A method for controlling the intelligent throttling pressure of an oil nozzle in a ground oil testing process

Abstract

The present invention discloses a method for intelligent throttling pressure control of a nozzle in a ground oil test process, the method comprising: using fluid mechanics software to model and simulate the nozzle flow field and pressure field, combining the throttling pressure data of the ground oil test site to obtain the pressure drop of nozzles of different specifications during the spraying, and determining the pressure ratio threshold before and after throttling of a single-stage nozzle; designing a fuzzy adaptive controller, using the adjustment experience of operators and experts as a knowledge base, and obtaining fuzzy rules to adjust PID parameters online in real time; designing a nozzle throttling series model, defining the real-time pressure range, and making an optimal series decision; after determining the throttling series, the control model adjusts the nozzle opening used through the controller to control the throttling pressure of nozzles of different levels to remain within the defined nozzle before and after pressure ratio threshold range. The present invention optimizes the throttling mode and reduces the possibility of cavitation and erosion that may be generated by the nozzle throttling manifold.

Description

Technical Field

[0001] The invention discloses an intelligent throttling pressure control method for an oil nozzle in a ground oil testing process, which belongs to a throttling oil nozzle pressure control method involving a process control model. Background Art

[0002] With the exploitation and utilization of oil and gas resources, the importance of oil testing in oil and gas field development has become increasingly prominent. The traditional nozzle throttling pressure control method mainly relies on the experience of on-site workers. Although it is relatively effective in terms of control effect, the control strategy is extensive. This type of method is incapable of coping with complex working conditions. Therefore, the nozzle throttling pressure control method is gradually transforming to intelligentization to better meet the new needs of oil and gas field development.

[0003] In oil testing operations, precise control of throttling nozzle pressure is crucial to improving the safety of complex well testing processes and reducing pressure fluctuations. The intelligent throttling nozzle pressure control method combines actual operating parameters with intelligent adaptive control methods to provide an advanced solution.

[0004] In the intelligent throttling pressure control of the nozzle, it is extremely important to combine the actual working parameters. These parameters include pressure, temperature, fluid viscosity and composition, etc., which have a direct impact on the final control effect. During the oil test, when the pressure difference before and after the nozzle increases, the flow rate of the solid, liquid and gas mixture increases accordingly, causing serious erosion and wear on the nozzle, and may even cause leakage. More precise parameters also mean a more accurate control model, thereby achieving higher control accuracy.

[0005] With the rapid development of artificial intelligence and machine learning technologies, intelligent control algorithms are increasingly being used in nozzle pressure control. The traditional PID control algorithm currently used in oil testing operations is favored because it is simple, intuitive, and easy to implement. However, the nozzle pressure control model is nonlinear, multivariable, and highly complex. In actual oil testing operations, multi-stage nozzles are not always used for pressure control at the same time. In the case of pressure changes and external disturbances, the existing control method lacks consideration of the nozzle throttling level, so the traditional PID control algorithm performs poorly.

[0006] The intelligent throttling pressure control method of the oil nozzle generally refers to a method of controlling the oil nozzle pressure by adjusting the throttling opening of the oil nozzle using a control method. However, the existing control methods firstly lack consideration of the operating parameters. Too large or too small pressure drop values ​​will lead to erosion and cavitation, which seriously threaten the safety of equipment and personnel; secondly, the speed, accuracy and stability are insufficient; thirdly, the existing control methods fail to give a reasonable throttling level according to the existing pressure range. Therefore, a method for intelligent throttling pressure control of the oil nozzle in the ground oil test process is proposed. Summary of the invention

[0007] In order to solve the above-mentioned problems, the present invention provides a method for intelligent throttling pressure control of oil nozzles in a ground oil test process, which combines the output data of the mechanism model with the throttling pressure data of the ground oil test, and can be used to optimize the precise regulation of the oil nozzle pressure at different levels, ensuring that the oil nozzle can adaptively adjust the pressure according to the working conditions, thereby achieving more precise, faster and more stable control performance.

