CN113266318B - Self-learning-based stroke frequency adjusting method for rod-type pumping unit well - Google Patents
Self-learning-based stroke frequency adjusting method for rod-type pumping unit well Download PDFInfo
- Publication number
- CN113266318B CN113266318B CN202110588435.2A CN202110588435A CN113266318B CN 113266318 B CN113266318 B CN 113266318B CN 202110588435 A CN202110588435 A CN 202110588435A CN 113266318 B CN113266318 B CN 113266318B
- Authority
- CN
- China
- Prior art keywords
- stroke
- pumping unit
- state
- frequency
- eta
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000005086 pumping Methods 0.000 title claims abstract description 106
- 238000000034 method Methods 0.000 title claims abstract description 56
- 238000004519 manufacturing process Methods 0.000 claims abstract description 55
- 239000003129 oil well Substances 0.000 claims abstract description 53
- 230000000694 effects Effects 0.000 claims abstract description 23
- 238000010586 diagram Methods 0.000 claims description 38
- HWKQNAWCHQMZHK-UHFFFAOYSA-N trolnitrate Chemical compound [O-][N+](=O)OCCN(CCO[N+]([O-])=O)CCO[N+]([O-])=O HWKQNAWCHQMZHK-UHFFFAOYSA-N 0.000 claims description 31
- 229960002485 trolnitrate Drugs 0.000 claims description 31
- 239000012530 fluid Substances 0.000 claims description 27
- 239000007788 liquid Substances 0.000 claims description 18
- 238000004364 calculation method Methods 0.000 claims description 8
- 238000007405 data analysis Methods 0.000 claims description 5
- 239000003795 chemical substances by application Substances 0.000 claims description 3
- 238000011084 recovery Methods 0.000 claims description 3
- 238000010276 construction Methods 0.000 abstract description 3
- 239000003921 oil Substances 0.000 description 27
- 238000005457 optimization Methods 0.000 description 6
- 230000001276 controlling effect Effects 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 4
- 238000011161 development Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- 230000001105 regulatory effect Effects 0.000 description 2
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 229910052799 carbon Inorganic materials 0.000 description 1
- 239000010779 crude oil Substances 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 238000004134 energy conservation Methods 0.000 description 1
- 238000005265 energy consumption Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000006386 neutralization reaction Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Images
Classifications
-
- E—FIXED CONSTRUCTIONS
- E21—EARTH OR ROCK DRILLING; MINING
- E21B—EARTH OR ROCK DRILLING; OBTAINING OIL, GAS, WATER, SOLUBLE OR MELTABLE MATERIALS OR A SLURRY OF MINERALS FROM WELLS
- E21B43/00—Methods or apparatus for obtaining oil, gas, water, soluble or meltable materials or a slurry of minerals from wells
- E21B43/12—Methods or apparatus for controlling the flow of the obtained fluid to or in wells
- E21B43/121—Lifting well fluids
- E21B43/126—Adaptations of down-hole pump systems powered by drives outside the borehole, e.g. by a rotary or oscillating drive
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F04—POSITIVE - DISPLACEMENT MACHINES FOR LIQUIDS; PUMPS FOR LIQUIDS OR ELASTIC FLUIDS
- F04B—POSITIVE-DISPLACEMENT MACHINES FOR LIQUIDS; PUMPS
- F04B49/00—Control, e.g. of pump delivery, or pump pressure of, or safety measures for, machines, pumps, or pumping installations, not otherwise provided for, or of interest apart from, groups F04B1/00 - F04B47/00
- F04B49/06—Control using electricity
- F04B49/065—Control using electricity and making use of computers
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P80/00—Climate change mitigation technologies for sector-wide applications
- Y02P80/10—Efficient use of energy, e.g. using compressed air or pressurized fluid as energy carrier
Landscapes
- Engineering & Computer Science (AREA)
- Geology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mining & Mineral Resources (AREA)
- Environmental & Geological Engineering (AREA)
- Fluid Mechanics (AREA)
- Physics & Mathematics (AREA)
- General Life Sciences & Earth Sciences (AREA)
- Geochemistry & Mineralogy (AREA)
- Computer Hardware Design (AREA)
- Mechanical Engineering (AREA)
- General Engineering & Computer Science (AREA)
- Control Of Positive-Displacement Pumps (AREA)
Abstract
A self-learning-based stroke frequency adjusting method for a rod pumping unit well belongs to the field of intelligent oil well collecting methods, and is characterized by comprising the following steps: firstly, extracting historical data of oil well production, and judging the current oil pumping unit state of an oil well; the pumping unit state comprises an adjusting state and a stable state; the adjustment state is as follows: when the oil pumping machine controller makes adjustment according to the new control parameters, the oil pumping machine enters an adjustment state; the stable state is: after the pumping unit operates for a period of time with new control parameters, the operating state of the pumping unit keeps stable and is switched into a stable state; then establishing a relation between the motor input frequency and the stroke frequency of the pumping unit in a stable state, establishing a model between the effect and the stroke frequency in the oil well production process, calculating the optimal stroke frequency, and completing the stroke frequency adjustment of the rod pumping unit well; the oil well cost reduction and efficiency improvement can be realized, and a technical basis is provided for realizing the construction of a green oil field.
