unit well working condition monitoring.

[0044] 4. The application provides a pumping unit well working condition monitoring knowledge base construction method based on theoretical electric power reconstruction, which automatically calculates feature parameters according to historical data sets of various working conditions, and avoids subjective error problems caused by expert experience and the like. BRIEF DESCRIPTION OF DRAWINGS

[0045] Figure 1 A pumping unit well working condition monitoring knowledge base construction method based on theoretical electric power reconstruction of the application embodiment has a flow chart;

[0046] Figure 2 A measured electric power curve diagram of the application embodiment;

[0047] Figure 3 A pumping unit size schematic diagram of the application embodiment;

[0048] Figure 4 A theoretical indicator diagram under various working conditions of the application embodiment and a generated theoretical indicator diagram;

[0049] Figure 5 A theoretical electric power curve diagram of the application embodiment;

[0050] Figure 6 A theoretical electric power curve feature extraction schematic diagram of the application embodiment;

[0051] Figure 7 A measured electric power curve feature extraction schematic diagram of the application embodiment;

[0052] Figure 8 A measured electric power curve under various working conditions of the application embodiment and a reconstructed theoretical electric power curve diagram. DETAILED DESCRIPTION

[0053] The specific embodiments of the application are described in further detail below in combination with the drawings and examples.

[0054] The application provides a pumping unit well working condition monitoring knowledge base construction method based on theoretical electric power reconstruction, which firstly realizes reconstruction of a theoretical electric power curve through establishment of an electric power to suspension point indicator diagram model, a suspension point indicator diagram to electric power model and a theoretical indicator diagram generation model, secondly formulates working condition monitoring rules by analyzing relationships between extracted theoretical electric power curve features, measured electric power curve features and various working conditions, and finally automatically calculates threshold values of the extracted features by using historical data sets of various working conditions, and establishes a pumping unit well working condition monitoring knowledge base.

[0055] Embodiment:

[0056] The embodiment proposes a pumping unit well working condition monitoring knowledge base construction method based on theoretical electric power reconstruction, as shown in Figure 1 The embodiment includes the following steps:

[0057] Step S1: Obtain the measured electric power curve of each working condition of the pumping unit well and the corresponding pumping unit parameters to form an electric power historical data set;

[0058] In the embodiment, the electric power curves and the corresponding pumping unit parameters of seven common working conditions, i.e., normal, insufficient liquid supply, gas influence, gas lock, rod break, fixed valve leakage, and traveling valve leakage, are collected to form an electric power historical data set, wherein the collected electric power curve is a discrete sequence obtained by equal interval sampling. The electric power historical data set contains n1 insufficient liquid supply working conditions, n2 gas influence working conditions, n3 gas lock working conditions, n4 traveling valve leakage working conditions, n5 fixed valve leakage working conditions, n6 rod break working conditions, and n7 normal working conditions. Let n = n1 + n2 + … + n7, and the electric power historical data set can be described as:

[0059]

[0060] T = {T1, T2, …, Tn} q}

[0061] wherein S m is the collected electric power historical data set, is the i-th electric power curve in the electric power historical data set, n is the number of the collected electric power curves, q is the number of sampling points of the electric power curve, and T is the sampling point set. The collected electric power curve is shown in Figure 2

[0062] Step S2: Establish the relationship between the suspension point displacement, the suspension point load, the electric power, and the crank angle to provide theoretical support for the electric power to suspension point indicator diagram model and the suspension point indicator diagram to electric power model;

[0063] In the embodiment, the relationship between the suspension point displacement and the crank angle is first established, and then the relationship between the suspension point load, the crank angle, and the electric power is established, and the specific process is as follows:

[0064] Step S2.1: Establish the relationship between the suspension point displacement and the crank angle;

[0065] In the embodiment, each crank angle of the pumping unit has a unique suspension point displacement corresponding thereto, but each suspension point displacement can have two crank angles corresponding thereto, i.e.,

[0066] PR = (ψ b - ψ) A (1) ​

[0067]

[0068] θ2=mod(2π-θ+κ,2π) (7)

[0069]

