A downhole working condition fault diagnosis method and device

Abstract

The application provides a downhole working condition fault diagnosis method and device, and belongs to the field of pipe string safety diagnosis. Factors affecting the sucker rod broken-off are summarized as pump working condition influence factors and wellbore working condition influence factors, the two influence factors are integrated into an existing mechanical model, the up and down stroke maximum hook load is adjusted to be consistent with the field working condition, the influence factor deviation is examined, and it is determined what kind of abnormality occurs downhole. The diagnosis method can diagnose the reason for the sucker rod broken-off according to the ground key parameters, and can not only diagnose the pump abnormality, but also diagnose the wellbore friction abnormality. Meanwhile, the diagnosis model trained through machine learning based on the BP neural network is used for intelligent diagnosis, the influence factors reflecting the actual working condition can be better estimated, the diagnosis efficiency is high, the prediction precision is good, the method has a very important reference significance for the field application of the beam pump oil production system, and the deficiency that the current domestic diagnosis is mostly dependent on the dynamometer card is made up.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of pipe string safety diagnosis, and particularly relates to a downhole working condition fault diagnosis method and device. BACKGROUND

[0002] Rod pump oil production is one of the traditional mechanical oil production methods in China, accounting for 80% of the production wells in various oilfields in China. At present, most of the oilfields in China have entered the high water cut stage, which not only increases the oil production cost, but also greatly increases the possibility of sucker rod breakage (abnormal downhole working condition). Once the sucker rod is broken, salvage operation is needed, and considering the salvage operation cost, production stoppage, possible damage to farmland caused by large equipment operation, rod pump re-entry into the well and other problems, the oilfield has suffered great economic losses. In order to avoid the impact caused by the sucker rod breakage, the downhole working condition fault should be predicted and diagnosed as much as possible before the problem occurs. The traditional diagnosis method mostly relies on the analysis of dynamometer card, which realizes the diagnosis of whether the pump has a problem, but cannot diagnose the friction influence.

[0003] Therefore, how to provide a more comprehensive downhole working condition fault diagnosis method is a problem to be solved at present. SUMMARY

[0004] The present application provides a downhole working condition fault diagnosis method and device, which is time-saving and labor-saving, has high diagnosis efficiency and good prediction accuracy, and makes up for the deficiency of the current domestic diagnosis method which mostly relies on dynamometer card. The specific scheme is as follows:

[0005] In a first aspect, the present application discloses a downhole working condition fault diagnosis method, comprising: establishing a sucker rod mechanical theory model, the theory model being used to calculate the theoretical value of the maximum surface load of the sucker rod of a target oil well in upstroke and downstroke; acquiring the actual value of the maximum surface load of the sucker rod of the target oil well in upstroke and downstroke collected on site; introducing a pump working condition influence factor and a wellbore working condition influence factor into the theory model, establishing the mapping relationship between the pump working condition influence factor, the wellbore working condition influence factor and the maximum surface load of the sucker rod in upstroke and downstroke; initializing the value of the pump working condition influence factor and the wellbore working condition influence factor, bringing the actual value into a fault diagnosis prediction model to obtain the pump working condition influence factor and the wellbore working condition influence factor matched with the actual value; comparing the matching result with the initialized value, and judging the abnormal situation in the downhole through the deviation of the influence factor.

[0006] Optionally, the step of calculating the theoretical value of the maximum surface load of the sucker rod of the target oil well in upstroke and downstroke specifically comprises: segmentally calculating the sucker rod by using the microelement method to obtain the theoretical value of the maximum surface load of the sucker rod in upstroke and downstroke.

[0007] Optionally, in the theoretical model, the associated factors of the pump operating condition influence factor include the resistance of the well fluid passing through the traveling valve, the friction between the pump plunger and the liner, and the pump mouth pressure; the associated factors of the wellbore operating condition influence factor include the contact force of the tubing on the sucker rod, the equivalent friction of the well fluid on the sucker rod, and the friction between the well fluid and the tubing. Optionally, the step of introducing the pump operating condition influence factor and the wellbore operating condition influence factor into the theoretical model, and establishing the mapping relationship between the pump operating condition influence factor and the wellbore operating condition influence factor and the maximum surface load of the upper and lower strokes of the sucker rod

[0008] , specifically includes: after the influence factor is introduced into the theoretical model, the calculation formula of the force of the upper stroke sucker rod microelement is: ; the calculation formula of the force of the lower stroke sucker rod microelement is: ; wherein, , are the axial forces of the upper, lower, and upper stroke sucker rod microelement nodes j, respectively, is the buoyancy of the unit length of the sucker rod, is the gravity of the unit length of the sucker rod, is the inertial force of the unit length of the sucker rod, is the frictional resistance of the unit length of the sucker rod to the well fluid, , are the forces of the upper and lower stroke well fluid flow on the unit length of the sucker rod, respectively, f is the friction coefficient of the tubing and the sucker rod, N is the support force of the tubing on the sucker rod microelement, is the mass of the sucker rod microelement, , are the accelerations of the upper and lower stroke sucker rod microelements, respectively;

[0009] The boundary condition calculation method is: ; ; wherein, is the pump operating condition influence factor, is the wellbore operating condition influence factor, is the force acting on the plunger at the bottom end of the sucker rod in the upper stroke, is the force acting on the plunger at the bottom end of the sucker rod in the lower stroke, is the liquid column load on the plunger, is the friction between the pump plunger and the liner, is the friction between the well fluid and the tubing, is the wellhead back pressure influence in the upper stroke, is the liquid column inertial load, is the load of the gas in the pump on the plunger, is the resistance of the well fluid passing through the traveling valve hole, is the wellhead back pressure influence.

