A method for co-optimizing the lifespan and oil and gas production of subsea oil production systems.

Abstract

This invention belongs to the field of marine petrochemical technology, and particularly relates to a method for co-optimizing the lifespan and oil and gas production of a marine subsea oil production system. This invention provides a method for co-optimizing the lifespan and oil and gas production of a marine subsea oil production system. By comprehensively assessing the health status of the equipment and production dynamics, it achieves real-time co-optimization control based on lifespan awareness. Furthermore, it is equipped with a steady-state operation and control switching mechanism, which ensures the safe transition of the system even when the health status deteriorates or operational disturbances occur, providing reliable technical support for improving system control accuracy and operational resilience. This method for co-optimizing the lifespan and oil and gas production of a marine subsea oil production system includes the following steps: assessing the health status of the marine subsea oil production system; predicting oil and gas production; co-optimizing the lifespan and oil and gas production of the marine subsea oil production system; and formulating a steady-state real-time control strategy.

Description

Technical Field

[0001] This invention belongs to the field of marine petrochemical technology, and in particular relates to a method for synergistic optimization of the lifespan and oil and gas production of a marine subsea oil production system. Background Technology

[0002] In the development of offshore oil and gas fields, subsea production systems are key equipment, and their performance directly affects the production efficiency and operational safety of oil and gas development. With increasing service life, critical components of the production system (such as subsea wellheads, production trees, and flow manifolds) experience varying degrees of performance degradation, thus impacting the overall system's capacity and reliability. Currently, most existing optimization methods focus on single-objective optimization, such as maximizing production or minimizing energy consumption, neglecting the dynamic synergy between equipment lifespan degradation and production capacity. Furthermore, due to the complex and variable marine environment and the uncertainty of production system operating conditions, technicians need to implement real-time response and steady-state control of the subsea production system's operation to achieve robust production output and sustainable equipment operation.

[0003] However, further research revealed that existing optimization methods lack integrated real-time collaborative optimization strategies, making it difficult to achieve optimal collaborative control of production capacity and equipment health status under multi-source constraints. Therefore, there is an urgent need for those skilled in the art to provide a novel collaborative optimization method that integrates equipment lifespan assessment, dynamic production control, and steady-state real-time optimization to improve the intelligence level and operational efficiency of offshore subsea oil production systems. Summary of the Invention

[0004] This invention provides a method for co-optimizing the lifespan and oil and gas production of a subsea oil and gas production system. This method achieves real-time co-optimization control based on lifespan awareness by comprehensively evaluating the health status and production dynamics of equipment in the subsea oil and gas production system. In addition, this method is equipped with a steady-state operation and control switching mechanism, which can still ensure the safe transition of the system when the health status of the subsea oil and gas production system deteriorates or the operation is disturbed. This provides strong technical support for improving the control accuracy and operational resilience of the subsea oil and gas production system.

[0005] To solve the above-mentioned technical problems, the present invention adopts the following technical solution:

[0006] A method for synergistic optimization of the lifespan and oil and gas production of subsea oil production systems includes the following steps:

[0007] Step S1: Construct a degradation model for the subsea oil production system; dynamically predict the degradation status of each piece of equipment in the subsea oil production system; predict the lifespan of the subsea oil production system.

[0008] Step S2: Identify the key variables affecting oil and gas production; construct an oil and gas production prediction model to obtain oil and gas production prediction results; predict the oil and gas production sequence.

[0009] Step S3: Construct a synergistic optimization objective function for the lifespan and oil and gas production of the marine subsea oil production system, and solve and output the multi-objective optimization results;

[0010] Step S4: Process the control strategy into commands; determine the state feedback mechanism based on model and data fusion; perform real-time rolling optimization and strategy updates; establish a steady-state operation constraint control and switching mechanism for the offshore oil subsea production system.

[0011] Preferably, the process of constructing a degradation model of the offshore oil subsea production system in step S1 is specifically described as follows:

[0012] The degradation characteristics of offshore subsea oil production systems are modeled using a Gamma process model, satisfying the following: ;

[0013] Where S(t) represents the degradation state of the device at time t; α is the shape parameter, representing the degradation rate; β is the scale parameter, representing the degradation fluctuation; and t is time.

[0014] The statistical properties of the Gamma distribution satisfy: ;

[0015] Where D is the failure threshold of the equipment; E[] is the mean of the distribution, and Var[] is the variance of the distribution;

[0016] When S(t)≥D, the device is considered to be faulty.

