A calculation method for the atmospheric modified refractive index profile under the effect of turbulence

Abstract

The present invention relates to the field of marine atmospheric boundary layer detection, and discloses a method for calculating the atmospheric modified refractive index profile under turbulent effects, which includes the following steps: obtaining meteorological observation data at different heights; obtaining different predicted temperature and humidity profiles through different dimensionless universal functions of temperature and humidity; conducting a sensitivity analysis on the dimensionless universal functions of temperature and humidity, merging or splitting the meteorological parameter intervals according to the analysis results, and selecting the dimensionless universal functions of temperature and humidity to be optimized; optimizing the coefficients by using a particle swarm intelligent optimization algorithm; calculating the refractive index structure constant based on the optimized dimensionless universal functions of temperature and humidity under this meteorological parameter interval, and finally obtaining the atmospheric modified refractive index profile under turbulent effects. The method disclosed by the present invention takes into account the influence of turbulent effects on the atmospheric modified refractive index, and at the same time selects the optimal dimensionless universal function for different meteorological parameter intervals, so as to achieve the purpose of improving the diagnostic accuracy of the model's atmospheric modified refractive index profile.

Description

Technical Field

[0001] The present invention relates to the field of marine atmospheric boundary layer detection, and particularly to a method for calculating the atmospheric modified refractive index profile under turbulent effects. Background Art

[0002] In a marine environment, an atmospheric duct is an important factor affecting electromagnetic propagation at sea, and it is of great significance to accurately predict the near-sea surface atmospheric duct. Due to phenomena such as ocean evaporation, near-sea surface turbulent motion, wave effects, and large-scale frontal subsidence, the characteristics of the near-sea surface atmospheric duct will change, affecting electromagnetic wave propagation at sea, resulting in over-the-horizon propagation phenomena and detection blind spots.

[0003] Currently, the main method for diagnosing the near-sea surface atmospheric duct is to use the Monin-Obukhov similarity theory to construct an evaporation duct model and predict the near-sea surface atmospheric modified refractive index profile by inputting the temperature, humidity, wind speed, air pressure, and sea surface temperature at a certain height. Atmospheric turbulence is one of the main motion forms of the near-sea surface atmosphere, playing a major role in the momentum transfer, heat transfer, water vapor exchange, and material transfer between the sea surface and the atmosphere, and is also an important way to form an atmospheric duct. In the air-sea interaction interface of the marine atmospheric boundary layer, the turbulent motion is complex. The current calculation methods for evaporation ducts do not consider the turbulent effects, which affects the diagnosis of the atmospheric modified refractive index profile by the duct model. Therefore, it is necessary to consider the influence of turbulent effects on the atmospheric duct to achieve a more accurate diagnosis of the atmospheric modified refractive index profile. Summary of the Invention

[0004] To solve the above technical problems, the present invention provides a method for calculating the atmospheric modified refractive index profile under turbulent effects, which considers the influence of turbulent effects on the atmospheric modified refractive index during the calculation process, and selects the optimal dimensionless universal functions of temperature and humidity for different meteorological parameter intervals to achieve the purpose of improving the diagnostic accuracy of the atmospheric modified refractive index profile of the model.

[0005] To achieve the above object, the technical solution of the present invention is as follows:

[0006] A method for calculating the atmospheric modified refractive index profile under turbulent effects includes the following steps:

[0007] Step 1: Use the meteorological and hydrological sensors installed at different heights on the sea surface observation platform to obtain meteorological observation data at different heights, and record the installation height of the sensors;

[0008] Step 2: Use the meteorological observation data, based on the Monin-Obukhov similarity theory and the COARE algorithm, obtain different predicted temperature and humidity profiles through different dimensionless universal functions of temperature and humidity, calculate the true specific humidity according to the temperature, humidity, and air pressure measured by the meteorological and hydrological sensors, and obtain the true temperature and humidity profile through non-linear least squares fitting;

[0009] Step 3: Conduct a sensitivity analysis on the dimensionless universal function of temperature and humidity. Based on the analysis results, merge or split the meteorological parameter intervals. For each meteorological parameter interval, select the dimensionless universal function of temperature and humidity with the smallest root mean square error between the predicted and true temperature and humidity profiles as the dimensionless universal function of temperature and humidity to be optimized.

[0010] Step 4: Use the particle swarm intelligence optimization algorithm to optimize the coefficients of the dimensionless universal function of temperature and humidity, and obtain the optimized dimensionless universal function of temperature and humidity.

[0011] Step 5: Collect meteorological observation data, determine which meteorological parameter interval it belongs to, calculate the refractive index structure constant based on the optimized dimensionless universal function of temperature and humidity under this meteorological parameter interval, calculate the atmospheric refractive index fluctuation index by combining with the turbulence spectrum, and finally obtain the atmospheric corrected refractive index profile under the influence of turbulence.