Description
Technical Field
The invention belongs to the field of intelligent oil well acquisition methods, and particularly relates to a self-learning-based method for adjusting the stroke frequency of a rod-type pumping unit.
Background
In recent years, with the national carbon neutralization policy requirements, energy conservation and consumption reduction of a sucker rod pumping system become important tasks for oil field production. Based on the digital oil field construction of the oil field site and the development of the domestic rod oil extraction technology, a large number of rod oil extraction system stroke frequency optimization technologies are provided: with the development of the internet of things technology and the edge computing technology, the data acquisition technology of an oil well indicator diagram and an electric diagram (three electric parameters, a power diagram, an electric diagram and the like) is basically mature, the computing algorithms of the oil well indicator diagram, the production capacity, the fullness, the system efficiency, the indicator diagram area and the like are preliminarily improved, and a foundation is laid for realizing intelligent control of an oil well.
With the change of international oil prices and the change of the actual production complexity of oil fields, the single liquid production amount and the power consumption are not enough to meet the production requirements of the oil fields. Meanwhile, as the control requirement of the oil well is improved, the relevant control theory is urgently needed to be combined with the actual working condition of oil well production. In the production process of an oil well, the change of the formation liquid supply capacity is a steady-state change process, the change of the page in a shaft is slow, and the like; meanwhile, aiming at different crude oil physical characteristics, the suction working conditions and the discharge working conditions of the sucker-rod oil pump are not completely the same under the same sinking degree, and oil well control needs to be realized by combining the working conditions of the oil well instead of simple regulation and control; at present, in the regulation and control process aiming at the oil well, the regulation and control parameters are mostly obtained according to experience or manually regulated according to on-site feedback, the labor cost is high, the efficiency is low, and the digitization level is low.
Disclosure of Invention
The invention aims to solve the problems and provides a self-learning-based method for adjusting the stroke frequency of a rod-type pumping well.
The invention relates to a self-learning-based method for adjusting the stroke frequency of a rod-type pumping well, which comprises the following steps: firstly, extracting historical data of oil well production, and judging the current oil pumping unit state of an oil well; the pumping unit state comprises an adjusting state and a stable state; the adjustment state is as follows: when the oil pumping machine controller makes adjustment according to the new control parameters, the oil pumping machine enters an adjustment state, and the working state of the oil pumping machine is changed during the adjustment state; the stable state is: after the pumping unit operates for a period of time according to the new control parameters, the operating state of the pumping unit is kept stable, and the operating state of the pumping unit is switched into a stable state;
then, adjusting the pumping unit in a stable state for the first time, and establishing the relationship between the input frequency F and the stroke number N of the motor; n ═ d × F + e; (1)
wherein N is the number of strokes, F is the frequency, d, e are constants.
Establishing a model between an effect E and a stroke number N in an oil well production process through self-learning to calculate the value of E:
E＝Q fluid production volume *W Q +Y*W Y ； (2)
In the formula, W is weight, E is effect, Q is liquid production amount, Y is pump efficiency (short for oil pump efficiency), and the maximum value of E is taken as an adjusting target; and calculating the optimal stroke number N to complete the stroke number regulation of the rod pumping well.
The invention relates to a self-learning-based method for adjusting the stroke frequency of a rod-type pumping well, which specifically comprises the following steps: setting the minimum filling degree eta min of the oil pumping unit; calculating the number of times of stroke Nmin which should be reached when the minimum fullness is met according to the minimum fullness eta min; after the pumping unit is started and operates to enter a stable state, calculating the stroke frequency N of the current pumping unit through real-time obtained indicator diagram data; acquiring the current motor frequency F through a frequency converter; and when N is equal to Nmin, judging that the pumping unit stops, sending a stop instruction, and waiting for production recovery.
Calculating the current approximate fullness eta of an oil well pump of the oil pumping unit; when N is greater than Nmin and eta is less than eta min, adjusting the frequency F of the motor to adjust the number of times of stroke N; entering an adjusting stage, and repeatedly adjusting for a plurality of times to enable the fullness eta to reach the minimum fullness eta min; continuing to adjust, and starting a self-learning process when eta is greater than eta min and N is greater than Nmin; after the pumping unit is started and operates to enter a stable state, simultaneously starting oil well data analysis, and analyzing the fullness eta, the liquid production quantity Q, the working fluid level H and the pump efficiency Y of the oil well under each stroke condition in the stable state; when the variation amplitude of the working fluid level H and the fullness eta in a certain period is smaller than the minimum limit value, judging that the oil well enters a new stable state;
and after the oil well enters a new stable state, entering the learning process again under the conditions that eta is greater than eta min and Nmin is less than N and less than Nmax, and counting the value of E under different stroke times according to a formula (2), so as to analyze the optimal stroke time when the value of E is maximum, namely completing the stroke time adjustment of the rod-type oil pumping well.
Further, the invention relates to a self-learning-based method for adjusting the stroke frequency of a rod-pumped well, wherein the self-learning process comprises the following steps: judging the adjusting times of the early-stage stroke adjusting stage; when the adjusting times are more than or equal to 1, the current stroke time is the maximum stroke time meeting eta > eta min, and the stroke time is reduced by controlling a frequency converter on the basis of the current stroke time; when the adjusting frequency is less than 1, increasing the stroke frequency by controlling the frequency converter on the basis of the current stroke frequency; and then judging whether the number of times of stroke needs to be reduced to continue adjustment according to the subsequent adjustment effect.