[0017] Preferably, the process of dynamically predicting the degradation status of each piece of equipment in the offshore oil and gas subsea production system in step S1 is specifically described as follows:

[0018] The particle filter algorithm is used to dynamically predict the degradation status of each piece of equipment in the marine subsea oil production system.

[0019] The system state equations satisfy: The observation equation satisfies: ;

[0020] in, ; This belongs to observation noise; Z t These are the sensor's observations;

[0021] Initialize the particle set using a priori distribution By sampling each particle according to its state transition, we obtain: ;

[0022] in, ;

[0023] Based on the observed values, calculate the observation likelihood number for each particle and update the weights accordingly: ;

[0024] Among them, R t To observe the noise covariance matrix;

[0025] Normalized weights and resampling are used to obtain the state prediction result at the current time step, satisfying: .

[0026] Preferably, the process of predicting the lifespan of the offshore subsea oil production system in step S1 is specifically described as follows:

[0027] Because the Gamma process has independent incremental properties, the lifetime of an offshore subsea oil production system satisfies: ;

[0028] The corresponding cumulative distribution function satisfies: ;

[0029] Where P() is the probability distribution operation, and FGamma is the cumulative distribution function of the Gamma distribution.

[0030] The preferred method, step S2, involves constructing an oil and gas production prediction model to obtain the oil and gas production prediction results. The specific description is as follows:

[0031] Using the random forest regression algorithm, predict oil and gas production in the next cycle;

[0032] Suppose the input feature vector at time t is: ;

[0033] The constructed oil and gas production prediction model satisfies: ;

[0034] Among them, f RF It is the random forest regression function, ε t It is the residual term;

[0035] Collect historical datasets {(x i Q i )} i=1 N The input variables and actual outputs at each time point are included and normalized.

[0036] M regression trees are randomly generated. Each tree is sampled with replacement from the training samples. Each tree node selects the optimal splitting feature from only the feature subset. Each tree is trained independently and outputs a sub-prediction value. ;

[0037] The average of the outputs from each regression tree is taken as the final oil and gas production prediction result, resulting in: .

[0038] More preferably, step S3 involves constructing a synergistic optimization objective function for the lifespan and oil and gas production of the subsea oil and gas production system, and solving and outputting the multi-objective optimization results. This process is specifically described as follows:

[0039] With the goal of maximizing oil and gas production and minimizing equipment lifespan loss within a future rolling time window, the resulting collaborative optimization objective function satisfies: ;

[0040] Where J1 and J2 are objective functions, u t and u k H is the control variable, C is the control sequence, and H is the control input. d ( ) represents the lifetime loss cost function;

[0041] The lifespan loss per unit time satisfies: ;

[0042] The two objectives in the collaborative optimization objective function are combined into a single weighted objective function, resulting in: ;

[0043] Where g1 + g2 = 1, and g1 and g2 are the objective weights; if maximizing oil and gas production is the priority, then let g1 > g2; ​​if minimizing equipment lifespan loss is the priority, then let g1 > g2. <g2;

[0044] Simulated annealing algorithm is used to perform global search optimization on the constructed weighted objective function; let the initial control sequence be U(0), the initial temperature be Te0, and the termination temperature be Te. min Let the cooling factor be b∈(0,1);

[0045] In each iteration, a new candidate control sequence U is generated by perturbing the current control sequence. new ,satisfy: ;

[0046] Among them, U current It is the control sequence at the current moment, and ∆U is the preset small change value;

[0047] Calculate the objective function value J of the candidate solution new Satisfying: J new =J(Unew );

[0048] If J new <J current If the solution is correct, then accept the solution; otherwise, accept the inferior solution with the following probability: ;

[0049] Where Te is the current temperature;

[0050] The search optimization process terminates when the temperature drops below the termination temperature or the maximum number of iterations is reached.

[0051] This invention provides a method for the coordinated optimization of the lifespan and oil and gas production of an underwater oil and gas production system, comprising the following steps: constructing a degradation model of the underwater oil and gas production system; dynamically predicting the degradation state of each piece of equipment in the underwater oil and gas production system; predicting the lifespan of the underwater oil and gas production system; identifying key variables affecting oil and gas production; constructing an oil and gas production prediction model and obtaining the prediction results; predicting the sequence of oil and gas production; constructing an objective function for the coordinated optimization of the lifespan and oil and gas production of the underwater oil and gas production system, solving and outputting the multi-objective optimization results; processing the control strategy into commands; determining a state feedback mechanism based on the model and data fusion; real-time rolling optimization and strategy updating; and establishing a steady-state operation constraint control and switching mechanism for the underwater oil and gas production system.