[0012] In the above solution, in Step 2, the dimensionless universal function of temperature and humidity includes HYQ92, SHEBA07, BH91, and CB05 under stable atmospheric conditions, and HYQ92, Edson04, Grachev00, and Fairall03 under unstable conditions.

[0013] According to the calculated bulk Richardson number Judge whether the current atmospheric state belongs to stable conditions or unstable conditions. The specific method is as follows:

[0014] The bulk Richardson number is calculated by the following formula:

[0015] ;

[0016] where is the acceleration due to gravity, is the installation height of the temperature and humidity sensor, is the temperature at the installation height of the temperature and humidity sensor, is the difference between the temperature at the installation height of the temperature and humidity sensor and the sea surface temperature, is the wind speed, is the profile coefficient;

[0017] When the bulk Richardson number , it is determined that the current atmospheric state is stable, and the dimensionless universal function of temperature and humidity under stable conditions is selected; when , it is determined that the current atmospheric state is unstable, and the dimensionless universal function of temperature and humidity under unstable conditions is selected.

[0018] In the above solution, in step 2, the predicted temperature and humidity profiles include a predicted temperature profile and a predicted specific humidity profile, and the calculation methods are as follows:

[0019] ;

[0020] ;

[0021] where is the height, and are the predicted temperature and predicted specific humidity at the height of , is the sea surface temperature, is the specific humidity at the sea surface, is the saturation specific humidity calculated based on the sea surface temperature, is the characteristic scale of potential temperature, is the characteristic scale of specific humidity, is the temperature roughness length, is the non - dimensional universal function of temperature, is the non - dimensional universal function of humidity, is the stability correction factor, is the dry adiabatic lapse rate, is the von Kármán constant.

[0022] In the above solution, in step 3, the Morris model is used for sensitivity analysis, and the process is as follows:

[0023] S1: First, it is determined that when the number of input parameters is 5, which are temperature, humidity, air pressure, wind speed, and sea surface temperature, one of the 5 input parameters is selected, and the remaining input parameters remain unchanged;

[0024] S2: Set the initial interval range of meteorological parameters and determine the input parameter spacing D;

[0025] S3: Normalize all input parameters, and the calculation method is as follows:

[0026] ;

[0027] where is the th input parameter, is the upper limit of the th input parameter, is the lower limit of the th input parameter;

[0028] S4: Discretize the input parameters, and determine the number of horizontal points p and the perturbation factor through the set parameter range and parameter spacing. The calculation method is as follows:

[0029] ;

[0030] ;

[0031] Among them, p is the number of horizontal points;

[0032] S5: Construct a single trajectory and determine the initial point of the model input , where each is randomly selected from the discretized data, and is randomly increased or decreased , to obtain a new set of input points. Only one parameter is changed each time. After randomly increasing or decreasing in sequence for 5 times, a total of 6 sets of input parameters are obtained together with the initial input point. The new set of input parameters is brought into the COARE algorithm to obtain the simulated temperature profile or the simulated temperature profile as the output result;

[0033] S6: Repeat step S5 times, and a total of sets of input parameters and output results are generated as the total sample size for sensitivity analysis;

[0034] S7: Calculate the variable sensitivity discrimination coefficient :

[0035] ;

[0036] Among them, is the output result of the Morris model at the th time, is the output result of the Morris model at the th time, is the model operation result after the input parameters are adjusted according to the parameter interval, is the change rate of the input parameters with respect to the initial input parameters after the model runs for the th time; is the change rate of the input parameters with respect to the initial input parameters after the model runs for the th time; is the number of runs of the Morris model.

[0037] In the above solution, in step 3, the meteorological parameter interval is divided according to the magnitude of the variable sensitivity discrimination coefficient . When the calculated by the temperature dimensionless universal function and the humidity dimensionless universal function in the same initial interval is inconsistent, the larger is taken as the of this interval. When When it is greater than 1, the interval is evenly split. When When it is less than 0.2, the interval is merged. When When it is between 0.2 and 1, the interval remains unchanged.