[0070] Where PR is the suspension point displacement, ψ is the angle between the line connecting the center of the gearbox output shaft and the center of the support bearing and the rear arm of the walking beam, ψ b ψ is the value when the suspension point is at the bottom dead center; γ is the angle between the line connecting the center of the crank pin and the center of the support bearing and the line connecting the center of the gearbox output shaft and the center of the support bearing; k is the angle between the line connecting the center of the gearbox output shaft and the center of the support bearing and the vertical direction; θ2 is the crank rotation angle in the positive direction (counterclockwise) starting from the line connecting the center of the gearbox output shaft and the center of the support bearing; θ is the crank rotation angle in the positive direction (clockwise) starting from the vertical direction; A is the length of the front arm of the pumping unit walking beam; C is the length of the rear arm of the pumping unit walking beam; P is the length of the connecting rod; L is the length from the center of the crank pin to the center of the support bearing; K is the length from the center of the gearbox output shaft to the center of the support bearing; I is the horizontal distance from the center of the gearbox output shaft to the center of the support bearing; R is the crank radius; J is a piecewise function, indicating that γ is positive when the crank angle is less than π and negative when the crank angle is greater than π. The pumping unit dimension diagram is shown below. Figure 3 As shown.

[0071] Step S2.2: Establish the relationship between suspension point load, crank angle, and suspension point load torque;

[0072] In this embodiment, the suspension point load torque can be calculated by multiplying the torque factor and the suspension point load at the same crank angle, i.e.

[0073] M r =YF(PRL-SUB) (9)

[0074]

[0075] α=π-(θ2-θ3) (11)

[0076] β=π-ψ-θ3 (12)

[0077]

[0078] Among them, M r θ is the suspension point load torque, PRL is the suspension point load, SUB is the structural unbalanced weight, TF is the torque factor of the pumping unit, α is the angle between the crank and the connecting rod, β is the angle between the walking beam and the connecting rod, and θ3 is the angle between the connecting rod and the line connecting the center of the gearbox output shaft to the center of the support bearing.

[0079] When the pumping unit parameters are determined, the relationship between the torque factor and the crank angle can be determined according to formula (10). The torque factor describes the force transmission process of the load at the suspension point to the load torque at the suspension point at the center of the crank shaft. The load at the suspension point transmits force to the center of the crank pin shaft through the beam, connecting rod and crank. The force transmitted to the center of the crank pin shaft generates a torque on the center of the crank shaft. The generated torque is the load torque at the suspension point.

[0080] Step S2.3: Establish the relationship formula between the load torque at the suspension point and the electric power.

[0081] In this embodiment, the relationship formula between the load torque at the suspension point and the electric power is as follows:

[0082]

[0083] M e = M net i g i b (15)

[0084] M net = M r -M b (16)

[0085] M b = M bmax sinθ (17)

[0086] Wherein, M b is the balance torque, M e is the motor load torque, M net is the net crank torque, n N is the rated speed, P m is the motor power, M r is the load torque at the suspension point, P m is the electric power, i g is the transmission ratio of the reduction gearbox, i b is the pulley reduction ratio, n N is the rated speed of the motor; M bmax is the maximum balance torque, which is the product of the sum of the weight of the balance block and the weight of the crank pin and the balance radius, the weight of the balance block, the weight of the crank pin and the crank radius are pumping unit parameters.

[0087] Step S3: According to the relationship formula among the suspension point displacement, the load at the suspension point, the electric power and the crank angle, an electric power to suspension point indicator diagram model is established, and the suspension point indicator diagram corresponding to the measured electric power curve is calculated.

[0088] In this embodiment, the i-th electric power curve data in the electric power data set is taken as an example, and the corresponding suspension point load value of each electric power value in the i-th electric power curve data is calculated. First, an electric power to suspension point load torque model is established, then a crank angle to pumping unit torque factor model and a suspension point load torque to suspension point load model are established, and finally a crank angle to suspension point displacement model is established, so as to complete the establishment of the suspension point indicator diagram, including:

[0089] Step S3.1: According to the relationship between the suspension point load torque and the electric power, an electric power to suspension point load torque model is established, and the corresponding suspension point load torque of the measured electric power is calculated.