[0052] The method for synergistic optimization of the lifespan and oil and gas production of subsea oil production systems, which has the above-mentioned steps, has at least the following technical advantages compared with existing technologies:

[0053] (1) The method for co-optimizing the lifespan and oil and gas production of the marine subsea oil production system provided by the present invention realizes real-time co-optimization control based on lifespan perception by integrating the health status and production dynamics of subsea equipment; based on lifespan modeling and health index assessment mechanism, it effectively quantifies the degradation level of key equipment in the marine subsea oil production system, and evaluates the future production trend in combination with the prediction model, providing a more constrained and forward-looking decision basis for the control strategy; using heuristic algorithms such as rolling optimization and simulated annealing, it efficiently searches for control strategies under multiple constraints and multiple variables, realizing a dynamic balance between production improvement and lifespan extension of the marine subsea oil production system.

[0054] (2) The method for co-optimizing the lifespan and oil and gas production of the marine subsea oil production system provided by the present invention can ensure the safe transition and restoration of stability of the system even when the health status of the marine subsea oil production system deteriorates or the operation is disturbed by setting a steady-state operating domain and a control switching mechanism. Therefore, it has significant practical value in improving the control accuracy, economic benefits and operational resilience of the marine subsea oil production system. Attached Figure Description

[0055] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the following drawings:

[0056] Figure 1 A flowchart illustrating the method for synergistic optimization of the lifespan and oil and gas production of an offshore subsea oil production system provided by this invention.

[0057] Figure 2 This is a schematic diagram of the obtained degradation state curve;

[0058] Figure 3 A schematic diagram of the lifespan loss prediction curve;

[0059] Figure 4 This is a diagram showing the comparison between oil and gas production forecasts and actual production figures.

[0060] Figure 5 A schematic diagram illustrating the decrease curve of the target value for simulated annealing optimization;

[0061] Figure 6 A diagram illustrating the strategy for adjusting control variables;

[0062] Figure 7 A schematic diagram illustrating the changes in overall benefits before and after implementing the method for synergistic optimization of the lifespan and oil and gas production of an offshore subsea production system provided by this invention. Detailed Implementation

[0063] This invention provides a method for co-optimizing the lifespan and oil and gas production of a subsea oil and gas production system. This method achieves real-time co-optimization control based on lifespan awareness by comprehensively evaluating the health status and production dynamics of equipment in the subsea oil and gas production system. In addition, this method is equipped with a steady-state operation and control switching mechanism, which can still ensure the safe transition of the system when the health status of the subsea oil and gas production system deteriorates or the operation is disturbed. This provides strong technical support for improving the control accuracy and operational resilience of the subsea oil and gas production system.

[0064] This invention provides a method for synergistic optimization of the lifespan and oil and gas production of subsea oil production systems, such as... Figure 1 As shown, it includes the following steps:

[0065] Step S1: Construct a degradation model for the marine subsea oil production system; dynamically predict the degradation status of each piece of equipment in the marine subsea oil production system; predict the lifespan of the marine subsea oil production system.

[0066] It is worth noting that, before implementing other steps, the present invention first assesses the health status of the subsea oil production system. Specifically, as a preferred embodiment of the present invention, the process of constructing a degradation model of the subsea oil production system in step S1 is described as follows:

[0067] The degradation characteristics of offshore subsea oil production systems are modeled using a Gamma process model, satisfying the following: ;

[0068] Where S(t) represents the degradation state of the device at time t; α is the shape parameter, representing the degradation rate; β is the scale parameter, representing the degradation fluctuation; and t is time.

[0069] The statistical properties of the Gamma distribution satisfy: ;

[0070] Where E[] is the mean of the distribution, Var[] is the variance of the distribution, D is the failure threshold of the equipment, and when S(t)≥D, the equipment is determined to be faulty.