[0038] In the above solution, in step 4, the process of optimization using the particle swarm intelligence optimization algorithm is as follows:

[0039] S1: Determine the number of coefficients m to be optimized in the dimensionless universal function of temperature or humidity in the dimensionless universal function of temperature and humidity to be optimized, and randomly generate E m-dimensional particles as the initial particle swarm;

[0040] S2: Calculate the particle fitness. For the dimensionless universal function of temperature, select the predicted temperature calculated by the COARE algorithm and the true temperature The root mean square error is used as the fitness function , for the dimensionless universal function of humidity, select the predicted specific humidity calculated by the COARE algorithm and the true specific humidity The root mean square error is used as the fitness function :

[0041] ;

[0042] ;

[0043] Among them, is the true temperature of the th particle, is the th particle's predicted temperature calculated by the COARE algorithm, is the fitness value of the temperature particle; The th particle's true specific humidity, is the th particle's predicted specific humidity calculated by the COARE algorithm, is the fitness value of the specific humidity particle;

[0044] S3: Determine the particle velocity and particle position according to the particle fitness value. The position of the th particle is , represents the position of the th particle in the th dimension. The velocity of the th particle is , represents the th particle's velocity in the th dimension; The The optimal position searched by a particle is , which represents the optimal position searched by the th particle in the th dimension. The globally optimal position searched by the particle swarm is , which represents the optimal position searched by the particle swarm in the th dimension;

[0045] The position and velocity of the th dimension after each iteration of the particle are calculated by the following formula:

[0046] ;

[0047] ;

[0048] where is the individual learning factor, is the swarm learning factor, and their value ranges are both [0, 2]; is the inertia weight; is the number of iterations; and are random numbers with a value range of [0, 1], is the position of the th particle in the th iteration in the th dimension; The position of the th particle in the th dimension when iterating to the next generation; is the velocity of the th particle in the th iteration in the th dimension; is the velocity of the th particle in the th dimension when iterating to the next generation; is the optimal position of the th particle in the th iteration in the th dimension; is the optimal position searched by the particle swarm in the th iteration in the th dimension;

[0049] S4: Repeat steps S2 and S3 until the preset number of iterations is reached, obtaining the velocity and position of the th particle after all iterations, and obtaining the optimal position of the th particle and the optimal position searched by the particle swarm , in is the optimal coefficient of the dimensionless universal function of temperature or the dimensionless universal function of humidity.

[0050] In the above solution, in step 5, the calculation method of the refractive index structure constant is as follows:

[0051] ;

[0052] ;

[0053] ;

[0054] wherein, is the temperature structure constant, is the humidity structure constant, is the temperature-humidity cross structure constant, is the height, is the height at the predicted temperature, is the atmospheric pressure, is the ratio of the gas constant of dry air to the gas constant of water vapor, is the height at the predicted specific humidity.

[0055] In the above solution, the calculation methods of the temperature structure constant , the humidity structure constant and the temperature-humidity cross structure constant are as follows:

[0056] ;

[0057] ;

[0058] ;

[0059] wherein, is the height, is the characteristic scale of the potential temperature, is the characteristic scale of the specific humidity, is the dimensionless universal function of temperature, is the dimensionless universal function of humidity, is the stability correction factor.

[0060] In the above solution, the calculation method of the atmospheric modified refractive index profile under the turbulent effect is as follows:

[0061] ;

[0062] ;

[0063] ;

[0064] wherein, is the height, is the height at the predicted temperature, is the height at the predicted atmospheric pressure, wherein, is the ratio of the gas constant of dry air to the gas constant of water vapor, is the height at the predicted specific humidity, is the height at the water vapor pressure, is the height at the atmospheric refractive index fluctuation index, is the height at the refractive index, is the height at the atmospheric modified refractive index under turbulent effects.

[0065] In the above solution, the atmospheric refractive index fluctuation index at height is calculated using a turbulence spectrum function , and the turbulence spectrum function includes the von Karman turbulence spectrum function, the Kolmogorov turbulence spectrum function, and the non-Kolmogorov turbulence spectrum function.

[0066] Through the above technical solution, a method for calculating the atmospheric modified refractive index profile under turbulent effects provided by the present invention has the following beneficial effects:

[0067] Based on the Monin-Obukhov similarity theory and the COARE algorithm, the present invention introduces the atmospheric refractive index fluctuation index on the basis of the traditional calculation of the atmospheric modified refractive index profile, so that the calculated atmospheric modified refractive index profile is closer to the true atmospheric modified refractive index profile in the near-sea area where the air-sea interface interaction is extremely complex.

[0068] The present invention uses the method of sensitivity analysis to perform sensitivity analysis under different meteorological conditions. By calculating the sensitivity discrimination coefficient, it accurately analyzes the sensitivity of different dimensionless universal functions of temperature and humidity to meteorological parameters, and optimizes the coefficients of the dimensionless universal functions of temperature and humidity through intelligent optimization algorithms related to artificial intelligence to make it more applicable to the target sea area. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art.