Further, the invention relates to a self-learning-based method for adjusting the stroke frequency of a rod-pumped well, which comprises the following specific steps of: calculating the value E according to formula (2); inserting the value E into a data queue, selecting the maximum value in the data queue after judging that the number of the data meets the minimum number, and judging whether the maximum value E is in the middle position of effective data; if so, returning the stroke value corresponding to the maximum value; if not, judging whether the mobile terminal is at the initial position; if yes, carrying out reverse adjustment to reduce the stroke frequency; if not, the positive adjustment is carried out to increase the stroke frequency.
Further, the method for adjusting the stroke frequency of the rod-pumped well based on self-learning comprises the following specific steps of: outputting the stroke data N according to the formula (1) and judging whether the calculation is the first calculation; if yes, directly adopting a preset relation parameter d for obtaining the frequency and the stroke frequency 0 And e 0 (ii) a If not, the relation parameter d is recalculated according to the data of the last two times i And e i And updating the preset parameter d 0 And e 0 (ii) a And then calculating the corresponding motor frequency F according to the formula (1).
Further, the self-learning-based stroke frequency adjusting method for the rod-pumped well comprises the following steps of: inserting the number of times of stroke value N into a historical data queue; calculating the average value of n times of stroke data in the historical data queue, calculating the difference value between the current data and the average value, and checking whether the difference value is within the allowable range of the equilibrium state; if so, the pumping unit reaches a stable state; if not, the pumping unit does not reach a stable state.
Further, the self-learning-based stroke frequency adjusting method for the rod-pumped well comprises the following steps: inserting the fullness value eta into a historical data queue; calculating the average value of n fullness data in the historical data queue, calculating the difference value between the current data and the average value, and checking whether the difference value is within the allowable range of the equilibrium state; if so, the oil well reaches a stable state; if not, the well does not reach a steady state.
Further, the self-learning-based stroke frequency adjusting method for the rod-pumped well comprises the following steps of: calculating each indicator diagram period by acquiring data of the pumping unit in real time; when the difference value between each period Ti and the average period T of the n periods is stabilized in a fixed range in the continuous n indicator diagram periods, the working state of the pumping unit enters a stable state;
the steady state is judged by the following index conditions:
1) stable condition of liquid production
Wherein: q AV The average value of the liquid production amount of the indicator diagram is shown; Δ X is the amount of change in fluid production; q i The fluid production amount in the ith period;
2) stable condition of work diagram area
Wherein: s AG Is the average value of the area of the ground indicator diagram, Deltay is the variation of the area of the ground indicator diagram, S i The area of the ground indicator diagram in the ith period;
and simultaneously, when the conditions 1) and 2) are met, the working state of the pumping unit is judged to enter a stable state.
Furthermore, the invention relates to a self-learning-based method for adjusting the stroke frequency of a rod-pumped well, which is characterized in that: after the working state of the pumping unit enters a stable state, the calculation formulas of the stable liquid production amount and the stable pump efficiency are as follows:
steady state fluid production:
steady state pump efficiency:
wherein Q is cv A steady state fluid production; y is cv A steady state pumping effect; n is a number.
The invention discloses a self-learning-based stroke frequency adjusting method for a rod-type pumping unit well, which has the technical effects that: (1) combining with the oil field production process, providing a set of parameter optimization method based on self-learning for oil field production, and realizing one-well one-strategy; (2) the effect after the automatic evaluation, verification and adjustment in the optimization process of the oil well parameters is realized, and the reliability is higher; (3) selecting the optimal value of the pulse times by combining historical data analysis, realizing optimization and a multi-objective optimization strategy, and being suitable for different production requirements of oil fields; (4) the oil well cost reduction and efficiency improvement can be realized, and a technical basis is provided for realizing the construction of a green oil field.
Drawings
FIG. 1 is a schematic flow chart of a self-learning based method for adjusting the stroke frequency of a sucker rod pumped well according to the present invention;
FIG. 2 is a flow chart of input data inspection and processing according to an embodiment of the present invention;
FIG. 3 is a flow chart of steady state determination for an oil pumping unit according to an embodiment of the present invention;
FIG. 4 is a flow chart of steady state determination of an oil well according to an embodiment of the present invention;
FIG. 5 is a schematic view of a process of analyzing a stroke number optimization target according to an embodiment of the present invention;
fig. 6 is a flowchart illustrating the relationship between the impulse and the frequency in the embodiment of the present invention.
Detailed Description
The self-learning based method for adjusting the stroke frequency of a rod-pumped well according to the present invention will be described in detail with reference to the accompanying drawings and examples.
The technical scheme of the invention mainly researches the influence of different factors on the oil well liquid production rate, and then obtains a calibration coefficient table of the oil well liquid production rate under the influence of each factor. And each indicator diagram collected by the oil well can obtain the indicator diagram after the diagnosis and analysis.
