Self-adaptive dynamic positioning control method for semi-submersible platform

Abstract

The invention relates to the technical field of self-adaptive control, in particular to a self-adaptive dynamic positioning control method for a semi-submersible platform, which establishes a control model for the motion characteristics of the semi-submersible platform, pays attention to the nearest error data through a sliding window, avoids the influence of long-term accumulated errors on a controller, and improves the positioning accuracy of the semi-submersible platform. The response speed and the stability of the control system are improved; the PSO algorithm combined with the real-time wave parameters can effectively restrain the search space, blind optimization is avoided, the control precision and adaptability of the control strategy under different sea conditions are improved, and the response speed and safety are greatly improved especially for extreme sea conditions.

Description

Technical Field

[0001] The present invention relates to the technical field of adaptive control, and in particular to an adaptive dynamic positioning control method for a semi-submersible platform. Background Art

[0002] In the development and progress of social economy, resources such as oil and natural gas are essential foundations for promoting development. However, with the surging demand for energy, the main onshore oil fields have successively entered the stage of production decline. Therefore, marching into the ocean and accelerating the development of offshore oil and gas resources is a major trend in the future energy strategy. Currently, countries around the world are methodically deploying strategic measures for the development of offshore oil and gas resources in shallow waters and gradually moving towards deep sea areas. Deep sea drilling production platforms are important equipment for entering deep sea oil and gas. A semi-submersible platform is a common deep sea drilling production platform, generally composed of a platform main body, columns, and a lower floating body. Its advantages are reflected in strong stability, small wave influence, strong wind resistance, large deck area and variable load, and a wide range of applicable water depths. The semi-submersible platform is a key research object for future deep sea oil and gas development.

[0003] However, since the semi-submersible platform will serve in harsh sea conditions for a long time, in a changing marine environment, mooring chains are extremely prone to failure events under environmental loads such as ocean currents and waves, as well as corrosion and wear. After the mooring fails, the platform will experience large drifts, which will have a serious impact on the overall platform system, resulting in production losses and economic losses. In order to resist the disasters brought by mooring failure events, semi-submersible platforms are generally equipped with a dynamic positioning system to assist the platform in remaining within a safe displacement in the event of extreme disasters. And with the increase in water depth, the installation and maintenance costs of traditional mooring positioning systems will be greatly increased, but the cost of the dynamic positioning system will not increase with water depth. Therefore, the mooring method of mooring plus power system will be the development direction of future deep water floating platforms.

[0004] At present, some scholars have studied and discussed the control strategies for dynamic positioning systems, but there is less research on the control strategies for the dynamic positioning systems of semi-submersible platforms. And traditional PID control has many defects and deficiencies, and it is difficult to provide optimal control effects for non-linear systems and parameters that change over time, and has poor dynamic responsiveness and low system stability. In view of the shortcomings and problems of traditional PID, combined with the motion response characteristics of semi-submersible platforms, an adaptive control strategy method for semi-submersible platforms is proposed. Summary of the Invention

[0005] Aiming at the deficiencies of the above-mentioned existing technologies, the purpose of the present invention is to provide an adaptive dynamic positioning control method for a semi-submersible platform, aiming to solve the problems that the traditional PID control strategy responds untimely or over-responds after the mooring failure of the semi-submersible platform and cannot make timely implementation strategy adjustments according to sea conditions.

[0006] According to the implementation scheme of the present invention, the provided scheme is as follows: An adaptive dynamic positioning control method for a semi-submersible platform, S1. Establish a hydrodynamic model to obtain the displacement and velocity data of the platform; S2. Apply a control strategy to the thrusters based on the platform state after mooring failure. The steps include: Through non-linear error feedback and sliding mode feedback, according to the current displacement or velocity error, use a non-linear saturation function to optimize the control output; Estimate external disturbances and model uncertainties through state observation and compensate them to update the system state of the platform; Receive the updated system state data and real-time sea condition data, and use the particle swarm optimization algorithm to dynamically adjust the sliding window size; Dynamically adjust the control parameters according to the error and error change rate within the window; S3. Apply thrust through the thrusters according to the dynamically adjusted control parameters to move the platform to the target position.

