Double-pid intelligent collaborative regulation method for oilfield production preparation station and injection station

Abstract

The application relates to the technical field of intelligent control, in particular to a double-PID intelligent collaborative regulation method for an oilfield production preparation station and an injection station, which comprises the following steps: obtaining initial proportional coefficients, initial integral times and dead zone parameters of PID control of the preparation station and PID control of the injection station, and based on all the obtained initial parameters, regulating the double-PID linkage of the injection station and the preparation station: analyzing the correlation degree between the liquid level height data of the injection station and the external delivery pressure data of the preparation station, and determining the double-station control disorder degree; filtering all the double-station control disorder degrees obtained at each collection time, analyzing the overall change of the double-station control disorder degrees before and after filtering, analyzing the influence degree of the proportional coefficient adjustment, determining the feedforward compensation factor, and adjusting the proportional coefficient of the PID control of the injection station. The application aims to realize intelligent control of the injection station flow regulator and precise regulation of the external delivery pump frequency of the preparation station.

Description

Technical Field

[0001] This application relates to the field of intelligent control technology, specifically to a dual PID intelligent coordinated regulation method for oilfield production preparation stations and injection stations. Background Technology

[0002] In oilfield development, the preparation station and injection station are key links in the polymer flooding production process, and the efficiency and quality of their coordinated operation directly affect the overall reservoir development results. As oilfield development enters the middle and late stages, the requirements for the refinement and intelligence of development technologies become increasingly stringent, and the drawbacks of traditional preparation and injection systems gradually become apparent.

[0003] In traditional oilfield production models, preparation stations and injection stations operate independently, lacking effective information exchange and coordination mechanisms. Production parameters change dynamically; the total injection volume at the injection station fluctuates due to reservoir demand, and the tank level changes with the injection and consumption of mother liquor. However, these changes cannot be transmitted to the preparation station in a timely manner, preventing the station from accurately adjusting production according to actual needs, resulting in resource waste and low production efficiency.

[0004] In the dual-PID coordinated regulation process between the production and preparation stations and the injection station, the intelligent control loop for liquid level in the injection station and the intelligent control loop for pressure in the preparation station are highly coupled. The intelligent control behavior of either control station will affect the controlled object of the other control station. For example, if the injection station increases the valve opening to raise the liquid level, it will lead to a decrease in the pressure of the outgoing pipeline of the preparation station. Conversely, if the preparation station increases the pump frequency to ensure pressure stability, it will increase the amount of mother liquor delivered to the injection station, thereby pushing up the liquid level in the injection station. If the intelligent control strategy and parameter tuning do not take into account the coupling problem between the preparation and injection stations, positive feedback problems are likely to occur, further causing oscillations in the control system or even instability, resulting in the system liquid level and pressure failing to converge around the set value. Summary of the Invention

[0005] In view of the above, it is necessary to provide a dual-PID intelligent coordinated regulation method for oilfield production preparation stations and injection stations to solve the above problems.

[0006] One embodiment of this application provides a dual-PID intelligent coordinated control method for oilfield production preparation stations and injection stations, the method comprising: S1: Obtain the initial proportional coefficient, initial integral time, and dead zone parameters of the PID control of the preparation station through simulation testing to control the output of mother liquor from the preparation station; S2: By combining the instantaneous flow rate data of the mother liquor from all single wells with the actual liquid level height, the ideal liquid level height, and the range height of the storage tank in the station, the demand for mother liquor is determined and substituted into the PID control of the injection station to obtain the initial proportional coefficient, initial integral time, and dead zone parameters. S3: Based on all obtained initial parameters, implement dual PID linkage regulation of the injection station and the preparation station: S301: Collects liquid level data from the injection station's storage tank and external pressure data from the preparation station; S302: Analyze the correlation between the liquid level data of the injection station storage tank and the external pressure data of the preparation station, and determine the dual-station control failure at each data acquisition time. S303: Filter all bi-station control misalignment obtained at each acquisition time, analyze the overall change of the bi-station control misalignment before and after filtering, and determine the feedforward compensation factor of the injection configuration at each acquisition time by combining the influence of bi-station control misalignment on the proportional coefficient adjustment. S304: Adjust the proportional coefficient of the PID control of the injection station based on the feedforward compensation factor.

[0007] Preferably, the initial proportional coefficient of the PID control in the preparation station is 4, the initial integral time is 0.01, and the dead zone parameter is 2.

