Ship folding fan type sail independent control method based on wind conditions

Abstract

The invention relates to the technical field of ship propulsion, and discloses a ship folding fan type sail independent control method based on a wind regime, and the method specifically comprises the steps: data collection, data analysis and fault early warning, wind direction compliance determination, safety evaluation, target angle execution, and angle error determination and compensation. By means of multi-sensor fusion and an intelligent algorithm, high-precision and high-safety autonomous control over the folding fan type sail of the ship is achieved, mechanical faults such as a bearing can be early warned in real time, the sail unfolding angle is dynamically optimized to maximize thrust efficiency, meanwhile, the wind direction effectiveness and structural stress safety are strictly verified, and operation under the dead wind or overload working condition is avoided; and the feedback of the incremental encoder and a grading error compensation mechanism are combined, so that the angle control precision is within + / -1 degree, the wind energy utilization efficiency is remarkably improved, the fuel consumption and carbon emission are reduced, and the safety of the sail and the ship body structure is effectively guaranteed.

Description

Technical Field

[0001] This invention relates to the field of ship propulsion technology, specifically to an independent control method for ship folding fan-shaped sails based on wind conditions. Background Technology

[0002] With the International Maritime Organization's increasingly stringent regulations on ship carbon emissions and the global shipping industry's push for "dual carbon" goals, wind energy, as a clean and renewable auxiliary propulsion energy source, is once again receiving widespread attention. Among them, folding fan sails, due to their advantages such as compact structure, flexible deployment and retraction, and strong adaptability, have been applied to various types of ocean-going merchant ships and research vessels, becoming an important technical path for energy conservation and emission reduction in ships. However, existing folding fan sail control systems generally suffer from problems such as low level of intelligence, slow response, and insufficient safety.

[0003] Currently, most systems still employ fixed-angle adjustment or open-loop control strategies based on simple lookup tables, failing to fully consider the impact of real-time wind conditions, wave disturbances, and changes in ship attitude on the aerodynamic performance of the sail, resulting in low wind energy capture efficiency. Simultaneously, the lack of online monitoring of mechanical status makes early fault warning difficult, which can easily lead to structural damage or even safety accidents due to overload or stall under strong crosswinds or gusts. Furthermore, traditional control methods do not incorporate closed-loop feedback mechanisms, and are affected by nonlinear factors such as transmission backlash and friction dead zones, resulting in a large deviation between the actual deployment angle and the target value, weakening the sail's propulsion efficiency. Therefore, a wind-condition-based independent control method for ship folding fan-shaped sails is proposed. Summary of the Invention

[0004] This invention provides a wind-condition-based independent control method for ship folding sails. Through multi-sensor fusion and intelligent algorithms, it achieves high-precision and high-safety autonomous control of ship folding sails. It can provide real-time early warning of mechanical failures such as bearings, dynamically optimize the sail deployment angle to maximize thrust efficiency, and strictly verify the effectiveness of wind direction and structural stress safety to avoid operation under headwind or overload conditions. This solves the problems of low efficiency, poor safety, and insufficient sail deployment adjustment precision mentioned in the background technology.

[0005] This invention provides the following technical solution: A method for independent control of a ship's folding fan-shaped sail based on wind conditions includes the following steps: Step 1: Data Acquisition: Environmental and mechanical parameters are collected synchronously through a multi-sensor array installed on the mast and hull; Step 2, Data Analysis and Fault Early Warning: The collected time-series data is trained and inferred using an LSTM neural network to achieve fault early warning and calculation of the optimal unfolding angle; Step 3: Wind Direction Compliance Determination: Determine whether the current wind direction is suitable for deploying sails; Step 4: Safety Assessment: Calculate the load on the sail and determine whether it is safe after the sail is deployed; Step 5, Target Angle Execution: Deploy the sails to the target angle; Step 6, Angle Error Judgment and Compensation: Real-time feedback on the current sail angle, and judgment and compensation adjustment of angle error.

