Calculation method of offshore roughness attenuation factor

Abstract

The invention relates to the field of marine atmospheric waveguide environment detection, and discloses a method for calculating a marine roughness attenuation factor, which comprises the following steps of: measuring meteorological and hydrological data of a target sea area; installing a microwave link in a target sea area, and calculating an electromagnetic wave propagation loss true value; calculating a roughness attenuation factor of the target sea area, and dividing the wind speed into a plurality of optimization intervals; a roughness attenuation factor prediction model is constructed in a weighted mode, for each divided optimization interval, a root-mean-square error between an electromagnetic wave propagation loss prediction value and a true value serves as a fitness function of a particle swarm optimization algorithm, and an optimal weight coefficient of the prediction model is obtained through iteration; and calculating roughness attenuation factors at different wind speeds by using the optimal prediction model combination. According to the method disclosed by the invention, different roughness attenuation factor models are weighted, the roughness attenuation factor prediction model with higher applicability is obtained through particle swarm optimization, and the method can be suitable for sea surface roughness conditions under different wind speed conditions.

Description

Technical Field

[0001] The invention relates to the field of ocean atmospheric waveguide environment detection, and in particular to a method for calculating an ocean roughness attenuation factor. Background Art

[0002] In recent years, radio technology has developed rapidly, and people's lifestyles have also been innovated. Only by mastering the propagation characteristics of radio waves can we provide theoretical and technical support for various communication, navigation and other systems. Electromagnetic wave propagation loss is an important characteristic of radio wave propagation characteristics, and electromagnetic wave propagation loss is of great significance for maritime communications and marine resource development. In the tropospheric waveguide, the propagation loss of radar electromagnetic waves is usually weaker than that in the normal atmospheric environment, and over-the-horizon propagation occurs, thereby expanding the detection range of the radar. However, since the electromagnetic waves are trapped in the waveguide layer, the electromagnetic wave energy is less distributed in the upper area of ​​the waveguide layer, resulting in unexpected holes, which reduces the target detection performance of the radar.

[0003] When calculating electromagnetic wave propagation loss at sea, the roughness attenuation factor is one of the very important parameters. At present, the roughness attenuation factor is mainly calculated by three calculation models, namely the Ament model, the Miller Brown model and the Shadowed model. Due to the complex and changeable conditions at sea, the roughness of the sea surface will change with the change of wind speed. Since different models are applicable to different sea surface roughness when calculating the sea surface roughness attenuation factor, the calculation of the roughness attenuation factor using a single model will be inaccurate, thus affecting the calculation of electromagnetic wave propagation loss. Summary of the invention

[0004] To solve the above technical problems, the present invention provides a method for calculating the offshore roughness attenuation factor, weights different roughness attenuation factor models, and obtains a roughness attenuation factor prediction model with higher applicability in the target sea area, which can be applied to sea surface roughness conditions under different wind speeds.

[0005] To achieve the above object, the technical solution of the present invention is as follows: A method for calculating a roughness attenuation factor at sea comprises the following steps: Step 1: Use a floating platform to install meteorological and hydrological sensors to measure meteorological and hydrological data of the target sea area; Step 2: Install a microwave link in the target sea area and calculate the true value of electromagnetic wave propagation loss through the power loss between the receiving end and the transmitting end; Step 3, using the Ament model, Miller Brown model, and Shadowed model to calculate the roughness attenuation factor of the target sea area, and dividing the wind speed into multiple optimization intervals based on the slope of the calculated roughness attenuation factor curve; Step 4, substituting the meteorological and hydrological data into the evaporation duct prediction model to obtain the evaporation duct height, and then using the evaporation duct height to calculate the evaporation duct atmospheric correction refractive index profile; Step 5, using the Ament model, Miller Brown model, and Shadowed model to weight and construct a roughness attenuation factor prediction model, for each divided optimization interval, using the roughness attenuation factor calculated by the prediction model combined with the atmospheric correction refractive index profile of the evaporation waveguide to calculate the predicted value of electromagnetic wave propagation loss, the root mean square error between the predicted value of electromagnetic wave propagation loss and the true value of electromagnetic wave propagation loss is used as the fitness function of the particle swarm optimization algorithm, and the optimal weight coefficient of the prediction model is iteratively obtained; for the multiple divided optimization intervals, the same method is used to iterate to find the optimal weight coefficient of each optimization interval, and the optimal prediction model combination is obtained; Step 6: Calculate the roughness attenuation factor at different wind speeds using the optimal prediction model combination.