[0008] Preferably, the formula for determining the required amount of mother liquor is as follows: In the formula, M represents the required amount of mother liquor; P represents the instantaneous total amount of mother liquor in the station; H represents the actual liquid level; h represents the ideal liquid level; and W represents the range height of the storage tank in the station.

[0009] Preferably, the required amount of mother liquor is used as the setpoint for the PID control of the injection station, representing the desired mother liquor flow rate standard.

[0010] Preferably, the initial proportional coefficient of the PID control at the injection station is 3, the initial integral time is 0.02, and the dead zone parameter is 2.

[0011] Preferably, determining the dual-station control misalignment at each acquisition time specifically involves: Obtain the grey relational degree and maximum wavelet coherence coefficient between the liquid level height data of the injection station storage tank at each acquisition time and all previous acquisition times and the external output pressure data of the preparation station; The result of forward fusion of the gray correlation degree and the maximum wavelet coherence coefficient at each acquisition time is used as the bi-station control outage scheduling at each acquisition time.

[0012] Preferably, the grey relational degree is obtained through grey relational degree analysis; the maximum wavelet coherence coefficient is obtained through wavelet coherence analysis.

[0013] Preferably, determining the feedforward compensation factor for the injection configuration at each acquisition time specifically involves: For PID control of the injection station, the predicted proportional coefficient is obtained based on the degree of influence, and the difference between the predicted proportional coefficient and the original proportional coefficient is recorded as the correction amount of the PID controller proportional coefficient. The mean value of the difference between the dual-station control outages at all the same acquisition times before and after filtering is obtained and recorded as the overall degree of change. The correction amount is positively integrated with the overall degree of change to obtain the feedforward compensation factor of the injection configuration at each acquisition time.

[0014] Preferably, the process of obtaining the predicted proportional coefficient is as follows: The difference between the proportional coefficient before and after the adjustment at each time point is taken as the proportional coefficient adjustment amount; Using all the bi-station control misalignments after filtering at each acquisition time as independent variables and all the proportional coefficient adjustments obtained at each acquisition time as dependent variables, a regression algorithm is used to obtain the predicted PID proportional coefficients.

[0015] Preferably, the proportional coefficient of the PID control of the injection station is adjusted using the following formula: ;in, B represents the initial proportional parameter for the PID control of the injection station; B is the feedforward compensation factor. This is a preset adjustment factor; e represents the natural constant. This represents the proportional parameter after adjustment by the PID control of the injection station.

[0016] This application has at least the following beneficial effects: This application first addresses the problem of deep coupling between the control loops of the injection station and the preparation station, which easily leads to positive feedback oscillations in the system. It uses grey relational analysis and wavelet coherence analysis to reflect the coupling strength characteristics of the system in terms of time-domain trend consistency and frequency-domain dominant scale, thus solving the problem that traditional single time-domain or frequency-domain analysis cannot fully capture the system dynamics, leading to continuous fluctuations in liquid level and pressure and failure to converge.

[0017] Secondly, to address the issues of control misadjustment and unnecessary energy consumption caused by sensor measurement errors and random interference, support vector regression and Kalman filtering methods are used to reflect the system coupling risk and the data reliability characteristics, thereby eliminating the impact of noise on real-time monitoring data.

[0018] Finally, by employing a modulation mechanism that converts the feedforward compensation factor into an adaptive proportional coefficient, the control parameters are intelligently and dynamically optimized, effectively eliminating system oscillations, reducing the number of pump and valve actions and energy consumption, improving the stability and control accuracy of the production system and the energy utilization efficiency, and achieving optimal energy efficiency throughout the entire process of energy-saving oil extraction. Attached Figure Description

[0019] Figure 1A flowchart of the dual PID intelligent coordinated regulation method for oilfield production preparation station and injection station provided in this application; Figure 2 The flow chart of the control function of the dispensing linkage system provided in this application; Figure 3 The response curves of the control device with P=2 and Ti=0.01 provided in this application; Figure 4 The response curves of the control device with P=4 and Ti=0.01 provided in this application; Figure 5 The response curves of the control device with P=5 and Ti=0.01 provided in this application; Figure 6 The response curves of the control device with P=4 and Ti=0.02 provided in this application; Figure 7 The response curves of the control device with P=4 and Ti=0.03 provided in this application; Figure 8 The response curves of the control device with P=4, Ti=0.01, and dead zone=1 provided for this application; Figure 9 The response curves of the control device with P=4, Ti=0.01, and dead zone=2 provided for this application; Figure 10 The response curves of the control device with P=4, Ti=0.01, and dead zone=3 provided for this application; Figure 11 The response curve of the dual PID linkage regulation provided in this application. Detailed Implementation

[0020] In the description of the embodiments in this application, the words "exemplary," "or," and "for example" are used to indicate examples, illustrations, or descriptions. Any embodiment or design scheme described as "exemplary" or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design schemes. Specifically, the use of the words "exemplary," "or," and "for example" is intended to present the relevant concepts in a specific manner.