The stroke frequency adjusting method of the rod-pumped well based on self-learning in the embodiment of the disclosure defines two working states of the pumping unit: a regulated state and a stable state. Adjusting the state: when the controller of the oil pumping unit makes adjustment according to the new control parameters, the oil pumping unit enters an adjustment state, and the working state is changed during the adjustment state. And (3) stable state: after the pumping unit operates for a period of time with new control parameters, the operation state of the pumping unit is kept stable, and the working state is switched to a stable state.
As shown in fig. 1, the method specifically comprises the following steps: firstly, data inspection is carried out; then judging the steady state of the oil well; setting the minimum filling degree eta min of the oil pumping unit; calculating the number of times of impact Nmin which should be reached when the minimum fullness is met according to the minimum fullness eta min; after the pumping unit is started and operates to enter a stable state, calculating the stroke frequency N of the current pumping unit through real-time obtained indicator diagram data; acquiring the current motor frequency F through a frequency converter; and when N is equal to Nmin, judging that the pumping unit stops, sending a stop instruction, and waiting for production recovery.
Calculating the current approximate fullness eta of the oil well pump; when N is greater than Nmin and eta is less than eta min, adjusting the frequency F of the motor to adjust the number of times of stroke N; entering an adjusting stage, and repeatedly adjusting for a plurality of times to enable the fullness eta to reach the minimum fullness eta min; continuing to adjust, and starting a self-learning process when eta is greater than eta min and N is greater than Nmin; after the pumping unit is started and operates to enter a stable state, simultaneously starting oil well data analysis, and analyzing the fullness eta, the liquid production quantity Q, the working fluid level H and the pump efficiency Y of the oil well under each stroke condition in the stable state; and when the variation amplitude of the working fluid level H and the fullness eta in a certain period is smaller than the minimum limit value, judging that the oil well enters a new stable state.
And after the oil well enters a new stable state, entering the learning process again under the conditions that eta is greater than eta min and Nmin is less than N and less than Nmax, and counting the value of E under different stroke times according to a formula (2), so as to analyze the optimal stroke time when the value of E is maximum, namely completing the stroke time adjustment of the rod-type oil pumping well.
In the embodiment of the present disclosure, as shown in fig. 2, the process of data checking is to check the validity of the input data, and check whether the data is within a set reasonable range; whether it is illegal data; after the data validity is checked, checking the validity of the data, removing invalid data with larger deviation according to comparison with historical data, and storing the valid data for subsequent calculation operation; as shown in fig. 5, the specific steps of analyzing the optimal E value are as follows: calculating the value E according to formula (2); inserting the value E into a data queue, selecting the maximum value in the data queue after judging that the number of the data meets the minimum number, and judging whether the maximum value E is in the middle position of effective data; if so, returning the stroke value corresponding to the maximum value; if not, judging whether the mobile terminal is at the initial position; if yes, carrying out reverse adjustment to reduce the stroke frequency; if not, the positive adjustment is carried out to increase the stroke frequency.
As shown in fig. 3, the process of determining the steady state of the pumping unit includes: inserting the number of times of stroke value N into a historical data queue; calculating the average value of n times of stroke data in the historical data queue, calculating the difference value between the current data and the average value, and checking whether the difference value is within the allowable range of the equilibrium state; if so, the pumping unit reaches a stable state; if not, the pumping unit does not reach a stable state.
As shown in fig. 4, the process of determining that the oil well enters the steady state includes: inserting the fullness value eta into a historical data queue; calculating the average value of n fullness data in the historical data queue, calculating the difference value of the current data and the average value, and checking whether the difference value is within the allowable range of the equilibrium state; if so, the oil well reaches a stable state; if not, the well does not reach a steady state.
When the pumping unit is started, the operation is started by using default control parameters, and the initial starting adjustment state is entered. And after waiting for the pumping unit to operate for a period of time, judging whether the pumping unit enters a stable state or not. In the embodiment of the present disclosure, the determination condition of the steady state is to calculate each indicator diagram period through data collected in real time. When there are n consecutive cycles of the diagram, and the difference between each cycle Ti and the average cycle T of the n cycles is within a certain range, the pumping unit can be considered to be in a stable state.
When the pumping unit enters a stable state, the stroke frequency N of the current pumping unit is calculated through the real-time acquired indicator diagram data 0 Obtaining the current motor frequency F through a frequency converter 0 (ii) a According to the calculation method in the prior art patent 'stroke frequency adjustment', the current approximate charge of the oil well pump is calculatedFull eta 0 When η 0 <Eta min and N 0 >Nmin, the number of strokes needs to be reduced, which is achieved by reducing the frequency F of the motor.
When Δ N is decreased, Δ N is 1, and the linear relationship between the impulse and the frequency is determined by parameters d and e, as can be seen from the equation of N-d × F + e. As shown in fig. 6, the specific steps of calculating the frequency F by the number of pulses N are as follows: outputting stroke frequency data N according to a formula (1) and judging whether the first calculation is carried out; if yes, directly adopting a preset relation parameter d for obtaining the frequency and the stroke frequency 0 And e 0 (ii) a If not, the relation parameter d is recalculated according to the data of the last two times i And e i And updating the preset parameter d 0 And e 0 (ii) a Then, the corresponding motor frequency F is calculated according to the formula (1).