[0006] In the above scheme, in step 1, the measured meteorological and hydrological data include wind speed, temperature, humidity, atmospheric pressure, sea surface temperature, and wavelength of the target sea area.

[0007] In the above scheme, in step 2, a receiving end of a microwave link is installed on the floating platform, and a transmitting end of the microwave link is installed on the shore 150 km away from the floating platform.

[0008] In the above scheme, in step 3, the true value of electromagnetic wave propagation loss Calculated by the following formula: ; in, is the transmission power, is the transmit antenna gain, is the received power, is the receiving antenna gain, is the loss of the feeder line.

[0009] In the above scheme, in step 5, the roughness attenuation factor prediction model constructed is as follows: ; in, Represents the roughness attenuation factor calculated by the Ament model, Represents the roughness attenuation factor calculated by the Miller Brown model, Represents the roughness attenuation factor calculated by the Shadowed model. Represents the weighted roughness attenuation factor.

[0010] In the above scheme, in step 5, the method for calculating the predicted value of electromagnetic wave propagation loss is as follows: (1) Roughness attenuation factor calculated using the prediction model Calculate the corrected reflection coefficient : ; in, is the Fresnel reflection coefficient, which represents the effect of the rough sea surface on the scattering of electromagnetic waves; (2) Calculation of surface impedance coefficient : ; in, is the local surface grazing angle, is the free space wave number, i is plural; (3) Determine the impedance boundary conditions of the parabolic equation: ; The parabolic equation expression under given evaporation waveguide conditions is as follows: ; in, represents the horizontally polarized electric field component and the vertically polarized electric field component, is the height, is the horizontal distance above sea level, Corrected refractive index profile for evaporative waveguide atmosphere; (4) The parabolic equation is solved using the step-by-step Fourier algorithm. In each step, the Fourier transform operation is performed in combination with the impedance boundary condition, and then multiplied by the atmospheric refractive index to obtain Finally, the predicted value of electromagnetic wave propagation loss is obtained : ; in, The frequency of the radio wave.

[0011] In a further technical solution, the atmospheric refractive index is calculated as follows: First, the atmospheric refractive index N is calculated based on the following relationship between the atmospheric corrected refractive index of the evaporation waveguide and the atmospheric refractive index N: ; in, For height The atmospheric corrected refractive index of the evaporation waveguide at a is the average radius of the Earth; z is the height; Then, the atmospheric refractive index n is calculated according to the following relationship between the atmospheric refractive index N and the atmospheric refractive index n: .

[0012] In the above scheme, in step 5, the method of iteratively obtaining the optimal weight coefficient of the prediction model using the particle swarm optimization algorithm is as follows: S1: Initialize particle swarm parameters: In the roughness attenuation factor prediction model, the weight coefficient is 3, and N 3D particles are randomly generated in the search space as the initial particle swarm; S2: Calculate the fitness value: ; in, is the fitness value, For the The true value of the electromagnetic wave propagation loss of each particle, For the The predicted value of electromagnetic wave propagation loss for each particle, is the number of particles; S3: Update particle speed and position according to fitness value. The position of a particle is , Indicates The position of a particle in the second dimension; The speed of a particle is , Indicates The speed of a particle in the second dimension; The optimal position searched by a particle is , Indicates The optimal position of a particle in the second dimension; the optimal position searched by the particle swarm is , Indicates the optimal position of the particle swarm in the second dimension; The speed and position of each particle after iteration are updated using the following formula: ; ; Where k is the number of iterations, is the inertia weight, , is the learning factor, and its value range is [0,2]. is a random number between 0 and 1, Representative The particle iterates k times to search for the optimal position in the second dimension; Representative The speed of a particle iterating k+1 times in the second dimension, Representative The position of a particle in the second dimension after iterating k times, Representative The position of a particle in the second dimension after iterating k+1 times; the calculation method of the position and velocity of the particle in the first and third dimensions after each iteration is the same as that of the second dimension; S4: Repeat steps S2 and S3 until the maximum number of iterations is reached and the The optimal position searched by a particle and the optimal position searched by a particle swarm , In Corresponding to the 3 optimal weight coefficients of the prediction model.

[0013] In the above scheme, the specific method of step 6 is as follows: first, the wind speed of the target sea area is obtained, and then the optimization interval is determined according to the wind speed, and then the optimized prediction model of the optimization interval is selected to calculate the roughness attenuation factor.