[0008] The specific steps of the intelligent throttling pressure control method for the oil nozzle in the surface oil testing process proposed by the present invention are as follows:

[0009] Step S10, using fluid mechanics software to model and simulate the flow field and pressure field of the nozzle, combining the throttling pressure data of the ground oil test site to obtain the pressure drop of nozzles of different specifications during the spraying, and determine the pressure ratio threshold before and after throttling of the single-stage nozzle;

[0010] Step S20, designing a fuzzy adaptive controller, taking the adjustment experience of operators and experts as a knowledge base, and obtaining fuzzy rules to adjust PID parameters online in real time;

[0011] Step S30, designing a nozzle throttling series decision model, and making an optimal series decision by defining a pressure range;

[0012] Step S40: After determining the throttling level, the control model adjusts the nozzle opening used through the controller to control the throttling pressure of different levels of nozzles to remain within the defined threshold range of the pressure ratio before and after the nozzle.

[0013] A further technical solution is that the specific process of step S10 is: perform pressure field-flow field modeling simulation on the nozzle manifold, and output pressure drop data of different nozzles through simulation computational fluid dynamics software, and combine the pressure drop of existing nozzles of different specifications, and obtain the threshold range of the upper and lower pressure ratio of the nozzle throttling through the package theorem:

[0014]

[0015] Among them, h f is the pressure difference between two adjacent nozzles, v1 and v2 are the flow rates of the upper and lower nozzles respectively, g is the gravitational acceleration, and k is the dimensionless empirical coefficient.

[0016] A further technical solution is that the specific process of step S20 is: determine the control target as the nozzle pressure, and the performance indicators as the steady-state error, adjustment time, and overshoot; use the Mamdani method for fuzzy reasoning, and the specific steps are: fuzzification - fuzzy rule definition - design reasoning mechanism - defuzzification.

[0017] A further technical solution is that the specific process of step S30 is: a new throttling level decision model is proposed, based on the known nozzle design parameters and pressure-throttling level data, a local weighted regression method is used to estimate the relationship between pressure and throttling level; the present invention also proposes a new multi-objective optimization function to find the optimal throttling level under the premise of considering operating conditions.

[0018] A further technical solution is that the specific process of step S40 is: designing a PID controller for basic flow and pressure control, which is responsible for handling the normal operation and slight disturbances of the system, and using the designed fuzzy controller to handle the dynamic changes of the system and the fuzzy PID controller for large-scale disturbances; the fuzzy controller dynamically adjusts the PID parameters according to the ground oil test throttling pressure data to improve the adaptability and sensitivity of the control system.

[0019] Compared with the prior art, the present invention has the following beneficial effects:

[0020] The present invention can more accurately determine the pressure ratio threshold before and after the nozzle by combining simulation computational fluid dynamics software simulation with actual nozzle throttling pressure data, thereby providing a scientific basis for intelligent throttling control of the nozzle and ensuring reliability.

[0021] The model of the present invention introduces a new nozzle throttling level decision model, which can dynamically select the optimal throttling level according to the real-time working conditions and the actual nozzle throttling pressure data, thereby realizing precise control of the nozzle pressure; the decision model takes different working conditions into consideration and determines the optimal throttling level through intelligent analysis, thereby optimizing the working performance of the nozzle.

[0022] The present invention proposes a new multi-stage nozzle intelligent pressure control method, which allows the system to finely adjust the nozzle opening online to maintain the nozzle throttling pressure within a relatively ideal range; the control system structure used has three unique subsystems to simulate multi-stage nozzle throttling, and this design allows the existence of three simultaneous / independent input real-time pressures, which is consistent with the actual working conditions of the oil test site operation; compared with the traditional method, this precise pressure ratio threshold setting and maintenance mechanism significantly improves the control accuracy and system stability, reduces overshoot and steady-state errors, and ensures the continuity and effectiveness of pressure regulation. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 It is a flow chart of the present invention;

[0024] Figure 2 It is the fuzzy rule knowledge graph;

[0025] Figure 3 The block diagram of the fuzzy controller used in the present invention;

[0026] Figure 4 This is a throttling level decision effect diagram of the present invention;

[0027] Figure 5 It is the structure diagram of the control model;