[0090] In this embodiment, according to the relationship between the suspension point load torque and the electric power (14)-(17), the corresponding suspension point load torque of each electric power is calculated, that is,

[0091]

[0092] wherein, is the suspension point load torque value corresponding to the j-th sampling point in the i-th electric power curve, is the crank angle corresponding to the j-th sampling point in the i-th electric power curve.

[0093] Step S3.2: According to the relationship between the crank angle and the suspension point load torque, a crank angle to torque factor model is established, and the corresponding torque factor of the crank angle is calculated.

[0094] In this embodiment, according to formula (7), (10)-(13), the corresponding pumping unit torque factor of each crank angle is calculated, that is,

[0095]

[0096] wherein, is the torque factor corresponding to the j-th sampling point in the i-th electric power curve, is the value of a corresponding to the j-th sampling point in the i-th electric power curve, is the value of β corresponding to the j-th sampling point in the i-th electric power curve, is the value of θ3 corresponding to the j-th sampling point in the i-th electric power curve, is the value of θ2 corresponding to the j-th sampling point in the i-th electric power curve, is the value of γ corresponding to the j-th sampling point in the i-th electric power curve, is the value of L corresponding to the j-th sampling point in the i-th electric power curve.

[0097] Step S3.3: According to the relationship between the suspension point load and the suspension point load torque, a suspension point load torque to suspension point load model is established, and the suspension point load corresponding to the suspension point load torque is calculated;

[0098] In this embodiment, according to the relationship between the suspension point load and the suspension point load torque (9), the least square method is used to calculate the suspension point load corresponding to each suspension point load torque, that is,

[0099]

[0100] wherein, is the optimal suspension point load value corresponding to the jth sampling point in the ith electric power curve, is the suspension point load variable corresponding to the jth sampling point in the ith electric power curve; respectively represent the upper and lower limits of the allowable suspension point load, and χ is a threshold parameter.

[0101] Step S3.4: According to the relationship between the suspension point displacement and the crank angle, a crank angle to suspension point displacement model is established, and the suspension point displacement corresponding to the crank angle is calculated.

[0102] In this embodiment, according to the relationship between the suspension point displacement and the crank angle (1)-(8), the suspension point displacement corresponding to each crank angle is calculated.

[0103]

[0104] wherein, is the suspension point displacement corresponding to the jth sampling point in the ith electric power curve, is the value of ψ corresponding to the jth sampling point in the ith electric power curve.

[0105] The suspension point indicator diagram corresponding to the measured electric power curve is drawn with the suspension point displacement as the abscissa and the suspension point load as the ordinate. The calculated suspension point indicator diagram is shown in Figure 4 .

[0106] Step S4: The theoretical indicator diagram generation method is used to convert the calculated suspension point indicator diagram into a theoretical indicator diagram;

[0107] In this embodiment, the theoretical indicator diagram is a parallelogram composed of an upper load line, a lower load line, a loading section and an unloading section, and is an important theoretical basis for judging the working condition of the pumping unit. Therefore, the calculated suspension point indicator diagram is converted into a theoretical indicator diagram, which includes:

[0108] Step S4.1: Calculate the slope of the loading section and the unloading section of the theoretical indicator diagram;

[0109] In this embodiment, the slopes of the loading section and the unloading section are equal in the theoretical dynamometer card, and the average slope of the calculated polished rod dynamometer card is calculated from the polished rod displacement of the lower dead point to the sampling point of the upper stroke n u u The number of sampling points of the polished rod displacement when the sucker rod column is deformed in the upper stroke.

[0110] Step S4.2: Calculate the upper load line of the theoretical dynamometer card.

[0111] In this embodiment, the upper load line of the theoretical dynamometer card is the average polished rod load value of all sampling points from the sampling point corresponding to the maximum polished rod load of the upper stroke to the sampling point corresponding to the upper dead point.

[0112] Step S4.3: Calculate the lower load line of the theoretical dynamometer card.

