A method and system for assessing a fouled hull of a ship

Abstract

This invention provides a method and system for assessing ship fouling. First, ship model test data is acquired. Ship navigation data, meteorological data, and navigation status data are collected at regular intervals. All collected data are divided into multiple datasets according to certain time intervals. Each dataset is then filtered to select those that meet preset conditions. Specific calculation methods are used to calculate wind resistance and water temperature resistance, and a backpropagation neural network is used to predict wave resistance. The corrected still-water main engine shaft power is then calculated. Based on the corrected still-water main engine shaft power, the original main engine shaft power in the ship model test data, and the total number of data entries in the dataset, an average fouling coefficient is calculated. This average fouling coefficient is compared with a preset threshold. When the average fouling coefficient is greater than the preset threshold, the ship's fouling is cleaned. This method effectively increases the accuracy and safety of the assessment.

Description

Technical Field

[0001] This invention relates to the field of ship fouling treatment technology, and specifically to a ship fouling assessment method and system. Background Technology

[0002] Because ships spend long periods of time in the water, fouling inevitably forms on their hulls. Fouling increases the roughness of the hull, significantly increasing drag and consequently boosting fuel consumption. Shipping companies need to pay close attention to fouling assessment and removal, as timely removal improves operational efficiency and increases profit margins. However, prematurely addressing fouling increases removal costs, while delaying it significantly increases operating costs. Therefore, the monitoring and removal of fouling has become a focal point for the global shipping industry.

[0003] Currently, there are three main methods and technologies for monitoring ship fouling both domestically and internationally: The first is to have divers take photos of the fouling bottom, but this method requires a large amount of manpower and resources, and the safety of divers cannot be guaranteed; the second is to take photos of the fouling bottom using fully submerged underwater robots or shore-based underwater high-definition cameras, but this method is not only expensive, but the safety and reliability of the robot's internal equipment underwater cannot be guaranteed, and remote operation is complex; the third is to use operational data from ship navigation for evaluation. This method is the most economical for ships with navigation data collection. However, most of the existing data evaluation methods use traditional calculation methods. While the structural models and parameters of this method are clear, and the calculations for wind resistance are relatively accurate, the accuracy of the traditional empirical formulas for calculating wave drag, which is highly nonlinear, falls far short of the requirements.

[0004] Therefore, there is an urgent need for a ship fouling monitoring and assessment method that is highly accurate, safe, economical, and reliable. Summary of the Invention

[0005] To address the issues of low accuracy, safety, economy, and reliability in current ship fouling monitoring and assessment processes, this invention provides a ship fouling assessment method. Based on model ship test data, ship data, meteorological data, and navigation status data, and employing a BP neural network and specific calculation methods, it calculates the average fouling coefficient. Combined with a specific judgment method, it assesses the ship's fouling condition and performs cleaning, effectively increasing the accuracy of the assessment. Simultaneously, it eliminates the need for personnel to go into the water to inspect and assess the ship's fouling condition, greatly enhancing safety. This invention also relates to a ship fouling assessment system.

[0006] The technical solution of the present invention is as follows:

[0007] A method for assessing ship fouling, characterized by comprising the following steps:

[0008] Data collection and filtering steps: Obtain the ship model test data obtained from the ship model test, and collect ship data, meteorological data and navigation status data during ship navigation at regular intervals. Divide all the collected data into multiple datasets according to a certain time interval, and then filter each dataset to select the datasets that meet the preset conditions.

[0009] Mean calculation steps: Calculate the average value of each parameter in the selected data set, including ship data, meteorological data, and navigation status data. Sort the datasets containing the average value of each parameter in chronological order. Then, combine several consecutive datasets in the sorted datasets to form a new dataset.

[0010] Wind drag calculation steps: In the new dataset, calculate the ship's speed relative to the water based on the ship's ground speed in the navigation status data and the seawater current speed in the meteorological data. Calculate the ship's actual draft and windward area based on the ship's design draft and windward area in the ship data, as well as the bow and stern drafts in the navigation status data. Calculate the wind force coefficient based on the wind direction angle in the meteorological data and the bow and stern drafts in the navigation status data. Calculate the wind drag increase based on the wind force coefficient, the wind speed in the meteorological data, and the calculated actual draft and windward area.

[0011] Steps for calculating water temperature drag increase: In the new dataset, calculate the water temperature drag increase based on the ship's wetted surface area in the ship data, the calculated ship speed in the water, and the ship's frictional resistance coefficient in the navigation status data.

[0012] Wave drag prediction steps: In the new dataset, calculate the propeller thrust based on wind drag and water temperature drag, and calculate the propeller advance speed based on the ship's speed over water. Calculate constant points based on propeller thrust, propeller advance speed, and propeller diameter from the ship data. Obtain the propeller advance speed coefficient from the fitted function based on the constant points. Then, calculate the propeller open-water efficiency under specific conditions based on the propeller open-water efficiency, as well as the hull efficiency, shaft transmission efficiency, and propeller relative rotation efficiency from the ship model test data. Finally, calculate the actual ship propulsion efficiency based on the bow draft, stern draft, and... The ship's static resistance is calculated based on its ground speed. Historical wave resistance is calculated based on the ship's static resistance, its speed over water, wind resistance, propulsion efficiency, and the ship's shaft power obtained from the ship model test data. Relevant feature parameters from meteorological data and navigation status data, along with historical wave resistance, are used as training set samples. A wave resistance prediction model is trained on the training set samples using a BP neural network. Relevant feature parameters from meteorological data and navigation status data are used as inputs to predict wave resistance for a future time period based on the wave resistance prediction model.

[0013] Main engine shaft power calculation steps: In the new dataset, calculate the corrected still water main engine shaft power based on wind drag increase, water temperature drag increase, predicted wave drag increase, ship speed in the water and actual ship propulsion efficiency.

[0014] Steps for calculating the average fouling coefficient: Calculate the average fouling coefficient based on the corrected still water main engine shaft power, the original main engine shaft power in the ship model test data, and the total number of data entries in the new dataset. Compare the average fouling coefficient with a preset threshold. When the average fouling coefficient is greater than the preset threshold, clean the fouling on the ship.