[0028] Figure 6 It is the output pressure curve of multi-stage nozzle;

[0029] Figure 7 This is a multi-stage nozzle pressure ratio curve. DETAILED DESCRIPTION

[0030] The present invention is further described below in conjunction with the accompanying drawings and implementation examples to make the purpose, technical solutions and advantages of the present invention more clear. Figure 1 For example, the process of the present invention is: combining the calculation data of the simulation computational fluid dynamics software and the existing nozzle pressure drop to obtain the threshold range of the pressure ratio before and after the nozzle throttling, passing the ground test oil throttling pressure data through the throttling series decision model, and then through the nozzle pressure control model online control, and finally outputting the real-time nozzle throttling pressure control response result, which specifically includes the following steps:

[0031] Step S10: Establish a flow field-pressure field model for the nozzle, simulate and calculate the nozzles of different specifications by using simulation computational fluid dynamics software, combine the simulation data with the existing nozzle-pressure relationship, and expand the existing nozzle throttling pressure data. Some of the data are shown below:

[0032]

[0033] It is then obtained that the threshold value of the pressure ratio before and after throttling of the multi-stage nozzle is 2.8 to 3.5 times.

[0034] Step S20: Figure 2 For example, through investigation and historical data integration, the adjustment experience of operators and experts is used as a knowledge base to obtain control rules; for the continuously input throttling pressure data, the Gaussian membership function is used to perform the first step of fuzzification:

[0035]

[0036] Where x is the value of the input variable, c is the mean of the center point of the membership function, and σ is the standard deviation of the membership function, which controls the width of the function.

[0037] Fuzzy rule definition: Determine the input variable as the nozzle throttling pressure P, and the output variable as the nozzle opening U i ; After analyzing historical data and collecting the experience and strategies of operators and experts in the adjustment process, the fuzzy variables are divided into five states: the first, second, third, fourth and fifth, from low to high. The pressure error is represented by e, and e crepresents the error change rate, The first derivative representing the rate of change of the error;

[0038] Among them, the state division of the nozzle pressure depends on the safety pressure P safe ; Under normal working conditions, if 0.05P safe <P≤0.15P safe , then determine that the nozzle pressure corresponds to the first state; if 0.15P safe <P≤0.35P safe , then determine that the nozzle pressure corresponds to the second state; if 0.35P safe <P≤0.5P safe , then determine that the nozzle pressure corresponds to the third state; if 0.5P safe <P≤0.65P safe , then determine that the nozzle pressure corresponds to the fourth state; if 0.65P safe <P≤0.8P safe , it is determined that the nozzle pressure corresponds to the fifth state;

[0039] If 0<|e|≤1, it is determined that the nozzle pressure error corresponds to the first state; if 1<|e|≤2, it is determined that the nozzle pressure error corresponds to the second state; if 2<|e|≤3, it is determined that the nozzle pressure error corresponds to the third state; if 3<|e|≤4, it is determined that the nozzle pressure error corresponds to the fourth state; if 4<|e|≤5, it is determined that the nozzle pressure error corresponds to the fifth state;

[0040] If 0<|e c |≤0.05, the nozzle pressure error change rate is determined to correspond to the first state; if 0.05<|e c |≤0.1, the nozzle pressure error change rate is determined to correspond to the second state; if 0.1<|e c |≤0.12, the nozzle pressure error change rate is determined to correspond to the third state; if 0.12<|e c |≤0.15, the nozzle pressure error change rate is determined to correspond to the fourth state; if 0.15<|e c |≤0.2, it is determined that the nozzle pressure error change rate corresponds to the fifth state;

[0041] like Determine the first-order derivative of the nozzle pressure error change rate corresponding to the first state; if Determine the first-order derivative of the nozzle pressure error change rate corresponding to the second state; if Determine the first-order derivative of the nozzle pressure error change rate corresponding to the third state; if Determine that the first-order derivative of the nozzle pressure error change rate corresponds to the fourth state; if Determine that the first derivative of the nozzle pressure error change rate corresponds to the fifth state;