[0007] Further, in step S1, the steps of establishing the hydrodynamic model include: Receive the structural parameters and environmental load data of the platform, and through hydrodynamic model calculation, obtain the displacement and velocity data in the mooring failure state to provide the initial state of the control strategy. The hydrodynamic model is established based on the time-domain motion equation and adopts the convolution integral form.

[0008] Further, the structural parameters include the mass matrix and the added mass matrix, and the environmental loads include wave force, wind force, current force, mooring force and thruster thrust.

[0009] Further, the time-domain motion equation includes the structural mass matrix, added mass matrix, damping matrix, stiffness matrix and external force terms. The external force terms include first-order wave force, second-order wave force, current force, wind force, mooring force and thruster thrust.

[0010] Further, in step S2, the steps of the non-linear error feedback and sliding mode feedback include: Receive the data of the error and error change rate between the current displacement and velocity and the target values, Through non-linear error feedback and sliding mode feedback, use a non-linear saturation function to optimize the output control signal, Nonlinear saturation function Adopt the following formula In the formula, represents the input error is the nonlinear coefficient, used to adjust the influence degree of the error on the control output is the set threshold parameter The optimized control signal avoids high-frequency chattering and improves the response speed

[0011] Furthermore, in step S2, the control system is subjected to disturbance estimation and system state compensation through the disturbance prediction and compensation module Receive the error and the control signal, estimate the external disturbance and update the state estimation through the established disturbance prediction and compensation model, and output the position and speed estimation values Estimate the unknown disturbance and model uncertainty by real-time observing the state of the system to ensure the control accuracy Update the state estimation through the recurrence formula, and add the information contained in the error and the control signal to the extended state equation The calculation formula for the updated estimated value of the position state is as follows , The formula is iteratively updated during the calculation process In the formula, represents the estimated value of the system at the current displacement state represents the estimated value of the displacement at the previous moment represents the compensation, the time discretization step size represents the estimated value of the speed at the previous moment represents the position observer gain parameter, used to control the convergence speed of the estimation error represents the error of the displacement at the current moment is the current actually measured displacement The difference between the displacement estimated at the previous moment The calculation formula for the updated estimated value of the speed state is as follows , The formula is iteratively updated during the calculation process In the formula, represents the estimated value of the speed of the system at the current moment represents the estimated value of the speed at the previous moment represents the estimated value of the total disturbance at the previous moment represents the speed observer gain parameter represents the control input coefficient Represents the control input at the previous moment.

[0012] Further, in S2, a sliding window model is established, and the particle swarm optimization algorithm is combined with real-time sea condition characteristics to dynamically adjust the size of the sliding window, improving environmental adaptability; The operation steps of the sliding window model include: Collect error values and sea condition data in real time. The sea condition data includes wave height, spectral peak period, and error data, and the errors are stored in the sliding window. The sliding window limits the number of stored errors to ensure that the stored errors are data within the latest time period. When the number of errors in the window reaches or exceeds the preset size, calculate the standard deviation of the errors in the window. Calculate the wave force and sea condition dynamic index according to the JONSWAP spectrum. Use the particle swarm optimization algorithm to optimize the window size and constrain the window range according to the wave period.

[0013] Further, the fitness function used in the particle swarm optimization algorithm includes the sum of squared errors, control force energy, and wave coupling term; The inertia weight is dynamically adjusted according to the wave energy flux density, and the inertia weight is used to update the particle velocity. The window size is constrained by the main wave period.

[0014] Further, the steps of dynamically adjusting the PID gain using fuzzy logic in step S2 include: Obtain the mean value and change rate of the errors in the window. Normalize the errors and change rate, and divide the fuzzy sets through triangular membership functions. Use the centroid method for defuzzification to adjust K p 、K i 、K d ; Use fuzzy logic to obtain the change value of the gain parameter and update the dynamically adjusted control parameters.

[0015] Further, set the upper and lower limits of the gain during the process of adjusting the control parameters to prevent slow response caused by too low gain and avoid oscillations caused by too high gain.

[0016] Compared with the prior art, the beneficial effects of the technical solution provided by this application.