[0113] In this embodiment, the lower load line of the theoretical dynamometer card is the average polished rod load value of all sampling points from the sampling point corresponding to the minimum polished rod load of the lower stroke to the sampling point corresponding to the upper dead point.

[0114] Step S4.4: According to the slopes of the loading section and the unloading section, the upper load line, and the lower load line of the theoretical dynamometer card, generate a theoretical dynamometer card corresponding to the calculated polished rod dynamometer card.

[0115] In this embodiment, according to the calculated slopes of the loading section and the unloading section, the upper load line, and the lower load line, a parallelogram can be determined, and the parallelogram is the generated theoretical dynamometer card. Taking the i-th data as an example, the j-th load data and the displacement data in the generated theoretical dynamometer card are and The polished rod dynamometer card of each working condition and the generated theoretical dynamometer card are as shown in Figure 4

[0116] Step S5: According to the relationship among the polished rod displacement, the polished rod load, the electric power, and the crank angle, establish a polished rod dynamometer card to electric power model, and calculate a theoretical electric power curve corresponding to the theoretical dynamometer card.

[0117] In this embodiment, taking the i-th data as an example, the theoretical electric power value corresponding to each sampling point in the theoretical dynamometer card is calculated. First, the least square method is used to calculate the theoretical crank angle corresponding to each theoretical polished rod displacement, then the torque factor corresponding to each theoretical crank angle and the theoretical polished rod load torque corresponding to each theoretical polished rod load are calculated, and finally the theoretical electric power corresponding to each theoretical polished rod load torque is calculated, including:

[0118] Step S5.1: According to the relationship between the polished rod displacement and the crank angle, establish a polished rod displacement to crank angle model, and calculate the theoretical crank angle corresponding to the theoretical polished rod displacement.​​

[0119] In this embodiment, the theoretical crank angle corresponding to each theoretical suspension point displacement is calculated by using the least square method according to the relationship between the suspension point displacement and the crank angle of formula (1)-(8), that is,

[0120]

[0121] wherein, is the optimal theoretical crank angle corresponding to the jth sampling point in the ith electric power curve, is the theoretical suspension point displacement corresponding to the jth sampling point in the ith electric power curve, is the crank angle variable corresponding to the jth sampling point in the ith electric power curve, and are the upper and lower boundaries of the allowable crank angle, respectively. When is located in the upper stroke sampling point, when is located in the lower stroke sampling point,

[0122] Step S5.2: According to the crank angle-torque factor model, the theoretical crank angle corresponding to the pumping unit torque factor is calculated.

[0123] In this embodiment, the relationship between the pumping unit torque factor and the crank angle is established according to formula (18), and the torque factor corresponding to each theoretical crank angle is calculated.

[0124] Step S5.3: According to the relationship between the suspension point load and the suspension point load torque, a suspension point load-suspension point load torque model is established, and the theoretical suspension point load torque corresponding to the theoretical suspension point load is calculated.

[0125] In this embodiment, the theoretical suspension point load torque corresponding to each theoretical suspension point load is calculated according to formula (9), that is,

[0126]

[0127] wherein, is the theoretical suspension point load torque corresponding to the jth sampling point in the ith electric power curve.

[0128] Step S5.4: According to the relationship between the suspension point load torque and the electric power, a suspension point load-electric power model is established, and the theoretical electric power corresponding to the theoretical suspension point load torque is calculated.

[0129] In this embodiment, the relationship between the suspension point load torque and the electric power is formula (14)-(17), and the theoretical electric power corresponding to each theoretical suspension point load torque is calculated, that is,

[0130]

[0131] wherein, is the theoretical electric power value corresponding to the jth sampling point in the ith electric power curve.

[0132] The theoretical electric power curve is established with the theoretical electric power value as the ordinate and the sampling time as the abscissa. The established theoretical electric power curve is shown in FIG. 2. Figure 5

[0133] Step S6: extracting the shape features of the theoretical electric power curve and the measured electric power curve, and establishing the pumping unit well working condition monitoring knowledge base.