[0010] Optionally, the initialization pump working condition influence factor and the wellbore working condition influence factor are valued, the actual value is brought into the fault diagnosis prediction model, and a pump working condition influence factor and a wellbore working condition influence factor matched with the actual value are obtained. The step specifically comprises:

[0011] The pump working condition influence factor and the wellbore working condition influence factor are both assigned an initial value 1.

[0012] The actual value is input into the fault diagnosis prediction model.

[0013] The fault diagnosis prediction model adjusts the values of the pump working condition influence factor and the wellbore working condition influence factor, so that the maximum surface load of the sucker rod upstroke and downstroke calculated by the adjusted pump working condition influence factor and the wellbore working condition influence factor matches the actual value.

[0014] The adjusted pump working condition influence factor and the wellbore working condition influence factor are output.

[0015] Optionally, the training method of the fault diagnosis prediction model comprises: the pump working condition influence factor and the wellbore working condition influence factor are respectively valued and combined at certain intervals in the interval (0, 1] to obtain a plurality of influence factor groups.

[0016] Each influence factor group is respectively brought into the theoretical model to calculate the maximum surface load of the sucker rod upstroke and downstroke corresponding to each influence factor group.

[0017] An influence factor group and the maximum surface load of the sucker rod upstroke and downstroke corresponding thereto are combined to obtain a data group.

[0018] The plurality of data groups obtained according to the plurality of influence factor groups are divided into a test set and a validation set, and are trained through a BP neural network model. The maximum surface load of the sucker rod upstroke and downstroke is taken as the input of the training model, and the influence factor group is taken as the output of the training model.

[0019] Optionally, the step of comparing the matching result with the initial value and judging the abnormal situation occurring in the downhole through the deviation of the influence factor specifically comprises:

[0020] The deviation degrees of the pump working condition influence factor and the wellbore working condition influence factor output by the fault diagnosis prediction model from the initial value 1 are compared respectively.

[0021] The higher the deviation degree of the influence factor from the initial value 1, the higher the probability that the corresponding reason leads to the working condition fault.

[0022] In a second aspect, the application discloses a downhole working condition fault diagnosis device, comprising:

[0023] a model establishing unit configured to establish a theoretical model of a sucker rod mechanism, the theoretical model being used to calculate a theoretical value of a maximum load of a sucker rod of a target oil well;

[0024] a data collecting unit configured to obtain an actual value of the maximum load of the sucker rod of the target oil well collected in the field;

[0025] a factor introducing unit configured to introduce a pump working condition factor and a wellbore working condition factor into the theoretical model, and establish a mapping relationship between the pump working condition factor and the wellbore working condition factor and the maximum load of the sucker rod;

[0026] a prediction calculating unit configured to initialize the pump working condition factor and the wellbore working condition factor, and bring the actual value into a fault diagnosis prediction model to obtain the pump working condition factor and the wellbore working condition factor matched with the actual value;

[0027] a fault diagnosing unit configured to compare the matching result with the initialized value, and judge an abnormal situation occurring in a downhole according to a deviation of the factors.

[0028] A third aspect of the embodiment of the present application provides an electronic device, comprising:

[0029] one or more processors; a memory; one or more application programs, wherein the one or more application programs are stored in the memory and configured to be executed by the one or more processors, and the one or more programs are configured to execute the method according to the first aspect.

[0030] A fourth aspect of the embodiment of the present application provides a computer readable storage medium, wherein the computer readable storage medium stores program codes, and the program codes can be called and executed by a processor to execute the method according to the first aspect.

[0031] In summary, the present application provides a downhole working condition fault diagnosis method and device, factors affecting the sucker rod broken are attributed to pump working condition influence factors and wellbore working condition influence factors, the two influence factors are integrated into the existing mechanical model, and it is assumed that the theoretical calculation model is an ideal working condition model, when the downhole working condition deviates from the design or the downhole environment deteriorates, the influence factor decreases. By adjusting the parameters, the maximum surface load of the upstroke and downstroke is consistent with the field working condition, and the influence factor deviation is examined, so that the specific abnormality of the downhole can be preliminarily judged. The diagnosis method can diagnose the cause of the sucker rod broken according to the key parameters on the ground, and can not only diagnose the abnormality of the pump, but also diagnose the abnormality of the wellbore friction. At the same time, the diagnosis model trained based on the machine learning of the BP neural network can better estimate the influence factor reflecting the actual working condition, has the advantages of high diagnosis efficiency and good prediction accuracy, and has a very important reference significance for the field application of the beam pump oil production system. BRIEF DESCRIPTION OF DRAWINGS

[0032] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments. It should be understood that the following drawings only show some embodiments of the present application, and therefore should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can also be obtained without creative labor on the basis of these drawings.

[0033] Figure 1 The method flow chart of the downhole working condition fault diagnosis method of the embodiments of the present application;

[0034] Figure 2 The schematic diagram of the micro-element form of the sucker rod in the inclined well of the embodiments of the present application;

[0035] Figure 3 The force analysis diagram of the upstroke sucker rod of the embodiments of the present application;

[0036] Figure 4 The force analysis diagram of the downstroke sucker rod of the embodiments of the present application;

[0037] Figure 5 The comparison diagram of the theoretical value and the actual value of the maximum surface load of the upstroke of the embodiments of the present application;

[0038] Figure 6 The comparison diagram of the theoretical value and the actual value of the maximum surface load of the downstroke of the embodiments of the present application;

[0039] Figure 7 The comparison diagram of the theoretical value and the actual value of the maximum surface load of the upstroke after matching of the embodiments of the present application;

[0040] Figure 8A comparison chart of the matched post-stroke maximum suspension point load theoretical value and actual value of the embodiment of the present application;

[0041] Figure 9 A functional module block diagram of the downhole working condition fault diagnosis device of the embodiment of the present application;

[0042] Figure 10 A structural block diagram of an electronic device for executing the downhole working condition fault diagnosis method according to the embodiment of the present application;

[0043] Figure 11 A structural block diagram of a computer readable storage medium for storing or carrying program codes for implementing the downhole working condition fault diagnosis method according to the embodiment of the present application.

