Cement-based offshore floating structure stability control method based on dynamic water tank adjustment algorithm

Abstract

The invention provides a cement-based marine floating structure stability control method based on a dynamic water tank adjustment algorithm, which is applied to a dynamic water tank adjustment system of a cement-based marine floating structure, and the system comprises a water tank module, a sensing network, an execution mechanism and a central control unit, the method comprises the following steps: monitoring the change of the water level in the water tank module and the shaking angle data of the cement-based offshore floating structure through a sensing network; the central control unit analyzes the data acquired by the sensing network in real time, predicts the wave load change trend in a period of time in the future, simulates the stress state of the structure, and generates a target water level value of the water tank module in combination with the characteristics of the structure material; according to the target water level value and the real-time water level value, the execution mechanism is controlled to adjust the water level of the water tank module so as to achieve stability control over the cement-based offshore floating structure, water tank adjustment and material characteristics are accurately matched, control failure or structural damage can be avoided, and then the structural stability is improved.

Description

Technical Field

[0001] The invention relates to the technical field of floating structure stability control, and in particular to a cement-based offshore floating structure stability control method based on a dynamic water tank adjustment algorithm. Background Art

[0002] As the development of marine resources extends to the deep sea, cement-based composite materials are widely used in offshore floating platforms, floating wind power foundations and other fields due to their corrosion resistance and low cost. However, this type of structure has the characteristics of uniform mass distribution and low inherent damping, and faces significant stability control challenges in complex sea conditions. Current stability control technologies are mainly divided into three categories: 1. Passive ballast adjustment technology: The most widely used stability control method is to adjust the center of gravity by pre-setting ballast water tanks. Although this method has a simple structure and high reliability, it has the inherent defect of slow response speed and is difficult to cope with the rapid changes in wave loads. Especially in extreme sea conditions such as typhoons, the lag in ballast water adjustment can easily lead to structural instability.

[0003] 2. Active dynamic balancing system: It uses hydraulically driven balance wings or mechanical actuators such as adjustable buoyancy units to compensate for wave disturbances through real-time actions. It can effectively shorten the response time in moderate sea conditions, but it relies on complex mechanical transmission devices, which leads to a significant increase in the difficulty of offshore operation and maintenance and energy consumption costs. Especially for large cement-based floating structures, when the structural mass exceeds 1,000 tons, the system energy efficiency ratio will drop sharply, resulting in limited stability adjustment effect.

[0004] 3. Application of intelligent control algorithms: In recent years, algorithms such as adaptive PID control and fuzzy logic control have been proposed to try to improve the accuracy of stability control through dynamic feedback. However, traditional control models are not adaptable enough to the nonlinear characteristics of wave loads, and control deviation accumulation is prone to occur in large structures of 100 meters or long-period irregular waves.

[0005] The shortcomings of the existing technical solutions mainly include the following aspects: First, the water tank control system of the structure using concrete-based materials still uses the steel structure adjustment logic, and the control parameters are not matched well enough; second, the passive system relying on the communicating vessel has a significant response delay, which makes it difficult to cope with sudden extreme sea conditions, and the pipeline network structure is complex and the maintenance cost is high; third, the control algorithm does not fully consider the time-varying characteristics of dynamic wave loads, and the control accuracy is poor; fourth, the existing stability control solutions are mostly aimed at triangular platforms, and lack optimization applications for floating platforms of other shapes. Summary of the invention

[0006] The object of the present invention is to provide a cement-based offshore floating structure stability control method based on a dynamic water tank adjustment algorithm, so as to solve or at least partially solve the above-mentioned problems existing in the prior art.