[0017] One or more of the above technical solutions in the embodiments of the present invention have at least one of the following technical effects: 1. Establish a dynamic positioning control method suitable for semi-submersible platforms according to the motion characteristics of semi-submersible platforms, providing an accurate state basis for the control strategy of semi-submersible platform dynamic positioning; 2. After adding a sliding window: Through the setting of the sliding window, the system only focuses on the recent error data, avoiding the impact of long-term accumulated errors on the controller, thereby improving the response speed and stability of the control system; 3. The PSO algorithm combined with real-time wave parameters can correctly constrain the search space, avoiding blind optimization; it improves the control accuracy and adaptability of the control strategy under different sea conditions, especially significantly enhancing the safety in extreme sea conditions; 4. Fuzzy logic replacing the ordinary standard deviation threshold method or the simple proportional adjustment parameter method is more in line with the dynamic reality. Especially for a semi-submersible platform in a multi-disturbance and strong non-linear environment, this method can significantly improve the control accuracy and adaptability; 5. Adaptive control makes the control strategy more in line with the actual operation process of the semi-submersible platform, improving the control accuracy and adaptability. The dynamic window also avoids ineffective thrust against wave inertia, reducing the energy consumption of the thruster and significantly enhancing the economy. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0019] Among them: Figure 1 It is a flowchart of an adaptive dynamic positioning control method for a semi-submersible platform system in one embodiment; Figure 2 It is a comparison diagram of the displacement recovery after the mooring failure of a semi-submersible platform in one embodiment; Figure 3 It is a comparison diagram of the error values between the actual displacement and the expected displacement of a semi-submersible platform applying an adaptive dynamic positioning control strategy and a traditional PID control strategy in one embodiment; Figure 4 It is a comparison diagram of the thrust of a semi-submersible platform applying an adaptive dynamic positioning control strategy and the thrust of a traditional PID control strategy in one embodiment. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0020] To enable those skilled in the art to better understand the technical solutions in this application, the following will clearly and completely describe the technical solutions in the embodiments of this application with reference to the accompanying drawings in the embodiments of this application. Obviously, the described embodiments are only a part of the embodiments of this application, rather than all the embodiments. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the scope of protection of this application.

[0021] The following further explains and illustrates the present invention in combination with specific implementation manners.

[0022] As Figure 1 shown, the control flow of a control strategy method for the adaptive dynamic positioning of a semi-submersible platform proposed in this application is shown.

[0023] The development of this method is based on the actual application scenario of a certain platform in the South China Sea. The steps of the adaptive dynamic positioning control strategy method include: S1, establish a hydrodynamic model and obtain the required displacement and velocity data, Establish the hydrodynamic model of the platform. Since the floating platform exhibits significant nonlinear motion and coupling effects in random waves, we select the time-domain motion equation to accurately simulate the transient dynamic behavior. The convolution integral method is used to convert the added water mass and potential flow damping varying with frequency into the impulse response function, expressed as the following formula: In the formula, is the mass matrix of the structure; is the fluid added mass matrix when the frequency approaches infinity; is the damping matrix other than the linear radiation damping generated by the diffraction unit; represents the hydrostatic stiffness matrix, is the external force acting on the structure, including the first-order wave force, second-order wave force, current force, wind force, mooring force, and thruster thrust.

[0024] refers to the acceleration impulse function matrix, and its specific formula is: In the formula, represents the added mass matrix, is the hydrodynamic radiation damping matrix.

[0025] is the angular frequency, is the high-frequency limit response.

[0026] Obtain the hydrodynamic response data of the platform through the hydrodynamic model, and obtain the response data of the semi-submersible platform in the mooring failure state and the response data after applying the adaptive thruster.

[0027] Accurate initial state data is required to ensure that the control input is based on the true physical response. Then, by simulating the multi-source external forces in the deep-sea environment, the motion characteristics of the semi-submersible platform can be accurately reflected, providing a theoretical basis for the control after mooring failure.

[0028] The model relied on in this application is specifically designed for semi-submersible platforms. Considering its strong stability and strong resistance to wind and waves, etc., the applicability of the control strategy is improved.