[0044] Icon:

[0045] Model establishment unit 110; data acquisition unit 120; factor introduction unit 130; prediction calculation unit 140; fault diagnosis unit 150; electronic device 300; processor 310; memory 320; computer readable storage medium 400; program code 410. DETAILED DESCRIPTION

[0046] Rod pump oil production is currently one of the traditional mechanical oil production methods in China, accounting for 80% of the production wells in China's oilfields. At present, most of China's oilfields have entered the high water cut stage, not only increasing the oil production cost, but also greatly increasing the possibility of sucker rod breakage (downhole working condition abnormality). Once the sucker rod is broken, salvage operations are needed, considering salvage operation costs, production impact, possible farmland damage compensation caused by large equipment on-site operations, rod pump re-entry into the well, etc., which brings great economic losses to the oilfield. To avoid the impact of the sucker rod breakage fault, the downhole working condition fault should be predicted and diagnosed as much as possible before the problem occurs. Traditional diagnosis methods mostly rely on analyzing the dynamometer card, realizing the diagnosis of whether the pump has a problem, but cannot diagnose the friction influence.

[0047] Traditional downhole working condition prediction and diagnosis methods often require a lot of manpower. Machine learning has been widely used in complex problems in various engineering applications and scientific fields. In recent years, machine learning has emerged in the oil field and has performed excellently in improving operation efficiency, reducing operation cost, providing safety measures, etc.

[0048] Therefore, how to provide a more comprehensive downhole working condition fault diagnosis method is a problem to be solved at present.

[0049] Based on the above problems, the application designs a neural network downhole working condition fault prediction and diagnosis method based on ground parameters, i.e., maximum suspension point load of upper and lower dead points, which can directly predict and diagnose downhole faults through field collected rod and tube pump parameters, can diagnose not only pump abnormalities but also wellbore friction abnormalities, saves time and effort, and has practical significance and good application prospect.

[0050] To make the objectives, technical solutions and advantages of the embodiments of the present application clearer, the following will combine the drawings in the embodiments of the present application

[0051] to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are some of the embodiments of the present application but not all the embodiments. The components of the embodiments of the present application described and shown in the accompanying drawings can be arranged and designed in various different configurations.

[0052] Therefore, the following detailed description of the embodiments of the present application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but only represents selected embodiments of the present application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present application without creative labor are within the scope of protection of the present application.

[0053] It should be noted that: similar reference numbers and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings.

[0054] In the description of the present application, it should be noted that the terms "top", "bottom", "inner", "outer" and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, or the orientation or positional relationship in which the product of the present application is usually placed, and are only for the convenience of describing the present application and simplifying the description, and therefore cannot be understood as indicating or implying that the indicated device or element must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as limiting the present application. In addition, the terms "first", "second" and the like are only used to distinguish the description and cannot be understood as indicating or implying relative importance.

[0055] In the description of the present application, it should also be noted that unless otherwise explicitly specified and limited, the terms "set", "mount", "connected", "connected" should be broadly understood, for example, it can be fixedly connected, or it can be detachably connected, or integrally connected; it can be mechanically connected, or it can be electrically connected; it can be directly connected, or it can be indirectly connected through an intermediate medium, or it can be the communication inside two elements. For those of ordinary skill in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0056] It should be noted that the embodiments in the present application and the features in the embodiments can be combined with each other without conflict.

[0057] As shown in the drawings, Figure 1 The downhole working condition fault diagnosis method of the present application comprises:

[0058] Step S101, a theoretical model of the pumping rod is established, and the theoretical model is used to calculate the theoretical value of the maximum load of the pumping rod in the target oil well.

[0059] Based on the specific situation of the downhole working condition, the theoretical model of the pumping rod is established, and the dynamic characteristics and mechanical behavior of the system can be analyzed by solving the model. By establishing the mathematical model of the relevant key factors and applying appropriate analytical techniques, the force, stress, deformation and vibration of the pumping rod system under different conditions can be predicted.

[0060] After the theoretical model is established, as a preferred embodiment, when calculating the theoretical value of the maximum load of the pumping rod in the target oil well, the pumping rod can be segmented and calculated by using the micro-element method, wherein the micro-element form of the pumping rod in the ordinary inclined well is as shown in Figure 2 The pumping rod in the working state is bent and deformed along the shape of the borehole trajectory.

[0061] The specific calculation method is:

[0062] In the ordinary inclined well, the force of a certain section of the pumping rod (pumping rod micro-element) is analyzed, and Newton's second law of motion is applied to the micro-element section, that is:

[0063] (1)

[0064] In formula (1):

[0065] - the mass of the pumping rod micro-element, kg;

[0066] a- acceleration of the pumping rod micro-element, N·kg-1;

[0067] - the sum of the forces of the pumping rod micro-element, N.

[0068] The force of the pumping rod is as shown in Figure 3 and Figure 4 On this basis, combined with formula (1), according to the principle of balance, we can get:

[0069] The force of the pumping rod micro-element in the upstroke is: ;

[0070] The force of the pumping rod micro-element in the downstroke is: ;

[0071] wherein: , - the axial force at the upper stroke Sucker rod micro-element node j, N;

[0072] , - the axial force at the lower stroke Sucker rod micro-element node j+1, N;

[0073] - the buoyancy per unit length of the sucker rod, ;

[0074] - the gravity per unit length of the sucker rod, ;

[0075] - the inertial force per unit length of the sucker rod, ;

[0076] - the well fluid friction resistance per unit length of the sucker rod, ;

[0077] , - the force of the upper and lower stroke well fluid flow on the unit length of the sucker rod, ;

[0078] f- the tubing and sucker rod friction coefficient; N- the support force of the tubing on the micro-element of the sucker rod, N; - the mass of the micro-element of the sucker rod, kg.