[0014] Through the above technical solution, the method for calculating the offshore roughness attenuation factor provided by the present invention has the following beneficial effects: 1. The present invention weights different roughness attenuation factor models to obtain a roughness attenuation factor prediction model with higher applicability in the target sea area. This model combines different models with weights and can be applicable to different sea surface roughness conditions caused by different wind speeds; 2. The present invention uses a floating platform to measure the air temperature, relative humidity, wind speed, sea surface temperature, and wavelength of the target sea area, uses an evaporation waveguide prediction model to obtain the atmospheric correction refractive index profile of the evaporation waveguide, determines the parabolic equation of given conditions, and calculates the predicted value of electromagnetic wave propagation loss in different wind speed intervals according to the weighted model. The roughness attenuation factor under different wind speed conditions can be measured and calculated in real time, thereby calculating the sea surface electromagnetic wave propagation loss value; 3. The measurement area of ​​the present invention is at sea, and the evaporation duct at sea is a very common abnormal phenomenon, which has a great influence on the propagation of electromagnetic waves. Therefore, when calculating the electromagnetic wave propagation loss, the electromagnetic wave propagation loss in the case of the evaporation duct is considered, which improves the applicability of the model at sea; 4. The present invention uses the roughness attenuation factor prediction model after weighted fusion and multiple rounds of iterations of the particle swarm optimization algorithm to finally obtain the roughness attenuation factor prediction model with the best weight combination, thereby improving the accuracy of the model's calculation of the roughness attenuation factor in the target sea area, thereby improving the accuracy of the calculation results of the electromagnetic wave propagation loss; 5. Since the roughness attenuation factor cannot be directly measured, the present invention uses the roughness attenuation factor to calculate the electromagnetic wave propagation loss as the predicted value in the fitness function of the particle swarm optimization algorithm, uses the power loss at the transmitting and receiving ends of the microwave link to calculate the true value of the electromagnetic wave propagation loss, and uses the root mean square error between the two as the fitness for optimization, thereby avoiding the situation where the roughness attenuation factor cannot be directly optimized due to the inability to measure the true value, and realizes the optimization of the roughness attenuation factor using an intermediate quantity. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art are briefly introduced below.

[0016] Figure 1 A schematic flow chart of a method for calculating a roughness attenuation factor at sea disclosed in an embodiment of the present invention; Figure 2 This is a schematic diagram of the roughness attenuation factor within a wind speed of 0-15M / S disclosed in an embodiment of the present invention; Figure 3 A schematic diagram of a particle swarm optimization algorithm disclosed in an embodiment of the present invention. DETAILED DESCRIPTION

[0017] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0018] The present invention provides a method for calculating the marine roughness attenuation factor, such as Figure 1 As shown, the following steps are included: Step 1: Use a floating platform to install meteorological and hydrological sensors to measure the meteorological and hydrological data of the target sea area.

[0019] Meteorological and hydrological sensors are installed on the buoy platform to measure the meteorological and hydrological data of the target sea area; wind speed sensors are installed to measure the wind speed of the target sea area, temperature and humidity sensors are installed to measure the temperature and relative humidity of the target sea area, atmospheric pressure sensors are installed to measure the atmospheric pressure of the target sea area, infrared sensors are installed to measure the sea surface temperature, and the wavelength is measured using the buoy's own wave sensor.

[0020] Step 2: Install a microwave link in the target sea area and calculate the true value of the electromagnetic wave propagation loss through the power loss between the receiving end and the transmitting end.

[0021] The present invention installs a receiving end of a microwave link on a floating platform, installs the receiving end of the microwave link on a shore within 150 km, and calculates the actual electromagnetic wave propagation loss through the power loss between the receiving end and the transmitting end.

[0022] True value of electromagnetic wave propagation loss Calculated by the following formula: ; in, is the transmission power, is the transmit antenna gain, is the received power, is the receiving antenna gain, is the loss of the feeder line.

[0023] Step 3: Use the Ament model, Miller Brown model, and Shadowed model to calculate the roughness attenuation factor of the target sea area, and divide the wind speed into multiple optimization intervals based on the slope of the calculated roughness attenuation factor curve.

