JP6951615B2 - Sea state information measurement buoy and sea state information measurement device - Google Patents

Description

The present invention relates to a sea state information measuring buoy and a sea state information measuring device.

The present invention is a sea state information measuring device that measures sea state information, that is, wave height (wave height), wave direction (wave direction), flow direction, and flow velocity. For example, ultrasonic type, hydraulic type and the like have been conventionally known as wave height measuring devices. The ultrasonic type and the hydraulic type have a problem that the measurable water depth is limited. Further, since an expensive sensor is mounted, there is a problem that the entire device becomes expensive.

To solve these problems, a wave height measuring device that measures the wave height by mounting a receiver that receives radio waves from GPS (Global Positioning System) satellites on the buoy and measuring the three-dimensional position of the buoy It has been proposed (see Patent Document 1). This wave height measuring device is equipped with a GPS antenna, a GPS receiver, a data recording device, a data processing device and a transmitter, and is a calculated data in which the wave height and flow direction flow velocity at the site where the buoy floats are calculated and observed by the GPS system. Is transmitted to a base station equipped with a receiver. Since the data processing device calculates the wave height and the flow direction flow velocity, the accuracy can be improved by processing the position data.

Further, Patent Document 2 describes that as a buoy that can be used as a marine platform for wave measurement, there are a relaxation mooring type surface buoy such as a cylindrical buoy and a tension mooring type spar buoy such as a retracting mooring type spar buoy. Patent Document 2 relates to an improvement of a tension mooring type spar buoy. In Patent Document 2, a GPS-type motion measurement sensor mounted on the upper part of the spar buoy measures the motion displacement due to the shaking of the upper part of the spar buoy in time series as three-dimensional position data, and the wave height is calculated from the three-dimensional position data obtained in the time series. It is described that the approximate value is obtained by the interpolation calculation method by table lookup. That is, in the measurement of the wave height wave direction and the flow velocity flow velocity of Patent Document 2, the wave height wave direction and the flow direction are taken into consideration in consideration of the sway characteristics depending on the shape and size of the buoy used, that is, the motion characteristics such as acceleration and attitude change. The flow velocity was calculated.

Japanese Unexamined Patent Publication No. 2003-30221 Japanese Unexamined Patent Publication No. 2007-327853

The wave height measuring device described in Patent Document 1 is equipped with a data processing device for obtaining the wave height from the measured three-dimensional position on the buoy, and it is necessary to perform complicated data processing in order to increase the measurement degree. was there. As a result, the cost of the measuring device may increase. In addition, it is necessary to measure the wave height with a buoy having a specific configuration, and it is difficult to measure the wave height using an existing ship, for example.

As described in Patent Document 2, in the case of a tension mooring type super buoy, the water level change near the draft of the buoy body is measured by a water level gauge fixed to the buoy by utilizing the small shaking of the buoy. The method is to convert the data to wave height. Further, as a motion measurement method, a method of acquiring three-dimensional information by a GPS-type motion measurement sensor is also described. The present invention relates to a relaxation mooring type surface buoy, and is different from the tension mooring type super buoy which is the subject of Cited Document 2. In the case of a relaxation mooring type surface buoy, it is a method in which the buoy is regarded as following the wave surface, its vertical motion is measured, and the data is converted into a crest value. Similar to Patent Document 1, it is necessary to measure the wave height of the one described in Patent Document 2 by using a buoy having a specific configuration, and it is difficult to measure the wave height using an existing ship, for example.

Therefore, an object of the present invention is that in a configuration in which a buoy moves up and down following a wave surface, it is possible to use a floating body such as an existing ship as a measuring device, not limited to the buoy, with a simple configuration and low cost. To provide a sea state information measuring buoy and a sea state information measuring device.