[0007] To achieve the above object, the technical solution provided by the present invention is as follows: A stability control method for a cement-based offshore floating structure based on a dynamic water tank regulation algorithm, the method is applied to a dynamic water tank regulation system of a cement-based offshore floating structure, the system includes a water tank module, a sensing network, an actuator, and a central control unit, and the method includes the following operations: Monitor the water level change in the water tank module and the sway angle data of the cement-based offshore floating structure through the sensing network; The central control unit performs real-time analysis on the data collected by the sensing network, predicts the change trend of wave loads in the next period of time, simulates the structural stress state, and combines the structural material characteristics to generate the target water level value of the water tank module; According to the target water level value and the real-time water level value, control the actuator to adjust the water level of the water tank module to achieve the stability control of the cement-based offshore floating structure.

[0008] Furthermore, the water tank module includes a plurality of independent compartments distributed longitudinally and transversely along the cement-based offshore floating structure. The independent compartments include a main ballast tank and an auxiliary compensation tank. The main ballast tank is arranged along the edge of the cement-based offshore floating structure and is used to undertake the main buoyancy adjustment task. The auxiliary compensation tank is distributed in the geometric center area of the cement-based offshore floating structure and is used to adjust the center of gravity offset. The compartment walls of the independent compartments are prefabricated with cement-based materials. A conical drainage port is provided at the bottom of the independent compartment, and an actuator is provided at the top. The actuator is used to realize the injection and discharge of seawater.

[0009] Furthermore, the actuator includes an electromagnetic valve group. The electromagnetic valve group includes a two-way control valve, forming channel 1 and channel 2. Channel 1 is used for the seawater inlet directly connected to the outside through the top, and channel 2 is used for the control channel connecting the drainage pipe at the bottom of the independent compartment.

[0010] Furthermore, when the actuator realizes seawater injection, it performs the following operations: The electromagnetic valve group opens the valve of channel 1 and closes the valve of channel 2, allowing seawater to flow naturally from the top seawater inlet under the action of gravity. The conical drainage port is used as an exhaust channel at this time, and the air in the compartment is accelerated to be discharged through the Venturi effect. When the sensing network monitors that the water level in the compartment rises to the target height, close the valve of channel 1; When the actuator realizes seawater discharge, it performs the following operations: The electromagnetic valve group closes the valve of channel 1 and opens channel 2, and the positive pressure in the compartment from the draft depth pressure of the cement-based offshore floating structure is used to press the water out from the conical drainage port. The valve of channel 1 is converted into an air intake compensation port at this time.

[0011] Furthermore, the sensing network includes a pressure sensor and an inclination sensor. The pressure sensor is arranged at the bottom of each independent compartment for real-time monitoring of the water level change in the compartment; the inclination sensor is installed at the four corners and the center point of the cement-based offshore floating structure for dynamically collecting roll angle data and pitch angle data.

[0012] Furthermore, the central control unit adopts a multi-threaded parallel computing architecture and incorporates a dynamic water tank regulation algorithm. The dynamic water tank regulation algorithm includes: A wave load prediction module for analyzing the real-time data of the sensing network based on the short-time Fourier transform, extracting the wave spectrum characteristics, and predicting the load change trend in the next period of time; A stability compensation calculation module for simulating the stress state of the cement-based offshore floating structure through a finite element model, combining the mechanical properties and micro-deformation characteristics of the cement-based material, and dynamically generating the target water level values of each independent compartment of the water tank module; A valve control strategy module for calculating the optimal valve opening combination and adjustment timing according to the difference between the target water level value and the real-time water level value, preferentially enabling the auxiliary compensation compartment for rapid fine-tuning, and coordinating the main ballast tank to achieve buoyancy compensation. The control instruction is updated every 0.1 seconds and sent to the actuator.

[0013] Furthermore, the wave load prediction module is specifically used to predict the wave load change trend in the next 5 - 10 seconds in real time, providing a leading control basis for stability compensation, and finally obtaining the wave force amplitude-phase matrix within the future time window, including the following operations: Short-time Fourier transform dynamic spectrum analysis: Collect real-time wave data through wave gauges and accelerometers arranged around the structure, perform short-time Fourier transform analysis on the real-time wave data using a sliding time window, extract the distribution characteristics of wave energy in the frequency domain, especially the significant wave frequency band of 0.1 - 0.3 Hz; combine the autoregressive moving average model to predict the future wave sequence, and dynamically update the model parameters through the recursive least squares method to adapt to the non-stationary characteristics of the waves; Deep learning algorithm correction: Train and deploy a time series algorithm through experimental data, input historical wave spectra, structural motion responses, and environmental data or networks, and output the time-domain envelope of the wave force.