[0079] , - the acceleration of the upper and lower stroke sucker rod micro-element, .

[0080] ; ;

[0081] Equations (3-a) and (3-b) are the force model of the micro-element of the sucker rod.

[0082] The connection point of the sucker rod and the plunger is defined as node 0, and and can be obtained according to the relevant boundary conditions, which are substituted into the above equation, and the axial force of each node on the sucker rod can be obtained by using the iterative algorithm.

[0083] The boundary condition calculation method is: ; ;

[0084] wherein: - the force acting on the bottom end of the sucker rod plunger in the upstroke, positive value, generally tensile force, N; - the force acting on the bottom end of the sucker rod plunger in the downstroke, negative value, generally pressure, N; - the liquid column load on the plunger, N; - the friction between the pump plunger and the bushing, N; - the friction between the well fluid and the tubing, N; - the wellhead back pressure influence in the upstroke, N; - the liquid column inertia load, N; - the load generated by the gas in the pump on the plunger, N; - the resistance generated by the well fluid through the traveling valve hole, N; - the wellhead back pressure influence, N.

[0085] wherein the calculation of the 12 forces in formula (4-a) and formula (3-b) is shown in Table (1): Table (1) Summary of Mechanical Calculation Model

[0086]

[0087] wherein: - the density of the sucker rod material, ; - the density of the coupling material, ; - the density of the centralizer material, ; L - the total length of the sucker rod, coupling and centralizer, m; - the total length of the sucker rod, m; - the total length of the coupling, m; - the total length of the centralizer, m; K - the influence coefficient of the traveling valve resistance increase due to the sucker rod string vibration; - the pressure drop generated by the well fluid through the traveling valve, ; - the pressure drop generated by the well fluid through the fixed valve, approximately equal to the pressure drop generated by the well fluid through the traveling valve, ; H - the pump hanging vertical depth, m; - the depth of the submergence in the vertical direction, m; - the cross-sectional area of the sucker rod, ; - the cross-sectional area of the coupling, ; - the cross-sectional area of the centralizer, ; - the cross-sectional area of the plunger, ; - the cross-sectional area of the tubing, ; - the density of the oil fluid, ; - total length of centralizer, m; - force generated by well fluid flow on the entire rod string in upstroke, N; - force generated by well fluid flow on the entire rod string in downstroke, N;a- acceleration of rod string, ; ε- acceleration reduction factor of oil column caused by enlarged flow cross section of tubing; - back pressure at well head, Pa; - number of traveling valves; g- acceleration of gravity, N / kg; - diameter of plunger, m; δ- radial clearance between plunger and liner of pump, m; - effective stroke of pumping unit, m; - liquid level in pump, m; - gas-oil ratio at surface, ; α- solubility coefficient, ; - gas pressure in pump, Pa; - standard pressure, Pa; - volume factor of crude oil; - water cut, decimal; T- temperature at pump inlet, K; - temperature at surface, K; Y- compressibility factor of natural gas; - frictional resistance of well fluid to unit length of rod, N / m; - frictional resistance of well fluid to unit length of coupling, N / m; - frictional resistance of well fluid to unit length of centralizer, N / m; - equivalent frictional resistance of well fluid to unit length of rod string, N / m; by substituting the formulas described in Table 1 into corresponding positions, combined with boundary conditions, the maximum hanging point load of the rod in upstroke and downstroke can be calculated by iteration from the bottom end of the rod to the well head.

[0088] Through the above steps, the theoretical value of the maximum hanging point load of the rod in upstroke and downstroke can be calculated based on the theoretical model. For different target wells, the corresponding parameter values are brought in, and the theoretical value of the maximum hanging point load of the rod in upstroke and downstroke for different target wells can be calculated.

[0089] It should be noted that, as other embodiments of the present application, the calculation of the theoretical value of the maximum hanging point load of the rod in upstroke and downstroke can be calculated by using any one of the disclosed models or algorithms, or by using improved or optimized models or algorithms, which are not specifically limited here.

[0090] Step S102, acquiring the actual value of the maximum surface load of the sucker rod of the target oil well in the field.

[0091] For the target oil well that needs to be diagnosed and predicted, the actual value of the maximum surface load of the sucker rod of the target oil well is collected in the field, which is used for subsequent matching.

[0092] It should be noted that the method of collecting the actual value of the maximum surface load of the sucker rod of the target oil well is not specifically limited here.

[0093] Step S103, introducing the pump working condition influence factor and the wellbore working condition influence factor into the theoretical model, and establishing the mapping relationship between the pump working condition influence factor and the wellbore working condition influence factor and the maximum surface load of the sucker rod.

[0094] The pump working condition influence factor and the wellbore working condition influence factor are introduced into the theoretical model, wherein the pump working condition influence factor corresponds to the pump working condition, and the wellbore working condition influence factor corresponds to the wellbore working condition.

[0095] Specifically, for the twelve forces involved in Table 1, the associated factors of the pump working condition influence factor include the resistance of the well fluid passing through the traveling valve , the friction between the oil pump plunger and the bushing , and the pump inlet pressure ; the associated factors of the wellbore working condition influence factor include the contact force of the tubing on the sucker rod , the equivalent friction of the well fluid on the sucker rod , and the friction between the well fluid and the tubing .