[0015] Preferably, the data acquisition and filtering steps include filtering each dataset based on stable navigation, shallow water effects, and stable wind and waves.

[0016] And / or, the ship data includes propeller diameter, design draft, design draft windward area, and ship wetted surface area; the meteorological data includes wind speed, wind direction, current direction, wind angle, sea current speed, average wave period, significant wave height, and main wave angle; the navigation status data includes ship bow draft, ship stern draft, rudder angle, heading angle, ship friction drag coefficient, and ground speed.

[0017] Preferably, in the wave drag prediction step, the relevant characteristic parameters include bow draft, stern draft, ground speed, heading angle, mean wave period, significant wave height, and main wave heading angle.

[0018] Preferably, the ship model test includes a ship model self-propulsion test, a ship model wind tunnel test, and a ship model propeller open water test.

[0019] Preferably, in the wave drag prediction step, calculating the propeller open-water efficiency under specific conditions based on the propeller advance coefficient includes:

[0020] In the propeller open-water performance curve constructed based on the propeller advance rate coefficient, constant point, and propeller open-water efficiency, when the propeller advance rate coefficient is located at the propeller open-water experimental data point, the propeller open-water efficiency is directly obtained; when the propeller advance rate coefficient is located between any two adjacent propeller open-water experimental data points, the propeller open-water experimental data is interpolated to obtain the propeller open-water efficiency.

[0021] A ship fouling assessment system, characterized in that it comprises, in sequence, a data acquisition and filtering module, a mean calculation module, a wind drag calculation module, a water temperature drag calculation module, a wave drag prediction module, a main engine shaft power calculation module, and a fouling average coefficient calculation module.

[0022] The data acquisition and filtering module acquires the ship model test data obtained from the ship model test, and collects ship data, meteorological data and navigation status data during ship navigation at regular intervals. It then divides all the collected data into multiple datasets according to a certain time interval, and filters each dataset to select the datasets that meet the preset conditions.

[0023] The mean calculation module calculates the average value of each parameter in the selected data set, including ship data, meteorological data, and navigation status data. It then sorts the datasets containing the average value of each parameter in chronological order and combines several consecutive datasets in the sorted datasets into a new dataset.

[0024] The wind resistance calculation module calculates the ship's speed relative to the water based on the ship's ground speed in the navigation status data and the seawater current speed in the meteorological data in the new dataset. It also calculates the ship's actual draft and windward area based on the design draft and windward area in the ship data, as well as the bow draft and stern draft in the navigation status data. Furthermore, it calculates the wind force coefficient based on the wind direction angle in the meteorological data and the bow draft and stern draft in the navigation status data. Finally, it calculates the wind resistance based on the wind force coefficient, the wind speed in the meteorological data, and the calculated actual draft and windward area.

[0025] The water temperature resistance calculation module calculates the water temperature resistance in the new dataset based on the ship's wet surface area in the ship data, the calculated ship speed in the water, and the ship's friction resistance coefficient in the navigation status data.

[0026] The wave drag prediction module calculates propeller thrust based on wind and water temperature drag in the new dataset, and propeller advance rate based on the ship's speed in the water. It then calculates a constant point based on propeller thrust, propeller advance rate, and propeller diameter from the ship data. The module obtains the propeller advance rate coefficient from the fitted function based on the constant point, and calculates the propeller open-water efficiency under specific conditions based on the propeller advance rate coefficient. Finally, it calculates the actual ship propulsion efficiency based on the propeller open-water efficiency, as well as the hull efficiency, shaft transmission efficiency, and propeller relative rotation efficiency from the ship model test data. Finally, it calculates the bow draft from the navigation status data. The static resistance of the actual ship is calculated based on the ship's stern draft and ground speed. Historical wave resistance is calculated based on the static resistance, ship speed, wind resistance, propulsion efficiency, and shaft power obtained from the ship model test data. Relevant feature parameters from meteorological data and navigation status data, as well as historical wave resistance, are used as training set samples. A wave resistance prediction model is trained on the training set samples using a BP neural network. Relevant feature parameters from meteorological data and navigation status data are used as inputs to predict wave resistance in a future time period based on the wave resistance prediction model.

[0027] The main shaft power calculation module calculates the corrected still water main shaft power in the new dataset based on wind resistance, water temperature resistance, predicted wave resistance, ship speed in the water, and actual ship propulsion efficiency.

[0028] The fouling average coefficient calculation module calculates the fouling average coefficient based on the corrected still water main engine shaft power, the original main engine shaft power in the ship model test data, and the total number of data entries in the new dataset. It then compares the fouling average coefficient with a preset threshold. When the fouling average coefficient is greater than the preset threshold, the ship's fouling is cleaned.

[0029] Preferably, the ship data includes propeller diameter, design draft, design draft windward area, and ship wetted surface area; the meteorological data includes wind speed, wind direction, current direction, wind angle, sea current speed, average wave period, significant wave height, and main wave angle; and the navigation status data includes bow draft, stern draft, rudder angle, heading angle, ship friction drag coefficient, and ground speed.

[0030] Preferably, the relevant characteristic parameters include bow draft, stern draft, ground speed, heading angle, mean wave period, significant wave height, and main wave heading angle.

[0031] Preferably, the ship model test includes a ship model self-propulsion test, a ship model wind tunnel test, and a ship model propeller open water test.

[0032] Preferably, in the wave drag prediction module, calculating the propeller open-water efficiency under specific conditions based on the propeller advance coefficient includes:

[0033] In the propeller open-water performance curve constructed based on the propeller advance rate coefficient, constant point, and propeller open-water efficiency, when the propeller advance rate coefficient is located at the propeller open-water experimental data point, the propeller open-water efficiency is directly obtained; when the propeller advance rate coefficient is located between any two adjacent propeller open-water experimental data points, the propeller open-water experimental data is interpolated to obtain the propeller open-water efficiency.