[0014] Furthermore, the stability compensation calculation module is specifically used to calculate the target water levels of each independent compartment based on the wave prediction results, so that the restoring moment of the cement-based offshore floating structure dynamically balances the wave overturning moment, and finally obtains the target water level vector of each independent compartment, specifically including the following operations: Construct a finite element-lumped parameter hybrid model: Use the reduced-order finite element method to quickly solve the structural motion equation, input the structural motion characteristics obtained from the wave force amplitude-phase matrix to obtain the instantaneous motion response of the cement-based offshore floating structure; Simplify the water body in the independent cabin into a spring-mass-damper system, and calculate the center of gravity offset at different water levels through the equivalent pendulum model; Based on the creep model of cement-based materials, update the elastic modulus decay coefficient online according to the service life of cement-based materials. Optimize the algorithm to solve the target water level: Construct an optimization function, and the expression is W = minimize the capsizing moment residual + energy consumption penalty term to suppress frequent valve actions, so that W is minimized; Use the model predictive control framework to solve the constrained optimization problem in each control cycle, and the expression is:

[0015] where, represents the prediction time step of model predictive control, reflecting the algorithm's rolling optimization of wave moments and restoring moments within the next control cycles, represents the wave capsizing moment, represents the restoring moment, represents the change in the flow rate of the water tank module, represents the absolute time stamp at the current time step, represents the moment at the i th future control cycle, represents the energy consumption penalty coefficient, which is used to adjust the weight of suppressing frequent valve actions in the optimization objective.

[0016] Furthermore, the valve control strategy module is specifically used to convert the target water level into a specific valve action sequence, including the following operations: Valve characteristic modeling and linearization: Calibrate the flow coefficient of each valve of the actuator through experiments, and establish the mapping relationship between valve opening and flow rate; Implementation of hierarchical control architecture: Calculate the target water level difference and total required flow rate based on the target water level vector of each independent cabin obtained by the stability compensation calculation module, and allocate flow quotas to each independent cabin according to the independent cabin priority.

[0017] Furthermore, when the sensing network detects that the roll angle of the cement-based offshore floating structure exceeds the safety threshold, perform the following operations: The central control unit starts the wave load prediction module, and combines historical data and real-time spectrum analysis to judge the main source direction of the wave capsizing moment; The stability compensation calculation module generates a water injection instruction for the main ballast tank on the corresponding side and a drainage instruction for the independent cabin on the opposite side according to the structural center of gravity offset, and at the same time activates the auxiliary compensation tank in the geometric center area for diagonal compensation; The valve control strategy module decomposes the control instructions into a specific valve opening and closing action sequence and sends them to the actuator.

[0018] Compared with the prior art, the beneficial effects of the present invention are as follows: The method provided by the present invention is applied to the dynamic water tank regulation system of a cement-based offshore floating structure. By monitoring the water level change of the water tank module and the structural sway angle data through a sensing network, and further analyzing the sensing network data through a central control unit, the change trend of wave loads in the next period of time is predicted, the structural stress state is simulated, the adaptability of the control algorithm to non-linear disturbances and time-varying characteristics is enhanced. At the same time, the control algorithm is optimized based on material parameters to ensure the precise matching of the water tank module regulation and the mechanical properties of the cement-based material, avoiding control failure or structural damage caused by parameter mismatch, and effectively improving the stability of the cement-based offshore floating structure. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only the preferred embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0020] Figure 1 It is a schematic diagram of the overall process of a stability control method for a cement-based offshore floating structure based on a dynamic water tank regulation algorithm provided by an embodiment of the present invention.