[0096] After the influence factors are integrated into the theoretical model of the sucker rod mechanics, the calculation formula of the microelement stress of the upstroke sucker rod is: ; the calculation formula of the microelement stress of the downstroke sucker rod is: ; the calculation method of the boundary condition after integrating the influence factors is: ; ; by integrating the influence factors into the calculation formula of the microelement stress of the upstroke and downstroke sucker rod, the mapping relationship between the influence factors and the maximum surface load of the sucker rod is established, that is, based on the acquired ground key parameters and the value of the influence factors, the theoretical value of the corresponding maximum surface load of the sucker rod can be calculated by the theoretical model.

[0097] Step S104, initializing the values of the pump working condition influence factor and the wellbore working condition influence factor, and taking the actual values to the

[0098] into the fault diagnosis prediction model, obtaining the pump working condition influence factor and the wellbore working condition influence factor matched with the actual values.

[0099] According to the mapping relationship between the influence factor and the maximum surface load of the sucker rod, the fault diagnosis prediction model can correspondingly determine the pump working condition influence factor and the wellbore working condition influence factor matched with the actual values according to the actual values of the maximum surface load of the sucker rod collected on site. The method process corresponding to step S104 specifically includes:

[0100] The pump working condition influence factor and the wellbore working condition influence factor are both assigned an initial value 1;

[0101] The actual values are input into the fault diagnosis prediction model;

[0102] The fault diagnosis prediction model adjusts the values of the pump working condition influence factor and the wellbore working condition influence factor, so that the maximum surface load of the sucker rod calculated by the adjusted pump working condition influence factor and the wellbore working condition influence factor matches the actual values;

[0103] The adjusted pump working condition influence factor and the wellbore working condition influence factor are output.

[0104] When the downhole working condition deviates from the design or the downhole environment deteriorates, the influence factor decreases. By adjusting the parameters to make the surface load consistent with the field working condition, and examining the deviation of the influence factor, it can be preliminarily judged what kind of abnormality occurs in the downhole.

[0105] On the basis of the above, as a preferred embodiment of the present application, in order to improve the accuracy of the calculation of the fault diagnosis prediction model, a BP neural network model is used for learning and training, and the specific training method includes:

[0106] The pump working condition influence factor and the wellbore working condition influence factor are respectively taken in the interval (0, 1] at certain intervals and combined to obtain a plurality of influence factor groups;

[0107] Each influence factor group is respectively taken into the theoretical model to calculate the maximum surface load value of the sucker rod corresponding to each influence factor group;

[0108] An influence factor group and its corresponding maximum surface load value of the sucker rod are combined to obtain a data group;

[0109] The multiple data sets obtained according to the multiple influence factor groups are divided into two parts of a test set and a verification set, and learning training is performed through a BP neural network model; wherein the maximum load values of the sucker rod upper and lower strokes are taken as the input of the training model, and the influence factor groups are taken as the output of the training model.

[0110] As a preferred embodiment, when constructing the test set and the verification set for training, the two influence factors are valued at intervals of 0.01 in the interval (0, 1], and are combined with each other in pairs to obtain ten thousand influence factor groups. Then the ten thousand influence factor groups are brought into the theoretical model respectively, and the maximum load values of the sucker rod upper and lower strokes corresponding to the ten thousand influence factor groups are calculated, and then are combined to obtain ten thousand data sets, and finally the ten thousand data sets are divided into two parts of a test set and a verification set, and a BP neural network model is selected for learning: the ten thousand data sets obtained from the target well are put into the BP neural network model for learning. Experimental data prove that the data trained by the BP neural network can obtain good precision, and the accuracy is as high as 99.9%, and the fault diagnosis prediction model obtained by training is very suitable for predicting and diagnosing the downhole working conditions of the target well.

[0111] Learning training using the BP neural network model can make the generated fault diagnosis prediction model more accurately estimate the influence factor group corresponding to the actual value by using a large number of data sets. According to the theoretical model introduced above, the influence factor with different values is substituted, and then the relevant parameter values collected on site are combined to calculate the corresponding maximum load values of the sucker rod upper and lower strokes. However, in actual fault prediction and diagnosis, the fault diagnosis prediction model needs to calculate and estimate an influence factor group with the highest possibility corresponding to the actual value of the maximum load of the sucker rod upper and lower strokes collected on site, as the output. Therefore, when learning and training, the larger the data set used, the higher the accuracy, and the more accurate the estimation of the fault diagnosis prediction model to the influence factor group.

[0112] In step S105, the matching result is compared with the initial value, and the abnormal situation downhole is judged through the deviation of the influence factor.

[0113] When the downhole working condition deviates from the design or the downhole environment deteriorates, the influence factor decreases. Therefore, based on the influence factor group output by the fault diagnosis prediction model, the deviation degrees of the pump working condition influence factor and the wellbore working condition influence factor output by the fault diagnosis prediction model from the initial value 1 are compared respectively; the higher the deviation degree of the influence factor from the initial value 1, the higher the probability that the corresponding reason leads to the working condition fault.

[0114] The pump working condition and the wellbore working condition corresponding to the two influence factors respectively can be judged simultaneously by comparison.

[0115] The downhole working condition fault diagnosis method is described below through a specific case:

[0116] Taking 16 fault wells in a block as target wells, the downhole working condition fault diagnosis method is applied respectively, and the specific process is as follows:

[0117] (1) First, a mechanical theoretical model of the pumping rod is established, and the theoretical values of the maximum suspension point load of the pumping rod in the upstroke and downstroke are calculated respectively, and the calculation results are shown in Figure 5 , Figure 6 .