[0071] One point that needs further explanation in this process is that by constructing a degradation model for the subsea oil production system and describing its performance degradation process over time, a theoretical foundation is provided for subsequent steps of state estimation and life prediction. Specifically, by collecting operation and maintenance log data during equipment operation (including cumulative equipment operating time, maintenance records at various time points, replacement time of key components, failure time and corresponding failure mode, etc.), a time-failure correlation is established based on the failure time and cumulative operating time (this correlation can be used to characterize the information when the degradation state reaches the failure threshold). Then, key equipment parameter data collected by various underwater sensors are collected, such as temperature, pressure, vibration intensity, flow rate, torque, current and voltage signals. After processing by feature extraction algorithms, a monotonically increasing degradation feature sequence reflecting the degree of equipment performance degradation can be obtained. In addition, by collecting performance evaluation data of the equipment during actual operation (which includes quantitative indicators reflecting the downward trend of equipment performance, such as the rate of change of liquid production, the delay of opening and closing response time, the energy consumption per unit cycle, and the overall system efficiency), and by normalizing and aligning the above quantitative indicators with time, characteristic variables that increase over time can be extracted and used as observation inputs for the degradation state.

[0072] Based on the completed degradation model of the subsea oil production system, the degradation status of each piece of equipment in the system is further dynamically predicted. It is worth noting that, based on real-time observation data and fully considering factors such as data noise, system nonlinearity, and operational uncertainties, dynamically estimating the degradation status of each piece of equipment in the subsea oil production system enables online tracking of equipment health status.

[0073] Specifically, a particle filter algorithm is used to dynamically predict the degradation state of each piece of equipment in the offshore oil and gas subsea production system. The Gamma process model constructed in the preceding steps of this algorithm serves as the framework for the state transition model, and is combined with three types of heterogeneous observation data (i.e., operation and maintenance log data, sensor monitoring data, and performance index data) for state correction.

[0074] The system state equations satisfy: The observation equation satisfies: ;

[0075] in, ; This belongs to observation noise; Z t These are the sensor's observations.

[0076] One additional point to note is that the Z t The sensor's observations include a fusion of degradation characteristic values ​​and performance index sequence values ​​obtained from the sensor, and state-related label data extracted from the operation and maintenance logs. Further research reveals that the sensor's observation value Z... t Specifically, it consists of the following sub-data vectors: namely, Z t (1) Real-time degradation features such as vibration acceleration and flow rate decrease were extracted from sensor monitoring data; Z t (2) Z represents dynamic performance indicators such as response time and liquid yield in the performance data; t (3) This refers to status events or maintenance tags in the operation and maintenance log data that correspond to the current time window, such as status warnings, minor fault records, etc.

[0077] The observation mapping relationship can be constructed by defining the following fusion function: ;

[0078] H1, H2, and H3 are mapping matrices corresponding to the three types of observation data, respectively. ω1, ω2, and ω3 ∈ [0, 1] are weight coefficients that satisfy ω1 + ω2 + ω3 = 1 and can be dynamically adjusted based on experience.

[0079] Then, the particle set is initialized using a priori distribution. By sampling each particle according to its state transition, we obtain: ;

[0080] in, .

[0081] Based on the observed values, calculate the observation likelihood number for each particle and update the weights accordingly: .

[0082] Among them, R t To observe the noise covariance matrix;

[0083] Normalized weights and resampling are used to obtain the state prediction result at the current time step, satisfying: .

[0084] Based on the dynamic prediction of the degradation status of various equipment in the subsea oil production system, the lifespan of the subsea oil production system is further predicted. It is worth noting that, due to the independent incremental nature of the Gamma process, the lifespan of the subsea oil production system satisfies: .

[0085] The corresponding cumulative distribution function satisfies: ;

[0086] Where P() is the probability distribution operation, and FGamma is the cumulative distribution function of the Gamma distribution.

[0087] Step S2: Identify the key variables affecting oil and gas production; construct an oil and gas production prediction model to obtain oil and gas production prediction results; predict the oil and gas production sequence.

[0088] Specifically, step S2 is used to predict oil and gas production. It is worth noting that identifying and determining the key variables affecting oil and gas production can provide a (data) input basis for building subsequent prediction models.

[0089] Further research reveals that the key variables affecting oil and gas production can be mainly divided into the following three categories:

[0090] (1) Equipment health status variables: such as the current lifespan (RUL) of the equipment. t And the estimated degradation status of the current equipment;

[0091] (2) Operating condition parameter variables: including but not limited to inlet pressure p in,t Export pressure p out,t Wellbore temperature T t Fluid density ρ t and moisture content W t；

[0092] (3) Control parameter variables: for example, the speed v of the downhole electric submersible pump. t and nozzle opening k t The electric submersible pump's rotational speed v t and nozzle opening k t The parameter can be uniformly represented as u t =[v t k t ] T Furthermore, the output variable can be represented as the oil and gas production Q for the current period. t .