[0034] The beneficial effects of this invention are as follows:

[0035] This invention provides a method for assessing ship fouling. First, ship model test data is acquired. Ship navigation data, meteorological data, and navigation status data are collected at regular intervals. All collected data are divided into multiple datasets according to certain time intervals. Each dataset is then filtered to select those meeting preset conditions. The average values ​​of various parameters in the ship, meteorological, and navigation status data are calculated within the selected datasets. The datasets containing the average values ​​of each parameter are then sorted chronologically. Several consecutive datasets from the sorted datasets are combined to form a new dataset. The ship's speed and actual draft windward area are calculated, followed by the calculation of wind resistance, water temperature resistance, and spiral drag. The propeller open-water efficiency and actual ship propulsion efficiency are analyzed. Then, a wave drag prediction model is obtained by training a BP neural network. Relevant feature parameters from meteorological data and navigation status data are used as inputs. The wave drag prediction model predicts the wave drag in a future time period, which can effectively improve the model's nonlinearity, prediction accuracy and reliability. It can also be applied to other target ships of the same type, greatly improving the model's generalization ability and engineering generalization capability. Then, the corrected still-water main engine shaft power is calculated. Based on the corrected still-water main engine shaft power, the original main engine shaft power in the ship model test data and the total number of data in the new dataset, the average fouling coefficient is calculated. The average fouling coefficient is compared with a preset threshold. When the average fouling coefficient is greater than the preset threshold, the ship's fouling is cleaned. The calculation steps for the mean value, wind resistance, water temperature resistance, wave resistance prediction, and main engine shaft power are essentially equivalent to a multi-feature parameter algorithm fusion calculation. This invention provides a multi-feature parameter algorithm fusion method for ship fouling assessment. It processes the collected and screened data, converts the current velocity into the ship's speed, and then calculates the ship's wind resistance and water temperature resistance. A BP neural network is used to predict nonlinear wave resistance, constructing a wave resistance prediction model. The resistance calculation results are then used to calculate the corrected still water resistance and main engine power (still water main engine shaft power). Based on the main engine power calculation for a ship without fouling in still water, the fouling condition of the ship is assessed. This overcomes the shortcomings of directly using traditional calculation methods and models with low prediction accuracy due to neglecting wave and meteorological conditions or using empirical formulas. This invention has the advantages of high accuracy and clear physical structure and parameters. It also reduces the need for personnel to go into the water to inspect and assess the ship's fouling condition, greatly enhancing safety.

[0036] This invention also relates to a ship fouling assessment system, which corresponds to the aforementioned ship fouling assessment method. This system can be understood as a system for implementing the aforementioned ship fouling assessment method, comprising a data acquisition and filtering module, a mean calculation module, a wind resistance calculation module, a water temperature resistance calculation module, a wave resistance prediction module, a main engine shaft power calculation module, and a fouling average coefficient calculation module connected in sequence. These modules work collaboratively. The data acquisition and filtering module is the input module. The mean calculation module, wind resistance calculation module, water temperature resistance calculation module, wave resistance prediction module, and main engine shaft power calculation module collectively function as an intermediate module for multi-feature parameter algorithm fusion calculation. The fouling average coefficient calculation module is the output module. By inputting... The system imports ship parameters, ship navigation status information, and corresponding meteorological data into an intermediate module that uses a multi-feature parameter algorithm for fusion calculation. This module processes the navigation data collected every second in real time, converting chronologically ordered data (e.g., 30-minute averages) into water speed, thus transforming the influence of current velocity into the ship's water speed. Next, it calculates the ship's wind resistance and water temperature drag using formulas, and uses a BP neural network to predict nonlinear wave drag. The resistance calculation results are then used for propulsion calculations to calculate the corrected still-water navigation resistance and main engine power. The corrected still-water main engine power result is imported into the output module. Finally, the output module calculates and assesses the ship's fouling status based on the main engine power of a ship navigating in still water without fouling. This invention can be understood as a multi-feature parameter algorithm fusion ship fouling assessment system. This system combines the advantages of existing white-box and black-box calculations, offering advantages such as high accuracy and clear physical structure and parameters. This multi-feature parameter algorithm-integrated ship fouling assessment system overcomes the shortcomings of traditional calculation methods and the low prediction accuracy caused by neglecting wave and meteorological conditions or relying on empirical formulas in white-box models. It also integrates machine learning algorithms, employing a BP neural network to predict wave drag, improving the model's nonlinearity, prediction accuracy, and reliability. Furthermore, the wave drag prediction module, trained for wave drag, can still be applied to other target ships of the same type, significantly enhancing the model's generalization and engineering capabilities. Therefore, this invention, based on ship model test data, ship data, meteorological data, and navigation status data, uses a BP neural network and specific calculation methods to calculate the average fouling coefficient, combined with specific judgment methods to assess the ship's fouling condition and perform cleaning, effectively increasing the accuracy of the assessment. Simultaneously, it reduces the need for personnel to go into the water to inspect and assess the ship's fouling condition, greatly enhancing safety. Attached Figure Description

[0037] Figure 1 This is a flowchart of the ship fouling assessment method of the present invention.

[0038] Figure 2 This is a structural block diagram of the data filtering method in this invention.

[0039] Figure 3 This is a curve showing the open-water performance characteristics of the propeller of this invention.

[0040] Figure 4 This is a structural block diagram of the wave resistance prediction method of the present invention.

[0041] Figure 5 This is a schematic diagram of the BP neural network structure of the present invention. Detailed Implementation

[0042] The present invention will now be described with reference to the accompanying drawings.

[0043] This invention relates to a method for assessing ship fouling, the flowchart of which is shown below. Figure 1 As shown, the steps are as follows:

[0044] 1. Data Acquisition and Filtering Steps: Acquire ship model test data obtained from the ship model experiment, and collect ship data, meteorological data, and navigation status data at regular intervals. Divide all collected data into multiple datasets according to certain time intervals, and then filter each dataset to select the datasets that meet preset conditions. Preferably, the ship data includes propeller diameter, design draft, design draft windward area, and ship wetted surface area; the meteorological data includes wind speed, wind direction, current direction, wind direction angle, seawater current speed, average wave period, significant wave height, and main wave direction angle; the navigation status data includes ship bow draft, ship stern draft, rudder angle, heading angle, ship friction drag coefficient, and ground speed.