[0024] The three roughness attenuation factor calculation models are: Roughness attenuation factor calculation model Ament model: ; Roughness attenuation factor calculation model Miller Brown model: ; Roughness attenuation factor calculation model Shadowed model: ; ; in, is the surface Rayleigh roughness parameter, represents the wave number, , is the wavelength; is the grazing angle, estimated using the method of geometric optics, is the angle between the radar emission wave and the sea level, is the root mean square height deviation of the sea surface, which can be calculated using the wind speed. The formula is ,in, represents the wind speed measured by the wind speed sensor in the floating platform, is a modified Bessel function of order zero.

[0025] Substitute wind speed, wavelength, and grazing angle into the above formula to obtain Figure 2The roughness attenuation factor curve shown can divide the optimization interval into multiple intervals, where the slope is not large when the wind speed is 0m / s-4m / s, so the wind speed 0m / s-4m / s can be divided into an optimization interval. The slope of the roughness attenuation factor curve is large when the wind speed is 4m / s-11m / s, so each wind speed can be regarded as an optimization interval. The slope of the roughness attenuation curve is small when the wind speed is 11m / s-15m / s, so 11m / s-15m / s is regarded as an optimization interval.

[0026] Step 4: Substitute the meteorological and hydrological data into the evaporation duct prediction model to obtain the evaporation duct height, and then use the evaporation duct height to calculate the evaporation duct atmospheric corrected refractive index profile.

[0027] Substitute the meteorological and hydrological parameters measured in step 1, wind speed, temperature, humidity, atmospheric pressure, and sea surface temperature, into the evaporation duct prediction model to obtain the evaporation duct height , and then the calculated evaporation duct height Substitute it into the mathematical expression of the atmospheric correction refractive index profile of the evaporation duct to obtain the atmospheric correction refractive index profile of the evaporation duct. The mathematical expression of the atmospheric correction refractive index profile of the evaporation duct is shown as follows: ; in, The atmospheric correction refractive index of the sea surface is generally considered to be 350. For height The atmospheric corrected refractive index of the evaporation waveguide at is the evaporation waveguide height, roughness length , the atmospheric correction refractive index profile of the evaporation waveguide in the target sea area can be expressed as , To measure the horizontal distance to the target sea area.

[0028] Step 5, use the Ament model, Miller Brown model, and Shadowed model to weight and construct a roughness attenuation factor prediction model. For each optimization interval divided, use the roughness attenuation factor calculated by the prediction model and the atmospheric corrected refractive index profile of the evaporation waveguide to calculate the predicted value of the electromagnetic wave propagation loss. The root mean square error between the predicted value of the electromagnetic wave propagation loss and the true value of the electromagnetic wave propagation loss is used as the fitness function of the particle swarm optimization algorithm, and the optimal weight coefficient of the prediction model is iteratively obtained. For the multiple optimization intervals divided, the same method is used to iterate to find the optimal weight coefficient of each optimization interval, and the optimal prediction model combination is obtained.

[0029] Step 1: Construct a roughness attenuation factor prediction model: The roughness attenuation factor prediction model constructed is as follows: ; in, Represents the roughness attenuation factor calculated by the Ament model, Represents the roughness attenuation factor calculated by the Miller Brown model, Represents the roughness attenuation factor calculated by the Shadowed model. Represents the weighted roughness attenuation factor.

[0030] Step 2: Calculate the predicted value of electromagnetic wave propagation loss: The method for calculating the predicted value of electromagnetic wave propagation loss is as follows: (1) Roughness attenuation factor calculated using the prediction model Calculate the corrected reflection coefficient : ; in, is the Fresnel reflection coefficient, which represents the effect of the rough sea surface on the scattering of electromagnetic waves; (2) Calculation of surface impedance coefficient : ; in, is the local surface grazing angle, is the free space wave number, i is plural; (3) Determine the impedance boundary conditions of the parabolic equation: ; The parabolic equation expression under given evaporation waveguide conditions is as follows: ; in, represents the horizontally polarized electric field component and the vertically polarized electric field component, is the height, is the horizontal distance above sea level, Corrected refractive index profile for evaporative waveguide atmosphere; (4) The parabolic equation is solved using the step-by-step Fourier algorithm. In each step, the Fourier transform operation is performed in combination with the impedance boundary condition, and then multiplied by the atmospheric refractive index to obtain Finally, the predicted value of electromagnetic wave propagation loss is obtained : ; in, The frequency of the radio wave.