[0021] Figure 2 It is a schematic diagram of the cross-sectional structure of a dynamic water tank regulation system provided by an embodiment of the present invention.

[0022] Figure 3 It is a schematic diagram of the sensor layout of a dynamic water tank regulation system provided by an embodiment of the present invention.

[0023] Figure 4 It is a schematic diagram of the system dynamic regulation process provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0024] The following describes the principles and features of the present invention with reference to the drawings. The listed embodiments are only used to explain the present invention and are not used to limit the scope of the present invention.

[0025] This embodiment provides a stability control method for a cement-based offshore floating structure based on a dynamic water tank regulation algorithm, and the method is implemented based on the dynamic water tank regulation system of the cement-based offshore floating structure. The dynamic water tank regulation system includes a water tank module, a sensing network, an actuator, and a central control unit. Referring to Figure 1 , the method includes the following operations: S1. Monitor the water level changes in the water tank module and the sway angle data of the cement-based offshore floating structure through the sensing network.

[0026] S2. The central control unit performs real-time analysis on the data collected by the sensing network, predicts the changing trend of wave loads in the next period of time, simulates the structural stress state, and combines the structural material characteristics to generate the target water level value of the water tank module.

[0027] S3. According to the target water level value and the real-time water level value, control the actuator to adjust the water level of the water tank module to achieve the stability control of the cement-based offshore floating structure.

[0028] Refer to Figures 2 - 3 , the water tank module includes a plurality of independent compartments distributed longitudinally and transversely along the cement-based offshore floating structure. The independent compartments include a main ballast tank and an auxiliary compensation tank. The main ballast tank is arranged along the edge of the cement-based offshore floating structure, and its volume accounts for more than 70% of the overall water tank, which is used to undertake the main buoyancy adjustment task. The auxiliary compensation tank is distributed in the geometric center area of the cement-based offshore floating structure, which is used to finely adjust the center of gravity offset. Its volume is small but the response speed is faster. The compartment walls of the independent compartments are prefabricated with ultra-high-strength and ultra-high-tensile cement-based materials. A conical drainage port is provided at the bottom of the independent compartment, and its main function is to ensure that the water flow forms a laminar flow rather than a turbulent flow when entering and leaving, reducing energy loss. An actuator is provided at the top of the compartment, and the actuator is used to achieve the injection and discharge of seawater.

[0029] In this embodiment, the actuator includes an electromagnetic valve group. The electromagnetic valve group includes a two-way control valve, forming channel 1 and channel 2. Channel 1 is used for the seawater inlet directly connected to the outside through the top, and channel 2 is used for the control channel connecting the drainage pipe at the bottom of the independent compartment. The opening and closing angle of the valve is precisely controlled by a stepper motor to achieve the flow rate grading adjustment of seawater injection and discharge.

[0030] When the actuator realizes seawater injection, it performs the following operations: The electromagnetic valve group opens the valve of channel 1 and closes the valve of channel 2, allowing seawater to flow naturally from the top seawater inlet under the action of gravity. At this time, the conical drainage port serves as an exhaust channel, and the air in the compartment is accelerated to be discharged through the Venturi effect. When the sensing network monitors that the water level in the compartment rises to the target height, the valve of channel 1 is closed.

[0031] When the actuator realizes seawater discharge, it performs the following operations: The electromagnetic valve group closes the valve of channel 1 and opens the valve of channel 2, and the positive pressure in the compartment from the draft depth pressure of the cement-based offshore floating structure is used to press the water out from the conical drainage port. At this time, the valve of channel 1 is converted into an air intake compensation port to avoid forming a vacuum to block drainage.