[0021] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used in this application's specification is for the purpose of describing particular embodiments only and is not intended to be limiting of the application.

[0022] It should also be noted that the terms "first" and "second" in this application and its accompanying drawings are used to distinguish similar objects, rather than to describe a specific order or sequence. The methods disclosed in the embodiments of this application or the methods shown in the flowcharts include one or more steps for implementing the method. Without departing from the scope of protection of this application, the execution order of multiple steps can be interchanged, and some steps can also be deleted.

[0023] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains.

[0024] Architecture of the Dual PID Intelligent Collaborative Regulation and Control System: (1) Overall architecture: The design revolves around the connection between the injection station and the preparation station, with dynamic regulation as the core. The key data of the injection station is monitored in real time to realize the intelligent control of automatic switching of the flow regulator and maintain the stability of the liquid level in the injection station. The preparation station and the injection station are connected through the external pipeline. When the injection station adjusts the mother liquor receiving valve, causing a change in the external pressure of the preparation station, the control system of the preparation station adjusts the frequency of the external pump according to the change in external pressure to ensure stable and efficient delivery of mother liquor and achieve coordinated operation.

[0025] (2) Monitoring and control point setting: On the injection station side, monitor the injection volume of mother liquor in a single well, the total injection volume of mother liquor in a single station, the demand of mother liquor, the liquid level of the storage tank, and the control quantity of the flow controller; on the preparation station side, control the start and stop of the external pump, adjust the frequency of the external pump, monitor the outlet pressure of the external pump, select the station under the jurisdiction of the external pump, and monitor and control key parameters in all aspects to provide data support for linkage operation and ensure that the production process is controllable.

[0026] (3) Control Method: A dual PID control method is adopted to regulate the frequency converter and flow regulator of the external pump. On the one hand, PID adjustment is performed based on the set pressure and the actual external pressure. The output of the external pump frequency converter is adjusted according to the deviation to change the speed of the external pump and make the external pressure closer to the set value. On the other hand, PID adjustment is performed based on the comparison between the demand and the actual inflow. The flow regulator increases or decreases its opening accordingly to achieve precise control of the preparation and injection system, ensure stable external flow, meet the needs of different injection stations, and improve the stability and reliability of the production system. The flow chart of the control function of the preparation and injection linkage system is as follows: Figure 2 As shown.

[0027] This application proposes a dual-PID intelligent coordinated control method for oilfield production preparation stations and injection stations, applied in the field of intelligent control technology. (See attached document.) Figure 1 The method includes the following steps: S1: Obtain the initial proportional coefficient, initial integral time, and dead zone parameters of the PID control of the preparation station through simulation testing, and control the output of mother liquor from the preparation station.

[0028] Taking the No. 3 external pump of the new preparation station and its three injection stations as pilot projects, based on the real-time changes in external pressure, the external pressure, external pressure setpoint, frequency, and frequency setpoint are accurately written into the corresponding positions of the PID control module to realize the automatic adjustment function, thereby precisely controlling the external output of the mother liquor of the preparation station.

[0029] First, the ranges of the proportional coefficient and integral time were initially determined through theoretical analysis and experience. During the debugging of the proportional controller, the proportional coefficient K was gradually adjusted, and the system's response to the signal input was observed. When the value of K was too small, the system response was slow and unable to react promptly to changes in the output pressure; when the value of K was too large, the system would exhibit significant overshoot, or even oscillation, severely affecting the system's stability. During the debugging of the integral controller (I), it was found that the integral time Ti also had a significant impact on the system performance. The larger Ti was, the better the system stability, but the slower the rate of steady-state error elimination; the smaller Ti was, the faster the rate of steady-state error elimination, but the system was prone to oscillation. After simulation testing, it was finally determined that the system's control effect was optimal when the initial proportional coefficient K=4 and the initial integral time Ti=0.01. At this time, the output pump could reach the target pressure in a shorter time, with a smaller overshoot, and the system operated relatively stably. However, due to the deep coupling between the control loops of the oilfield production preparation station and the injection station, as well as problems such as sensor measurement errors and random interference, the system still faces the challenges of oscillation risk and energy waste. Therefore, it is necessary to adaptively adjust the obtained proportional coefficient K based on the feedforward compensation factor to achieve dynamic optimization control and ensure the stability and energy efficiency of the system under complex operating conditions.