[0029] S2. Implementation of the thruster control strategy after mooring failure When the platform offset exceeds the set threshold, it is determined that the mooring has failed.

[0030] S21. Nonlinear error feedback and sliding mode feedback After applying control, there is an error between the displacement or velocity of the platform under the current control and the expected value. The nonlinear error feedback module will adjust based on the error and the change rate of the error. Through the nonlinear control strategy, the nonlinear error feedback module responds more sensitively to the change of the error, so as to make effective compensation when the system error is large.

[0031] In specific implementation, selecting the position-type PID can better meet the requirements of system stability and effectively adjust the system error, with strong adaptability. Its formula is: In the formula, represents the control quantity at the current moment; represents the displacement or velocity error value of the semi-submersible platform generated by the current control; represents the displacement or velocity error value of the semi-submersible platform generated by the control at the previous moment.

[0032] K P is the proportional gain, K I is the integral gain, K D is the derivative gain.

[0033] is a non-linear function, and its meaning is as follows: To avoid the high-frequency chattering phenomenon, a non-linear saturation function is defined, which focuses on the adjustment of the instantaneous response. Improve the response characteristics of the error in the PID control. When the error is large, non-linear adjustment is adopted, and when the error is small, linear adjustment is adopted. So that the controller can respond faster when the error is large, and reduce the change of the control quantity when the error is small to avoid over-adjustment and ensure smoothness.

[0034] In the formula, represents the input error, that is, the difference between the expected value and the actual value, indicating the error of the system at the current moment. is the nonlinear coefficient, which is used to adjust the influence degree of the error on the control output. is the set threshold parameter.

[0035] When the error whose absolute value is greater than the threshold, at this time the output value is proportional to the power of the absolute value of the error. When the absolute value of the error is less than the threshold, at this time the output value has a linear relationship with the error, and the error will be scaled to reduce the influence of the error on the control output.

[0036] Elastically adjust the nonlinear feedback strength according to the size of the error. The nonlinear error feedback provided by the nonlinear error feedback module can play a supplementary role in PID control, ensuring that when PID control is insufficient to handle large nonlinear errors, the system performance can be improved through nonlinear feedback.

[0037] S22, the disturbance prediction and compensation module performs error prediction and compensation. The disturbance prediction and compensation module plays the role of disturbance estimation and system state compensation in the control system.

[0038] By observing the state of the system in real time, estimate the unknown disturbance and model uncertainty. This method helps the system to cope with uncertainty, ensure the control accuracy, and improve the system robustness.

[0039] If the system is subject to external disturbances or there are errors in the model.

[0040] Set the state equation of the system as: represents the extended state vector, which includes the state of the system and the estimate of the disturbance. represents the estimate of the disturbance.

[0041] z(t): The state vector of the system, which includes all independent variables describing the internal dynamics of the system.

[0042] u(t): The input vector of the system, representing the externally applied control or disturbance signal.

[0043] A state matrix, which describes the dynamic evolution relationship of the internal state of the system.

[0044] The B input matrix describes the influence of the input on the state change.

[0045] Update the state estimate through a recurrence formula, adding the information contained in the input signal to the extended state equation based on the error observed by this module. The calculation formula for the estimated value of the position state after update is as follows: This formula is iteratively updated during use. In the formula, represents the estimated value of the system at the current displacement state. represents the estimated value of the displacement at the previous moment. represents the compensation, the step size of time discretization. represents the estimated value of the velocity at the previous moment. represents the position observer gain parameter, which is used to control the convergence speed of the estimation error. represents the error of the displacement at the current moment. , the currently actually measured displacement is the difference between the displacement estimated at the previous moment.

[0046] The calculation formula for the estimated value of the velocity state after update is as follows: This formula is iteratively updated during the calculation process. In the formula, represents the estimated value of the velocity of the system at the current moment. represents the estimated value of the velocity at the previous moment. represents the estimated value of the total disturbance at the previous moment. represents the velocity observer gain parameter, and fal is the non - linear saturation function defined in step S2, which enhances the robustness of the system. represents the control input coefficient. represents the control input at the previous moment.