[0032] As a possible implementation, the actuator adopts a redundant design. The electromagnetic valve group in each independent compartment includes a main valve and a standby valve. The main valve is a large-diameter butterfly valve with a maximum flow rate of up to 50 m³ / min, which is used for rapid injection and drainage of water. The standby valve is a precision ball valve with a flow regulation accuracy of ±2%, which is used for high-precision water level correction. The motor used to drive the valve is equipped with a torque feedback device, which is used to monitor the valve state in real time and feedback whether the valve is stuck or leaks. At the same time, to reduce the temperature rise effect inside the cavity of the cement-based mechanism, the connection between the electromagnetic valve group and the pipeline is coated with a fiber heat insulation layer to ensure the reliability of long-term operation.

[0033] The sensing network includes high-precision pressure sensors and tilt sensors, which are embedded in key parts of the structure in an array form. The pressure sensors are arranged at the bottom of each independent compartment to monitor the water level change inside the compartment in real time. The tilt sensors are installed at the four corners and the center point of the cement-based offshore floating structure to dynamically collect roll angle data and pitch angle data. The data collected by all sensors are transmitted to the central control unit through an optical fiber network, and the sampling frequency is not less than 100 Hz to ensure a millisecond-level response to wave loads.

[0034] The central control unit adopts a multi-threaded parallel computing architecture and is built-in with a dynamic water tank regulation algorithm. The dynamic water tank regulation algorithm includes a wave load prediction module, a stability compensation calculation module, and a valve control strategy module.

[0035] The wave load prediction module is used to analyze the real-time data of the sensing network based on the short-time Fourier transform, extract the wave spectrum characteristics, and predict the load change trend in the next 5 - 10 seconds.

[0036] The stability compensation calculation module is used to simulate the stress state of the cement-based offshore floating structure through a finite element model, and combine the mechanical properties and micro-deformation characteristics of the cement-based material to dynamically generate the target water level values of each independent compartment of the water tank module.

[0037] The valve control strategy module is used to calculate the optimal valve opening combination and adjustment timing according to the difference between the target water level value and the real-time water level value, give priority to enabling the auxiliary compensation compartment for rapid fine-tuning, and at the same time coordinate the main ballast tank to achieve large-range buoyancy compensation. The control instruction is updated every 0.1 seconds and sent to the actuator.

[0038] Among them, the core task of the wave load prediction module is to predict the wave load change trend in the next 5 - 10 seconds in real time, provide a leading control basis for stability compensation, and finally obtain the wave force amplitude-phase matrix in the future time window, including the following operations: Short-time Fourier transform dynamic spectrum analysis: Real-time wave data is collected by wave gauges and acceleration sensors arranged around the structure. The short-time Fourier transform analysis is performed on the real-time wave data using a sliding time window to extract the distribution characteristics of wave energy in the frequency domain, especially the significant wave frequency band of 0.1 - 0.3 Hz. The autoregressive moving average model is combined to predict the future wave sequence, and the model parameters are dynamically updated by the recursive least squares method to adapt to the non-stationary characteristics of the waves. Deep learning algorithm correction: The time series algorithm is trained and deployed with experimental data. The historical wave spectrum, structural motion response, and environmental data or network are input, and the time-domain envelope of the wave force is output. Exemplarily, the time series algorithm can be a long short-term memory network or a Transformer.

[0039] The core task of the stability compensation calculation module is to calculate the target water levels of each independent compartment based on the wave prediction results, so that the restoring moment of the cement-based offshore floating structure dynamically balances the wave overturning moment, and finally obtains the target water level vector of each independent compartment, which specifically includes the following operations: Construct a finite element-lumped parameter hybrid model: The reduced-order finite element method is used to quickly solve the structural motion equation, and the structural motion characteristics obtained from the wave force amplitude-phase matrix, such as rolling, pitching, and heaving, are input to obtain the instantaneous motion response of the cement-based offshore floating structure. The water body in the independent compartment is simplified as a spring-mass-damper system, and the center of gravity offset at different water levels is calculated through an equivalent pendulum model. Based on the creep model of the cement-based material, the elastic modulus decay coefficient is updated online according to the service duration of the cement-based material to achieve result correction. Optimization algorithm to solve the target water level: Construct an optimization function W, and the expression is W = minimize the overturning moment residual + energy consumption penalty term for suppressing frequent valve actions, so that W is minimized. The model predictive control framework is adopted to solve the constrained optimization problem in each control cycle, and the expression is:

[0040] Among them, represents the prediction time step of the model predictive control, reflecting the algorithm's rolling optimization of the wave moment and restoring moment within the next control cycles, represents the wave overturning moment, represents the restoring moment, represents the change in the flow rate of the water tank module, represents the absolute timestamp of the current time step, represents the i th moment of the future control cycle, The larger it is, the more the system tends to reduce the flow rate change; conversely, it gives priority to tracking the torque balance. and both need to be calibrated in advance. In this embodiment, , .

[0041] The core task of the valve control strategy module is to convert the target water level into a specific valve action sequence, including the following operations: Valve characteristic modeling and linearization: By experimentally calibrating the flow coefficients of each valve of the actuator, establish the mapping relationship between the valve opening and the flow rate; Implementation of hierarchical control architecture: Based on the target water level vectors of each independent compartment obtained by the stability compensation calculation module, calculate the target water level difference and the total required flow rate, and allocate flow quotas to each independent compartment according to the priority of the independent compartments. Exemplarily, the main valve is used for large-flow coarse adjustment, adopting Bang-Bang control (fully open / fully closed) to shorten the response time; the standby valve is used for precise adjustment, adopting PID control to suppress overshoot.

[0042] Refer to Figure 4 , the system realizes the dynamic stability adjustment of the offshore floating structure according to the monitoring data of the sensing network. Specifically, when the sensing network detects that the roll angle of the cement-based offshore floating structure exceeds the safety threshold, the following operations are performed: The central control unit starts the wave load prediction module, combines historical data with real-time spectrum analysis, and judges the main source direction of the wave overturning moment; The stability compensation calculation module generates the water injection instruction for the main ballast tank on the corresponding side and the drainage instruction for the independent compartment on the opposite side according to the structure's center of gravity offset, and at the same time activates the auxiliary compensation tank in the geometric center area for diagonal compensation; The valve control strategy module decomposes the control instruction into a specific valve opening and closing action sequence and issues it to the actuator.

[0043] For example, when a storm comes, the main valve at the bottom of the target compartment is preferentially opened for large-flow water injection, and the standby valve is switched to for fine adjustment when the water level approaches the target value. The whole process takes a short time from data acquisition to execution completion, and can effectively offset the periodic disturbance caused by the waves. In addition, the central control unit can optimize the algorithm parameters according to the historical adjustment records to gradually improve the control efficiency under different sea conditions.

[0044] As another alternative implementation, to adapt to cement-based floating structures of different forms, the water tank module adopts a design that combines standardized prefabrication and flexible configuration. For platforms with any regular graphic shape, the main ballast tanks are symmetrically arranged along their periphery, and the auxiliary compensation tanks are set near the center of gravity of the platform structure in the inner circle, and the initial mass is appropriately increased. The volume and position parameters of all compartments can be dynamically adjusted according to the structural dimensions. And the optimal control strategy is automatically matched through the topology recognition algorithm of the central control unit.

[0045] As yet another alternative implementation, a high-pressure air chamber is added beside the main ballast tank of the water tank module, and the main ballast tank and the high-pressure air chamber are connected through an electromagnetic reversing valve. When rapid drainage is required, the electromagnetic reversing valve is opened to inject the compressed air in the high-pressure air chamber into the main ballast tank, and the sea water is accelerated to be discharged from the conical drainage port by using the air pressure to increase the drainage speed.

[0046] Meanwhile, the hydraulic control channels of the main valve / standby valve are retained for steady-state regulation. The pneumatic mode is only used for emergency response under sudden large wave impacts. The pressure difference between the air chamber and the water tank is monitored in real time through a pressure sensor to avoid over-compensation.