[0045] Specifically, such as Figure 2 As shown, first, ship model test data were obtained from self-propelled tests, wind tunnel tests, and propeller open-water tests. Ship data, meteorological data, and navigation status data were collected every second during ship navigation. All data collected every second during ship navigation were divided into multiple 10-minute datasets according to a 10-minute time interval. Then, calculations and filtering were performed on each 10-minute dataset. Figure 2 The data processing shown includes stable navigation filtering, shallow water effect filtering, and stable wind and wave filtering. A 10-minute dataset meeting the preset conditions is selected, including:

[0046] 1.1 Stable navigation filtering: For each 10-minute time interval dataset, the following calculations and filtering are performed:

[0047] abs(rudder angle max - rudder angle min) < 5 deg

[0048] (Max RPM - Min RPM) < 3 rpm

[0049] Mean square error of rotational speed < 1 rpm

[0050] The mean square error of the water speed is < 0.5 kN

[0051] Average speed over water > 9 knots

[0052] The average flow velocity measured by the flow meter is < 2kN.

[0053] The mean square error of the flow velocity measured by the flow meter is < 0.2 kN.

[0054] 1.2 Shallow water effect screening: For each 10-minute time interval dataset, the following calculations and screenings are performed:

[0055] Water depth > 80 m

[0056] Alternatively, the distance between the bottom of the ship and the seabed is > 50 m.

[0057] 1.3 Stable wind and wave filtering: For each 10-minute time interval dataset, the following calculations are performed for filtering:

[0058] The ship's anemometer measured wind speed < 25 km / h.

[0059] The mean square error of the wind direction angle measured by the ship's anemometer is <30 deg

[0060] Average wave height < 3 m

[0061] Wave height mean square error <0.3 m

[0062] The mean square error of the main wave direction angle is <30 deg

[0063] 2. Mean calculation steps: Calculate the average value of each parameter in the selected data set, including ship data, meteorological data, and navigation status data. Sort the datasets containing the average value of each parameter in chronological order. Then, combine several consecutive datasets in the sorted datasets to form a new dataset.

[0064] Specifically, after filtering each dataset, the average values ​​of each parameter in the ship data, meteorological data, and navigation status data are calculated from the selected multiple 10-minute datasets, such as... Figure 2 The temporary 10-minute data in the data processing shown is then filtered through continuous and stable navigation data. The datasets containing the average value of each parameter are sorted in chronological order. Then, the average value of the parameter values ​​of the three consecutive 10-minute datasets in the sorted dataset is calculated to form a new dataset, which is then transformed into a 30-minute dataset and output.

[0065] 3. Wind drag calculation steps: In the new dataset, calculate the ship's speed relative to the ground from the navigation status data and the seawater current speed from the meteorological data. Calculate the ship's actual draft and windward area from the ship's design draft and design draft windward area from the ship's data, as well as the ship's bow and stern drafts from the navigation status data. Calculate the wind force coefficient of the ship during navigation from the wind direction angle from the meteorological data and the ship's bow and stern drafts from the navigation status data. Calculate the wind drag increase based on the wind force coefficient, the wind speed from the meteorological data, and the calculated actual windward area.

[0066] Specifically, the ship's speed over water is first calculated based on the ship's land speed from the navigation status data and the seawater current speed from the meteorological data. The ship's speed over water is calculated using the following formula:

[0067] Vs = Vg + V_cur (1)

[0068] In the above formula, Vs is the ship's speed over water, Vg is the ship's speed over land, and V_cur is the current velocity relative to the ship's direction of travel, as measured by the ship's current meter.

[0069] Then, based on the design draft and design draft windward area in the ship data, and the bow draft and stern draft in the navigation status data, the actual draft windward area of ​​the ship during navigation is calculated. The actual draft windward area A is calculated according to the following formula:

[0070] A = ((Ta+Tf) / 2+2 (h_design_draft-(Ta+Tf) / 2)) A_design_draft / h_design_draft (2)

[0071] In the above formula, A is the actual draft and windward area, h_design_draft is the design draft, A_design_draf is the design draft and windward area, Ta is the stern draft, and Tf is the bow draft.

[0072] Then, based on the wind direction angle in the meteorological data and the ship's bow and stern drafts in the navigation status data, the wind force coefficient during actual navigation is calculated. The wind force coefficient is calculated according to the following formula:

[0073] Coefficient_wind = G(Wind_direction,Tf,Ta) (3)

[0074] In the above formula, Coefficient_wind is the wind force coefficient during actual navigation of the ship, G(Wind_direction,Tf,Ta) is the calculation function of the wind force coefficient, which is obtained by polynomial fitting of the ship model test data measured by the ship model wind tunnel test, Wind_direction is the wind direction angle measured by the ship's anemometer, Ta is the stern draft, and Tf is the bow draft.

[0075] Finally, the wind resistance increase F_wind is calculated based on the wind force coefficient, wind speed from meteorological data, and the calculated actual draft windward area. The wind resistance increase F_wind is calculated according to the following formula:

[0076] F_wind=0.5 p_airG(Wind_direction,Tf,Ta) A Vwind (4)

[0077] In the above formula, F_wind represents wind resistance, p_air represents air density, Vwind represents the wind speed measured by the ship's anemometer, A represents the actual draft and windward area, and G(Wind_direction,Tf,Ta) represents the wind force coefficient function during actual navigation of the ship.

[0078] 4. Steps for calculating water temperature drag increase: In the new dataset, calculate the water temperature drag increase based on the ship's wetted surface area from the ship data, the calculated ship speed relative to water, and the ship's frictional resistance coefficient from the navigation status data. The water temperature drag increase Fwater is calculated according to the following formula:

[0079] (5)

[0080] In the above formula, Fwater is the drag increase due to water temperature, p_s is the water density, S is the wetted surface area of ​​the ship, Vs is the speed over water, and C f and C f0 These are the ship friction resistance coefficients corresponding to the actual water temperature and a water temperature of 15 degrees Celsius, respectively.