[0118] (2) The theoretical values calculated in step (1) are compared with the actual values of the maximum suspension point load in the upstroke and downstroke collected on site, and the comparison results are shown in Figure 5 , Figure 6 . As can be seen from the figure, the theoretical value of the maximum suspension point load in the upstroke is generally lower than the calculated value of the maximum suspension point load in the upstroke, and the theoretical value of the maximum suspension point load in the downstroke is generally higher than the calculated value of the maximum suspension point load in the downstroke.

[0119] (3) The pump working condition influence factor and the wellbore working condition influence factor are introduced.

[0120] (4) The two influence factors in 3) are integrated into the existing mechanical model, and it is assumed that the theoretical calculation model is an ideal working condition model, and the initial parameters are given as 1. When the downhole working condition deviates from the design or the downhole environment deteriorates, the influence factor decreases. The influence factor group that matches the calculated values of the maximum suspension point load in the upstroke and downstroke with the actual values collected on site is obtained through the fault diagnosis prediction model, and the specific prediction results are shown in Figure 7 , Figure 8 . The deviation of the obtained influence factor group can preliminarily determine what kind of abnormality occurs in the downhole.

[0121] The judgment results for the 16 fault wells are shown in Table (2):

[0122] Table (2) Influence factor matching result table

[0123]

[0124] From the data in Table (2), it can be seen that the lower the number of the influence factor, the higher the probability of the corresponding reason causing the working condition fault. Generally, when the value of a certain influence factor is lower than 0.5, it means that the corresponding reason causes the working condition fault.

[0125] In order to further illustrate the downhole working condition fault diagnosis method provided by the present application, a specific fault well is taken as an example, and the specific application steps are as follows:

[0126] 1) Establish the mechanical model of sucker rod, calculate the theoretical value of the maximum load of the target well's upper and lower stroke: the maximum load of the target well's upper stroke is 55949; the maximum load of the target well's lower stroke is 39468.

[0127] 2) Compare the values obtained in step 1) with the actual values of the maximum load of the upper and lower stroke collected on site: according to the field test data, the maximum load of the target well's upper stroke is 60640; the maximum load of the target well's lower stroke is 36650.

[0128] 3) Introduce the pump working condition influence factor and the wellbore working condition influence factor .

[0129] 4) Put the two influence factors between 0 and 1 at intervals of 0.01 into the mechanical model for calculation, obtain ten thousand groups of data composed of influence factor groups and the maximum load of the upper and lower stroke. Divide the ten thousand data groups into test set and validation set two parts, select BP neural network model for learning.

[0130] 5) Put the target well's upper and lower stroke maximum load on-site value into the fault diagnosis prediction model for prediction, the wellbore working condition influence factor The prediction result is 0.51; the pump working condition influence factor The prediction result is 0.94. The result shows that the cause of the target well's downhole working condition failure is related to the wellbore.

[0131] 6) The target well's field work summary observes that the sucker rod has waxing, scaling and eccentric wear, which shows that the cause of the target well's downhole working condition failure is related to the wellbore condition, and further proves that the prediction and diagnosis method is effective.

[0132] In summary, the downhole working condition fault diagnosis method provided by the embodiment of the present application attributes the factors affecting the sucker rod broken-off to pump working condition influence factors and wellbore working condition influence factors, integrates the two influence factors into the existing mechanical model, assumes that the theoretical calculation model is an ideal working condition model, and when the downhole working condition deviates from the design or the downhole environment deteriorates, the influence factors decrease. The upstroke and downstroke maximum surface loads are adjusted to be consistent with the field working condition, and the deviation of the influence factors is examined, so that it can be preliminarily judged what kind of abnormality occurs in the downhole. The diagnosis method can diagnose the reasons for the sucker rod broken-off according to the ground key parameters, and can not only diagnose the pump abnormality, but also diagnose the wellbore friction abnormality. Meanwhile, the diagnosis model trained based on the BP neural network is combined for intelligent diagnosis, the influence factors reflecting the actual working condition can be better estimated, the diagnosis efficiency is high, the prediction accuracy is good, and the method has a very important reference significance for the field application of the rod pump oil production system. The method saves time and labor, has high diagnosis efficiency, and makes up for the deficiency of the current domestic diagnosis and prediction mainly relying on the dynamometer card.

[0133] As shown in Figure 9 The downhole working condition fault diagnosis device provided by the embodiment of the present application comprises:

[0134] A model establishing unit 110 is configured to establish a sucker rod mechanical theoretical model, and the theoretical model is configured to calculate a theoretical value of the upstroke and downstroke maximum surface loads of the sucker rod of a target oil well;

[0135] A data acquisition unit 120 is configured to acquire actual values of the upstroke and downstroke maximum surface loads of the sucker rod of the target oil well collected in the field;

[0136] A factor introducing unit 130 is configured to introduce pump working condition influence factors and wellbore working condition influence factors into the theoretical model, and establish a mapping relationship between the pump working condition influence factors and the wellbore working condition influence factors and the upstroke and downstroke maximum surface loads of the sucker rod;

[0137] A prediction calculation unit 140 is configured to initialize the values of the pump working condition influence factors and the wellbore working condition influence factors, and bring the actual values into a fault diagnosis prediction model to obtain pump working condition influence factors and wellbore working condition influence factors matched with the actual values;

[0138] A fault diagnosis unit 150 is configured to compare the matching results with the initialized values, and judge the abnormal conditions in the downhole through the deviation of the influence factors.

[0139] The downhole working condition fault diagnosis device provided by the embodiment of the present application is configured to realize the downhole working condition fault diagnosis method described above, and therefore the specific embodiments and the method are the same, and will not be described herein again.

[0140] As shown in Figure 10As shown, an embodiment of the present application provides a structural block diagram of an electronic device 300. The electronic device 300 can be a smart phone, a tablet computer, an electronic book, or the like, which can run an application program. The electronic device 300 in the present application can include one or more of the following components: a processor 310, a memory 320, and one or more application programs, wherein the one or more application programs can be stored in the memory 320 and configured to be executed by the one or more processors 310, and the one or more programs are configured to perform the method as described in the foregoing method embodiments.