[0042] Under normal working conditions, the five critical values ​​corresponding to the nozzle opening are: U1=18, U2=43, U3=50, U4=75, U5=100;

[0043] The fuzzy rule table constructed with the fuzzy language set {NB, NS, ZO, PS, PB} is as follows:

[0044]

[0045] e and e c , U i The four fuzzy variables are represented by NB, NM, NS, ZO, PS, PM, and PB to represent the degree of membership. The centroid method is used to defuzzify and transform them into parameter values ​​that need to be adjusted:

[0046]

[0047] Among them, V out Indicates the center of gravity of the fuzzy output, representing the relative importance of the parameters, μ i is the membership function value of the output variable, v i Describe the central value of fuzzy set i;

[0048] by Figure 3 For example, the input pressure of each nozzle is P(0), and the corresponding e and e are calculated. c , Input the fuzzy PID controller to obtain the controller parameters. The output pressure is changed by adjusting the nozzle opening, and the output pressure is fed back to the fuzzy controller as a feedback signal. The fuzzy controller adjusts the pressure control at the next moment according to the feedback signal and the controller parameters calculated previously. The output pressure P(t) is:

[0049]

[0050] where k p , k i , k d , k ff , k ff ′ is the control parameter, Δ′P(t) represents the rate of change of pressure error, and Δ″P(t) represents the first-order derivative of the rate of change of pressure error; k p is the proportional gain, which determines the sensitivity of the controller output to the current pressure error e(t); k i is the integral gain, which is used to eliminate the steady-state error from time 0 to the current time t; k dis the differential gain, which is used to predict the trend of pressure error; compared with the ordinary PID control parameters, an adaptive adjustment parameter k is added ff , k ff ′,k ff is the first-order feedforward gain, which adjusts the control input in advance according to the rate of change of the pressure error, k ff ′ is the second-order feedforward gain. The control input is applied in advance according to the first-order derivative of the pressure error change rate to improve the tracking performance of the controller. The controller parameter k is dynamically adjusted by the real-time input pressure error, error change rate and the first-order derivative of the error change rate. p , k i , k d , k ff , k ff ′, determine the output of the fuzzy controller as parameter k p , k i , k d , k ff , k ff ′Increment.

[0051] Step S30: The throttling pressure data of the ground oil test is first input into the throttling level decision model. According to the performance of the current oil nozzle, the optimal throttling mode is decided after defining the pressure range. The specific formula of local weighted regression is as follows:

[0052]

[0053] P is the design matrix, which contains each surface oil test throttling pressure data point p n The corresponding pressure value;

[0054]

[0055] Y is the response variable, y i Indicates the throttling level corresponding to the i-th data point;

[0056]

[0057] W is a weighted matrix, which is a diagonal matrix. Each diagonal element is a weight calculated based on the distance. This weight reflects the influence of the data point on the predicted point. h is a bandwidth parameter used to control the degree of weight. The predicted value of the throttling level at a given pressure value p is:

[0058]

[0059] The decision system also needs to take into account the working condition constraints of the nozzle: minimum and maximum flow limits, minimum and maximum pressure limits, and the optional range of throttling stages. In order to balance the relationship between flow and pressure and take into account the selection of stages, a new multi-objective optimization function is proposed:

[0060] Minimize w l ·P(Q,n)+w2·|P ri -P ti |+α·n

[0061] The multi-objective optimization function is divided into two parts: maximizing flow and minimizing pressure deviation; P(Q, n) is the flow function, which represents the flow output of the nozzle under a given pressure when the throttling level n is selected; here n represents the selected throttling level, and this function is modeled based on the historical data and simulation data of the nozzle; P ri is the actual output pressure, P ti is the target pressure, α is the penalty coefficient of the throttling level, which is used to consider the impact of the throttling level on the pressure; the purpose of stabilizing the pressure is achieved by minimizing the absolute deviation between the actual output pressure and the target pressure; Figure 4 For example, the critical value of the throttling pressure of the first-stage nozzle is P t1 The critical value of the throttling pressure of the secondary nozzle is 38.25 MPa. t2 The critical value of the throttling pressure of the three-stage nozzle is 14.29 MPa. t3 is 9.15MPa; at this time P safe is 52.5MPa, the input real-time pressure is 8Mpa, between 0 and P t3 The throttling stage decision system selects the three-stage nozzle for throttling pressure control, and the first and second stage nozzles have no input.