[0047] On the basis of retaining the core function of water tank regulation, this implementation introduces a pneumatic compensation mechanism to improve the response speed through the synergistic effect of compressed air and seawater medium. In addition, during the pneumatic drainage stage, the discharged seawater can also be used to drive a micro-turbine generator to recover part of the energy for compressed air storage, reducing the overall energy consumption.

[0048] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A stability control method for a cement-based offshore floating structure based on a dynamic water tank regulation algorithm, characterized in that, The method is applied to the dynamic water tank regulation system of a cement-based offshore floating structure. The system includes a water tank module, a sensing network, an actuator, and a central control unit. The method includes the following operations: Monitoring the water level change in the water tank module and the sway angle data of the cement-based offshore floating structure through the sensing network; The central control unit performs real-time analysis on the data collected by the sensing network, predicts the change trend of wave loads in the next period of time, simulates the stress state of the structure, and generates the target water level value of the water tank module in combination with the structural material characteristics; According to the target water level value and the real-time water level value, controlling the actuator to adjust the water level of the water tank module to achieve the stability control of the cement-based offshore floating structure; The water tank module includes a plurality of independent compartments distributed longitudinally and transversely along the cement-based offshore floating structure. The independent compartments include a main ballast tank and an auxiliary compensation tank. The main ballast tank is arranged along the edge of the cement-based offshore floating structure and is used to undertake the main buoyancy adjustment task. The auxiliary compensation tank is distributed in the geometric center area of the cement-based offshore floating structure and is used to adjust the center of gravity offset. The compartment walls of the independent compartments are prefabricated with cement-based materials. A conical drainage port is provided at the bottom of the independent compartment, and an actuator is provided at the top. The actuator is used to realize the injection and discharge of seawater; The central control unit adopts a multi-threaded parallel computing architecture and is built-in with a dynamic water tank regulation algorithm. The dynamic water tank regulation algorithm includes: A wave load prediction module, which is used to analyze the real-time data of the sensing network based on the short-time Fourier transform, extract the wave spectrum characteristics, and predict the change trend of the load in the next period of time; A stability compensation calculation module, which is used to simulate the stress state of the cement-based offshore floating structure through a finite element model, and dynamically generate the target water level values of each independent compartment of the water tank module in combination with the mechanical properties and micro-deformation characteristics of the cement-based material; A valve control strategy module, which is used to calculate the optimal valve opening combination and adjustment timing according to the difference between the target water level value and the real-time water level value, preferentially enable the auxiliary compensation tank for rapid fine-tuning, and at the same time coordinate the main ballast tank to achieve buoyancy compensation. The control instruction is updated every 0.1 second and sent to the actuator; 2. The method according to claim 1, wherein The actuator includes an electromagnetic valve group. The electromagnetic valve group includes a two-way control valve, forming channel 1 and channel 2. Channel 1 is used for the seawater inlet directly connected to the outside at the top, and channel 2 is used for the control channel connecting the drainage pipe at the bottom of the independent compartment.

3. The method according to claim 2, wherein When the actuator realizes seawater injection, it performs the following operations: The electromagnetic valve group opens the valve of channel 1 and closes the valve of channel 2, allowing seawater to flow naturally from the top seawater inlet under the action of gravity. The conical drainage port is used as an exhaust channel at this time, and the air in the compartment is accelerated to be discharged through the Venturi effect. When the sensing network monitors that the water level in the compartment rises to the target height, the valve of channel 1 is closed; When the actuator realizes seawater discharge, it performs the following operations: The electromagnetic valve group closes the valve of channel 1 and opens channel 2, and the positive pressure in the compartment from the draft depth pressure of the cement-based offshore floating structure is used to press the water out from the conical drainage port. The valve of channel 1 is converted into an air intake compensation port at this time.