[0093] Furthermore, as a preferred embodiment of the present invention, the process of constructing an oil and gas production prediction model and obtaining oil and gas production prediction results in step S2 is specifically described as follows:

[0094] Using the random forest regression algorithm, predict oil and gas production in the next cycle;

[0095] Suppose the input feature vector at time t is: ;

[0096] The constructed oil and gas production prediction model satisfies: ;

[0097] Among them, f RF It is the random forest regression function, ε t It is the residual term;

[0098] Collect historical datasets {(x i Q i )} i=1 N The input variables and actual outputs at each time point are included and normalized.

[0099] M regression trees are randomly generated. Each tree is sampled with replacement from the training samples. Each tree node selects the optimal splitting feature from only the feature subset. Each tree is trained independently and outputs a sub-prediction value. ;

[0100] The average of the outputs from each regression tree is taken as the final oil and gas production prediction result, resulting in: .

[0101] Step S3: Construct a synergistic optimization objective function for the lifespan and oil and gas production of the marine subsea oil production system, and solve and output the multi-objective optimization results.

[0102] Specifically, step S3 is used to synergistically optimize the lifespan and oil and gas production of the offshore subsea oil production system. It should be noted that the synergistic optimization objectives have two key indicators: maximizing oil and gas production and minimizing equipment lifespan loss within a future rolling time window. Therefore, with maximizing oil and gas production and minimizing equipment lifespan loss within a future rolling time window as the synergistic optimization objectives, the resulting synergistic optimization objective function satisfies: ;

[0103] Where J1 and J2 are objective functions, u t and u k H is the control variable, C is the control sequence, and H is the control input. d ( ) represents the lifetime loss cost function;

[0104] The lifespan loss per unit time satisfies: .

[0105] Then, after obtaining the collaborative optimization objective function, a set of optimal control strategies that satisfy engineering constraints are further solved through a multi-objective optimization algorithm.

[0106] Specifically, the two objectives in the collaborative optimization objective function are combined into a single weighted objective function, resulting in: ;

[0107] Where g1 + g2 = 1, and g1 and g2 are the objective weights; if maximizing oil and gas production is the priority, then let g1 > g2; ​​if minimizing equipment lifespan loss is the priority, then let g1 > g2. <g2;

[0108] Simulated annealing algorithm is used to perform global search optimization on the constructed weighted objective function; let the initial control sequence be U(0), the initial temperature be Te0, and the termination temperature be Te. min Let the cooling factor be b∈(0,1);

[0109] In each iteration, a new candidate control sequence U is generated by perturbing the current control sequence. new ,satisfy: ;

[0110] Among them, U current It is the control sequence at the current moment, and ∆U is the preset small change value;

[0111] Calculate the objective function value J of the candidate solution new Satisfying: J new =J(U new );

[0112] If J new <Jcurrent If the solution is correct, then accept the solution; otherwise, accept the inferior solution with the following probability: ;

[0113] Where Te is the current temperature;

[0114] The search optimization process terminates when the temperature drops below the termination temperature or the maximum number of iterations is reached.

[0115] It should be noted that, considering the actual operational characteristics of offshore subsea oil production systems, the system's working environment is characterized by deep water, high pressure, high corrosion, high cost, and long cycles. The core equipment has complex coupling relationships, and the control strategy must simultaneously consider both equipment degradation suppression and production efficiency improvement. In the actual optimization process, the collected data includes at least: the remaining life prediction value output from equipment health assessments, real-time production curves, wellhead and manifold pressures, pump speeds, bottom hole flowing pressures, energy consumption indicators, equipment opening and closing frequencies, and control valve state sequences. Therefore, the collaborative optimization objective function constructed based on the above multi-source data should at least cover two core objectives: maximizing production per unit time and minimizing the cumulative lifespan loss of key equipment (such as ESPs and SCM control valve assemblies).

[0116] In addition, it should be noted that the optimal control strategy output by the simulated annealing algorithm described above can be expressed as: This can be specifically used to guide the generation of real-time control strategies for subsequent steps. The purpose of doing so is not only to meet the multi-objective coordinated optimization requirements of oil and gas production and equipment life, but also to directly guide subsequent real-time operation control, ensuring that the subsea production system achieves the comprehensive operation goals of reasonable energy consumption, equipment steady state, and high production capacity in extreme marine environments.