[0134] In this embodiment, firstly, the key features of the theoretical electric power curve and the measured electric power curve are determined, secondly, the relationship between the theoretical electric power curve features, the measured electric power curve features and the working conditions is determined, and finally, the feature parameters of the working condition monitoring knowledge base are determined according to the electric power historical data set, including:

[0135] Step S6.1: determining the key features of the theoretical electric power curve;

[0136] In this embodiment, the extracted theoretical electric power curve features are shown in Table 1. The schematic diagram of the extracted theoretical electric power curve features is shown in FIG. 3. Figure 6

[0137] Table 1: the extracted theoretical electric power curve features

[0138]

[0139] Step S6.2: determining the key features of the measured electric power curve;

[0140] In this embodiment, the extracted measured electric power curve features are shown in Table 2. The schematic diagram of the extracted measured electric power curve features is shown in FIG. 4. Figure 7

[0141] Table 2: the extracted measured electric power curve features

[0142]

[0143]

[0144] Step S6.3: determining the relationship between the theoretical electric power curve features, the measured electric power curve features and the working conditions;

[0145] In this embodiment, the relationship between the theoretical electric power curve features, the measured electric power curve features and the working conditions is shown in Table 3, wherein, wherein x1, x2…x 10 are the feature parameters. The measured electric power curve of each working condition and the reconstructed theoretical electric power curve are shown in FIG. 5.​​​Figure 8 Theoretical electric power curve characteristics, measured electric power curve characteristics and working conditions are shown in Table 3.

[0146] Table 3 Relationship between theoretical electric power curve characteristics, measured electric power curve characteristics and working conditions

[0147]

[0148] Step S6.4: Establishing the working condition monitoring knowledge base of the pumping unit well by using the electric power historical data set according to the key characteristics of the theoretical electric power curve, the key characteristics of the measured electric power curve, the relationship between the theoretical electric power curve characteristics, the measured electric power curve characteristics and the working conditions.

[0149] In this embodiment, the feature parameters x1, x2…x 10 of the working condition monitoring knowledge base are calculated by using the constructed electric power data set and the determined relationship between the theoretical electric power curve characteristics, the measured electric power curve characteristics and the working conditions.

[0150] For the feature parameter x1, the minimum value of the in the historical electric power data set of n1 liquid-lacking working conditions is determined, that is

[0151]

[0152] wherein, i∈[1,n1] represents the value of the i-th electric power curve in the liquid-lacking working condition data set.

[0153] For the feature parameters x2, x3, x4, x5, the minimum value and the maximum value of the and in the historical electric power data set of n2 gas-affected working conditions are determined, that is

[0154]

[0155] wherein, i∈[1,n2] represents the value of the i-th electric power curve in the gas-affected working condition data set, i∈[1,n2] represents the value of the i-th electric power curve in the gas-affected working condition data set.

[0156] For the feature parameters x6, x7, the minimum value of the and in the historical electric power data set of n3 gas lock working conditions is determined, that is

[0157]

[0158] wherein, i∈[1, n3] represents the value of the i-th electric power curve in the gas lock working condition data set , i∈[1, n3] represents the value of the i-th electric power curve in the gas influence working condition data set .

[0159] For the characteristic parameters x8, x9, the minimum value and the maximum value in the n4 electric power data sets of the floating valve leakage working condition are determined, that is , ,

[0160]

[0161]

[0162] wherein, i∈[1, n4] represents the value of the i-th electric power curve in the floating valve leakage working condition data set , i∈[1, n4] represents the value of the i-th electric power curve in the floating valve leakage working condition data set .

[0163] For the characteristic parameters x 10 , the maximum value of the n5 electric power data sets of the fixed valve leakage working condition is determined, that is ,

[0164]

[0165] wherein, i∈[1, n5] represents the value of the i-th electric power curve in the floating valve leakage working condition data set .

[0166] Based on the above established pumping unit working condition monitoring knowledge base, the pumping unit working condition is monitored in real time. First, the measured electric power curve is calculated with the theoretical electric power curve, then the features of the measured electric power curve and the theoretical working condition electric power curve are extracted, and finally the extracted features are input into the computer loaded with the established knowledge base, and the current working condition type is output by the computer.