[0030] To further improve system stability and reduce unnecessary adjustments, a dead-band setting was added to the PI settings. Simulation tests showed that when the dead-band is set to 2, the system adjustment time is the shortest and the operation is the most stable, effectively avoiding frequent system adjustments and improving system reliability and lifespan.

[0031] The simulation test process for setting specific proportional coefficient, integral time, and dead zone parameters is as follows: Figures 3-10 As shown. The response curves of the control device with P=2 and Ti=0.01 are as follows. Figure 3 As shown; P=4, Ti=0.01, control time is 54s, the response curve of the control device with P=4 and Ti=0.01 is as follows. Figure 4 As shown; with P=5, Ti=0.01, and a control time of 89s, the system exhibits significant oscillations. The response curve of the control device with P=5 and Ti=0.01 is shown below. Figure 5 As shown, the system oscillates when P=5, and the comparison of three sets of experimental data shows that when Ti is set to the minimum value, the oscillation is minimal and the control time is shortest.

[0032] Furthermore, P was set to 4, and Ti was gradually increased from 0.01 until the system oscillated. Specific experimental data are as follows: With P=4 and Ti=0.02, and a control time of 72s, the response curve of the control device with P=4 and Ti=0.02 is shown below. Figure 6 As shown; with P=4, Ti=0.03 and a control time of 83s, the system exhibits significant oscillations. The response curve of the control device with P=4 and Ti=0.03 is shown below. Figure 7 As shown in the figure; the experimental data above shows that the system oscillates when Ti>0.01, and the external pump control time is shortest when P=4 and Ti=0.01.

[0033] To avoid excessively frequent system adjustments, a dead-zone setting is incorporated into the PI control to improve system stability. With P=4, Ti=0.01, and the dead-zone gradually increasing from 1, the specific experimental data is as follows: When P=4, Ti=0.01, and the dead-zone is 1, the system adjusts relatively frequently. The response curve of the control device with P=4, Ti=0.01, and dead-zone=1 is shown below. Figure 8 As shown; when P=4, Ti=0.01, and dead zone=2, the system operates relatively smoothly. The response curve of the control device with P=4, Ti=0.01, and dead zone=2 is shown below. Figure 9 As shown; when P=4, Ti=0.01, and dead zone=3, the system amplitude is relatively large. The response curve of the control device with P=4, Ti=0.01, and dead zone=3 is shown below. Figure 10 As shown.

[0034] S2: By combining the instantaneous flow rate data of the mother liquor from all individual wells with the actual liquid level, ideal liquid level, and the range height of the storage tanks in the station, the demand for mother liquor is determined, and the initial proportional coefficient, initial integral time, and dead zone parameters of the PID control of the injection station are determined.

[0035] To address the shortcomings of the injection station control system, targeted secondary development was carried out.

[0036] First, the key parameter of the instantaneous total injection volume per well is obtained. Specifically, the instantaneous flow rate data of the mother liquor from each individual well is collected and organized, and the instantaneous flow rates of the mother liquor from all individual wells are summed up using a summation calculation method to obtain the total instantaneous injection volume of the entire single station. Obtaining this parameter provides accurate data support for comprehensively understanding the mother liquor injection situation at the injection station per unit time, which helps to accurately control the real-time status of the injection process. Accurate calculation of the mother liquor demand is crucial for maintaining stable liquid levels at the injection station and rationally allocating mother liquor resources. Using the obtained instantaneous total injection volume, combined with relevant data such as the instantaneous total injection volume of a single station and the liquid level height of the mother liquor tank, a specific calculation formula is used. This formula comprehensively considers multiple factors and can accurately calculate the mother liquor demand based on the actual situation of the injection station, providing an important basis for subsequent flow rate adjustment. The specific formula is as follows:

[0037] In the formula, M represents the required amount of mother liquor; P represents the instantaneous total amount of mother liquor in the station; H represents the actual liquid level; h represents the ideal liquid level; and W represents the range height of the storage tank in the station.