[0047] By capturing unknown disturbances such as waves and wind in real - time, improve the system's adaptability to uncertainties. By iteratively updating the state estimate, reduce the influence of model error on control accuracy; use the non - linear function fal to enhance the system's anti - interference ability in complex environments. The non - linear error feedback will cooperate with the subsequent PSO dynamic window to provide the state data after disturbance compensation and optimize the control decision.

[0048] S23. Use the Particle Swarm Optimization (PSO) algorithm to optimize the selection of the dynamic window. Since the operating location of the semi-submersible platform is in the deep sea, the environment is harsh and uncertain. Therefore, the influence of waves should be considered in the control process and incorporated into the control strategy to better adapt to different sea conditions and improve the survival ability in extreme sea conditions.

[0049] In this application, the sea conditions are combined with the dynamic window size, and the dynamic window senses the real-time sea condition characteristics, thus significantly improving the environmental adaptability. The specific operation steps are as follows: Collect error values in real time: Observe the current displacement value and velocity value of the platform in real time, and collect the sea condition values where the platform is located at present, including the wave spectral peak period, wave height, etc., and store them in the sliding window. The sliding window limits the number of stored errors to ensure that the stored errors are data in the latest time period; The working logic of the sliding window is that when the number of errors in the window reaches the preset maximum value, the window will slide and clear the old data, and store the core data sets of control decisions such as tracking errors, control forces, time, wave force estimates, etc. When the number of errors in the window reaches or exceeds the preset size, the standard deviation of the error values will be calculated at this time to measure the data fluctuation of the errors.

[0050] The energy density model of the wave spectrum adopts the formula of the JONSWAP spectrum as follows: where represents the wave frequency; represents the spectral peak frequency, generally determined by the wind speed and the fetch length ; , represents the energy scale parameter related to the wave height ; , represents the spectral peak enhancement factor, represents the spectral peak width.

[0051] Therefore, the equivalent wave force received by the platform is: represents the equivalent wave force received by the platform; represents the energy spectrum of the wave, which describes the energy distribution of the wave at each frequency. represents the hydrodynamic coefficient of the platform; represents the seawater density; represents the gravitational acceleration, represents the added mass coefficient.

[0052] Define the sea condition dynamic index based on the above wave spectrum energy and direction distribution: In the formula is the distribution function representing the wave direction, is the angle of the direction; represents the sea state dynamic index.

[0053] Using PSO data-driven collaborative optimization to dynamically adjust the window size. First, use the existing physical model to guide the initialization, and use the wave parameters to initialize the particle swarm of PSO, thereby reducing the search space and improving the efficiency.

[0054] The fitness function can be designed as: The first term is the error term, representing the integral square sum of the errors within the window , the second term is the energy term, representing the energy of the control force output, , the third term is the wave coupling term, whose meaning is to penalize the time period of the wave force change rate and the error phase within the window, .

[0055] represents the error weight, represents the energy weight, represents the wave coupling weight. is the time step index, is the window size. The error term represents the cumulative square sum of the system tracking errors within the specified time window, and its formula is , where represents the error at time step . is the energy term, representing the energy consumption of the control force, and its formula is , represents the energy output at time step . is the wave coupling term, to penalize the time period of the wave force change rate and the error phase within the window, and its formula is , is the equivalent wave force on the platform at time

[0056] The inertia weight is dynamically adjusted according to the wave energy flux density: ω represents the inertia weight, ω min and ω max are the minimum and maximum values of the inertia weight respectively. P represents the wave energy flux density. P min and P max are the minimum and maximum values of the wave energy flux density P respectively, used to normalize the value of P.

[0057] Calculation formula for wave energy flux density P: Density of seawater; Acceleration due to gravity; Significant wave height, which usually represents the effective height of the wave; Peak period of the spectrum, indicating the periodicity of the wave.

[0058] Particle velocity update: represents the velocity of the i-th particle at time t + 1, represents the velocity of the i-th particle at time t, represents the inertia weight, which can control the global search ability, , represents the learning factor, , represents a random number used to introduce the randomness of the search; and represents the wave coupling coefficient; represents the sensitivity of the window size to the sea state index.