[0117] Step S4: Process the control strategy into commands; determine the state feedback mechanism based on model and data fusion; perform real-time rolling optimization and strategy updates; establish a steady-state operation constraint control and switching mechanism for the offshore oil subsea production system.

[0118] Specifically, step S4 is used to formulate a steady-state real-time control strategy.

[0119] Specifically, the instruction-based processing of the control strategy refers to mapping the optimized output control vector into executable control commands for the execution units of the offshore oil and gas subsea production system. These control vectors include standardized pump speed commands, valve opening commands, and other key control variables.

[0120] Furthermore, the optimized standardized pump speed value is mapped to actual voltage and current control signals through a motor drive model to control the operating frequency and start-up curve of the downhole electric submersible pump. The valve opening percentage is converted into the angular displacement signal of the actuator through a nonlinear position-angle conversion model, which is used to control key fluid control components such as nozzles, throttle valves, and air lift valves.

[0121] Determining a state feedback mechanism based on model and data fusion means that in each control cycle, the current control input, current observation output, current degradation state prediction value, and current lifetime prediction value are collected in real time; then, based on the model and data fusion (feedback situation), the production prediction trend (error) and lifetime loss situation are re-evaluated, and finally the state feedback mechanism is generated.

[0122] Real-time rolling optimization and strategy update refers to further adopting a rolling time-domain control method, which moves the rolling window forward in each control cycle; then updates the health assessment and yield prediction model inputs.

[0123] In this process, only the control quantity at the first optimization moment is applied to the system, while the rest are reserved as alternatives. The above process is repeated to form a closed-loop control system, which ensures that the marine oil subsea production system of the present invention can still operate in a steady state under conditions of health degradation, external interference or sensor drift.

[0124] Establishing a steady-state operation constraint control and switching mechanism for offshore subsea oil production systems refers to further developing such a mechanism. Subsea production systems are characterized by long-distance operation, unattended operation, high pressure and corrosion, and strong environmental disturbances. Core system equipment includes wellhead Christmas trees, subsea control modules, subsea pump systems, multiphase flow metering devices, and valve actuators. During operation, continuous monitoring and analysis of the system's operating status is achieved by collecting real-time control data such as pump speed, current, voltage, valve opening, wellhead pressure, and production rate, as well as health indices and remaining life estimates output by the equipment health assessment model, and key physical quantities monitored by sensors such as temperature, vibration, and flow rate.

[0125] Based on the aforementioned multi-source data, during the optimization and control strategy execution process, the system's operation is continuously evaluated to ensure it remains within the preset steady-state operating range. This means that production is within the target range, system efficiency is above the threshold, equipment degradation rate is controlled, and there are no potential risk warning indicators exceeding limits. If monitoring results indicate that the system deviates from this steady-state range—for example, abnormal production fluctuations, pump current overload, or health indices falling outside the safe range—a control switching mechanism is immediately triggered to correct the control strategy online. This automatically adjusts the control parameters of key equipment, such as reducing the pump speed of the electric submersible pump and the variation in the opening of the compressor nozzle, and limiting the valve opening and closing frequency. This allows the system to quickly return to the steady-state operating range, preventing equipment damage, drastic production fluctuations, or operational failures caused by control imbalances, thus ensuring the safety and continuity of deep-sea oil and gas production.

[0126] To facilitate understanding of the present invention by those skilled in the art, a set of simulation-based production optimization examples are further provided to verify the effectiveness and reliability of the method for synergistic optimization of the lifespan and oil and gas production of the marine subsea production system provided by the present invention.

[0127] Specifically, the following table provides the various parameters required for the simulation process. The names and values ​​of these parameters can be found in the table below.

[0128] ;

[0129] Then, a degradation model for offshore subsea oil production systems is constructed. (See reference...) Figure 2 As shown, a degradation state curve is generated using the Gamma process model; the quantified indicators are normalized and time-series aligned, where the normalization yields the lifetime loss index Cd, which represents the decline in equipment health over time, as shown in the figure. Figure 3 As shown.

[0130] Then, an oil and gas production prediction model was constructed to predict oil and gas production. Specifically, a production prediction model Q was constructed that couples control variables (pump speed, valve opening) with degradation status and environmental factors. t This captures the combined effect of system operating status on oil and gas production. A comparison chart of the predicted and actual oil and gas production results can be found, for reference... Figure 4 As shown.

