A method and device for estimating sea surface wind speed based on radar echo center line spectrum intensity

Abstract

The invention discloses a method and a device for estimating sea surface wind speed based on line spectrum intensity in radar echoes. The method comprises the steps of establishing an experiential model in which intensities of a first-order peak and a secondary harmonic peak are changed along air speed, firstly estimating model parameters by fitting by taking air speed data measured in real time by virtue of a site buoy and air speed estimated on a radar close range unit by virtue of a conventional second-order spectrum method as standards, then estimating air speed by virtue of the intensities of the first-order peak and the secondary harmonic peak in an echo Doppler spectrum, weighting and combining the estimating model parameters with the air speed, so as to obtain a final air speed estimation value. According to the method and the device, the problem of high-frequency ground wave radar that an existing air speed estimation method based on an echo second-order spectrum is not accurate enough is solved; the intensities of a first-order Bragg peak and the secondary harmonic peak in the echo Doppler spectrum are related with the air speed value, so that the air speed can be directly extracted from line spectrums including the first-order peak, the secondary harmonic peak and the like, and the estimation precision of the sea surface wind speed by the high-frequency ground wave radar is substantially improved.

Description

technical field [0001] The invention relates to the technical field of high-frequency ground wave radar marine environment monitoring, in particular to a method and device for estimating sea surface wind speed based on radar echo midline spectrum intensity. Background technique [0002] Sea surface wind speed is a very important sea state parameter, which has an important impact on many marine activities. However, remote sensing of wind speed is still a difficult problem. At present, all oceanic countries in the world have built or are building marine environment monitoring networks, among which shore-based high-frequency sea state radar plays an important role. The existing mainstream method of wind speed retrieval by high-frequency radar relies on the acquisition of wave height parameters, that is, the effective wave height and wave period are estimated first, and then the wind speed is estimated according to a certain model. Since there are certain errors in the estimati...

AI Technical Summary

Problems solved by technology

In particular, portable high-frequency ground wave radars that use compact receiving antennas such as monopoles / cross-rings cannot obtain high spatial gain, and can only extract wind and wave parameters at several relatively close distance elements, and are vulnerable to The impact of external interference, while the spatial aliasing of the second-order spectrum will also have an adverse effect

Since the second-order spectrum is mainly a continuous spectrum and occupies a wide frequency range, incoming waves in various directions are mixed together and cannot be separated

In practical applications, due to the cost of radar station construction, the difficulty of promotion and the consideration of radio management, it is not possible to simply expand the array aperture and increase the transmission power to improve the accuracy of wind speed measurement

Therefore, under the existing framework of extracting wind and waves from the second-order spectrum, high-frequency ground wave radar, especially portable radar, can only measure the wind speed in the offshore area, and it is difficult to achieve high accuracy

[0004] Literature (Shen, W., K.-W. Gurgel, G. Voulgaris, T. Schlick, and D. Stammer. 2012. "Wind-speed inversion from HF radar first-order backscatter signal." OceanDynamics 62(1): 105-121.) reported a method of directly estimating wind speed using the first-order spectrum. However, this method uses the training of buoy data and uses the neural network to perceive the relationship between the spectral value and wind speed, and does not clearly give the first-order Mathematical Model of Spectrum Response with Wind Speed

The performance of the neural network is also affected by the network form and the selection of the training sample set, which is not conducive to engineering applications.

Images