4. The method according to claim 1, characterized in that, The sensing network includes a pressure sensor and an inclination sensor. The pressure sensor is arranged at the bottom of each independent cabin for real-time monitoring of the water level change in the cabin. The inclination sensor is installed at the four corners and the center point of the cement-based offshore floating structure for dynamically collecting roll angle data and pitch angle data.

5. The method according to claim 1, wherein The wave load prediction module is specifically used for real-time predicting the change trend of wave loads within the next 5 to 10 seconds, providing a leading control basis for stability compensation, and finally obtaining the wave force amplitude-phase matrix within the future time window, including the following operations: Short-time Fourier transform dynamic spectrum analysis: Collect real-time wave data through wave height gauges and acceleration sensors arranged around the structure. Perform short-time Fourier transform analysis on the real-time wave data using a sliding time window to extract the distribution characteristics of wave energy in the frequency domain, especially the significant wave frequency band of 0.1 to 0.3 Hz. Combine the autoregressive moving average model to predict the future wave sequence, and dynamically update the model parameters through the recursive least squares method to adapt to the non-stationary characteristics of the waves. Deep learning algorithm correction: Train and deploy a time series algorithm through experimental data. Input the historical wave spectrum, structural motion response, and environmental data or network, and output the time-domain envelope of the wave force.

6. The method according to claim 5, characterized in that, The stability compensation calculation module is specifically used for calculating the target water level of each independent cabin based on the wave prediction result, making the restoring moment of the cement-based offshore floating structure dynamically balance the wave overturning moment, and finally obtaining the target water level vector of each independent cabin, including the following operations: Construct a finite element-lumped parameter hybrid model: Use the reduced-order finite element to quickly solve the structural motion equation, input the structural motion characteristics obtained from the wave force amplitude-phase matrix to obtain the instantaneous motion response of the cement-based offshore floating structure. Simplify the water body in the independent cabin into a spring-mass-damper system, and calculate the center of gravity offset at different water levels through the equivalent pendulum model. Based on the creep model of the cement-based material, update the elastic modulus decay coefficient online according to the service duration of the cement-based material. Optimal algorithm to solve the target water level: Construct an optimization function W, and the expression is W = minimize the overturning moment residual + energy consumption penalty term to suppress frequent valve actions, making W minimum. Adopt a model predictive control framework to solve the constrained optimization problem in each control cycle, and the expression is: Among them, represents the prediction time domain step of the predictive control, reflecting that the algorithm performs rolling optimization on the wave moment and the restoring moment within the next control cycles, represents the wave capsizing moment, represents the restoring moment, represents the change in the flow rate of the water tank module, represents the absolute timestamp of the current time step, represents the moment of the i th future control cycle, represents the energy consumption penalty coefficient, which is used to adjust the weight of suppressing frequent valve actions in the optimization objective.

7. The method according to claim 6, wherein The valve control strategy module is specifically used for converting the target water level into a specific valve action sequence, including the following operations: Valve characteristic modeling and linearization: Calibrate the flow coefficient of each valve of the actuator through experiments, and establish the mapping relationship between the valve opening and the flow rate. Implementation of hierarchical control architecture: Calculate the target water level difference and the total required flow rate based on the target water level vector of each independent cabin obtained by the stability compensation calculation module, and allocate flow quotas to each independent cabin according to the priority of the independent cabin.

8. The method according to claim 7, characterized in that, When the sensing network detects that the roll angle of the cement-based offshore floating structure exceeds the safety threshold, perform the following operations: The central control unit starts the wave load prediction module, combines historical data and real-time spectrum analysis to judge the main source direction of the wave overturning moment. The stability compensation calculation module generates the water injection command for the main ballast tank on the corresponding side and the drainage command for the independent compartment on the opposite side according to the offset of the structural center of gravity, and at the same time activates the auxiliary compensation tank in the geometric center area for diagonal compensation; The valve control strategy module decomposes the control command into a specific valve opening and closing action sequence and sends it to the actuator.

Images