[0070] Figure 1 Schematic diagram of a method for calculating the atmospheric modified refractive index profile under turbulent effects disclosed in the embodiments of the present invention;

[0071] Figure 2 Schematic diagram of the comparison between the temperature profile calculated by the COARE algorithm and the true temperature profile;

[0072] Figure 3 Schematic diagram of the comparison between the specific humidity profile calculated by the COARE algorithm and the true specific humidity profile;

[0073] Figure 4 Schematic diagram of the flow of the Morris global sensitivity analysis method;

[0074] Figure 5 Schematic diagram of the process of optimizing the dimensionless universal function of temperature and humidity by the particle swarm intelligence optimization algorithm. Specific implementation manners

[0075] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention.

[0076] The present invention provides a method for calculating the atmospheric modified refractive index profile under turbulent effects, as Figure 1 shown, including the following steps:

[0077] Step 1, obtaining meteorological observation data at different heights by using meteorological and hydrological sensors installed at different heights on a sea surface observation platform, and recording the installation heights of the sensors.

[0078] Install meteorological and hydrological sensors at positions at different heights from the sea surface on the sea surface observation platform, specifically including temperature and humidity sensors installed at different heights. The temperature and humidity sensors are installed starting from the bottom of the observation platform, one every two meters. When installing the temperature and humidity sensors, they are installed successively on the left and right sides of the observation platform, and a barometric pressure sensor is installed on the observation platform, and a wind speed sensor and an infrared temperature sensor are installed to obtain temperature , relative humidity , barometric pressure , wind speed and sea surface temperature data, and recording the installation heights of the temperature and humidity sensors, the wind speed sensor, and the barometric pressure sensor.

[0079] When installing sensors on the observation platform, the occlusion of the observation platform itself should be considered. One infrared temperature sensor should be installed on each of the left and right sides of the observation platform to avoid the shadow generated by the observation platform under sunlight, which may affect the data accuracy. To ensure the data validity, the infrared sensors should be installed as close as possible to a position within 15 meters above the sea surface. To obtain the temperature and humidity data at different heights, the observation platform should cover a height range of 0 - 40 meters as much as possible, and a marine platform such as a meteorological observation tower that meets the sensor installation conditions should be adopted.

[0080] Step 2: According to the Monin - Obukhov similarity theory and the COARE algorithm, using the meteorological observation data, different predicted temperature and humidity profiles are obtained through different dimensionless universal functions of temperature and humidity. The true specific humidity is calculated based on the temperature, humidity, and air pressure measured by the meteorological and hydrological sensors, and the true temperature and humidity profiles are obtained through non - linear least - squares fitting.

[0081] The dimensionless universal functions of temperature and humidity include HYQ92, SHEBA07, BH91, and CB05 under stable atmospheric conditions, and HYQ92, Edson04, Grachev00, and Fairall03 under unstable conditions.

[0082] According to the calculated bulk Richardson number Judge whether the current atmospheric state belongs to stable conditions or unstable conditions. The specific method is as follows:

[0083] The bulk Richardson number is calculated by the following formula:

[0084] ;

[0085] where is the acceleration due to gravity, is the installation height of the temperature and humidity sensor, is the temperature at the installation height of the temperature and humidity sensor, is the difference between the temperature at the installation height of the temperature and humidity sensor and the sea surface temperature, is the wind speed, is the profile coefficient;

[0086] When the bulk Richardson number the current atmospheric state is determined to be stable, and the dimensionless universal function of temperature and humidity under stable conditions is selected; when the current atmospheric state is determined to be unstable, and the dimensionless universal function of temperature and humidity under unstable conditions is selected.

[0087] The predicted temperature and humidity profiles include the predicted temperature profile and the predicted specific humidity profile. The calculation methods are as follows:

[0088] ;

[0089] ;

[0090] Among them, is the height, and are the predicted temperature and predicted specific humidity at the height of , is the sea surface temperature, is the specific humidity at the sea surface, is the saturation specific humidity calculated according to the sea surface temperature, is the characteristic scale of the potential temperature, is the characteristic scale of the specific humidity, is the temperature roughness length, is the dimensionless universal function of temperature, is the dimensionless universal function of humidity, is the stability correction factor, is the dry adiabatic lapse rate, is the von Kármán constant.

[0091] The true specific humidity data is calculated from the temperature, humidity and air pressure data obtained by the sensor. The calculation method is as follows:

[0092] ;

[0093] ;

[0094] ;

[0095] Among them, is the installation height of the temperature and humidity sensor, is the temperature at the installation height of the temperature and humidity sensor, is the humidity at the installation height of the temperature and humidity sensor, is the air pressure at the installation height of the temperature and humidity sensor, is the saturation water vapor pressure at the installation height of the temperature and humidity sensor, is the water vapor pressure at the installation height of the temperature and humidity sensor, is the true specific humidity at the installation height of the temperature and humidity sensor.

[0096] The true temperature profile and true specific humidity profile are fitted according to the temperature and true specific humidity at different heights. The calculation method is as follows:

[0097] ;

[0098] ;

[0099] Among them, a, b, and c are all fitting coefficients.