[0006] As a preferred embodiment of the present invention, the environmental parameters include: wind speed. ,wind direction Wave height and heading The mechanical parameters include: the current angle of the sail deployment. Drive motor torque and bearing vibration frequency .

[0007] As a preferred embodiment of the present invention, the fault warning is achieved by comparing vibration frequencies. Compared with normal threshold ,when > hour( The vibration standard deviation under normal operating conditions triggers a wind turbine equipment fault warning and sends it to the ship's central control system; the optimal deployment angle is determined by the ship's speed. With wind speed The optimal unfolding angle is solved using a Bayesian optimization algorithm. This makes the sail thrust coefficient Maximize, the calculation formula is as follows.

[0008] in, This is the critical angle of attack for stall.

[0009] As a preferred embodiment of the present invention, the wind direction compliance determination criteria are as follows: like If the angle is ≤90°, it is considered a valid wind-receiving area and proceeds to the subsequent safety assessment process; if If the wind angle is greater than 90°, it is determined to be a headwind or crosswind, and the emergency folding procedure is initiated.

[0010] As a preferred embodiment of the present invention, the formula for calculating the load on the sail is as follows:

[0011] in, air density (taken as 1.225 kg / m³) 3 ), The area of ​​the sail when it is deployed. The lift coefficient (a polynomial function fitted by wind tunnel tests). when ≤ If the condition is met, it is deemed safe, and the subsequent target-oriented execution process will proceed. when > If this occurs, it is deemed unsafe, triggering an emergency folding mechanism and an alarm.

[0012] As a preferred embodiment of the present invention, the sail deployment step is as follows: a target angle execution signal, after being determined to be safe, is sent to the servo motor; according to the above signal instruction, the servo motor starts to operate, thereby adjusting the sail to the target angle. .

[0013] As a preferred embodiment of the present invention, the angle error determination specifically includes: The current angle is fed back in real time through an incremental encoder. , according to, Calculate the angle error, if If the temperature is greater than 1°, then the compensation process will begin; When 1 < When the angle is less than 5°, a gap compensation fine-tuning method is used: the mechanical gap is compensated by the number of motor pulses, and the number of pulses is adjusted each time. , ( (This is the pulse-to-angle conversion coefficient, taken as 10 pulses / °). When the angle is ≥5°, the drive actuator (folding) is triggered, and the unfolding process is re-executed; like If the angle is ≤1°, the current angle is maintained and steady-state monitoring is entered.

[0014] As a preferred embodiment of the present invention, it also includes marine meteorological fusion prediction, the specific steps of which are as follows: Access to shipborne weather radar and satellite short-term forecast data; Construct an LSTM-Transformer hybrid model to predict the arrival time of gusts, the probability of sudden changes in wind direction, and the cycle and direction of surges within the next 30 minutes. Pre-emptive triggering of pre-adjustment strategies to avoid structural impact from emergency actions. Compared with the prior art, the present invention provides an independent control method for ship folding fan-shaped sails based on wind conditions, which has the following beneficial effects: This wind-based independent control method for ship folding sails achieves high-precision and high-safety autonomous control of ship folding sails through multi-sensor fusion and intelligent algorithms. It can provide real-time early warning of mechanical failures such as bearings, dynamically optimize the sail deployment angle to maximize thrust efficiency, and strictly verify the effectiveness of wind direction and structural stress safety to avoid operation under headwind or overload conditions. Combined with incremental encoder feedback and graded error compensation mechanism, the angle control accuracy reaches within ±1°, which significantly improves wind energy utilization efficiency, reduces fuel consumption and carbon emissions, and effectively ensures the safety of sails and ship structure. Attached Figure Description

[0015] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the accompanying drawings used in the description of the specific embodiments or the prior art will be briefly introduced below.