[0038] A flow rate PID control module was added to the programmable logic controller (PLC) of the injection station to achieve automatic liquid level adjustment. The mother liquor inflow rate was written as a measured value into the PID control module, reflecting the actual inflow in real time. The mother liquor demand was written as a setpoint, representing the desired mother liquor flow rate standard. The flow controller setpoint was written as a feedback value, constructing a complete feedback adjustment mechanism. Multiple comparative experiments were conducted to determine the PI value of the flow control PID program. Different combinations of P and I values ​​were tried, and the system's operating effect was observed. Ultimately, a P value of 3, an I value of 0.02, and a dead zone setpoint of 2 resulted in the most ideal PID curve. At this setting, the system has a short adjustment time and rapid response, quickly adjusting the flow controller opening according to liquid level changes, ensuring the liquid level in the injection station remains within a reasonable range. This effectively avoids production problems caused by excessively high or low liquid levels, ensuring the safe and stable operation of the injection station.

[0039] S3: Based on all the obtained initial parameters, the injection station and the preparation station are adjusted by dual PID linkage.

[0040] When the liquid level at the injection station rises, a valve closure operation is executed, and the PID control program for the station's external output pressure reduces the output frequency and decreases the amount of mother liquor output. When the liquid level at the injection station drops, a valve opening operation is executed, and the PID control program for the station's external output pressure increases the output frequency and increases the amount of mother liquor output, thereby maintaining the mother liquor flow balance in the system and ensuring the stability of parameters such as pressure and liquid level in each stage. The response curve of the dual-PID linkage regulation is shown below. Figure 11 As shown.

[0041] S301: Collects liquid level data from the injection station's storage tank and external pressure data from the preparation station.

[0042] A high-precision radar level sensor is installed in the storage tank area of ​​the injection station to achieve real-time acquisition of the tank level data; and an intelligent pressure transmitter is installed at the outlet pipeline of the external pump of the preparation station to complete the acquisition of external pressure data.

[0043] To achieve coordinated intelligent control between the injection station and the preparation station, the data acquisition clock source for both types of data is kept consistent to ensure strict synchronization in the time domain. The acquisition frequency for both types of data is set to 1Hz to avoid excessively complex data information due to excessively fast acquisition, while effectively capturing details such as liquid level changes and pressure fluctuations. After each acquisition, the Min-Max normalization algorithm is used to normalize the data based on the acquired data and historical data, eliminating the adverse effects of the different dimensions of liquid level and pressure, and converting the two types of data to the range of [0,1].

[0044] This application adopts a closed-loop feedback adjustment mechanism for the proportional coefficient, that is, the proportional coefficient is adjusted according to the calculation in this application, and the adjustment amount of the proportional coefficient at each acquisition time is obtained and fed back into the calculation to realize the closed-loop feedback adjustment of the proportional coefficient.

[0045] For each acquisition time, sort each type of data collected at all previous acquisition times in ascending order of time to obtain the liquid level height sequence of the injection station storage tank, the external pressure sequence of the preparation station, and the proportional coefficient adjustment sequence for each acquisition time.

[0046] S302: Analyze the correlation between the liquid level data of the injection station storage tank and the external pressure data of the preparation station to determine the dual-station control outage at each data acquisition time.

[0047] In the process of energy-saving oil extraction, the intelligent control systems of the injection station and the preparation station form a strongly coupled closed-loop system, which causes mutual interference between the intelligent control loops of the injection station and the preparation station, resulting in continuous system oscillation. This oscillation not only wastes energy, but also seriously restricts the efficiency of energy-saving oil extraction.

[0048] Therefore, using the sequence of liquid level height in the injection station's storage tank and the sequence of external pressure at the preparation station at each acquisition time as input, a grey relational analysis algorithm is used. A resolution coefficient of 0.5 is set to balance the discriminative power and stability of the correlation degree. The grey relational degree between the sequence of liquid level height in the injection station's storage tank and the sequence of external pressure at the preparation station is output. This output represents the degree of consistency in the changing trends of the two sequences. The closer its value is to 1, the stronger the coupling of the system. In energy-saving oil extraction, the larger its value, the more serious the mutual interference of the intelligent control loop.

[0049] In the dual-PID intelligent coordinated control system of the oilfield production preparation station and injection station, the liquid level changes relatively slowly, while the pressure fluctuates rapidly, causing the intelligent control actions to interfere with each other at different time scales. This multi-scale coupling makes it difficult for traditional single time-domain or frequency-domain analysis methods to fully capture the system dynamics, easily triggering positive feedback oscillations. This causes the liquid level and pressure to fluctuate continuously around the set value, failing to converge. This not only reduces the stability of the production system and the oil displacement effect, but also leads to frequent operation of pumps and valves and energy waste, severely restricting the energy-saving extraction efficiency of oil.