[0100] The predicted temperature profile calculated by the COARE algorithm and the true temperature profile are as shown in Figure 2 Figure [0000440]. The predicted specific humidity profile calculated by the COARE algorithm and the true specific humidity profile are as shown in Figure 3 Figure [0000441].

[0101] Step 3: Conduct a sensitivity analysis on the dimensionless universal function of temperature and humidity. Based on the analysis results, merge or split the meteorological parameter intervals. For each meteorological parameter interval, select the dimensionless universal function of temperature and humidity with the smallest root mean square error between the predicted and true temperature and humidity profiles as the dimensionless universal function of temperature and humidity to be optimized.

[0102] As shown in Figure 4 Figure [0000446], the sensitivity analysis is carried out using the Morris model, and the process is as follows:

[0103] S1: First, it is determined that when the number of input parameters is 5, namely temperature, humidity, air pressure, wind speed, and sea surface temperature, select one of the 5 input parameters , and keep the other input parameters unchanged;

[0104] S2: Set the initial interval range of meteorological parameters and determine the input parameter spacing D;

[0105] In setting the initial interval range of meteorological parameters, the temperature parameter range is set to [20, 40] with the unit of °C and the parameter spacing of 2, the humidity range is set to [50, 90] with the unit of %, the parameter spacing of 10, the air pressure parameter range is set to [900, 1100] with the unit of hPa and the parameter spacing of 100, the wind speed parameter range is set to [0, 16] with the unit of m / s and the parameter spacing of 4, and the air-sea temperature difference parameter range is set to [-5, 5] with the unit of °C and the parameter spacing of 1.

[0106] S3: Normalize all input parameters, and the calculation method is as follows:

[0107] ;

[0108] where is the th input parameter, is the upper limit of the th input parameter, is the th input parameter's lower limit;

[0109] S4: Discretize the input parameters. Determine the number of horizontal points p and the perturbation factor through the set parameter range and parameter spacing. The calculation method is as follows:

[0110] ;

[0111] ;

[0112] Among them, p is the number of horizontal points;

[0113] S5: Construct a single trajectory and determine the initial point of the model input , where each is randomly selected from the discretized data, and is randomly increased or decreased , obtaining a new set of input points. Only one parameter is changed each time. After sequentially increasing or decreasing for 5 parameters, a total of 6 sets of input parameters are obtained together with the initial input point. The new set of input parameters is brought into the COARE algorithm to obtain the simulated temperature profile or the simulated temperature profile as the output result;

[0114] S6: Repeat step S5 times, generating a total of sets of input parameters and output results as the total sample size for sensitivity analysis;

[0115] S7: Calculate the variable sensitivity discrimination coefficient :

[0116] ;

[0117] Among them, is the output result of the Morris model at the th time, is the output result of the Morris model at the th time, is the model operation result after the input parameters are adjusted according to the parameter interval, is the change rate of the input parameters with respect to the initial input parameters after the model runs for the th time; is the change rate of the input parameters with respect to the initial input parameters after the model runs for the th time; is the number of runs of the Morris model.

[0118] According to the magnitude of the variable sensitivity discrimination coefficient , divide the meteorological parameter interval. When the calculated by the temperature dimensionless universal function and the humidity dimensionless universal function in the same initial interval is inconsistent, take the larger as the of this interval. When is greater than 1, evenly split this interval. When is less than 0.2, merge this interval. When When it is between 0.2 and 1, the interval remains unchanged.

[0119] Step 4: Use the particle swarm intelligence optimization algorithm to optimize the coefficients of the dimensionless universal function of temperature and humidity, and obtain the optimized dimensionless universal function of temperature and humidity.

[0120] Such as Figure 5 shown, taking the dimensionless universal function of temperature of CB05 under stable conditions as an example, the process of optimization using the particle swarm intelligence optimization algorithm is as follows:

[0121] S1: The dimensionless universal function of temperature of CB05 under stable conditions is expressed as , and this dimensionless universal function of temperature contains two coefficients and . When initializing, determine that the number of parameters to be optimized is 2, and generate E 2-dimensional particles as the initial particle swarm;

[0122] S2: Calculate the particle fitness. Select the root mean square error between the predicted temperature calculated by the COARE algorithm and the true temperature as the fitness function , and calculate the fitness value:

[0123] ;

[0124] Among them, is the true temperature of the th particle, is the predicted temperature calculated by the COARE algorithm of the th particle, is the fitness value of the temperature particle;

[0125] S3: Determine the particle velocity and particle position according to the particle fitness value. The position of the th particle is , represents the position of the th particle in the first dimension. The velocity of the th particle is , represents the velocity of the th particle in the first dimension; The best position searched by the th particle is , represents the best position searched by the th particle in the first dimension. The globally best position searched by the particle swarm is , represents the best position searched by the particle swarm in the first dimension;