[0016] Figure 1 This is a schematic diagram of the method flow of the present invention. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] Example: Reference Figure 1 A method for independent control of a ship's folding fan-shaped sail based on wind conditions includes the following steps: Step 1: Data Acquisition: Environmental and mechanical parameters are collected synchronously through a multi-sensor array installed on the mast and hull; Step 2, Data Analysis and Fault Early Warning: The collected time-series data is trained and inferred using an LSTM neural network to achieve fault early warning and calculation of the optimal unfolding angle; Step 3: Wind Direction Compliance Determination: Determine whether the current wind direction is suitable for deploying sails; Step 4: Safety Assessment: Calculate the load on the sail and determine whether it is safe after the sail is deployed; Step 5, Target Angle Execution: Deploy the sails to the target angle; Step 6, Angle Error Judgment and Compensation: Real-time feedback on the current sail angle, and judgment and compensation adjustment of angle error.

[0019] Environmental parameters include: wind speed ,wind direction Wave height and heading Mechanical parameters include: the current angle of the sail deployment. Drive motor torque and bearing vibration frequency .

[0020] Fault warning: By comparing vibration frequencies Compared with normal threshold ,when > hour( (The vibration standard deviation under normal operating conditions) triggers a wind turbine equipment fault warning and pushes it to the ship's central control system; Optimal deployment angle: combined with the ship's speed With wind speed The optimal unfolding angle is solved using a Bayesian optimization algorithm. This makes the sail thrust coefficient Maximize, the calculation formula is as follows.

[0021] in, This is the critical angle of attack for stall.

[0022] The criteria for determining wind direction compliance are as follows: like If the angle is ≤90°, it is considered a valid wind-receiving area and proceeds to the subsequent safety assessment process; if If the wind angle is greater than 90°, it is determined to be a headwind or crosswind, and the emergency folding procedure is initiated.

[0023] The formula for calculating the load on a sail is as follows:

[0024] in, air density (taken as 1.225 kg / m³) 3 ), The area of ​​the sail when it is deployed. The lift coefficient (a polynomial function fitted by wind tunnel tests). when ≤ If the condition is met, it is deemed safe, and the subsequent target-oriented execution process will proceed. when > If this occurs, it is deemed unsafe, triggering an emergency folding mechanism and an alarm.

[0025] The sail deployment process is as follows: After determining that the target angle is safe, a signal is sent to the servo motor. Based on the signal command, the servo motor starts running, thereby adjusting the sail to the target angle. .

[0026] The angle error determination specifically includes: The current angle is fed back in real time through an incremental encoder. , according to, Calculate the angle error, if If the temperature is greater than 1°, then the compensation process will begin; When 1 < When the angle is less than 5°, a gap compensation fine-tuning method is used: the mechanical gap is compensated by the number of motor pulses, and the number of pulses is adjusted each time. , ( (This is the pulse-to-angle conversion coefficient, taken as 10 pulses / °). When the angle is ≥5°, the drive actuator (folding) is triggered, and the unfolding process is re-executed; like If the angle is ≤1°, the current angle is maintained and steady-state monitoring is entered.

[0027] As described above, by deploying a multi-sensor array on the mast and hull, environmental parameters such as wind speed, wind direction, wave height, and heading, as well as mechanical status data such as the current sail deployment angle, drive motor torque, and bearing vibration frequency, are collected in real time and synchronously. LSTM neural networks are used to analyze the time-series data. On the one hand, by monitoring whether the vibration frequency exceeds the standard deviation of normal operating conditions, early fault warnings for the wind turbine equipment are achieved and pushed to the central control system. On the other hand, combining ship speed and wind speed, a Bayesian optimization algorithm is used to solve for the optimal deployment angle that maximizes the sail thrust coefficient without stalling (angle of attack less than the critical stall angle). The system first determines whether the relative wind direction angle is not greater than 90° to confirm that it is in the effective windward range; if not, emergency folding is immediately initiated. If the conditions are met... For wind conditions, the lift load on the sail is further calculated based on aerodynamic formulas (considering air density, deployment area, and lift coefficient fitted by wind tunnel tests), and compared with the structural safety threshold. The target angle deployment command is executed only when the load does not exceed the limit, and the sail is driven by a servo motor. During deployment, the actual angle is fed back in real time through an incremental encoder, and a graded compensation strategy is implemented according to the angle error: when the error is ≤1°, the sail remains in a steady state; when the error is 1 < 5°, the sail is finely adjusted with a conversion coefficient of 10 pulses / ° to eliminate the influence of mechanical backlash; when the error is ≥5°, the sail is judged as a serious deviation, and the actuator is triggered to fold the sail and then re-deploy it. This ensures high-precision and high-reliability autonomous control of the sail, maximizing wind energy utilization efficiency while ensuring navigation safety.