In the embodiment of the disclosure, a preset d is configured according to experience before the controller of the pumping unit is started 0 And e 0 Parameter, therefore, from N to d 0 *F+e 0 The frequency F1 to be adjusted can be calculated, and the frequency amplitude Δ F to be adjusted is F 0 -F 1 . Due to the setting of d 0 And e 0 Different from the actual situation, therefore, when adjusting for the first time, to
As the amplitude of the adjustment, set the frequency to
When is expressed as F 1 After the frequency is operated to a stable state, the actual N is calculated 1 And degree of fullness η 1 Then through F 0 、N 0 And F 1 、N 1 The actual d1 and e1 parameters of the current pumping unit system are determined.
Judgment of eta 1 And eta min, N 1 Relation to Nmin, if still requiring readjustment, using d 1 And e 1 The parameters are used as the linear relation of frequency and impulse frequency to calculate N 2 ＝N 0 Frequency F to be set at- Δ N 2 Again with adjustment of the motor by F2Frequency; after the adjustment state is finished to be operated to the stable state, the fullness eta is calculated again through the indicator diagram data 2 And current number of strokes N 2 Judgment of eta 2 And eta min, N 2 In relation to Nmin, a determination is made as to whether further adjustments are made.
If adjustments are still needed, the filling level is adjusted according to the relationship eta of the stroke and the filling degree 0 、η 0 ，N 1 、η 1 And N 2 、η 2 And determining parameters a and c.
And calculating the number of strokes N which should be adjusted according to the minimum fullness eta min, checking the relation between N and Nmin, stopping the pumping unit when N is less than Nmin, sending a stop instruction, and waiting for recovering production. Otherwise, adjusting the motor frequency to reach eta min according to the calculated N and F.
When the stroke frequency adjusting effect reaches eta > eta min and N > Nmin, the automatic learning process is started, so that the oil pumping unit system reaches the optimal state.
The self-learning process is started on the premise that eta is greater than eta min and N is greater than Nmin; the latter self-learning process must satisfy this condition or the self-learning process will be exited.
After the self-learning process is started, the adjusting times of the early-stage stroke adjusting stage are judged, because the stroke adjusting is always adjusted from high to low, if the stroke adjusting times are more than or equal to 1, the condition that eta is greater than eta min can be met only when the low stroke adjusting is performed, and the optimal efficiency and yield are analyzed by reducing the stroke under the condition that N is greater than Nmin. If the stroke frequency adjusting times is less than 1, the condition that eta is greater than eta min is met when the pumping unit starts to work, and the optimal yield and efficiency are analyzed by respectively increasing the stroke frequency and reducing the stroke frequency under the condition that N is greater than Nmin.
When the adjusting times are more than or equal to 1, the current stroke time is the maximum stroke time meeting eta > eta min, and the stroke time is reduced by controlling a frequency converter on the basis of the current stroke time; and when the adjusting times are less than 1, increasing the number of times of stroke by delta N by controlling the frequency converter on the basis of the current number of times of stroke. And judging whether the number of times of stroke needs to be reduced to continuously search for the optimal solution according to the subsequent adjustment effect.
After the frequency converter is controlled to change the stroke frequency of the oil well, the pumping unit is waited to run to a stable state, then the oil well data analysis stage is started, and the filling degree eta i, the liquid production amount Qi, the working fluid level data Hi and the pump efficiency Y of the oil well working to enter the stable state under each stroke frequency condition are analyzed i . And when the variation amplitude of the working fluid level and the fullness in a certain period is smaller than the minimum limit value, the oil well enters a new balance state.
When entering a new equilibrium state, pair Q Fluid production volume And normalizing the data with Y according to the formula of Q Fluid production *W Q +Y*W η The value of E is calculated.
And after the new balance is established, entering the learning process again under the conditions that eta is greater than eta min and Nmin is less than N < Nmax, and counting the value of E under different stroke times so as to analyze the optimal stroke time when the value of E is maximum.
Example two
On the basis of the first embodiment, the working condition parameter table of a pumping well is taken as an example in the embodiment of the disclosure: (1) a CYJ10-3-37HB type pumping unit is selected, the corresponding stroke is 3.0m, the current stroke frequency is 2.27 times/min, and the data of 6 work diagrams produced continuously are shown in the following table 1:
TABLE 1
| Stroke-type | Number of strokes | Pump efficiency | Fluid production volume | Area of ground diagram | |
| Work diagram one | 3 | 2.27 | 16 | 0.8 | 25821.26 |
| Second diagram | 3 | 2.27 | 16 | 0.8 | 25409.21 |
| Third diagram | 3 | 2.27 | 14 | 0.71 | 28869.28 |
| Diagram four | 3 | 2.27 | 16 | 0.78 | 28692.85 |
| Five diagram | 3 | 2.27 | 14 | 0.73 | 30518.21 |
| Six power diagrams | 3 | 2.27 | 14 | 0.7 | 28378.98 |
(2) And (3) steady state judgment: the variation of the liquid production amount is not more than 20%, and the variation of the ground work diagram area is not more than 20%.