[0059] The dynamic window constraint range should be based on the real-time wave characteristics: represents the main period of the wave, represents the empirical coefficient, represents the control period, is the wave height at time t, is the peak frequency of the spectrum at time t.

[0060] Through the real-time operation and iteration of PSO, the fitness function integrates the wave coupling term, updates the window size and stores the error data in a rolling manner.

[0061] The window is dynamically adjusted so that it can be shrunk or expanded when the sea state changes to better match the wave period. When the sea state is severe, under large wave heights and high-frequency waves, the error changes rapidly and the window is shrunk for the controller to respond quickly, while when the sea state is in a calm period, the error changes slowly and the window should be enlarged to suppress noise.

[0062] Suppose the semi-submersible platform is in an extremely harsh environment. At this time, the dynamic adjustment window senses the harsh sea conditions and will adopt a strategy of shrinking the window. The system tracks changes within a shorter time window, avoiding environmental mutations in old data. In the case of a fixed window, it takes 10 steps to fully refresh the data, and at this time, the gain adjustment is delayed. However, with a dynamic window, it only takes 3 seconds to update all the data, and the response speed is greatly improved.

[0063] S24, Dynamically adjust the gain of PID based on fuzzy logic. In S23, the size of the dynamic window and the errors and error change rates stored in the window are given to the control system as signals, indicating the current severity of the environment and the change situation, so as to help the system select the next control parameters.

[0064] When the mean value of the error and the error change rate stored in the window are large, the error is in a state of rapid change, indicating that the system is undergoing large fluctuations. This usually means that the system is greatly affected by external disturbances or the response of the controller needs to be accelerated. It can be achieved by increasing the proportional gain (Kp) and the derivative gain (Kd). Kp can help adjust the output more quickly and reduce the system deviation. Kd can predict future error changes in advance by increasing the response to the error change rate and reduce overshoot or oscillation.

[0065] When the mean value of the error and the error change rate stored in the window are small, the error is in a state of slow change, indicating that the error of the system has tended to be stable or fluctuates within a small range. At this time, rapid response is not required, but rather the overreaction of the controller should be minimized as much as possible. The adjustment method is to reduce the proportional gain (Kp) and the derivative gain (Kd) to avoid unnecessary fluctuations in the system. And the integral gain (Ki) can be increased synchronously to gradually eliminate the residual static error and avoid the system deviating from the target. The error change rate is determined by calculating the error change rate of the platform movement without adding dynamic parameter adjustment, and the PID control parameters are gradually updated.

[0066] Fuzzy logic gain adjustment dynamically adjusts PID parameters according to the analysis of the sliding window error results.

[0067] The calculation formula for the error change rate is: represents the error of displacement or velocity at the current moment. represents the error of displacement or velocity at the previous moment.

[0068] The mean value of the error within the window and the change rate of the error calculated from the data stored in the dynamic adjustment window are used to set the gain adjustment factor. The basic rules of fuzzy logic are shown in the table:

[0069] First, normalize the error and the rate of change of error. The error range is determined by the platform tolerance. Assume the error range is Normalize it to the range . The rate of change of error range is determined by the platform dynamic response. Assume the rate of change of error range is : Use the triangular membership function to divide the fuzzy set. Then the membership function of the error and the membership function of the rate of change of error are: Negative represents negative large, zeor represents zero, Positive represents positive large. Use the centroid method to defuzzify: represents the proportional gain, is the value for adjusting the proportional gain (increase, decrease, or remain unchanged), represents the membership degree corresponding to this point. Similarly, for the derivative gain and the integral gain also have the same formula: represents the derivative gain, is the value for adjusting the derivative gain, represents the integral gain, represents the value for adjusting the integral gain.

[0070] After obtaining the change values of the gain parameters through fuzzy logic, update the three parameters of the PID respectively using the following formula: In the formula represents the original proportional gain value, represents the updated proportional gain value, represents the finally determined proportional gain value after limited selection.

[0071] In the formula represents the original derivative gain value, represents the updated derivative gain value, Represents the final determined differential gain value after limited selection.

[0072] In the formula Represents the original integral gain value, Represents the updated integral gain value, Represents the final determined integral gain value after limited selection.