[0062] Step S40: After the decision model determines the throttling mode, the ground test oil throttling pressure data is input into the nozzle pressure controller to throttle the corresponding nozzle. Figure 5 For example, the real-time pressure input module is responsible for receiving the test oil throttling pressure data through the API interface; the pressure data provided by the pressure input module is used as the input of the throttling level decision model to determine the required nozzle level; once the decision model determines the required nozzle level, the data will enter the nozzle pressure controller of the corresponding level, and the nozzle pressure controller is responsible for adaptively adjusting the opening of each nozzle according to the result of the decision model to keep the output pressure within the target range. The controller adjusts the flow rate through the opening adjustment module according to the feedback signal, thereby controlling the output pressure;

[0063] Under normal working conditions, the nozzle throttling pressure is used as the input variable and the nozzle opening is used as the output variable. c , It is only used as a reference parameter for adjusting the throttling opening of the auxiliary nozzle. The nozzle opening is adjusted mainly to meet the single-stage front and rear nozzle pressure ratio threshold.

[0064] by Figure 6 , Figure 7For example, the throttling pressure of the nozzle is monitored in real time, and this pressure value is used as the input variable of the system. The pressure ratio of the current single-stage front and rear nozzles is compared with the set pressure ratio threshold. If the current pressure ratio of the single-stage front and rear nozzles is not within the set threshold range, the fuzzy PID controller in the pressure controller intervenes and e and e c , The three variables are fuzzified, and the controller performs reasoning calculation according to the designed fuzzy rule set. The calculation result is defuzzified to obtain the nozzle opening that needs to be adjusted, and the output pressure is controlled through the opening adjustment module;

[0065] When the multi-stage nozzle pressure ratio is in the range of 2.8 to 3.5 times, the nozzle opening remains unchanged; when it is less than 2.8 times and greater than 3.5 times, the control model adaptively adjusts the nozzle opening in the working state to ensure that the pressure ratio is always within the threshold range.

[0066] The model described in this application includes a nozzle pressure control model and a nozzle throttling level decision model. The nozzle pressure control model is used to maintain pressure control when the nozzle pressure suddenly changes and there is strong interference. The nozzle throttling level decision model is used to define the pressure range and make the best throttling method decision.

[0067] The above description is not intended to impose any form of limitation on the present invention. Although the present invention has been disclosed through the above-mentioned implementation cases, it is not intended to limit the present invention. Any technician familiar with the profession can make some changes or modifications to equivalent implementation examples with equivalent changes by using the technical contents disclosed above without departing from the scope of the technical solution of the present invention. However, any simple modification, equivalent change and modification made to the above implementation examples based on the technical essence of the present invention without departing from the content of the technical solution of the present invention still falls within the scope of the technical solution of the present invention.

Claims

1. A method for intelligent throttling pressure control of a nozzle in a ground oil testing process, characterized in that: After combining the nozzle modeling simulation output pressure data with the ground oil test throttling pressure data, the control algorithm is used to perform intelligent throttling pressure control of the nozzle, including the following steps: Step S10, using fluid mechanics software to model and simulate the flow field and pressure field of the nozzle, combining the throttling pressure data of the ground oil test site to obtain the pressure drop of nozzles of different specifications during the spraying, and determine the pressure ratio threshold before and after throttling of the single-stage nozzle; Step S20, designing a fuzzy adaptive controller, which uses the adjustment experience of operators and experts as a knowledge base and uses the Mamdani method for fuzzy reasoning to adjust PID parameters online in real time; Step S30, designing a nozzle throttling series decision model, based on known nozzle design parameters and pressure-throttling series data, using a local weighted regression method to estimate the relationship between pressure and throttling series, defining a real-time nozzle range, and making an optimal throttling series decision; Step S40: After determining the throttling level, the control model adjusts the nozzle opening used through the controller to control the throttling pressure of different levels of nozzles to remain within the defined threshold range of the pressure ratio before and after the nozzle.