[0167] Each embodiment in the present application is described in a progressive manner, and the same and similar parts between each embodiment can be referred to each other, and each embodiment mainly explains the difference from other embodiments.

[0168] The protection scope of the present application is not limited to the above-mentioned embodiments, and obviously, those skilled in the art can make various modifications and changes to the present disclosure without departing from the scope and spirit of the present disclosure. If these modifications and changes belong to the scope of the claims of the present disclosure and its equivalent technologies, the present disclosure also intends to include these modifications and changes.

Claims

1. A pumping unit well operating condition monitoring knowledge base construction method based on theoretical electric power reconstruction, characterized in that, The method comprises the following steps: obtaining measured electric power curves of each working condition of the pumping unit well and corresponding pumping unit parameters to form an electric power historical data set; establishing a relationship between the suspension point displacement, the suspension point load, the electric power and the crank angle; establishing an electric power to suspension point dynamometer card model according to the relationship between the suspension point displacement, the suspension point load, the electric power and the crank angle, and calculating a suspension point dynamometer card corresponding to the measured electric power curve; using a theoretical dynamometer card generation method to convert the calculated suspension point dynamometer card into a theoretical dynamometer card; establishing a suspension point dynamometer card to electric power model according to the relationship between the suspension point displacement, the suspension point load, the electric power and the crank angle, and calculating a theoretical electric power curve corresponding to the theoretical dynamometer card; extracting shape features of the theoretical electric power curve and the measured electric power curve, and establishing a pumping unit well working condition monitoring knowledge base by using the electric power historical data set; establishing an electric power to suspension point dynamometer card model according to the relationship between the suspension point displacement, the suspension point load, the electric power and the crank angle, and calculating a suspension point dynamometer card corresponding to the measured electric power curve, which comprises: calculating a suspension point load torque corresponding to each electric power according to the relationship between the suspension point load torque and the electric power, i.e. wherein, is the suspension load torque value corresponding to the jth sampling point in the ith electric power curve, is the crank angle corresponding to the jth sampling point in the ith electric power curve; n N is the rated speed, M bmax is the maximum balance torque, i g is the reduction gear ratio, i b is the pulley reduction ratio, is the jth value of the ith electric power curve in the electric power history data set; establishing a crank angle to torque factor model according to the relationship between the crank angle and the suspension point load torque, and calculating a torque factor corresponding to the crank angle, i.e.: wherein, is a torque factor corresponding to the jth sampling point in the ith electric power curve, is an a value corresponding to the jth sampling point in the ith electric power curve, is a β value corresponding to the jth sampling point in the ith electric power curve, is a θ3 value corresponding to the jth sampling point in the ith electric power curve, is a θ2 value corresponding to the jth sampling point in the ith electric power curve, is a γ value corresponding to the jth sampling point in the ith electric power curve, is an L value corresponding to the jth sampling point in the ith electric power curve, A is a beam fore-arm length, C is a beam back-arm length, P is a connecting rod length, L is a length from a crank pin shaft center to a bracket bearing center, γ is an included angle between the crank pin shaft center and the bracket bearing center and a line between a reduction gearbox output shaft center and the bracket bearing center, θ is a crank rotation angle from a vertical direction, clockwise rotation being a positive direction, is a θ value of the jth sampling point in the ith electric power curve, κ is an included angle between the reduction gearbox output shaft center and the bracket bearing center and a vertical direction, J is a piecewise function, K is a length from the reduction gearbox output shaft center to the bracket bearing center, R is a crank radius, θ2 is a crank rotation angle from a line between the reduction gearbox output shaft center and the bracket bearing center, counterclockwise rotation being a positive direction, and θ3 is an included angle between a connecting rod and a line between the reduction gearbox output shaft center and the bracket bearing center. calculating a suspension point load corresponding to each suspension point load torque according to the relationship between the suspension point load and the suspension point load torque by using the least square method, i.e.: wherein, is the optimal suspension load value corresponding to the jth sampling point in the ith electric power curve, is the suspension load variable corresponding to the jth sampling point in the ith electric power curve; respectively represent the upper and lower bounds of the allowable suspension load, χ is a threshold parameter, is the torque factor corresponding to the jth sampling point in the ith electric power curve, is the suspension load of the jth sampling point in the ith electric power curve, SUB is the structural unbalance weight; calculating a suspension point displacement corresponding to each crank angle according to the relationship between the suspension point displacement and the crank angle; wherein, is the suspension point displacement corresponding to the jth sampling point in the ith electric power curve, is the value of ψ corresponding to the jth sampling point in the ith electric power curve, ψ being the angle between the line connecting the centre of the output shaft of the reduction gearbox and the centre of the bearing of the support and the back arm of the walking beam, ψ b is the value of ψ when the suspension point is at the bottom dead point; establishing a suspension point dynamometer card to electric power model according to the relationship between the suspension point displacement, the suspension point load, the electric power and the crank angle, and calculating a theoretical electric power curve corresponding to the theoretical dynamometer card, which comprises: calculating a theoretical crank angle corresponding to each theoretical suspension point displacement according to the relationship between the suspension point displacement and the crank angle by using the least square method, i.e.: wherein, is the optimal theoretical crank angle corresponding to the jth sampling point in the ith electric power curve, is the theoretical suspension displacement corresponding to the jth sampling point in the ith electric power curve, is the crank angle variable corresponding to the jth sampling point in the ith electric power curve, and are the upper and lower bounds of the allowed crank angle, respectively, when is located in the upper stroke sampling point, is located in the lower stroke sampling point, is located in the lower stroke sampling point, establishing a suspension point load to suspension point load torque model according to the relationship between the suspension point load and the suspension point load torque, and calculating a theoretical suspension point load torque corresponding to the theoretical suspension point load, i.e.: wherein, Tth(i,j) is the theoretical load torque of the suspension point corresponding to the jth sampling point in the ith electric power curve, establishing a suspension point load to electric power model according to the relationship between the suspension point load torque and the electric power, and calculating a theoretical electric power corresponding to the theoretical suspension point load torque, i.e.: wherein, Pth(i,j) is the theoretical electric power value corresponding to the jth sampling point in the ith electric power curve.