[0050] To address this issue, the sequence of liquid level in the injection station's storage tanks and the sequence of external pressure at the preparation station at each acquisition time were used as inputs. Wavelet coherence analysis was employed, with the Morlet wavelet as the mother wavelet, and the maximum wavelet coherence coefficient was output. This output represents the maximum correlation strength between two sequences on the dominant time scale. In energy-efficient oil extraction, its value directly reflects the coupling strength of the system in the most sensitive frequency band.

[0051] Based on the above analysis, the bi-station control misalignment A at each data acquisition time is calculated, specifically as the result of forward fusion between the gray correlation degree and the maximum wavelet coherence intensity coefficient. In this embodiment, the forward fusion adopts a multiplication calculation method. In other embodiments, the forward fusion can also adopt an addition or averaging calculation method.

[0052] It should be understood that grey relational analysis is used to assess the changing trends of two sequences; the greater the grey relational degree, the more similar the changing trends of the two sequences. In existing technologies, wavelet coherence analysis is used to determine the multi-scale correlation characteristics of two sequences in the time-frequency domain; the larger the maximum wavelet coherence coefficient, the higher the degree of correlation between the two sequences in the time scale. On the other hand, for energy-saving oil extraction, the coupling between the intelligent control loops of the injection station and the preparation station is not only about the similarity or convergence of macro trends, but also about the influence of many time scales. Therefore, relying solely on time-domain trend correlation may ignore local coupling behavior in the frequency domain, while relying solely on frequency-domain coherence may fail to capture the coordinated changes in the overall trend. Based on this, combining grey relational analysis and wavelet coherence analysis, the resulting dual-station control misalignment can simultaneously characterize the consistency of sequence trends and the coupling strength of the dominant frequency band, ultimately comprehensively quantifying the system coupling degree. The closer the value is to 1, the greater the disturbance in both the trend and frequency domain aspects of the intelligent control loop, making it easier to identify the oscillation risk of the system.

[0053] For each acquisition time, the bistation control misalignment A calculated from all acquisition times prior to each acquisition time is sorted in ascending order of time to obtain the bistation control misalignment sequence for each acquisition time, in order to identify the dynamic disturbance mode of the control loop.

[0054] S303: Filter all bi-station control misalignment data obtained at each acquisition time, analyze the overall change in bi-station control misalignment before and after filtering, and determine the feedforward compensation factor for injection configuration at each acquisition time by combining the degree of influence of bi-station control misalignment on proportional coefficient adjustment.

[0055] Sensor measurement errors and random disturbances can distort the observations of the dual-station control misalignment A, leading to control misadjustment, increased unnecessary energy consumption, and hindering the precise implementation of energy-saving oil extraction.

[0056] To address this issue, the dual-station control misalignment sequence at each acquisition time is used as input. A Kalman filter algorithm is employed, with the state transition matrix F=1.0 representing the stable evolution of the system state, the observation matrix H=1.0 representing the direct observation relationship, the process noise covariance Q=0.01 reflecting small internal disturbances, and the observation noise covariance R=0.1. The filtered dual-station control misalignment sequence is output. The mean value of the difference between the dual-station control misalignment sequences at all the same acquisition times before and after filtering is obtained and denoted as the overall degree of change. In this embodiment, the difference between variables is calculated using the absolute value of the difference. The overall degree of change reflects the influence of sensor measurement errors and external interference on coupling strength monitoring. In the process of energy-saving oil extraction, when... A high value indicates that the current monitoring system is severely affected by noise pollution, and the reliability of the liquid level and pressure data decreases.

[0057] Based on the above analysis, changing the valve opening during the liquid level control process at the injection station will cause changes in the external pressure of the preparation station; adjusting the frequency of the external pump during the pressure stabilization process at the preparation station will also affect the liquid level at the injection station. If poor coupling occurs, positive feedback will occur, leading to system instability. This will cause the liquid level and pressure to fluctuate significantly around the set value without being able to converge, reducing production efficiency and energy utilization, and affecting energy-saving oil extraction.

[0058] Therefore, using the filtered bistation control misalignment sequence at each acquisition time as the independent variable and the proportional coefficient adjustment sequence at each acquisition time as the dependent variable, support vector regression (SVR) is used for modeling. The first 30% of historical data is used as the training set, and a penalty parameter C=1.5 is set to balance fitting accuracy and generalization ability. The parameter is insensitive. To control the regression error tolerance, the predicted value of the PID proportional coefficient correction for the dependent variable is output and used as the proportional coefficient adjustment factor. This value represents the correction amount of the control parameters predicted by the model based on historical data of coupling strength. If the value tends to 0, it indicates that the system coupling strength is low, and the control sensitivity needs to be increased. Conversely, it indicates that the coupling is strong, and the control sensitivity should be reduced to avoid oscillation. This helps to achieve more stable and energy-saving oil extraction.