2. The method for controlling the intelligent throttling pressure of the oil nozzle in the surface oil testing process according to claim 1, characterized in that: In step S10, a flow field-pressure field model is established for the nozzle, and nozzles of different specifications are simulated and calculated using simulation computational fluid dynamics software. The simulation data is combined with the existing nozzle-pressure relationship to expand the existing nozzle pressure data, and it is concluded that the threshold value of the pressure ratio before and after throttling of the multi-stage nozzle is 2.8 to 3.5 times.

3. The method for controlling the intelligent throttling pressure of the oil nozzle in the surface oil testing process according to claim 1, characterized in that: In step S20, the input variable is the nozzle throttling pressure P, and the output variable is the nozzle opening U i ; After analyzing historical data and collecting the experience and strategies of operators and experts in the adjustment process, the fuzzy variables are divided into five states: the first, second, third, fourth and fifth, from low to high. The pressure error is represented by e, and e c represents the error change rate, The first derivative representing the rate of change of the error; Among them, the state division of the nozzle pressure depends on the safety pressure P safe ; Under normal working conditions, if 0.05P safe <P≤0.15P safe , then determine that the nozzle pressure corresponds to the first state; if 0.15P safe <P≤0.35P safe , then determine that the nozzle pressure corresponds to the second state; if 0.35P safe <P≤0.5P safe , then determine that the nozzle pressure corresponds to the third state; if 0.5P safe <P≤0.65P safe , then determine that the nozzle pressure corresponds to the fourth state; if 0.65P safe <P≤0.8P safe , it is determined that the nozzle pressure corresponds to the fifth state; If 0<|e|≤1, it is determined that the nozzle pressure error corresponds to the first state; if 1<|e|≤2, it is determined that the nozzle pressure error corresponds to the second state; if 2<|e|≤3, it is determined that the nozzle pressure error corresponds to the third state; if 3<|e|≤4, it is determined that the nozzle pressure error corresponds to the fourth state; if 4<|e|≤5, it is determined that the nozzle pressure error corresponds to the fifth state; If 0<|e c |≤0.05, the nozzle pressure error change rate is determined to correspond to the first state; if 0.05<|e c |≤0.1, the nozzle pressure error change rate is determined to correspond to the second state; if 0.1<|e c |≤0.12, the nozzle pressure error change rate is determined to correspond to the third state; if 0.12<|e c |≤0.15, the nozzle pressure error change rate is determined to correspond to the fourth state; if 0.15<|e c |≤0.2, it is determined that the nozzle pressure error change rate corresponds to the fifth state; like Determine the first-order derivative of the nozzle pressure error change rate corresponding to the first state; if Determine the first-order derivative of the nozzle pressure error change rate corresponding to the second state; if Determine the first-order derivative of the nozzle pressure error change rate corresponding to the third state; if Determine that the first-order derivative of the nozzle pressure error change rate corresponds to the fourth state; if Determine that the first derivative of the nozzle pressure error change rate corresponds to the fifth state; Under normal working conditions, the five critical values ​​corresponding to the nozzle opening are: U1=18, U2=43, U3=50, U4=75, U5=100; The output pressure P(t) of each stage nozzle is: where k p , k i , k d , k ff , k ff ' is the control parameter, Δ′P(t) represents the rate of change of pressure error, and Δ”P(t) represents the first-order derivative of the rate of change of pressure error; k p is the proportional gain, which determines the sensitivity of the controller output to the current pressure error e(t); k i is the integral gain, which is used to eliminate the steady-state error from time 0 to the current time t; k d is the differential gain, which is used to predict the trend of pressure error; compared with the ordinary PID control parameters, an adaptive adjustment parameter k is added ff , k ff ', k ff is the first-order feedforward gain, which adjusts the control input in advance according to the rate of change of the pressure error, k ff ' is the second-order feedforward gain, and the control input is applied in advance according to the first-order derivative of the pressure error change rate to improve the tracking performance of the controller; the controller parameter k is dynamically adjusted by the real-time input pressure error, error change rate and the first-order derivative of the error change rate p , k i , k d , k ff , k ff ', determine the output of the fuzzy controller as parameter k p , k i , k d , k ff , k ff 'Increment.