[0081] 5. Wave drag prediction steps: In the new dataset, calculate the propeller thrust based on wind drag and water temperature drag, and calculate the propeller advance speed based on the ship's speed in the water. Calculate a constant point based on the propeller thrust, propeller advance speed, and propeller diameter from the ship data. Obtain the propeller advance speed coefficient from the calculation function fitted to the constant point. Then, calculate the propeller open-water efficiency under specific conditions based on the propeller advance speed coefficient. Calculate the actual ship propulsion efficiency based on the propeller open-water efficiency, as well as the hull efficiency, shaft transmission efficiency, and propeller relative rotation efficiency from the ship model test data. Finally, calculate the bow draft from the navigation status data. The static resistance of the actual ship is calculated based on the ship's stern draft and ground speed. Historical wave resistance is calculated based on the static resistance, ship speed, wind resistance, propulsion efficiency, and shaft power obtained from the ship model test data. Relevant feature parameters from meteorological data and navigation status data, as well as historical wave resistance, are used as training set samples. A wave resistance prediction model is trained on the training set samples using a BP neural network. Relevant feature parameters from meteorological data and navigation status data are used as inputs to predict wave resistance in a future time period based on the wave resistance prediction model.

[0082] Specifically, the propeller thrust unaffected by waves is first calculated based on wind drag and water temperature drag. The propeller thrust T S Calculate according to the following formula:

[0083] T S = (F_wind+Fwater) / (1-t) (6)

[0084] In the above formula, F_wind represents wind resistance, Fwater represents water temperature resistance, and t represents the propeller thrust reduction factor.

[0085] Then, the propeller advance speed is calculated based on the ship's speed over water. The propeller advance speed VA is calculated according to the following formula:

[0086] VA = (1 - w) Vs (7)

[0087] In the above formula, w is the wake coefficient, which can be calculated using empirical formulas or obtained through self-propulsion tests on a ship model, and Vs is the ship's speed over water.

[0088] The constant point K is calculated based on propeller thrust, propeller advance speed, and propeller diameter from ship data. T / J 2 .

[0089] K T / J 2 = Ts 1000 / (p_s VA2 D 2 (8)

[0090] In the above formula, K T / J 2 Let p_s be a constant point, D be the propeller diameter, and K be the water density. T is the thrust coefficient, and J is the propeller advance coefficient.

[0091] Then, through the calculated constant point K T / J 2 And the propeller advance coefficient calculation function F, from which the propeller advance coefficient J is obtained, is calculated according to the following formula:

[0092] J=F(K T _J 2 (9)

[0093] In the above formula, J is the propeller advance coefficient, and F is the calculation function for the propeller advance coefficient. This calculation function is obtained by plotting J and K on the propeller open-water characteristic curve. T / J 2 Obtained by curve.

[0094] Finally, the propeller open-water efficiency under specific conditions is calculated based on the propeller advance coefficient, such as... Figure 3 As shown, when the propeller advance coefficient J is exactly located at the propeller open-water test data point (J, K) T / J 2 When the propeller advance coefficient J is located between any two adjacent propeller open water test data points, the propeller open water test data is interpolated to obtain the propeller open water efficiency η0.

[0095] Then, as Figure 4 As shown, the actual ship propulsion efficiency is calculated based on the propeller open-water efficiency, as well as the hull efficiency, shaft transmission efficiency, and propeller relative rotation efficiency from the ship model test data. The actual ship propulsion efficiency ηDs is calculated according to the following formula:

[0096] ηDs=η0 ηH ηS ηR (10)

[0097] In the above formula, η0 is the propeller open-water efficiency, ηH is the hull efficiency, ηS is the shaft transmission efficiency, and ηR is the propeller relative rotation efficiency. These data can be obtained through open-water experiments on a ship model or from a series of propeller diagrams.

[0098] Then, based on the ship's bow draft, stern draft, and ground speed from the navigation status data, the static water resistance of the actual ship at different speeds under different bow and stern drafts is calculated. The static water resistance F_still of the actual ship is calculated according to the following formula:

[0099] F_still = F (Tf,Ta,Vg) (11)

[0100] In the above formula, F_still is the static water resistance of the actual ship, F (Tf,Ta,Vg) is the static water resistance function fitted based on ship model experimental data, Tf is the bow draft, Ta is the stern draft, and Vg is the ground speed.

[0101] Then, based on the ship's still water resistance, ship speed relative to water, wind resistance, ship propulsion efficiency, and the ship's shaft power converted from the ship model shaft power in the ship model test data, the historical wave resistance (i.e., wave resistance) of the actual ship is calculated. The historical wave resistance is calculated according to the following formula:

[0102] F_wave_train = Pact / (Vs) ηDs)-(F_wind+ F_still) (12)

[0103] In the above formula, F_wave_train represents the historical wave drag, Pact represents the actual ship shaft power, Vs represents the ship's speed over water, F_wind represents the wind drag, F_still represents the actual ship's still water resistance, and ηDs represents the actual ship's propulsion efficiency.

[0104] Finally, the average wave period, significant wave height, main wave direction angle, and heading angle from meteorological data, and the bow draft, stern draft, and ground speed from navigation status data were used as training input data. Historical ship wave drag was used as training output data. The training input and output data were used as training set samples. A wave drag prediction model was trained on the training set samples using a BP neural network. The average wave period, significant wave height, main wave direction angle, and heading angle from meteorological data, and the bow draft, stern draft, and ground speed from navigation status data were used as inputs to the wave drag prediction model. Based on the wave drag prediction model, the wave drag for a future time period was predicted. The prediction of wave drag can be calculated using the following simulation function:

[0105] F_wave =F(Tf,Ta,Vg,T_wave,H_wave,Direction_wave,Heading) (13)

[0106] In the above formula, F_wave is the predicted wave drag, F(Tf,Ta,Vg,T_wave,H_wave,Direction_wave,Heading) is the prediction function after training the BP artificial neural network, Tf is the bow draft, Ta is the stern draft, Vg is the ground speed, T_wave is the average wave period, H_wave is the significant wave height, Direction_wave is the main wave direction angle, and Heading is the heading angle.

[0107] Among them, such as Figure 5 As shown, the input layer of a BP neural network has a total of There are 10 node variables, and the hidden layer has a total of 10 nodes. There are [number] nodes in the output layer. indivual. For the first The input value of each node, The hidden layer is represented by the first... The node and the first node of the input layer The weights between nodes Indicates the hidden layer number 1 The threshold of each node, This represents the activation function of the hidden layer. Indicates the output layer number The node and the hidden layer The weights between nodes Then it is the output layer. The threshold of each node, This represents the activation function of the output layer. This indicates the node output of the final output layer. For the wave drag prediction module, the number of input layer nodes is 7, the number of output layer nodes is 1, and the intermediate layer and solver parameter settings need to be appropriately modified according to the target vessel.