2. The method for constructing a knowledge base of oil pumping unit well operating condition monitoring based on theoretical electric power reconstruction according to claim 1, characterized in that, obtaining measured electric power curves of each working condition of the pumping unit well and corresponding pumping unit parameters to form an electric power historical data set, which comprises: collecting electric power curves and corresponding pumping unit parameters under normal, insufficient liquid supply, gas influence, gas lock, rod break, fixed valve leakage and traveling valve leakage working conditions to form the electric power historical data set.

3. The method for constructing a knowledge base of oil pumping unit well operating condition monitoring based on theoretical electric power reconstruction according to claim 1, characterized in that, using a theoretical dynamometer card generation method to convert the calculated suspension point dynamometer card into a theoretical dynamometer card, which comprises: calculating loading segment and unloading segment slopes of the theoretical dynamometer card; calculating an upper load line of the theoretical dynamometer card; calculating a lower load line of the theoretical dynamometer card; generating a theoretical dynamometer card corresponding to the calculated suspension point dynamometer card according to the loading segment and unloading segment slopes, the upper load line and the lower load line of the theoretical dynamometer card.

4. The method for constructing a knowledge base of oil pumping unit well operating condition monitoring based on theoretical electric power reconstruction according to claim 1, characterized in that, The shape characteristics of the extracted theoretical electric power curve and the measured electric power curve, the pumping well working condition monitoring knowledge base is established by using the electric power historical data set, including: Determine the key features of the theoretical electric power curve; Determine the key features of the measured electric power curve; Determine the relationship between the features of the theoretical electric power curve, the features of the measured electric power curve and the working condition; According to the key features of the theoretical electric power curve, the key features of the measured electric power curve, the relationship between the features of the theoretical electric power curve, the features of the measured electric power curve and the working condition, the pumping well working condition monitoring knowledge base is established by using the electric power historical data set.

Images