[0126] The position and velocity of the particle in the first dimension after each iteration are calculated by the following formula:

[0127] ;

[0128] ;

[0129] where, is the individual learning factor, is the swarm learning factor, and their value ranges are both [0, 2]; is the inertia weight; is the number of iterations; and are random numbers, and their value ranges are [0, 1], is the position of the th particle in the first dimension at the number of iterations of ; The position of the th particle in the first dimension when iterating to the next generation; is the th particle's velocity in the first dimension at the number of iterations of ; is the th particle's velocity in the first dimension when iterating to the next generation; is the th particle's optimal position in the first dimension at the number of iterations of ; is the optimal position searched by the particle swarm in the first dimension at the number of iterations of ; The calculation method of the position and velocity of the particle in the second dimension after each iteration is the same as that in the first dimension;

[0130] S4: Repeat steps S2 and S3 until the preset number of iterations is reached, obtain the velocity and position of the th particle after all iterations, obtain the optimal position of the th particle and the optimal position searched by the particle swarm , in and is and 's optimal parameters.

[0131] The particle swarm optimization method for the remaining dimensionless universal functions of temperature and humidity is the same as above.

[0132] Step 5: Collect meteorological observation data, determine which meteorological parameter interval it belongs to, calculate the refractive index structure constant based on the optimized dimensionless universal function of temperature and humidity in this meteorological parameter interval, calculate the atmospheric refractive index fluctuation index by combining the turbulence spectrum, and finally obtain the atmospheric modified refractive index profile under the influence of turbulence effect.

[0133] The first step is to calculate the temperature structure constant , the humidity structure constant and the temperature-humidity cross structure constant , and the calculation methods are as follows:

[0134] ;

[0135] ;

[0136] ;

[0137] where is the height, is the characteristic scale of potential temperature, is the characteristic scale of specific humidity, is the dimensionless universal function of temperature, is the dimensionless universal function of humidity, is the stability correction factor.

[0138] The second step is to calculate the refractive index structure constant, and the calculation method is as follows:

[0139] ;

[0140] ;

[0141] ;

[0142] where is the temperature structure constant, is the humidity structure constant, is the temperature-humidity cross structure constant, is the height, is the height at which the predicted temperature is, is the atmospheric pressure, is the ratio of the gas constant of dry air to the gas constant of water vapor, is the height at which the predicted specific humidity is.

[0143] The third step is to calculate the atmospheric refractive index fluctuation index :

[0144] Use the turbulence spectrum function to calculate the atmospheric refractive index fluctuation index at height where , the turbulence spectrum functions include von Karman turbulence spectrum function, Kolmogorov turbulence spectrum function, and non-Kolmogorov turbulence spectrum function.

[0145] Taking the von Karman turbulence spectrum function as an example, the calculation method is as follows:

[0146] ;

[0147] ;

[0148] ;

[0149] ;

[0150] Among them, is a random number uniformly distributed between 0 and 1, is the refractive index structure constant, is the turbulence spectrum function, represents the wave number, is the inner scale wave number parameter, is the outer scale wave number parameter, is the demarcation value of small and medium scale fluctuations, is the demarcation value of medium and large scale fluctuations.

[0151] Fourthly, calculate the atmospheric modified refractive index profile under the turbulence effect. The calculation method is as follows:

[0152] ;

[0153] ;

[0154] ;

[0155] Among them, is the height, is the height at the predicted temperature, is the height at the predicted pressure, among which, is the ratio of the gas constant of dry air to the gas constant of water vapor, is the height at the predicted specific humidity, is the height at the water vapor pressure, is the atmospheric refractive index pulsation index at the height , is the refractive index at the height , Height under turbulent effects The atmospheric modified refractive index at

[0156] Among them, the predicted pressure profile The specific calculation method is as follows:

[0157] ;

[0158] Among them, is the height, is the installation height of the temperature and humidity sensor, is the gas constant under dry conditions, is the atmospheric pressure, is at the height and the average virtual temperature at the location, is the acceleration of gravity.

[0159] For the gradient meteorological data of the sea surface observation platform in the embodiment, using the calculation method of the atmospheric modified refractive index profile under turbulent effects proposed by the present invention, the optimal dimensionless universal function of temperature and humidity is selected according to the sensitivity analysis results. After algorithm optimization, the refractive index structure constant and the atmospheric refractive index pulsation index are calculated, and the atmospheric modified refractive index profile under turbulent effects is obtained. Theoretically, it is more reasonable compared with not considering turbulent effects.