[0108] During the forward propagation of a BP neural network signal, the output value of the output layer... The following formula is used to calculate:

[0109] (14)

[0110] Backpropagation of neural network errors will adjust and optimize the weights and thresholds of each layer. The error correction of the prediction result is calculated using the following formula:

[0111] (15)

[0112] The corrected weights and thresholds for each node are calculated using the following formulas:

[0113] (16)

[0114] (17)

[0115] (18)

[0116] (19)

[0117] in, The number of nodes in the input layer. The number of hidden layer nodes. This represents the number of nodes in the output layer. For the first The input value of each node, The hidden layer is represented by the first... The node and the first node of the input layer The weights between nodes Indicates the hidden layer number 1 The threshold of each node, This represents the activation function of the hidden layer. Indicates the output layer number The node and the hidden layer The weights between nodes Then it is the output layer. The threshold of each node, This represents the activation function of the output layer. This indicates the output of the node in the final output layer. This is the actual value.

[0118] 6. Main engine shaft power calculation steps: Calculate the corrected still water main engine shaft power based on wind resistance increase, water temperature resistance increase, predicted wave resistance increase, ship speed in the water, and actual ship propulsion efficiency.

[0119] Specifically, the corrected still-water main engine shaft power is calculated based on wind drag, water temperature drag, predicted wave drag, ship speed over water, and actual ship propulsion efficiency. The corrected still-water main engine shaft power PDs is calculated according to the following formula:

[0120] PDs = (F_wave + F_wind + Fwater) Vs / ηDs (20)

[0121] Where PDs is the corrected still water main shaft power, F_wave is the predicted wave drag, F_wind is the wind drag, Fwater is the water temperature drag, Vs is the ship's speed over water, and ηD is the actual ship propulsion efficiency under the propeller advance coefficient J.

[0122] After calculating the corrected still water main shaft power, the still water main shaft power PDs is compared with the original main shaft power Pori obtained in the self-propelled test of the ship model under still water conditions without bottom fouling. The ship's draft, trim, and speed in still water conditions should be the same as in actual sea conditions. In addition, the original main shaft power data that cannot be directly read can be obtained by interpolation calculation.

[0123] 7. Steps for calculating the fouling average coefficient: Calculate the fouling average coefficient based on the corrected still water main engine shaft power (PDs), the original main engine shaft power (Pori) from the ship model test data, and the total number of data entries in the new dataset. The fouling average coefficient (percentage) is calculated using the following formula:

[0124] (twenty one)

[0125] Where percentage is the average soil staining coefficient, n is the number of data points in 30 minutes, and PDs i This is the corrected static turbine shaft power of the i-th 30-minute data point, Pori i It is the original main engine shaft power in still water, which is the same as the actual sea conditions, bow draft, stern draft, and speed.

[0126] Then, the calculated average fouling coefficient is compared with a preset threshold. When the average fouling coefficient is greater than the preset threshold, for example, when the average fouling coefficient is greater than 20, a suggestion is given to clean the fouling of the ship.

[0127] This invention also relates to a ship fouling assessment system, which corresponds to the aforementioned ship fouling assessment method and can be understood as a system for implementing the aforementioned method. The system includes, in sequence, a data acquisition and filtering module, a mean calculation module, a wind resistance calculation module, a water temperature resistance calculation module, a wave resistance prediction module, a main engine shaft power calculation module, and a fouling average coefficient calculation module. Specifically,

[0128] The data acquisition and filtering module acquires the ship model test data obtained from the ship model test, and collects ship data, meteorological data and navigation status data during ship navigation at regular intervals. It then divides all the collected data into multiple datasets according to a certain time interval, and filters each dataset to select the datasets that meet the preset conditions.

[0129] The mean calculation module calculates the average value of each parameter in the selected data set, including ship data, meteorological data, and navigation status data. It then sorts the datasets containing the average value of each parameter in chronological order and combines several consecutive datasets in the sorted datasets into a new dataset.

[0130] The wind resistance calculation module calculates the ship's speed relative to the water based on the ship's ground speed in the navigation status data and the seawater current speed in the meteorological data in the new dataset. It also calculates the ship's actual draft and windward area based on the design draft and windward area in the ship data, as well as the bow draft and stern draft in the navigation status data. Furthermore, it calculates the wind force coefficient based on the wind direction angle in the meteorological data and the bow draft and stern draft in the navigation status data. Finally, it calculates the wind resistance based on the wind force coefficient, the wind speed in the meteorological data, and the calculated actual draft and windward area.

[0131] The water temperature resistance calculation module calculates the water temperature resistance in the new dataset based on the ship's wet surface area in the ship data, the calculated ship speed in the water, and the ship's friction resistance coefficient in the navigation status data.

[0132] The wave drag prediction module calculates propeller thrust based on wind and water temperature drag in the new dataset, and propeller advance rate based on the ship's speed in the water. It then calculates a constant point based on propeller thrust, propeller advance rate, and propeller diameter from the ship data. The module obtains the propeller advance rate coefficient from the fitted function based on the constant point, and calculates the propeller open-water efficiency under specific conditions based on the propeller advance rate coefficient. Finally, it calculates the actual ship propulsion efficiency based on the propeller open-water efficiency, as well as the hull efficiency, shaft transmission efficiency, and propeller relative rotation efficiency from the ship model test data. Finally, it calculates the bow draft from the navigation status data. The static resistance of the actual ship is calculated based on the ship's stern draft and ground speed. Historical wave resistance is calculated based on the static resistance, ship speed, wind resistance, propulsion efficiency, and shaft power obtained from the ship model test data. Relevant feature parameters from meteorological data and navigation status data, as well as historical wave resistance, are used as training set samples. A wave resistance prediction model is trained on the training set samples using a BP neural network. Relevant feature parameters from meteorological data and navigation status data are used as inputs to predict wave resistance in a future time period based on the wave resistance prediction model.