4. The method for controlling the intelligent throttling pressure of the oil nozzle in the surface oil testing process according to claim 1, characterized in that: In step S30, the specific method for designing the nozzle throttling series decision model is: based on the known nozzle design parameters and pressure-throttling series data, the relationship between pressure and throttling series is estimated using the local weighted regression method. The specific formula of the local weighted regression is as follows: P is the design matrix, which contains each surface oil test throttling pressure data point p n The corresponding pressure value; <h2 style=";text-align:left;direction:ltr">Y = [y1,y2,y3,...,y<h2 style=";text-align:left;direction:ltr"> i <h2 style=";text-align:left;direction:ltr"> ]<h2 style=";text-align:left;direction:ltr"> T Y is the response variable, y i Indicates the throttling level corresponding to the i-th data point; W is a weighted matrix, which is a diagonal matrix. Each diagonal element is a weight calculated based on the distance. This weight reflects the influence of the data point on the predicted point. h is a bandwidth parameter used to control the degree of weight. The predicted value of the throttling level at a given pressure value p is: The decision system also needs to take into account the working condition constraints of the nozzle: minimum and maximum flow limits, minimum and maximum pressure limits, and the optional range of throttling stages. In order to balance the relationship between flow and pressure and take into account the selection of stages, a new multi-objective optimization function is proposed: Minimize w1·P(Q,n)+w2·|P ri -P ti |+α·n Among them, the multi-objective optimization function is divided into two parts: maximizing the flow rate and minimizing the pressure deviation; P(Q,n) is the flow function, which represents the flow output of the nozzle under a given pressure when the throttling level n is selected; here n represents the selected throttling level, and this function is modeled based on the historical data and simulation data of the nozzle; P ri is the actual output pressure, P ti is the target pressure, α is the penalty coefficient of the throttling level, which is used to consider the impact of the throttling level on the pressure; the purpose of stabilizing the pressure is achieved by minimizing the absolute deviation between the actual output pressure and the target pressure.

5. The method for controlling the intelligent throttling pressure of the oil nozzle in the surface oil testing process according to claim 1, characterized in that: In step S40, after the decision model determines the throttling mode, the ground test oil throttling pressure data is input into the nozzle pressure controller to throttle the corresponding nozzle, and the real-time pressure input module is responsible for receiving the test oil throttling pressure data through the API interface; the pressure data provided by the pressure input module is used as the input of the throttling level decision model to determine the required nozzle level; Once the decision model determines the required number of nozzles, the data will enter the nozzle pressure controller of the corresponding number of nozzles. The nozzle pressure controller is responsible for adaptively adjusting the opening of each nozzle according to the result of the decision model to keep the output pressure within the target range. The controller adjusts the flow rate through the opening adjustment module according to the feedback signal, thereby controlling the output pressure; Under normal working conditions, the nozzle throttling pressure is used as the input variable and the nozzle opening is used as the output variable. c , It is only used as a reference parameter for adjusting the throttling opening of the auxiliary nozzle. The nozzle opening is adjusted mainly to meet the pressure ratio threshold of the single-stage front and rear nozzles. The throttling pressure of the nozzle is monitored in real time, and this pressure value is used as the input variable of the system. The current pressure ratio of the single-stage front and rear nozzles is compared with the set pressure ratio threshold. If the current pressure ratio of the single-stage front and rear nozzles is not within the set threshold range, the fuzzy PID controller in the pressure controller intervenes and e and e c , The three variables are fuzzified, and the controller performs reasoning calculation according to the designed fuzzy rule set. The calculation result is defuzzified to obtain the nozzle opening that needs to be adjusted; When the multi-stage nozzle pressure ratio is in the range of 2.8 to 3.5 times, the nozzle opening remains unchanged; when it is less than 2.8 times and greater than 3.5 times, the control model adaptively adjusts the nozzle opening in the working state to ensure that the pressure ratio is always within the threshold range.

Images