[0141] The processor 310 can include one or more processing cores. The processor 310 connects various parts within the entire electronic device 300 by various interfaces and lines, performs various functions of the electronic device 300 and processes data by running or executing instructions, programs, code sets or instruction sets stored in the memory 320, and calling data stored in the memory 320. Optionally, the processor 310 can be implemented in at least one of a hardware form of a digital signal processing (DSP), a field-programmable gate array (FPGA), and a programmable logic array (PLA). The processor 310 can integrate a combination of one or several of a central processing unit (CPU), a graphics processing unit (GPU), and a modem. Among them, the CPU is mainly used to process an operating system, a user interface, and an application program, etc.; the GPU is used to be responsible for rendering and drawing of display content; and the modem is used to process wireless communication. It can be understood that the above-mentioned modem can also not be integrated into the processor 310, but be implemented by a separate communication chip.

[0142] The memory 320 can include a random access memory (RAM) and can also include a read-only memory (ROM). The memory 320 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 320 can include a program storage area and a data storage area, wherein the program storage area can store instructions for implementing an operating system, instructions for implementing at least one function (such as a touch function, a sound playing function, an image playing function, etc.), instructions for implementing each of the following method embodiments, etc. The data storage area can also store data created by the terminal in use (such as a phone book, audio and video data, chat record data), etc.

[0143] As Figure 11As shown, the embodiment of the present application provides a structural block diagram of a computer readable storage medium 400. The computer readable medium stores program code 410, which can be called by a processor to execute the method described in the above method embodiment.

[0144] The computer readable storage medium 400 can be an electronic memory such as a flash memory, an EEPROM (electrically erasable programmable read-only memory), an EPROM, a hard disk, or a ROM. Alternatively, the computer readable storage medium 400 includes a non-transitory computer readable medium. The computer readable storage medium 400 has a storage space for the program code 410 to execute any of the above methods. These program codes 410 can be read from or written to one or more computer program products. The program code 410 can be compressed in an appropriate form, for example.

[0145] To sum up, the present application provides a downhole working condition fault diagnosis method and device. Factors affecting the sucker rod breakage are attributed to pump working condition influence factors and wellbore working condition influence factors. These two influence factors are integrated into the existing mechanical model. The theoretical calculation model is assumed to be an ideal working condition model. When the downhole working condition deviates from the design or the downhole environment deteriorates, the influence factor decreases. By adjusting the parameters, the maximum surface load of the up and down strokes is consistent with the field working condition, and the deviation of the influence factor is examined, so that the specific abnormality occurring in the downhole can be preliminarily judged. The diagnosis method can diagnose the cause of the sucker rod breakage according to the key parameters on the ground. It can not only diagnose the abnormality of the pump, but also diagnose the abnormality of the wellbore friction. At the same time, it combines the diagnosis model trained by machine learning based on the BP neural network to intelligently diagnose, which can better estimate the influence factor reflecting the actual working condition, has the advantages of high diagnosis efficiency and good prediction accuracy, and has a very important reference significance for the field application of the rod pump oil production system. The method saves time and effort, has high diagnosis efficiency, and makes up for the deficiency of the current domestic prediction diagnosis mainly relying on the dynamometer card.

[0146] In several embodiments disclosed in the present application, it should be understood that the disclosed apparatus and method can also be implemented in other manners. The described apparatus embodiments are merely illustrative. For example, the flowchart and block diagram in the accompanying drawings show the possible implementation architecture, function and operation of the apparatus, method and computer program product according to the embodiments of the present application. In this regard, each block in the flowchart or block diagram can represent a module, a segment or a portion of code which comprises one or more executable instructions for implementing the specified logic function. It should also be noted that in some alternative implementations, the functions shown in the blocks can occur in a different order than that shown in the figure. For example, two blocks shown in succession can in fact be executed substantially concurrently or in the reverse order, depending on the functionality involved. It should also be noted that each block in the block diagrams and / or flowchart illustrations, and combinations of blocks in the block diagrams and / or flowchart illustrations, can be implemented by dedicated hardware-based systems which perform the specified functions or acts or can be implemented by a combination of dedicated hardware and computer instructions.

[0147] In addition, the various functional modules in the embodiments of the present application can be integrated together to form a separate part, or each module can exist independently, or two or more modules can be integrated to form a separate part.

[0148] If the functions are implemented in the form of software function modules and sold or used as an independent product, they can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application essentially or the part that makes a contribution to the prior art or the part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a number of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various other media that can store program codes.