[0073] During the process of parameter substitution, the upper and lower limits of the gain can also be limited to prevent slow response caused by too low gain and avoid oscillations caused by too high gain.

[0074] S3. According to the optimized control parameters, apply thrust through the thrusters to make the semi-submersible platform return to the target position.

[0075] Such as Figure 2 As shown, the semi-submersible platform will have a failure event under extreme wind and wave conditions. The failure time is 1900s. At this time, the offset displacement of the semi-submersible platform gradually rises and there is a risk of out-of-position, resulting in a complete interruption of production of the semi-submersible platform. After applying the adaptive PID control at 1980s, the displacement of the platform quickly returns to the state before the mooring failure, the recovery time is fast, and the offset during the recovery process is smaller than that of the traditional control strategy.

[0076] Such as Figure 3 As shown, compared with the traditional PID control strategy, the error value between the actual displacement and the expected displacement of the platform after applying the adaptive dynamic positioning control strategy shows that the displacement deviation of the adaptive one is smaller and more accurate.

[0077] Such as Figure 4 As shown, it shows the comparison of the thrust of the thrusters. Due to the influence of wind, wave and current, the semi-submersible platform is in six-degree-of-freedom motion. The thrust can also quickly respond to changes in the environment and change the magnitude and direction of the thrust. Moreover, the thrust of the adaptive PID is much smaller than that of the traditional PID. The optimization of the control strategy enables the platform to maintain its displacement without a large thrust, reducing the overall energy consumption and increasing the economy.

[0078] The inventive concept is reasonable. The dynamic positioning control strategy method of the semi-submersible platform constructed is different from the traditional PID control strategy. It adds a non-linear error feedback module, a disturbance prediction and compensation module, a non-linear error feedback module, and a dynamic parameter tuning module, further optimizing the control strategy and avoiding the influence of long-term accumulated errors on the controller; increasing the adaptability of the controller and having a more flexible response method when facing different environmental changes and external disturbances; enhancing the control accuracy of the controller. At the same time, this method is also a dynamic positioning control strategy method established based on the motion characteristics of the semi-submersible platform and adapted to the semi-submersible platform; it fully combines the actual offshore operation status and complex ocean conditions, and realizes an efficient and energy-saving adaptive positioning strategy by specifically obtaining and optimizing the collected parameters, providing a reference for the control strategy of the dynamic positioning of the semi-submersible platform.

[0079] The technical features of the above embodiments can be combined arbitrarily. For the sake of concise description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope described in this specification.

[0080] The above-described embodiments only represent several implementation manners of the present application. The description is relatively specific and detailed, but it should not be construed as a limitation on the patent scope of the present application. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several deformations and improvements can still be made, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application should be subject to the appended claims.

Claims

1. An adaptive dynamic positioning control method for a semi - submersible platform, characterized in that, S1, establish a hydrodynamic model and obtain the displacement and velocity data of the platform; S2, apply a control strategy to the thrusters based on the platform state after mooring failure. The steps include, Through non - linear error feedback and sliding - mode feedback, according to the current displacement or velocity error, use a non - linear saturation function to optimize the control output; Estimate external disturbances and model uncertainties through state observation and compensate, and update the system state of the platform; Receive the updated system state data and real - time sea condition data, and use the particle swarm optimization algorithm to dynamically adjust the sliding window size; Dynamically adjust the control parameters according to the error and error change rate within the window; S3, according to the dynamically adjusted control parameters, apply thrust through the thrusters to move the platform to the target position.

2. The adaptive dynamic positioning control method for a semi - submersible platform according to claim 1, characterized in that, In step S1, the steps of establishing the hydrodynamic model include: Receive the structural parameters and environmental load data of the platform, and through hydrodynamic model calculation, obtain the displacement and velocity data in the mooring - failure state to provide the initial state of the control strategy, The hydrodynamic model is established based on the time - domain motion equation and adopts the form of convolution integral.

3. The adaptive dynamic positioning control method for a semi - submersible platform according to claim 2, characterized in that, The structural parameters include the mass matrix and the added - mass matrix, and the environmental loads include wave force, wind force, current force, mooring force and thruster thrust.