Q AV ＝0.75，Max(Q i )＝0.8，Min(Q i )＝0.7
S AG ＝27948.30，Max(S i )＝30518.21，Min(S i )＝25409.21
Substituting into the steady state decision condition formula as follows:
0.75-0.8/0.75 |, 6.67% < 20% and 0.75-0.7|/0.75 |, 6.67% < 20%
-27948.30-30518.21/27948.30 | -9.20% < 20% and
|27948.30-25409.21|/27948.30＝9.08％<20％
the formula is satisfied, so the pumping unit has reached a steady state, at which time:
Q cv ＝0.75，Y cv ＝15％。
(3) the weights for setting yield and efficiency each account for 50%;
effect E0.75 × 50% +0.15 × 50% + 0.45
The data are recorded as table 2 below:
TABLE 2
| Stroke-type | Number of strokes | Steady state pumping efficiency | Steady state fluid production | Effect E | |
| Scheme one | 3 | 2.3 | 15 | 0.75 | 0.45 |
(4) Controlling a frequency converter and adjusting the frequency of impact; the increased stroke frequency can be adopted for the first time, the effect E is calculated according to the above process when the stable state is reached, and the data records are as follows in the following table 3:
TABLE 3
| Stroke-type | Number of strokes | Steady state pumping efficiency | Steady state fluid production | Effect E | |
| Scheme two | 3 | 2.5 | 16 | 0.82 | 0.49 |
Comparing and finding that the E value of the effect is increased after adjusting the stroke frequency, continuing to adjust the stroke frequency in the positive direction, and obtaining the data when reaching the stable state as the following table 4:
TABLE 4
| Stroke-type | Number of strokes | Steady state pumping efficiency | Steady state fluid production | Effect E | |
| Scheme three | 3 | 2.8 | 18 | 0.86 | 0.52 |
The data for steady state with increasing stroke frequency with continued forward adjustment are shown in table 5 below:
TABLE 5
| Stroke-type | Number of strokes | Steady state pumping efficiency | Steady state fluid production | Effect E | |
| Scheme four | 3 | 3.0 | 17 | 0.83 | 0.50 |
At this point the value of E is reduced and the inverse adjustment reduces the number of strokes and the data for reaching steady state is as follows:
TABLE 6
| Stroke-type | Number of strokes | Steady state pumping efficiency | Steady state fluid production | Effect E | |
| Scheme five | 3 | 2.7 | 18.5 | 0.88 | 0.5325 |
(5) The data queue pairs for different burst conditions are as follows:
TABLE 7
| Stroke-type | Number of strokes | Steady state pumping efficiency | Steady state fluid production | Effect E | |
| Scheme one | 3 | 2.3 | 15 | 0.75 | 0.45 |
| Scheme two | 3 | 2.5 | 16 | 0.82 | 0.49 |
| Scheme three | 3 | 2.7 | 18.5 | 0.88 | 0.5325 |
| Scheme four | 3 | 2.8 | 18 | 0.86 | 0.52 |
| Scheme five | 3 | 3.0 | 17 | 0.83 | 0.50 |
According to the effect E, the value E of the third scheme is the maximum, the corresponding stroke frequency 2.7 is the optimal value, and the pumping efficiency and the liquid production amount of the oil well reach the maximum value under the stroke frequency.
Claims (8)
1. A self-learning-based stroke frequency adjusting method for a rod-pumped well is characterized by comprising the following steps: firstly, extracting historical data of oil well production, and judging the current oil pumping unit state of an oil well; the pumping unit state comprises an adjusting state and a stable state;
the adjustment state is as follows: when the oil pumping machine controller makes adjustment according to the new control parameters, the oil pumping machine enters an adjustment state, and the working state of the oil pumping machine is changed during the adjustment state;
the stable state is: after the pumping unit operates for a period of time according to the new control parameters, the operating state of the pumping unit is kept stable, and the operating state of the pumping unit is switched into a stable state;
then, adjusting the pumping unit in a stable state for the first time, and establishing the relationship between the input frequency F and the stroke number N of the motor; n ═ d × F + e; (1)
in the formula, N is the stroke frequency, F is the frequency, and d and e are constants;
establishing a model between an effect E and a stroke number N in an oil well production process through self-learning to calculate the value of E:
E＝Q fluid production *W Q +Y*W Y ； (2)
Wherein W is weight and E is effect; q is the liquid production amount; y is the pump efficiency; taking the maximum E as an adjusting target; calculating the optimal stroke number N to complete the stroke number regulation of the rod-type pumping well;
the method specifically comprises the following steps: setting the minimum filling degree eta min of the oil pumping unit;
calculating the number of times of stroke Nmin which should be reached when the minimum fullness is met according to the minimum fullness eta min;
after the pumping unit is started and operates to enter a stable state, calculating the stroke frequency N of the current pumping unit through real-time obtained indicator diagram data; acquiring the current motor frequency F through a frequency converter;
when N is equal to Nmin, judging that the pumping unit stops, sending a stop instruction, and waiting for production recovery;
calculating the current approximate fullness eta of an oil well pump of the oil pumping unit;
when N is greater than Nmin and eta is less than eta min, adjusting the frequency F of the motor to adjust the number of times of stroke N;
entering an adjusting stage, and repeatedly adjusting for a plurality of times to enable the fullness eta to reach the minimum fullness eta min;
continuing to adjust, and starting a self-learning process when eta is greater than eta min and N is greater than Nmin;
after the pumping unit is started and operates to enter a stable state, simultaneously starting oil well data analysis, and analyzing the fullness eta, the liquid production quantity Q, the working fluid level H and the pump efficiency Y of the oil well under each stroke condition in the stable state; when the variation amplitude of the working fluid level H and the fullness eta in a certain period is smaller than the minimum limit value, judging that the oil well enters a new stable state;
and after the oil well enters a new stable state, entering the learning process again under the conditions that eta is greater than eta min and Nmin is less than N and less than Nmax, and counting the value of E under different stroke times according to a formula (2), so as to analyze the optimal stroke time when the value of E is maximum, namely completing the stroke time adjustment of the rod-type oil pumping well.