[0028] LSTM fault prediction algorithms include: Input feature: vibration frequency Motor torque Ambient temperature ; Training set: Contains 1000 hours of time-series data under normal and fault conditions; Warning trigger condition: Fault probability output by the model A value greater than 0.85 triggers a warning.

[0029] As mentioned above, early fault identification based on vibration spectrum can issue warnings at potential stages such as bearing wear and motor eccentricity, reducing unplanned downtime, extending equipment life, and lowering operation and maintenance costs.

[0030] The optimal unfolding angle Bayesian optimization algorithm includes: Objective function: Maximize thrust coefficient , Constraint: 0°≤ ≤90° ≤

[0031] Optimization step size: 0.5°, convergence threshold: <0.001.

[0032] As mentioned above, the optimal deployment angle is dynamically solved based on Bayesian optimization, which ensures that the sail always operates in the high thrust coefficient range. Compared with fixed angle or empirical adjustment, the thrust contribution rate is increased by 18% to 25%, significantly reducing the fuel consumption of the main engine.

[0033] It also includes marine meteorological fusion forecasting, with the following specific steps: Access to shipborne weather radar and satellite short-term forecast data; Construct an LSTM-Transformer hybrid model to predict the arrival time of gusts, the probability of sudden changes in wind direction, and the cycle and direction of surges within the next 30 minutes. Pre-emptive adjustment strategies are triggered to avoid structural impact from emergency actions.

[0034] The algorithm flow of the LSTM-Transformer hybrid model is as follows: S1. Input variable acquisition: wind speed obtained from meteorological sensors. ,wind direction air pressure and wind speed Data; wave height acquired by ocean sensors Wave period And the waves Data; ship's speed in its current state ,course and roll angle Data; 10-minute average wind field data provided by weather radar or satellite.

[0035] S2, Data Preprocessing The variables are normalized using the following formula:

[0036] The following sine and cosine formulas are used to encode cyclic variables such as wind direction and wave direction:

[0037] S3. Use an LSTM encoder to extract local dynamics. Use bidirectional LSTM to capture the preceding and following context: , t=1, ..., L Output: Hidden state sequence , =128 This enables the extraction of local transient patterns such as sudden increases in wind speed and sudden drops in air pressure.

[0038] S4. Perform global dependency modeling using the Transformer encoder. Embed the LSTM output as input to the Transformer; The position is encoded using the following formula:

[0039] Calculate multi-head self-attention using the following formula:

[0040] Feedforward Network (FFN) Calculation:

[0041] Output: .

[0042] S5, Task-Specific Decoding Header from Feature vectors are extracted at the last time step: Gust arrival time:

[0043] Probability of sudden wind direction change:

[0044] Surge cycle:

[0045] Surge direction:

[0046] MLP is a type of feedforward artificial neural network.

[0047] S6, Multi-task Joint Training The total loss function is a weighted sum:

[0048] Default weights: =1, =1.5, =1, =1.2 (adjusted according to task difficulty). Using the AdamW optimizer, the learning rate... batchsize=64, training epoch=100.

[0049] As mentioned above, LSTM is adept at capturing local nonlinear dynamics such as wind speed fluctuations and pressure gradients; Transformer is used to capture cross-variable correlations (such as "sudden pressure drop → gusts → wave direction deflection"); multi-task learning shares representations to improve the generalization ability for small sample tasks (such as sudden wind direction changes); real-time rolling prediction meets the needs of ship decision-making systems and improves the safety of sail system deployment.

[0050] Components not described in detail in this article are existing technologies.