[0131] Then, considering both yield and lifetime objectives, the multi-objective optimization results are solved and output. (See reference...) Figure 5 As shown, Figure 5 The curve showing the decrease in the target value after simulated annealing optimization is illustrated. By optimizing the sequence of control variables through simulated annealing, the objective function (negative output plus weighted loss) is minimized, thereby obtaining the optimal control strategy for pump speed and valve.

[0132] Then, a state feedback mechanism based on model and data fusion was determined. When the equipment health index falls below the safety threshold, a corresponding emergency adjustment mechanism is triggered to gently lower the control variables to reduce further wear and tear on the equipment's lifespan. A schematic diagram of the specific control variable adjustment strategy can be found in [reference needed]. Figure 6 As shown.

[0133] Finally, a comparative analysis is conducted on the changes in benefits before and after implementing the method for synergistic optimization of the lifespan and oil and gas production of the offshore subsea production system provided by this invention. (See references...) Figure 7 As shown, the overall benefit before implementing the optimization method (without optimization) was 0.397, and the overall benefit after implementing the optimization method became 0.663. It can be seen that implementing the synergistic optimization method for the lifespan and oil and gas production of the marine subsea oil production system provided by this invention maintains the overall profitability of the marine subsea oil production system while maintaining production (with a slight decrease).

[0134] In summary, the above examples, through numerical simulation, demonstrate the method for synergistic optimization of the lifespan and oil and gas production of the marine subsea oil production system provided by this invention. They show how the optimization strategy achieves adaptive optimization of control variables (pump speed, valve opening) and how emergency adjustments are made through a feedback mechanism, thereby improving the overall efficiency of the marine subsea oil production system.

[0135] This invention provides a method for the coordinated optimization of the lifespan and oil and gas production of an underwater oil and gas production system, comprising the following steps: constructing a degradation model of the underwater oil and gas production system; dynamically predicting the degradation state of each piece of equipment in the underwater oil and gas production system; predicting the lifespan of the underwater oil and gas production system; identifying key variables affecting oil and gas production; constructing an oil and gas production prediction model and obtaining the prediction results; predicting the sequence of oil and gas production; constructing an objective function for the coordinated optimization of the lifespan and oil and gas production of the underwater oil and gas production system, solving and outputting the multi-objective optimization results; processing the control strategy into commands; determining a state feedback mechanism based on the model and data fusion; real-time rolling optimization and strategy updating; and establishing a steady-state operation constraint control and switching mechanism for the underwater oil and gas production system.

[0136] The method for synergistic optimization of the lifespan and oil and gas production of subsea oil production systems, which has the above-mentioned steps, has at least the following technical advantages compared with existing technologies:

[0137] (1) The method for co-optimizing the lifespan and oil and gas production of the marine subsea oil production system provided by the present invention realizes real-time co-optimization control based on lifespan perception by integrating the health status and production dynamics of subsea equipment; based on lifespan modeling and health index assessment mechanism, it effectively quantifies the degradation level of key equipment in the marine subsea oil production system, and evaluates the future production trend in combination with the prediction model, providing a more constrained and forward-looking decision basis for the control strategy; using heuristic algorithms such as rolling optimization and simulated annealing, it efficiently searches for control strategies under multiple constraints and multiple variables, realizing a dynamic balance between production improvement and lifespan extension of the marine subsea oil production system.

[0138] (2) The method for co-optimizing the lifespan and oil and gas production of the marine subsea oil production system provided by the present invention can ensure the safe transition and restoration of stability of the system even when the health status of the marine subsea oil production system deteriorates or the operation is disturbed by setting a steady-state operating domain and a control switching mechanism. Therefore, it has significant practical value in improving the control accuracy, economic benefits and operational resilience of the marine subsea oil production system.