2. The self-learning based rod-pumped well stroke rate adjustment method of claim 1, wherein the self-learning process comprises: judging the adjusting times of the early-stage stroke adjusting stage; when the adjusting times are more than or equal to 1, the current stroke is the maximum stroke meeting eta > eta min, and the stroke is reduced by controlling a frequency converter on the basis of the current stroke; when the adjusting frequency is less than 1, increasing the stroke frequency by controlling the frequency converter on the basis of the current stroke frequency; and then judging whether the number of times of stroke needs to be reduced to continue adjustment according to the subsequent adjustment effect.
3. The self-learning based stroke rate adjustment method for a rod-pumped well according to claim 2, wherein the step of analyzing the optimal E value comprises the following steps: calculating the value E according to formula (2); inserting the value E into a data queue, selecting the maximum value in the data queue after judging that the number of the data meets the minimum number, and judging whether the maximum value E is in the middle position of effective data; if so, returning the stroke value corresponding to the maximum value; if not, judging whether the mobile terminal is at the initial position; if yes, carrying out reverse adjustment to reduce the stroke frequency; if not, the positive adjustment is carried out to increase the stroke frequency.
4. The self-learning based stroke frequency adjustment method for the rod-pumped well according to claim 3, wherein the specific steps of calculating the frequency F according to the stroke frequency N are as follows: outputting stroke frequency data N according to a formula (1) and judging whether the first calculation is carried out; if yes, directly adopting a preset relation parameter d for obtaining the frequency and the stroke frequency 0 And e 0 (ii) a If not, the relation parameter d is recalculated according to the data of the last two times i And e i And updating the preset parameter d 0 And e 0 (ii) a Then, the corresponding motor frequency F is calculated according to the formula (1).
5. The self-learning based stroke frequency adjusting method for the sucker-rod pumping unit according to claim 4, wherein the steady state judgment process of the pumping unit comprises: inserting the number of times of stroke value N into a historical data queue; calculating the average value of n times of stroke data in the historical data queue, calculating the difference value between the current data and the average value, and checking whether the difference value is within the allowable range of the equilibrium state; if so, the pumping unit reaches a stable state; if not, the pumping unit does not reach a stable state.
6. The self-learning based rod-pumped well stroke rate adjustment method of claim 4, wherein the determination process of the well entering a steady state comprises: inserting the fullness value eta into a historical data queue; calculating the average value of n fullness data in the historical data queue, calculating the difference value of the current data and the average value, and checking whether the difference value is within the allowable range of the equilibrium state; if so, the oil well reaches a stable state; if not, the well does not reach a steady state.
7. The self-learning based stroke frequency adjusting method for the sucker-rod pumping unit according to claim 4, wherein the steady state judgment process of the pumping unit comprises: calculating each indicator diagram period by acquiring data of the pumping unit in real time; when the difference value between each period Ti and the average period T of the n periods is stabilized in a fixed range in the continuous n indicator diagram periods, the working state of the pumping unit enters a stable state;
the steady state is judged by the following index conditions:
1) stable condition of liquid production
Wherein: q AV The average value of the liquid production amount of the indicator diagram is shown; Δ X is the amount of change in fluid production; q i The fluid production amount in the ith period;
2) stable condition of work diagram area
Wherein: s AG Is the average value of the area of the ground indicator diagram, Deltay is the variation of the area of the ground indicator diagram, S i The area of the ground indicator diagram in the ith period;
and simultaneously, when the conditions 1) and 2) are met, the working state of the pumping unit is judged to enter a stable state.