[0133] The main shaft power calculation module calculates the corrected still water main shaft power in the new dataset based on wind resistance, predicted wave resistance, water temperature resistance, ship speed in the water, and actual ship propulsion efficiency.

[0134] The fouling average coefficient calculation module calculates the fouling average coefficient based on the corrected still water main engine shaft power, the original main engine shaft power in the ship model test data, and the total number of data entries in the new dataset. It then compares the fouling average coefficient with a preset threshold. When the fouling average coefficient is greater than the preset threshold, the ship's fouling is cleaned.

[0135] Preferably, the ship data includes propeller diameter, design draft and design draft windward area, and ship wetted surface area; the meteorological data includes wind speed, wind direction, current direction, seawater current speed, average wave period, significant wave height, and main wave direction angle; the navigation status data includes ship bow draft, ship stern draft, rudder angle, heading angle, ship friction drag coefficient, and ground speed.

[0136] Preferably, the relevant characteristic parameters include bow draft, stern draft, ground speed, heading angle, mean wave period, significant wave height, and main wave heading angle.

[0137] Preferably, the ship model test includes the ship model self-propulsion test, the ship model wind tunnel test, and the ship model propeller open water test.

[0138] Preferably, in the wave drag prediction module, calculating the propeller open-water efficiency under specific conditions based on the propeller advance coefficient includes:

[0139] In the propeller open-water performance curve constructed based on the propeller advance rate coefficient, constant point, and propeller open-water efficiency, when the propeller advance rate coefficient is located at the propeller open-water experimental data point, the propeller open-water efficiency is directly obtained; when the propeller advance rate coefficient is located between any two adjacent propeller open-water experimental data points, the propeller open-water experimental data is interpolated to obtain the propeller open-water efficiency.

[0140] This invention provides an objective and scientific method and system for assessing ship fouling. It is also a method and system for assessing ship fouling by fusing multiple feature parameters. Based on ship model test data, ship data, meteorological data, and navigation status data, it uses a BP neural network and specific calculation methods to calculate the average fouling coefficient. Combined with specific judgment methods, it assesses the ship's fouling condition and performs cleaning, effectively increasing the accuracy of the assessment. At the same time, it can reduce the need for personnel to go into the water to inspect and assess the ship's fouling condition, greatly enhancing safety.

[0141] It should be noted that the specific embodiments described above enable those skilled in the art to more fully understand the present invention, but do not limit the present invention in any way. Therefore, although the present invention has been described in detail with reference to the accompanying drawings and embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the present invention. In short, all technical solutions and improvements that do not depart from the spirit and scope of the present invention should be covered within the protection scope of the present invention patent.

Claims

1. A method for assessing ship fouling, characterized in that, Includes the following steps: Data collection and filtering steps: Obtain the ship model test data obtained from the ship model test, and collect ship data, meteorological data and navigation status data during ship navigation at regular intervals. Divide all the collected data into multiple datasets according to a certain time interval, and then filter each dataset to select the datasets that meet the preset conditions. Mean calculation steps: Calculate the average value of each parameter in the selected data set, including ship data, meteorological data, and navigation status data. Sort the datasets containing the average value of each parameter in chronological order. Then, combine several consecutive datasets in the sorted datasets to form a new dataset. Wind drag calculation steps: In the new dataset, calculate the ship's speed relative to the water based on the ship's ground speed in the navigation status data and the seawater current speed in the meteorological data. Calculate the ship's actual draft and windward area based on the ship's design draft and windward area in the ship data, as well as the bow and stern drafts in the navigation status data. Calculate the wind force coefficient based on the wind direction angle in the meteorological data and the bow and stern drafts in the navigation status data. Calculate the wind drag increase based on the wind force coefficient, the wind speed in the meteorological data, and the calculated actual draft and windward area. Steps for calculating water temperature drag increase: In the new dataset, calculate the water temperature drag increase based on the ship's wetted surface area in the ship data, the calculated ship speed in the water, and the ship's frictional resistance coefficient in the navigation status data. Wave drag prediction steps: In the new dataset, calculate the propeller thrust based on wind drag and water temperature drag, and calculate the propeller advance speed based on the ship's speed above water. Calculate constant points based on propeller thrust, propeller advance speed, and propeller diameter from the ship data. Obtain the propeller advance speed coefficient from the fitted function based on the constant points. Then, calculate the propeller open-water efficiency under specific conditions based on the propeller open-water efficiency, as well as the hull efficiency, shaft transmission efficiency, and propeller relative rotation efficiency from the ship model test data. Finally, calculate the ship's static water resistance based on the bow draft, stern draft, and ship's speed above ground. Historical wave drag is calculated using water resistance, ship speed relative to water, wind drag, actual ship propulsion efficiency, and actual ship shaft power obtained from ship model shaft power in ship model test data. Relevant characteristic parameters from meteorological data and navigation status data, along with historical wave drag, are used as training set samples. A wave drag prediction model is obtained by training the training set samples using a BP neural network. Relevant characteristic parameters from meteorological data and navigation status data are then used as inputs to predict wave drag for a future time period based on the wave drag prediction model. These relevant characteristic parameters include bow draft, stern draft, ground speed, heading angle, mean wave period, significant wave height, and main wave heading angle. Main engine shaft power calculation steps: In the new dataset, calculate the corrected still water main engine shaft power based on wind drag, water temperature drag, predicted wave drag, ship speed over water, and actual ship propulsion efficiency. The corrected still water main engine shaft power PDs is calculated according to the following formula: PDs =(F_wave+F_wind+ Fwater)* Vs / ηDs, Where PDs is the corrected still water main shaft power, F_wave is the predicted wave drag, F_wind is the wind drag, Fwater is the water temperature drag, Vs is the ship's speed over water, and ηDs is the actual ship propulsion efficiency under the propeller advance coefficient J. Steps for calculating the average fouling coefficient: Calculate the average fouling coefficient based on the corrected still water main engine shaft power, the original main engine shaft power in the ship model test data, and the total number of data entries in the new dataset. Compare the average fouling coefficient with a preset threshold. When the average fouling coefficient is greater than the preset threshold, clean the fouling on the ship.