[0160] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present invention. Various modifications to these embodiments will be obvious to those skilled in the art, and the general principles defined herein can 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 these embodiments shown herein, but will be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for calculating the atmospheric corrected refractive index profile under turbulence effect, characterized in that: The steps include: Step 1: Use meteorological and hydrological sensors installed at different heights on the sea surface observation platform to obtain meteorological observation data at different heights, and record the sensor installation height; Step 2: Using meteorological observation data, according to the Monin-Obukhov similarity theory and the COARE algorithm, different predicted temperature and humidity profiles are obtained through different temperature and humidity dimensionless universal functions, and the true specific humidity is calculated according to the temperature, humidity and air pressure measured by the meteorological and hydrological sensors, and the true temperature and humidity profile is obtained through nonlinear least squares fitting; Step 3, perform sensitivity analysis on the temperature and humidity dimensionless universal function, merge or split the meteorological parameter intervals according to the analysis results, and for each meteorological parameter interval, select the temperature and humidity dimensionless universal function with the smallest root mean square error between the predicted temperature and humidity profile and the actual temperature and humidity profile as the temperature and humidity dimensionless universal function to be optimized; Step 4, using a particle swarm intelligent optimization algorithm to optimize the coefficients of the temperature and humidity dimensionless universal function to obtain an optimized temperature and humidity dimensionless universal function; Step 5: Collect meteorological observation data, determine which meteorological parameter interval it belongs to, calculate the refractive index structure constant based on the optimized temperature and humidity dimensionless universal function in the meteorological parameter interval, calculate the atmospheric refractive index pulsation index in combination with the turbulence spectrum, and finally obtain the atmospheric corrected refractive index profile under the turbulence effect.

2. The method for calculating the atmospheric corrected refractive index profile under turbulence effect according to claim 1, characterized in that: In step 2, the temperature and humidity dimensionless universal functions include HYQ92, SHEBA07, BH91 and CB05 under stable atmospheric conditions, and HYQ92, Edson04, Grachev00 and Fairall03 under unstable conditions; According to the block Richardson number The specific method to determine whether the current atmospheric state is stable or unstable is as follows: Block Richardson number Calculated by the following formula: ； in, is the acceleration due to gravity, The installation height of the temperature and humidity sensor. is the temperature at the installation height of the temperature and humidity sensor, is the difference between the temperature at the installation height of the temperature and humidity sensor and the sea surface temperature, is the wind speed, is the section coefficient; When the block Richardson number When , the current atmospheric state is determined to be stable, and the dimensionless universal function of temperature and humidity under stable conditions is selected; when When , the current atmospheric state is determined to be unstable, and the dimensionless universal function of temperature and humidity under unstable conditions is selected.

3. The method for calculating the atmospheric corrected refractive index profile under turbulence effect according to claim 1, characterized in that: In step 2, the predicted temperature and humidity profile includes a predicted temperature profile and a predicted specific humidity profile, and the calculation method is as follows: ； ； in, is the height, and For at a height of The predicted temperature and predicted specific humidity at is the sea surface temperature, is the specific humidity at the sea surface, is the saturated specific humidity calculated based on the sea surface temperature, is the characteristic scale of potential temperature, is the characteristic scale of specific humidity, is the temperature roughness length, is the dimensionless universal function of temperature, is the dimensionless universal function of humidity, is the stability correction factor, is the dry adiabatic lapse rate, is the Karman constant.

4. The method for calculating the atmospheric corrected refractive index profile under turbulence effect according to claim 1, characterized in that: In step 3, the Morris model is used for sensitivity analysis, and the process is as follows: S1: First, determine that when the number of input parameters is 5, namely temperature, humidity, air pressure, wind speed and sea surface temperature, select one of the 5 input parameters , the rest of the input parameters remain unchanged; S2: Set the initial range of meteorological parameters and determine the input parameter interval D; S3: Normalize all input parameters. The calculation method is as follows: ； in, For the input parameters, For the The upper limit of input parameters, For the The lower limit of the input parameters; S4: Discretize the input parameters, and determine the number of horizontal points p and the disturbance factor by setting the parameter range and parameter spacing , the calculation method is as follows: ； ； Where p is the number of horizontal points; S5: Construct a single trajectory and determine the model input initial point , where each Randomly select from the discretized data Random increase or decrease , get a new set of input points, change only one parameter at a time, and change 5 Random increase or decrease After that, a total of 6 groups of input parameters are obtained together with the input initial point, and the new group of input parameters is brought into the COARE algorithm to obtain the simulated temperature profile or the simulated temperature profile as the output result; S6: Repeat step S5 times, a total of Group input parameters and output results as the total sample size for sensitivity analysis; S7: Calculate variable sensitivity discriminant coefficient : ； in, Morris model The output result of this time is Morris model Output results, The model running results after the input parameters are adjusted according to the parameter spacing. For the model After the first run, the rate of change of the input parameter relative to the initial input parameter; For the model After the first run, the rate of change of the input parameter relative to the initial input parameter; is the number of runs of the Morris model.