[0059] Based on this, a feedforward compensation factor B for the injection configuration at each acquisition time is constructed. Specifically, the difference between the adjusted proportional coefficient and the original proportional coefficient is obtained and denoted as the correction amount of the PID controller proportional coefficient. The correction amount is then positively fused with the overall degree of change to obtain the feedforward compensation factor for the injection configuration at each acquisition time. In this embodiment, the difference between variables is calculated using the absolute value of the difference, and the positive fusion between variables is calculated using a multiplication method.

[0060] In existing technologies, Support Vector Regression (SVR) is a method that uses historical data to learn and predict continuous variables. The predicted value of the dependent variable output by SVR represents the expected output value corresponding to the best-fit relationship learned based on historical data under given independent variables. Kalman Filtering (KF) is an algorithm used to optimally estimate the true state of a system in noisy environments. Its mean output observation error is used here to characterize the degree of contamination of sensor data; the larger the value, the lower the reliability of the real-time data. On the other hand, in the dual-PID intelligent coordinated regulation of oilfield production preparation stations and injection stations, a perfect control strategy should be combined with the actual situation. It should not only pre-adjust various parameters but also pay attention to the adjustment of the current environmental conditions. Otherwise, ignoring the current data situation when predicting control based on previous data will lead to deviations in feedforward control. If only the actual situation is considered and future prediction and adjustment are ignored, it is difficult to avoid coupled oscillations in the system. Based on this, this application multiplies the predicted output of SVR and the error evaluation result of KF to obtain the feedforward compensation factor for injection preparation, which represents the comprehensive difficulty of overcoming data distortion in the control system during the energy-saving extraction process of oil.

[0061] S304: Adjust the proportional coefficient of the PID control of the injection station based on the feedforward compensation factor.

[0062] Based on the above analysis, an adaptive strategy for the feedback parameters of the flow PID control module is designed again according to the feedforward compensation factor B. The specific adaptive design of the feedback parameters is as follows: Based on the feedforward compensation factor B of the injection preparation at each acquisition time, the proportional coefficient of the pressure control loop of the preparation station is designed as an adaptive parameter that is negatively correlated with the value of B. The specific formula is as follows:

[0063] in: This represents the proportional parameter after adjustment by the PID control of the injection station; The initial proportional parameter is 4 in this embodiment; B is the feedforward compensation factor. The adjustment factor is a constant with a value ranging from 0.4 to 0.5, and in this embodiment, it is specifically set to 0.5; e represents the natural constant.

[0064] The exponential function in the formula describes the decay relationship; its value decreases monotonically as the independent variable increases, making it suitable for constructing inhibitory regulation mechanisms; the constant term... As a bias term, it is used to ensure the baseline level of the proportional coefficient adjustment and prevent excessive weakening of the system response. On the other hand, in the dual PID coordinated regulation of the oilfield injection station and the preparation station, the feedforward compensation factor B comprehensively quantifies the coupled oscillation risk and data reliability crisis faced by the system. The larger the value of B, the greater the comprehensive threat to the system stability. Based on this, this application uses the decay characteristics of the negative exponential function to construct a negative correlation with the value of B. When the value of B is low, When the value of B approaches 1, the system appropriately enhances control sensitivity to improve response speed; when the B value increases, The system automatically reduces the control gain to suppress oscillations, effectively avoiding energy waste caused by over-adjustment of control under complex operating conditions.

[0065] By converting the feedforward compensation factor B into real-time parameter adjustments, the control system possesses intelligent resilience. Under the premise of ensuring the stability of the energy-saving oil extraction process, it can dynamically optimize control performance, directly reducing the ineffective actions and energy losses of actuators such as pumps and valves. This provides key technical support for achieving the ultimate goal of energy-saving oil extraction and ultimately ensures optimal energy efficiency throughout the entire energy-saving oil extraction process.