4. The adaptive dynamic positioning control method for a semi - submersible platform according to claim 2, characterized in that, The time - domain motion equation includes the structural mass matrix, added - mass matrix, damping matrix, stiffness matrix and external - force term. The external - force term includes first - order wave force, second - order wave force, current force, wind force, mooring force and thruster thrust.

5. The adaptive dynamic positioning control method for a semi - submersible platform according to claim 1, characterized in that, In step S2, the steps of the non - linear error feedback and sliding - mode feedback include: Receive the data of the error and error change rate between the current displacement and velocity and the target values, Through non - linear error feedback and sliding - mode feedback, use a non - linear saturation function to optimize the output control signal, Nonlinear saturation function Adopt the following formula wherein, represents the input error, The non - linear coefficient is used to adjust the influence degree of the error on the control output. is the set threshold parameter, The optimized control signal avoids high - frequency chattering and improves the response speed.

6. The adaptive dynamic positioning control method for a semi - submersible platform according to claim 5, characterized in that, In step S2, estimate the disturbance and compensate the system state of the control system through a disturbance prediction and compensation module; Receive the error and control signal, estimate the external disturbance through the established disturbance prediction and compensation model and update the state estimation, and output the position and velocity estimation values, Estimate the unknown disturbance and model uncertainty through real - time observation of the system state to ensure the control accuracy, Update the estimation of the state through a recurrence formula, and add the information contained in the error and control signal to the extended state equation, The calculation formula for the updated estimated value of the position state is as follows: ， The formula is iteratively updated during the calculation process, In the formula, represents the estimated value of the system at the current displacement state, represents the estimated value of the displacement at the previous moment, represents the compensation, the step size of time discretization, represents the estimated value of the velocity at the previous moment, represents the position observer gain parameter, which is used to control the convergence speed of the estimation error, represents the error of the displacement at the current moment, is the displacement actually measured at the current moment and the difference between the estimated displacement at the previous moment; The calculation formula for the updated estimated value of the velocity state is as follows: ， The formula is iteratively updated during the calculation process. In the formula, represents the estimated value of the system speed at the current moment, represents the estimated value of the speed at the previous moment, represents the estimated value of the total disturbance at the previous moment, represents the speed observer gain parameter, represents the control input coefficient, represents the control input at the previous moment.

7. The adaptive dynamic positioning control method for a semi-submersible platform according to claim 1, characterized in that In S2, a sliding window model is established, and the particle swarm optimization algorithm is used to dynamically adjust the size of the sliding window in combination with real-time sea state characteristics to improve environmental adaptability. The operation steps of the sliding window model include: Collect error values and sea state data in real time. The sea state data includes wave height, spectral peak period, and error data. The errors are stored in the sliding window. The sliding window limits the number of stored errors to ensure that the stored errors are data within the latest time period. When the number of errors in the window reaches or exceeds the preset size, calculate the standard deviation of the errors in the window. Calculate the wave force and sea state dynamic index according to the JONSWAP spectrum. Use the particle swarm optimization algorithm to optimize the window size and constrain the window range according to the wave period.

8. The adaptive dynamic positioning control method for a semi-submersible platform according to claim 7, characterized in that The fitness function used in the particle swarm optimization algorithm includes the sum of squared errors, control force energy, and wave coupling term. The inertia weight is dynamically adjusted according to the wave energy flux density, and the inertia weight is used to update the particle velocity. The window size is constrained by the main wave period.

9. The adaptive dynamic positioning control method for a semi-submersible platform according to claim 1, characterized in that In step S2, the steps of dynamically adjusting the PID gain using fuzzy logic include: Obtain the error mean and change rate within the window. Normalize the error and change rate, and divide the fuzzy set through the triangular membership function. Use the centroid method for defuzzification to adjust K p 、K i 、K d ; Use fuzzy logic to obtain the change value of the gain parameter and update the dynamically adjusted control parameter.

10. The adaptive dynamic positioning control method for a semi-submersible platform according to claim 1, characterized in that Upper and lower limits of the gain are set during the process of adjusting the control parameter to prevent slow response caused by too low gain and avoid oscillation caused by too high gain.

Images