5. The method for calculating the atmospheric corrected refractive index profile under turbulence effect according to claim 4, characterized in that: In step 3, according to the variable sensitivity discriminant coefficient The size of the meteorological parameter interval is divided. When the temperature dimensionless universal function and the humidity dimensionless universal function are calculated in the same initial interval, If they are inconsistent, take the larger As the interval ,when When it is greater than 1, the interval is split evenly. When it is less than 0.2, the interval is merged. Between 0.2 and 1, the interval remains unchanged.

6. The method for calculating the atmospheric corrected refractive index profile under turbulence effect according to claim 1, characterized in that: In step 4, the optimization process using the particle swarm intelligent optimization algorithm is as follows: S1: Determine the number m of coefficients to be optimized of the temperature dimensionless universal function or the humidity dimensionless universal function in the temperature and humidity dimensionless universal function to be optimized, and randomly generate E m-dimensional particles as the initial particle group; S2: Calculate the particle fitness. For the temperature dimensionless universal function, select the predicted temperature calculated by the COARE algorithm. and real temperature The root mean square error is used as the fitness function For the dimensionless universal function of humidity, the predicted specific humidity calculated by the COARE algorithm is selected. and real specific humidity The root mean square error is used as the fitness function : ； ； in, For the The true temperature of a particle, For the The predicted temperature calculated by the COARE algorithm for each particle, is the fitness value of the temperature particle; No. The real specific humidity of each particle is For the The predicted specific humidity calculated by the COARE algorithm for each particle is is the fitness value of the specific humidity particle; S3: Determine the particle speed and particle position according to the particle fitness value. The position of a particle is , Indicates The particle in The position in dimension, The speed of a particle is , Indicates The particle in The speed in the dimension; The optimal position of a particle is , Indicates The particle in The optimal position searched in dimension, the global optimal position searched by the particle swarm is , Indicates that the particle swarm is The optimal position searched in dimension; After each iteration, the particle The position and velocity in the dimension are calculated by the following formula: ； ； in, is the individual learning factor, is the group learning factor, and its value range is [0, 2]; is the inertia weight; is the number of iterations; and is a random number, ranging from [0, 1], For the The number of iterations for a particle is At the Position in dimension; No. The particle in The position when iterating to the next generation in the dimension; For the The number of iterations for a particle is At the Speed ​​in dimension; For the The particle in The speed of iterating to the next generation in dimension; For the The number of iterations for a particle is At the Optimal position in dimension; The particle swarm is At the The optimal position searched in dimension; S4: Repeat steps S2 and S3 until the preset number of iterations is reached and the first The speed and position of the particle after all iterations are obtained. The optimal position of a particle and the optimal position searched by the particle swarm , In It is the optimal coefficient of the dimensionless universal function of temperature or the dimensionless universal function of humidity.

7. The method for calculating the atmospheric corrected refractive index profile under turbulence effect according to claim 1, characterized in that: In step 5, the refractive index structure constant is calculated as follows: ； ； ； in, is the temperature structure constant, is the humidity structure constant, is the temperature and humidity cross structure constant, is the height, For height The predicted temperature at is the atmospheric pressure, is the ratio of the gas constant of dry air to the gas constant of water vapor, For height Predicted specific humidity at.

8. The method for calculating the atmospheric corrected refractive index profile under turbulence effect according to claim 7, characterized in that: Temperature structure constant , humidity structure constant and the temperature-humidity cross-structure constant The calculation method is as follows: ； ； ； in, is the height, is the characteristic scale of potential temperature, is the characteristic scale of specific humidity, is the dimensionless universal function of temperature, is the dimensionless universal function of humidity, is the stability correction factor.

9. The method for calculating the atmospheric corrected refractive index profile under turbulence effect according to claim 7, characterized in that: The calculation method of the atmospheric corrected refractive index profile under turbulence effect is as follows: ； ； ； in, is the height, For height The predicted temperature at For height The predicted pressure at , where is the ratio of the gas constant of dry air to the gas constant of water vapor, For height The predicted specific humidity at For height The water vapor pressure at For height The atmospheric refractive index pulsation index at For height The refractive index at is the height under turbulence effect The atmospheric corrected refractive index at .

10. The method for calculating the atmospheric corrected refractive index profile under turbulence effect according to claim 9, characterized in that: Calculate the height using the turbulence spectrum function Atmospheric refractive index pulsation index at The turbulence spectrum function includes von Karman turbulence spectrum function, Kolmogorov turbulence spectrum function, and non-Kolmogorov turbulence spectrum function.

Images