[0066] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions marked in the blocks may occur in a different order than that shown in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. In the descriptions corresponding to the flowcharts and block diagrams in the accompanying drawings, the operations or steps corresponding to different blocks may also occur in a different order than disclosed in the description; sometimes there is no specific order between different operations or steps. For example, two consecutive operations or steps may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. Each block in a block diagram and / or flowchart, and combinations of blocks in a block diagram and / or flowchart, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0067] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A dual-PID intelligent coordinated control method for oilfield production preparation stations and injection stations, characterized in that, The method includes the following steps: S1: Obtain the initial proportional coefficient, initial integral time, and dead zone parameters of the PID control of the preparation station through simulation testing to control the output of mother liquor from the preparation station; S2: By combining the instantaneous flow rate data of the mother liquor from all single wells with the actual liquid level height, the ideal liquid level height, and the range height of the storage tank in the station, the demand for mother liquor is determined and substituted into the PID control of the injection station to obtain the initial proportional coefficient, initial integral time, and dead zone parameters. S3: Based on all obtained initial parameters, implement dual PID linkage regulation of the injection station and the preparation station: S301: Collects liquid level data from the injection station's storage tank and external pressure data from the preparation station; S302: Analyze the correlation between the liquid level data of the injection station storage tank and the external pressure data of the preparation station, and determine the dual-station control failure at each data acquisition time. S303: Filter all bi-station control misalignment obtained at each acquisition time, analyze the overall change of the bi-station control misalignment before and after filtering, and determine the feedforward compensation factor of the injection configuration at each acquisition time by combining the influence of bi-station control misalignment on the proportional coefficient adjustment. S304: Adjust the proportional coefficient of the PID control of the injection station based on the feedforward compensation factor; The determination of the dual-station control misalignment at each acquisition moment is specifically as follows: Obtain the grey relational degree and maximum wavelet coherence coefficient between the liquid level height data of the injection station storage tank at each acquisition time and all previous acquisition times and the external output pressure data of the preparation station; The result of forward fusion of the gray correlation degree and the maximum wavelet coherence coefficient at each acquisition time is used as the bi-station control outage scheduling at each acquisition time.

2. The dual-PID intelligent coordinated control method for oilfield production preparation station and injection station as described in claim 1, characterized in that, The initial proportional coefficient of the PID control in the preparation station is 4, the initial integral time is 0.01, and the dead zone parameter is 2.

3. The dual-PID intelligent coordinated control method for oilfield production preparation station and injection station as described in claim 1, characterized in that, The formula for determining the required amount of mother liquor is as follows: In the formula, M represents the required amount of mother liquor; P represents the instantaneous total amount of mother liquor in the station; H represents the actual liquid level; h represents the ideal liquid level; and W represents the range height of the storage tank in the station.

4. The dual-PID intelligent coordinated control method for oilfield production preparation station and injection station as described in claim 1, characterized in that, The required amount of mother liquor is used as the setpoint for the PID control of the injection station, representing the desired standard for mother liquor flow rate.

5. The dual-PID intelligent coordinated control method for oilfield production preparation station and injection station as described in claim 1, characterized in that, The initial proportional coefficient of the PID control for the injection station is 3, the initial integral time is 0.02, and the dead zone parameter is 2.

6. The dual-PID intelligent coordinated control method for oilfield production preparation station and injection station as described in claim 1, characterized in that, The grey relational degree is obtained through grey relational degree analysis; the maximum wavelet coherence coefficient is obtained through wavelet coherence analysis.

7. The dual-PID intelligent coordinated control method for oilfield production preparation station and injection station as described in claim 1, characterized in that, The determination of the feedforward compensation factor for the injection configuration at each acquisition time is specifically as follows: For PID control of the injection station, the predicted proportional coefficient is obtained based on the degree of influence, and the difference between the predicted proportional coefficient and the original proportional coefficient is recorded as the correction amount of the PID controller proportional coefficient. The mean value of the difference between the dual-station control outages at all the same acquisition times before and after filtering is obtained and recorded as the overall degree of change. The correction amount is positively integrated with the overall degree of change to obtain the feedforward compensation factor of the injection configuration at each acquisition time.

8. The dual-PID intelligent coordinated control method for oilfield production preparation station and injection station as described in claim 7, characterized in that, The process of obtaining the predicted scaling factor is as follows: The difference between the proportional coefficient before and after the adjustment at each time point is taken as the proportional coefficient adjustment amount; Using all the bi-station control misalignments after filtering at each acquisition time as independent variables and all the proportional coefficient adjustments obtained at each acquisition time as dependent variables, a regression algorithm is used to obtain the predicted PID proportional coefficients.

9. The dual-PID intelligent coordinated control method for oilfield production preparation station and injection station as described in claim 1, characterized in that, The proportional coefficient of the PID control of the injection station is adjusted using the following formula: ;in, B is the initial proportional coefficient for the PID control of the injection station; B is the feedforward compensation factor. This is a preset adjustment factor; e represents the natural constant. This represents the proportional coefficient after the PID control adjustment at the injection station.

Images