[0051] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for independent control of a ship's folding fan-shaped sail based on wind conditions, characterized in that, Includes the following steps: Step 1: Data Acquisition: Environmental and mechanical parameters are collected synchronously through a multi-sensor array installed on the mast and hull; Step 2, Data Analysis and Fault Early Warning: The collected time-series data is trained and inferred using an LSTM neural network to achieve fault early warning and calculation of the optimal unfolding angle; Step 3: Wind Direction Compliance Determination: Determine whether the current wind direction is suitable for deploying sails; Step 4: Safety Assessment: Calculate the load on the sail and determine whether it is safe after the sail is deployed; Step 5, Target Angle Execution: Deploy the sails to the target angle; Step 6, Angle Error Judgment and Compensation: Real-time feedback on the current sail angle, and judgment and compensation adjustment of angle error.

2. The independent control method for a ship's folding fan-shaped sail based on wind conditions according to claim 1, characterized in that, The environmental parameters include: wind speed. ,wind direction Wave height and heading ; The mechanical parameters include: the current angle of the sail deployment. Drive motor torque and bearing vibration frequency .

3. The independent control method for a ship's folding fan-shaped sail based on wind conditions according to claim 1, characterized in that, The fault warning is based on comparing vibration frequencies. Compared with normal threshold ,when > hour( (The vibration standard deviation under normal operating conditions) triggers a fault warning for the wind turbine equipment and pushes it to the ship's central control system; The optimal deployment angle: combined with the ship's speed With wind speed The optimal unfolding angle is solved using a Bayesian optimization algorithm. This makes the sail thrust coefficient Maximize, the calculation formula is as follows. ； in, This is the critical angle of attack for stall.

4. The independent control method for a ship's folding fan-shaped sail based on wind conditions according to claim 1, characterized in that, The criteria for determining wind direction compliance are as follows: like If the angle is ≤90°, it is considered a valid wind-receiving area and proceeds to the subsequent safety assessment process; if If the wind angle is greater than 90°, it is determined to be a headwind or crosswind, and the emergency folding procedure is initiated.

5. The independent control method for a ship's folding fan-shaped sail based on wind conditions according to claim 3, characterized in that, The formula for calculating the load on the sail is as follows: ； in, air density (taken as 1.225 kg / m³) 3 ), The area of ​​the sail when it is deployed. The lift coefficient (a polynomial function fitted by wind tunnel tests). when ≤ If the condition is met, it is deemed safe, and the subsequent target-oriented execution process will proceed. when > If this occurs, it is deemed unsafe, triggering an emergency folding mechanism and an alarm.

6. The independent control method for a ship's folding fan-shaped sail based on wind conditions according to claim 1, characterized in that, The sail deployment steps are as follows: A target angle execution signal, determined to be safe, is sent to the servo motor. Based on this signal, the servo motor begins operation, thereby adjusting the sail to the target angle. .

7. The independent control method for a ship's folding fan-shaped sail based on wind conditions according to claim 6, characterized in that, The angle error determination specifically includes: The current angle is fed back in real time through an incremental encoder. , according to, Calculate the angle error, if If the temperature is greater than 1°, then the compensation process will begin; When 1 < When the angle is less than 5°, a gap compensation fine-tuning method is used: the mechanical gap is compensated by the number of motor pulses, and the number of pulses is adjusted each time. , ( (This is the pulse-to-angle conversion coefficient, taken as 10 pulses / °). When the angle is ≥5°, the drive actuator (folding) is triggered, and the unfolding process is re-executed; like If the angle is ≤1°, the current angle is maintained and steady-state monitoring is entered.

8. The independent control method for a ship's folding fan-shaped sail based on wind conditions according to claim 1, characterized in that, It also includes marine meteorological fusion forecasting, with the following specific steps: Access to shipborne weather radar and satellite short-term forecast data; Construct an LSTM-Transformer hybrid model to predict the arrival time of gusts, the probability of sudden changes in wind direction, and the cycle and direction of surges within the next 30 minutes. Pre-emptive adjustment strategies are triggered to avoid structural impact from emergency actions.

Images