8. The self-learning based rod-pumped well stroke frequency adjusting method of claim 7, wherein: after the working state of the pumping unit enters a stable state, the calculation formulas of the stable liquid production amount and the stable pump efficiency are as follows: steady state fluid production:
steady state pump efficiency:
wherein Q is cv A steady state fluid production; y is cv A steady state pumping effect; n is a number.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110588435.2A CN113266318B (en) | 2021-05-28 | 2021-05-28 | Self-learning-based stroke frequency adjusting method for rod-type pumping unit well |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110588435.2A CN113266318B (en) | 2021-05-28 | 2021-05-28 | Self-learning-based stroke frequency adjusting method for rod-type pumping unit well |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113266318A CN113266318A (en) | 2021-08-17 |
| CN113266318B true CN113266318B (en) | 2022-08-09 |
Family
ID=77233483
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110588435.2A Active CN113266318B (en) | 2021-05-28 | 2021-05-28 | Self-learning-based stroke frequency adjusting method for rod-type pumping unit well |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113266318B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115774420A (en) * | 2022-12-14 | 2023-03-10 | 安徽物迅科技有限公司 | Near and remote control system and method for oil well lifting |
| CN118092369A (en) * | 2024-04-29 | 2024-05-28 | 山东合能科技有限责任公司 | Pumping unit intermittent pumping control method adopting wireless measurement |
Family Cites Families (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101174813B (en) * | 2007-10-15 | 2012-12-12 | 李晓军 | Energy-saving control method of oil pumping machine |
| CN102213084B (en) * | 2011-05-23 | 2014-07-23 | 中国石油天然气股份有限公司 | Stroke frequency adjusting device of oil pumping unit |
| CN102865053B (en) * | 2011-07-04 | 2015-06-03 | 中国石油天然气股份有限公司 | Method and device for judging optimal stroke frequency of oil pumping unit |
| US8892372B2 (en) * | 2011-07-14 | 2014-11-18 | Unico, Inc. | Estimating fluid levels in a progressing cavity pump system |
| CN103161430B (en) * | 2011-12-14 | 2016-08-31 | 中国石油天然气股份有限公司 | Real-time intelligent adjustment method for stroke frequency and balance of oil pumping unit |
| CN103823424B (en) * | 2013-09-18 | 2015-05-20 | 大连仲德能源与环保科技有限公司 | Method for adjusting stroke times of beam pumper in automation oil extraction system |
| CN104612631B (en) * | 2014-12-31 | 2018-07-27 | 新疆维吾尔自治区第三机床厂 | Power-balance digital automatic control oil pumping method and oil recovery robot |
| CN106286255B (en) * | 2016-10-27 | 2017-12-12 | 北京安控科技股份有限公司 | A kind of oil pumper Intelligent air takes out control device and its control method |
| CN108223343A (en) * | 2016-12-22 | 2018-06-29 | 中国科学院沈阳自动化研究所 | A kind of pumpingh well method for controlling frequency conversion based on Dynamic Control Chart and indicator card |
| CN108805215B (en) * | 2018-06-19 | 2021-06-11 | 东北大学 | Dynamic liquid level soft measurement method for sucker-rod pump pumping well based on improved drosophila algorithm |
| CN109085438A (en) * | 2018-09-06 | 2018-12-25 | 东北大学 | Intelligent monitoring device and method based on electrical parameter real-time monitoring oil pumping system efficiency |
-
2021
- 2021-05-28 CN CN202110588435.2A patent/CN113266318B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN113266318A (en) | 2021-08-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113266318B (en) | Self-learning-based stroke frequency adjusting method for rod-type pumping unit well | |
| CN107060695B (en) | Beam-pumping unit energy-saving control system and method | |
| CN113107432B (en) | Automatic control method for oil pumping well | |
| CN103177155A (en) | Oilfield pumping unit oil pumping energy saving and production increasing optimization method based on back propagation neural network (BPNN) and strength Pareto evolutionary algorithm 2 (SPEA2) | |
| CN108757648B (en) | Hydraulic pile hammer control method and system | |
| CN106948796A (en) | Pumping control device and method for pumping unit during non-stop | |
| CN103195698A (en) | Energy-saving control method for synchronous variable-speed regulation of water pumps in water-level-variable water intake pump station | |
| CN113006188B (en) | Excavator staged power matching method based on LSTM neural network | |
| CN103488132A (en) | Automatic optimized parameter-tuning method based on indicator diagram | |
| CN108086965B (en) | Oil pumping control system and method based on working fluid level and casing pressure | |
| CN106014999A (en) | Efficient running control method for water ring vacuum pump unit and control system | |
| CN102155405A (en) | Intelligent revolving speed control system and control method for submersible oil screw pump | |
| CN113536710B (en) | Energy efficiency visual monitoring method for pump and pump set | |
| CN107305055B (en) | A kind of frequency-changing control system of water pump of central air conditioner | |
| CN103389743B (en) | Method for considering real-time operation cleanness factors of condenser and improving efficiency of generating set | |
| CN111594139A (en) | Oil well testing and equal pump filling stroke-number synchronous numerical control oil pumping method | |
| CN206376832U (en) | Energy-saving control system of beam-pumping unit | |
| CN110705130B (en) | Water pump model selection method for low-lift pump device of large pump station | |
| CN202023738U (en) | Rotating speed intelligent control system for oil-immersed screw pump | |
| CN206957665U (en) | Pumping control device of pumping unit during non-stop | |
| CN111008731A (en) | Flow regulation and control method for blade mechanism of intelligent pump station | |
| CN106354980B (en) | External characteristics mathematical model construction method is pumped under a kind of rotating forward pump operating condition | |
| CN111884557B (en) | Pumping unit control method based on motor power torque | |
| CN115596412A (en) | Automatic pumping unit stroke frequency adjusting method based on continuous liquid production of pumping unit wellhead | |
| CN112610263B (en) | Fully mechanized coal mining face-based follow machine control method and device and electronic equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| PE01 | Entry into force of the registration of the contract for pledge of patent right | ||
| PE01 | Entry into force of the registration of the contract for pledge of patent right |
Denomination of invention: A self-learning based method for adjusting the stroke rate of rod pumping wells Effective date of registration: 20230419 Granted publication date: 20220809 Pledgee: Xi'an innovation financing Company limited by guarantee Pledgor: Shaanxi effik Energy Technology Co.,Ltd. Registration number: Y2023610000285 |