2. The ship fouling assessment method according to claim 1, characterized in that, In the data collection and filtering steps, each dataset is filtered, including stable navigation filtering, shallow water effect filtering, and stable wind and wave filtering. And / or, the ship data includes propeller diameter, design draft, design draft windward area, and ship wetted surface area; the meteorological data includes wind speed, wind direction, current direction, wind angle, sea current speed, average wave period, significant wave height, and main wave angle; the navigation status data includes ship bow draft, ship stern draft, rudder angle, heading angle, ship friction drag coefficient, and ground speed.

3. The ship fouling assessment method according to claim 1, characterized in that, The ship model tests include ship model self-propulsion tests, ship model wind tunnel tests, and ship model propeller open water tests.

4. The ship fouling assessment method according to claim 1, characterized in that, In the wave drag prediction step, calculating the propeller open-water efficiency under specific conditions based on the propeller advance coefficient includes: In the propeller open-water performance curve constructed based on the propeller advance rate coefficient, constant point, and propeller open-water efficiency, when the propeller advance rate coefficient is located at the propeller open-water experimental data point, the propeller open-water efficiency is directly obtained; when the propeller advance rate coefficient is located between any two adjacent propeller open-water experimental data points, the propeller open-water experimental data is interpolated to obtain the propeller open-water efficiency.

5. A ship fouling assessment system, characterized in that, It includes, in sequence, a data acquisition and filtering module, a mean calculation module, a wind resistance calculation module, a water temperature resistance calculation module, a wave resistance prediction module, a main shaft power calculation module, and a sludge average coefficient calculation module. The data acquisition and filtering module acquires the ship model test data obtained from the ship model test, and collects ship data, meteorological data and navigation status data during ship navigation at regular intervals. It then divides all the collected data into multiple datasets according to a certain time interval, and filters each dataset to select the datasets that meet the preset conditions. The mean calculation module calculates the average value of each parameter in the selected data set, including ship data, meteorological data, and navigation status data. It then sorts the datasets containing the average value of each parameter in chronological order and combines several consecutive datasets in the sorted datasets into a new dataset. The wind resistance calculation module calculates the ship's speed relative to the water based on the ship's ground speed in the navigation status data and the seawater current speed in the meteorological data in the new dataset. It also calculates the ship's actual draft and windward area based on the design draft and windward area in the ship data, as well as the bow draft and stern draft in the navigation status data. Furthermore, it calculates the wind force coefficient based on the wind direction angle in the meteorological data and the bow draft and stern draft in the navigation status data. Finally, it calculates the wind resistance based on the wind force coefficient, the wind speed in the meteorological data, and the calculated actual draft and windward area. The water temperature resistance calculation module calculates the water temperature resistance in the new dataset based on the ship's wet surface area in the ship data, the calculated ship speed in the water, and the ship's friction resistance coefficient in the navigation status data. The wave drag prediction module calculates propeller thrust based on wind and water temperature drag in the new dataset, and propeller advance rate based on the ship's speed over water. It then calculates a constant point based on propeller thrust, propeller advance rate, and propeller diameter from the ship data. The module obtains the propeller advance rate coefficient from the fitted function of the constant point, calculates the propeller open-water efficiency under specific conditions based on the propeller advance rate coefficient, and calculates the actual ship propulsion efficiency based on the propeller open-water efficiency, as well as the hull efficiency, shaft transmission efficiency, and propeller relative rotation efficiency from the ship model test data. Finally, it calculates the actual ship's static water resistance based on the ship's bow draft, stern draft, and ground speed from the navigation status data. Historical wave drag was calculated based on the ship's still water resistance, ship speed above water, wind drag, ship propulsion efficiency, and ship shaft power obtained from ship model test data. Relevant characteristic parameters from meteorological and navigation data, along with historical wave drag, were used as training set samples. A wave drag prediction model was trained using a BP neural network on the training set samples. Relevant characteristic parameters from meteorological and navigation data were then used as inputs to predict wave drag for a future time period. These relevant characteristic parameters included bow draft, stern draft, ground speed, heading angle, mean wave period, significant wave height, and main wave heading angle. The main engine shaft power calculation module calculates the corrected still water main engine shaft power based on wind drag, water temperature drag, predicted wave drag, ship speed in the water, and actual ship propulsion efficiency in the new dataset. The corrected still water main engine shaft power PDs is calculated according to the following formula: PDs =(F_wave+F_wind+ Fwater)* Vs / ηDs, Where PDs is the corrected still water main shaft power, F_wave is the predicted wave drag, F_wind is the wind drag, Fwater is the water temperature drag, Vs is the ship's speed over water, and ηDs is the actual ship propulsion efficiency under the propeller advance coefficient J. The fouling average coefficient calculation module calculates the fouling average coefficient based on the corrected still water main engine shaft power, the original main engine shaft power in the ship model test data, and the total number of data entries in the new dataset. It then compares the fouling average coefficient with a preset threshold. When the fouling average coefficient is greater than the preset threshold, the ship's fouling is cleaned.

6. The ship fouling assessment system according to claim 5, characterized in that, The ship data includes propeller diameter, design draft, design draft windward area, and ship wetted surface area; the meteorological data includes wind speed, wind direction, current direction, wind angle, sea current speed, average wave period, significant wave height, and main wave angle; the navigation status data includes bow draft, stern draft, rudder angle, heading angle, ship friction drag coefficient, and ground speed.

7. The ship fouling assessment system according to claim 5, characterized in that, The ship model tests include ship model self-propulsion tests, ship model wind tunnel tests, and ship model propeller open water tests.

8. The ship fouling assessment system according to claim 5, characterized in that, In the wave drag prediction module, the propeller open-water efficiency under specific conditions is calculated based on the propeller advance coefficient, including: In the propeller open-water performance curve constructed based on the propeller advance rate coefficient, constant point, and propeller open-water efficiency, when the propeller advance rate coefficient is located at the propeller open-water experimental data point, the propeller open-water efficiency is directly obtained; when the propeller advance rate coefficient is located between any two adjacent propeller open-water experimental data points, the propeller open-water experimental data is interpolated to obtain the propeller open-water efficiency.

Images