Claims

1. A method of diagnosing a downhole workstring failure, comprising: The method comprises the following steps: a theoretical model of a sucker rod is established, and the theoretical model is used to calculate a theoretical value of a maximum hanging point load of the sucker rod of a target oil well in upstroke and downstroke; an actual value of the maximum hanging point load of the sucker rod of the target oil well in upstroke and downstroke is obtained through field collection; a pump working condition influence factor and a wellbore working condition influence factor are introduced into the theoretical model, and a mapping relationship between the pump working condition influence factor, the wellbore working condition influence factor and the maximum hanging point load of the sucker rod in upstroke and downstroke is established; values of the pump working condition influence factor and the wellbore working condition influence factor are initialized, the actual value is brought into a fault diagnosis prediction model, and the pump working condition influence factor and the wellbore working condition influence factor matched with the actual value are obtained; the matching result is compared with the initialized values, and an abnormal situation occurring in a downhole is judged according to a deviation of the influence factors; the step of calculating the theoretical value of the maximum hanging point load of the sucker rod of the target oil well in upstroke and downstroke specifically comprises the following steps: a micro-element method is used to segmentally calculate the sucker rod, and the theoretical value of the maximum hanging point load of the sucker rod in upstroke and downstroke is obtained; in the theoretical model, the pump working condition influence factor is related to a resistance generated by well fluid passing through a traveling valve, a friction force between a plunger of the oil pump and a bushing, and a pump port pressure; the wellbore working condition influence factor is related to a contact force of a tubing on the sucker rod, an equivalent friction drag of the well fluid on the sucker rod, and a friction force between the well fluid and the tubing; the step of introducing the pump working condition influence factor and the wellbore working condition influence factor into the theoretical model and establishing the mapping relationship between the pump working condition influence factor, the wellbore working condition influence factor and the maximum hanging point load of the sucker rod in upstroke and downstroke specifically comprises the following steps: The calculation formula of the micro-element stress of the upstroke sucker rod is: The calculation formula of the micro-element stress of the downstroke sucker rod is: Wherein, , are the axial forces of the micro-element of the upstroke, downstroke and stroke sucker rods respectively at node j, is the buoyancy of the unit length of the sucker rod, is the gravity of the unit length of the sucker rod, is the inertial force of the unit length of the sucker rod, is the frictional resistance of the unit length of the sucker rod to the well fluid, , are the forces of the upstroke and downstroke well fluid flow to the unit length of the sucker rod respectively, f is the friction coefficient of the tubing and the sucker rod, N is the support force of the tubing to the micro-element of the sucker rod, is the mass of the micro-element of the sucker rod, , are the accelerations of the micro-element of the upstroke and downstroke sucker rods respectively; The boundary condition calculation method is: ; ; ; wherein, is a pump operating factor, is a wellbore operating factor, is the force acting on the bottom end of the sucker rod during the upstroke, is the force acting on the bottom end of the sucker rod during the downstroke, is the liquid column load on the plunger, is the friction between the pump plunger and the liner, is the friction between the well fluid and the tubing, is the wellhead back pressure effect during the upstroke, is the liquid column inertial load, is the load on the plunger due to gas in the pump, is the resistance to flow through the traveling valve bore, is the wellhead back pressure effect.

2. The downhole working condition fault diagnostic method of claim 1, wherein, the step of initializing the values of the pump working condition influence factor and the wellbore working condition influence factor, bringing the actual value into the fault diagnosis prediction model, and obtaining the pump working condition influence factor and the wellbore working condition influence factor matched with the actual value specifically comprises the following steps: the pump working condition influence factor and the wellbore working condition influence factor are both assigned an initial value 1; the actual value is input into the fault diagnosis prediction model; the fault diagnosis prediction model adjusts the values of the pump working condition influence factor and the wellbore working condition influence factor, so that the maximum hanging point load of the sucker rod in upstroke and downstroke calculated by the adjusted pump working condition influence factor and the wellbore working condition influence factor matches the actual value; the adjusted pump working condition influence factor and the wellbore working condition influence factor are output.

3. The downhole working condition fault diagnostic method of claim 2, wherein, the training method of the fault diagnosis prediction model comprises the following steps: the pump working condition influence factor and the wellbore working condition influence factor are respectively taken in a certain interval in a (0, 1] interval and combined to obtain a plurality of influence factor groups; each influence factor group is respectively brought into the theoretical model, and a maximum hanging point load value of the sucker rod in upstroke and downstroke corresponding to each influence factor group is calculated; one influence factor group and the corresponding maximum hanging point load value of the sucker rod in upstroke and downstroke are combined to obtain one data group; The multiple data groups obtained according to the multiple influence factor groups are divided into a test set and a verification set, and learning training is performed through a BP neural network model; wherein the maximum polished rod load values of the upstroke and downstroke of the sucker rod are taken as the input of the training model, and the influence factor groups are taken as the output of the training model.

4. The downhole working condition fault diagnostic method of claim 3, wherein, The matching result is compared with the initial value, and the deviation of the influence factor is used to determine the abnormal situation in the well. The deviation of the pump working condition influence factor and the wellbore working condition influence factor output by the fault diagnosis prediction model from the initial value 1 is compared respectively. The higher the deviation of the influence factor from the initial value 1, the higher the probability of the corresponding reason causing the working condition fault.

5. The downhole working condition fault diagnostic method of claim 4, wherein, It comprises: A model establishment unit is configured to establish a theoretical model of the sucker rod mechanics, which is used to calculate the theoretical value of the maximum polished rod load of the upstroke and downstroke of the sucker rod of the target oil well; A data acquisition unit is configured to acquire the actual value of the maximum polished rod load of the upstroke and downstroke of the sucker rod of the target oil well collected in the field; A factor introduction unit is configured to introduce the pump working condition influence factor and the wellbore working condition influence factor into the theoretical model, and establish the mapping relationship between the pump working condition influence factor and the wellbore working condition influence factor and the maximum polished rod load of the upstroke and downstroke of the sucker rod; A prediction calculation unit is configured to initialize the value of the pump working condition influence factor and the wellbore working condition influence factor, and input the actual value into the fault diagnosis prediction model to obtain the pump working condition influence factor and the wellbore working condition influence factor matched with the actual value; A fault diagnosis unit is configured to compare the matching result with the initial value, and determine the abnormal situation in the well through the deviation of the influence factor. It comprises:

6. An electronic device, comprising: One or more processors; Memory; One or more application programs, wherein the one or more application programs are stored in the memory and configured to be executed by the one or more processors, and the one or more programs are configured to execute the method of any one of claims 1-5. The computer readable storage medium stores program code, and the program code can be called and executed by the processor to execute the method of any one of claims 1-5.

7. A computer readable storage medium characterized in that, ​

Images