[0139] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for synergistic optimization of the lifespan and oil and gas production of subsea oil production systems, characterized in that, The steps include the following: Step S1: Construct a degradation model for the subsea oil production system; dynamically predict the degradation status of each piece of equipment in the subsea oil production system; predict the lifespan of the subsea oil production system. Step S2: Identify the key variables affecting oil and gas production; construct an oil and gas production prediction model to obtain oil and gas production prediction results; predict the oil and gas production sequence. Step S3: Construct a synergistic optimization objective function for the lifespan and oil and gas production of the marine subsea oil production system, and solve and output the multi-objective optimization results; Step S4: Process the control strategy into instructions; determine the state feedback mechanism based on model and data fusion; Real-time rolling optimization and strategy updates; establishing a steady-state operation constraint control and switching mechanism for offshore subsea oil production systems; The process of constructing a degradation model of the offshore oil subsea production system in step S1 is specifically described as follows: The degradation characteristics of offshore subsea oil production systems are modeled using a Gamma process model, satisfying the following: ; Where S(t) represents the degradation state of the device at time t; α is the shape parameter, representing the degradation rate; β is the scale parameter, representing the degradation fluctuation; and t is time. The statistical properties of the Gamma distribution satisfy: ; Where E[] is the mean of the distribution, Var[] is the variance of the distribution, and D is the failure threshold of the equipment; When S(t)≥D, the device is considered to be faulty; The process of constructing the oil and gas production prediction model and obtaining the oil and gas production prediction results in step S2 is specifically described as follows: Using the random forest regression algorithm, predict oil and gas production in the next cycle; Suppose the input feature vector at time t is: ; The constructed oil and gas production prediction model satisfies: ; Among them, f RF It is the random forest regression function, ε t It is the residual term; Collect historical datasets The input variables and actual outputs at each time point are included and normalized. M regression trees are randomly generated. Each tree is sampled with replacement from the training samples. Each tree node selects the optimal splitting feature from only the feature subset. Each tree is trained independently and outputs a sub-prediction value. ; The average of the outputs from each regression tree is taken as the final oil and gas production prediction result, resulting in: .

2. The method for synergistic optimization of the lifespan and oil and gas production of a marine subsea oil production system according to claim 1, characterized in that, The process of dynamically predicting the degradation status of each piece of equipment in the offshore oil and gas subsea production system in step S1 is specifically described as follows: The particle filter algorithm is used to dynamically predict the degradation status of each piece of equipment in the marine subsea oil production system. The system state equations satisfy: ; The observation equation satisfies: ; in, ; This belongs to observation noise; Z t These are the sensor's observations; Initialize the particle set using a priori distribution By sampling each particle according to its state transition, we obtain: ; in, ; Based on the observed values, calculate the observation likelihood number for each particle and update the weights accordingly: ; Among them, R t To observe the noise covariance matrix; Normalized weights and resampling are used to obtain the state prediction result at the current time step, satisfying: .

3. The method for synergistic optimization of the lifespan and oil and gas production of an underwater oil production system according to claim 1, characterized in that, The process of predicting the lifespan of the offshore subsea oil production system in step S1 is specifically described as follows: Because the Gamma process has independent incremental properties, the lifetime of an offshore subsea oil production system satisfies: ; The corresponding cumulative distribution function satisfies: ; Where P() is the probability distribution operation, and FGamma is the cumulative distribution function of the Gamma distribution.

4. The method for synergistic optimization of the lifespan and oil and gas production of a marine subsea oil production system according to claim 1, characterized in that, Step S3 involves constructing a co-optimization objective function for the lifespan and oil and gas production of an offshore subsea oil production system, and solving and outputting the multi-objective optimization results. The specific description is as follows: With the goal of maximizing oil and gas production and minimizing equipment lifespan loss within a future rolling time window, the resulting collaborative optimization objective function satisfies: ; Where J1 and J2 are objective functions, u t and u k H is the control variable, C is the control sequence, and H is the control input. d ( ) represents the lifetime loss cost function; The lifespan loss per unit time satisfies: ; The two objectives in the collaborative optimization objective function are combined into a single weighted objective function, resulting in: ; Where g1 + g2 = 1, and g1 and g2 are the objective weights; if maximizing oil and gas production is the priority, then let g1 > g2; ​​if minimizing equipment lifespan loss is the priority, then let g1 > g2. <g2; Simulated annealing algorithm is used to perform global search optimization on the constructed weighted objective function; let the initial control sequence be U(0), the initial temperature be Te0, and the termination temperature be Te. min Let the cooling factor be b∈(0,1); In each iteration, a new candidate control sequence U is generated by perturbing the current control sequence. new ,satisfy: ; Among them, U current It is the control sequence at the current moment, and ∆U is the preset small change value; Calculate the objective function value J of the candidate solution new Satisfying: J new =J(U new ); If J new <J current If the solution is correct, then accept the solution; otherwise, accept the inferior solution with the following probability: ; Where Te is the current temperature; The search optimization process terminates when the temperature drops below the termination temperature or the maximum number of iterations is reached.

Images