[0031] The atmospheric refractive index is calculated as follows: First, the atmospheric refractive index N is calculated based on the following relationship between the atmospheric corrected refractive index of the evaporation waveguide and the atmospheric refractive index N: ; in, For height The atmospheric corrected refractive index of the evaporation waveguide at a is the average radius of the Earth; z is the height; Then, the atmospheric refractive index n is calculated according to the following relationship between the atmospheric refractive index N and the atmospheric refractive index n: .

[0032] Step 3: Use the particle swarm optimization algorithm to iteratively obtain the optimal weight coefficient of the prediction model: like Figure 3 As shown in the figure, the method of iteratively obtaining the optimal weight coefficient of the prediction model using the particle swarm optimization algorithm is as follows: S1: Initialize particle swarm parameters: In the roughness attenuation factor prediction model, the weight coefficient is 3, and N 3D particles are randomly generated in the search space as the initial particle swarm; S2: Calculate the fitness value: ; in, is the fitness value, For the The true value of the electromagnetic wave propagation loss of each particle, For the The predicted value of electromagnetic wave propagation loss for each particle, is the number of particles; S3: Update particle speed and position according to fitness value. The position of a particle is , Indicates The position of a particle in the second dimension; The speed of a particle is , Indicates The speed of a particle in the second dimension; The optimal position searched by a particle is , Indicates The optimal position of a particle in the second dimension; the optimal position searched by the particle swarm is , Indicates the optimal position of the particle swarm in the second dimension; The speed and position of each particle after iteration are updated using the following formula: ; ; Where k is the number of iterations, is the inertia weight, , is the learning factor, and its value range is [0,2]. is a random number between 0 and 1, Representative The particle iterates k times to search for the optimal position in the second dimension; Representative The speed of a particle iterating k+1 times in the second dimension, Representative The position of a particle in the second dimension after iterating k times, Representative The position of a particle in the second dimension after iterating k+1 times; the calculation method of the position and velocity of the particle in the first and third dimensions after each iteration is the same as that of the second dimension; S4: Repeat steps S2 and S3 until the maximum number of iterations is reached and the The optimal position searched by a particle and the optimal position searched by a particle swarm , In Corresponding to the 3 optimal weight coefficients of the prediction model.

[0033] Step 6: Calculate the roughness attenuation factor at different wind speeds using the optimal prediction model combination.

[0034] The specific method is as follows: first obtain the wind speed of the target sea area, then determine which optimization interval it belongs to based on the wind speed, and then select the optimized prediction model of the optimization interval to calculate the roughness attenuation factor.

[0035] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for calculating the marine roughness attenuation factor, characterized in that: The steps include: Step 1: Use a floating platform to install meteorological and hydrological sensors to measure meteorological and hydrological data of the target sea area; Step 2: Install a microwave link in the target sea area and calculate the true value of electromagnetic wave propagation loss through the power loss between the receiving end and the transmitting end; Step 3, using the Ament model, Miller Brown model, and Shadowed model to calculate the roughness attenuation factor of the target sea area, and dividing the wind speed into multiple optimization intervals based on the slope of the calculated roughness attenuation factor curve; Step 4, substituting the meteorological and hydrological data into the evaporation duct prediction model to obtain the evaporation duct height, and then using the evaporation duct height to calculate the evaporation duct atmospheric correction refractive index profile; Step 5, using the Ament model, Miller Brown model, and Shadowed model to weight and construct a roughness attenuation factor prediction model, for each divided optimization interval, using the roughness attenuation factor calculated by the prediction model combined with the atmospheric correction refractive index profile of the evaporation waveguide to calculate the predicted value of electromagnetic wave propagation loss, the root mean square error between the predicted value of electromagnetic wave propagation loss and the true value of electromagnetic wave propagation loss is used as the fitness function of the particle swarm optimization algorithm, and the optimal weight coefficient of the prediction model is iteratively obtained; for the multiple divided optimization intervals, the same method is used to iterate to find the optimal weight coefficient of each optimization interval, and the optimal prediction model combination is obtained; Step 6: Calculate the roughness attenuation factor at different wind speeds using the optimal prediction model combination.

2. The method for calculating the marine roughness attenuation factor according to claim 1, characterized in that: In step 1, the measured meteorological and hydrological data include wind speed, temperature, humidity, atmospheric pressure, sea surface temperature, and wavelength of the target sea area.

3. The method for calculating the marine roughness attenuation factor according to claim 1, characterized in that: In step 2, a receiving end of the microwave link is installed on the floating platform, and a transmitting end of the microwave link is installed on the shore 150 km away from the floating platform.

4. The method for calculating the marine roughness attenuation factor according to claim 1, characterized in that: In step 3, the true value of electromagnetic wave propagation loss Calculated by the following formula: ； in, is the transmission power, is the transmit antenna gain, is the received power, is the receiving antenna gain, is the loss of the feeder line.

5. The method for calculating the marine roughness attenuation factor according to claim 1, characterized in that: In step 5, the roughness attenuation factor prediction model constructed is as follows: ； in, Represents the roughness attenuation factor calculated by the Ament model, Represents the roughness attenuation factor calculated by the Miller Brown model, Represents the roughness attenuation factor calculated by the Shadowed model. Represents the weighted roughness attenuation factor.

6. The method for calculating the marine roughness attenuation factor according to claim 1, characterized in that: In step 5, the method for calculating the predicted value of electromagnetic wave propagation loss is as follows: (1) Roughness attenuation factor calculated using the prediction model Calculate the corrected reflection coefficient : ； in, is the Fresnel reflection coefficient, which represents the effect of the rough sea surface on the scattering of electromagnetic waves; (2) Calculation of surface impedance coefficient : ； in, is the local surface grazing angle, is the free space wave number, i is plural; (3) Determine the impedance boundary conditions of the parabolic equation: ； The parabolic equation expression under given evaporation waveguide conditions is as follows: ； in, represents the horizontally polarized electric field component and the vertically polarized electric field component, is the height, is the horizontal distance above sea level, Corrected refractive index profile for evaporative waveguide atmosphere; (4) The parabolic equation is solved using the step-by-step Fourier algorithm. In each step, the Fourier transform operation is performed in combination with the impedance boundary condition, and then multiplied by the atmospheric refractive index to obtain , and finally get the predicted value of electromagnetic wave propagation loss : ； in, The frequency of the radio wave.

7. A method for calculating the marine roughness attenuation factor according to claim 6, characterized in that: The atmospheric refractive index is calculated as follows: First, the atmospheric refractive index N is calculated based on the following relationship between the atmospheric corrected refractive index of the evaporation waveguide and the atmospheric refractive index N: ； in, For height The atmospheric corrected refractive index of the evaporation waveguide at a is the average radius of the Earth; z is the height; Then, the atmospheric refractive index n is calculated according to the following relationship between the atmospheric refractive index N and the atmospheric refractive index n: 。 8. The method for calculating the marine roughness attenuation factor according to claim 1, characterized in that: In step 5, the method of iteratively obtaining the optimal weight coefficient of the prediction model using the particle swarm optimization algorithm is as follows: S1: Initialize particle swarm parameters: In the roughness attenuation factor prediction model, the weight coefficient is 3, and N 3D particles are randomly generated in the search space as the initial particle swarm; S2: Calculate the fitness value: ； in, is the fitness value, For the The true value of the electromagnetic wave propagation loss of each particle, For the The predicted value of electromagnetic wave propagation loss for each particle, is the number of particles; S3: Update particle speed and position according to fitness value. The position of a particle is , Indicates The position of a particle in the second dimension; The speed of a particle is , Indicates The speed of a particle in the second dimension; The optimal position searched by a particle is , Indicates The optimal position of a particle in the second dimension; the optimal position searched by the particle swarm is , Indicates the optimal position of the particle swarm in the second dimension; The speed and position of each particle after iteration are updated using the following formula: ； ； Where k is the number of iterations, is the inertia weight, , is the learning factor, and its value range is [0,2]. is a random number between 0 and 1, Representative The particle iterates k times to search for the optimal position in the second dimension; Representative The speed of a particle iterating k+1 times in the second dimension, Representative The position of a particle in the second dimension after k iterations, Representative The position of a particle in the second dimension after iterating k+1 times; the calculation method of the position and velocity of the particle in the first and third dimensions after each iteration is the same as that of the second dimension; S4: Repeat steps S2 and S3 until the maximum number of iterations is reached and the The optimal position searched by a particle and the optimal position searched by a particle swarm , In Corresponding to the 3 optimal weight coefficients of the prediction model.

9. The method for calculating the marine roughness attenuation factor according to claim 1, characterized in that: The specific method of step 6 is as follows: first, obtain the wind speed of the target sea area, then determine which optimization interval it belongs to based on the wind speed, and then select the optimized prediction model of the optimization interval to calculate the roughness attenuation factor.

Images