The present invention is a structure floating on the sea surface, is used in a state affected by almost the same sea conditions and other floating body, sea state information measuring buoys have application to enable other floating body as Oceanographic information measurement device And
With the antenna and GNSS receiver,
A processing device that calculates sea state information from the three-dimensional position information obtained by the GNSS receiver and the sea state information estimation model previously obtained by machine learning.
It is equipped with a communication device that transmits the sea state information required by the processing device.
The sea state information estimation model floats a buoy that has almost the same shape as the sea state information measurement buoy, and is equipped with a sea state information measuring device that measures the sea state information received by the buoy.
Obtained by supplying the 3D position information received by the buoy's GNSS receiver and the measured sea state information measured by the sea state information measuring device to the computer, and performing machine learning using the 3D position information and the measured sea state information as training data. It is a sea state information measurement buoy.

According to the present invention, sea state information such as wave height and flow velocity can be acquired only by using the three-dimensional position information received by the GNSS receiver. The effects described here are not necessarily limited, and may be any of the effects described in the present specification.

FIG. 1 is a schematic diagram showing an outline of a configuration at the time of machine learning according to an embodiment of the present invention. FIG. 2 is a block diagram showing a configuration at the time of machine learning according to an embodiment of the present invention. FIG. 3 is a block diagram of an embodiment of the present invention. FIG. 4 is a schematic diagram used for explaining an example of how to use the test buoy. FIG. 5 is a block diagram of a mooring buoy used with a test buoy. FIG. 6 is a block diagram showing a configuration for obtaining a modified wave height estimation model for a mooring buoy. 7A and 7B are schematic diagrams used to explain other examples of how to use the test buoy and still other examples. FIG. 8 is a schematic diagram showing an outline of the configuration at the time of machine learning for measuring the flow velocity.

Hereinafter, embodiments of the present invention will be described. It should be noted that the embodiments described below are suitable specific examples of the present invention and are provided with various technically preferable limitations. Unless otherwise stated, it is not limited to these embodiments.

In order to facilitate understanding of one embodiment of the present invention, an outline of a sea state information measurement system using the present invention will be described. This system can easily realize a wave height measuring device by simply mounting a GNSS (Global Navigation Satellite System) receiver on every floating body, not just a buoy. GNSS is a general term for satellite positioning systems such as GPS, GLONASS, Galileo, and Quasi-Zenith Satellite (QZSS). For example, recently, with the start of operation of the quasi-zenith satellite system (MICHIBIKI), GPS is supplemented and more accurate positioning is becoming possible.

In this system, a wave height measuring device is manufactured by combining a GNSS receiver and a learning model (wave height estimation model) that estimates sea state information, for example, wave height from the shaking characteristics of a floating body by machine learning. Then, the learning model is reconstructed by using the sway characteristics and the estimated wave height obtained from this wave height measuring device as the output of the teacher data. That is, the sway characteristics of the target buoy, ship, for example, a small fishing boat are added as new input data, and the parameters of the wave height estimation model of each floating body are adjusted so that all the floating bodies can be used as the wave height measuring device. The present invention provides, for example, a sea state information measuring buoy and a wave height measuring device in such a system.

First, the creation of the wave height estimation model will be described with reference to FIG. In FIG. 1, 1 indicates a buoy. The buoy 1 has a structure that floats on the sea surface SF (or on a lake or a river), and has a dome-shaped device storage portion that is not easily affected by wind and a ring-shaped floating ring provided around the dome-shaped device storage portion. More specifically, the buoy 1 has a small size (about 70 cm) that floats on the surface of the sea, a disk shape, and an isotropic shape that easily follows the movement on the surface of the sea. As a result, the influence of wind and flow on the buoy body is extremely small compared to the conventional cylindrical buoy such as a super buoy, and the behavior is almost synchronized with the fluctuation of the sea level. By combining this unique buoy structure and machine learning, it is possible to estimate sea level fluctuations, in other words, fluctuation patterns of three-dimensional position information, which are different from the conventional method of analyzing the buoy's sway characteristics from acceleration and attitude. Become. That is, when the estimation model is generated by machine learning, it is possible to learn the sea level fluctuation characteristics peculiar to waves and estimate the wave height and the flow velocity from only the fluctuation pattern of the three-dimensional position information.

Although not shown, the buoy 1 is moored in a sinker or the like on the seabed. The buoy 1 has an antenna 2 and is provided with a GNSS receiver 3 that receives radio waves from the satellite by the antenna 2. The GNSS receiver 3 also receives the error correction information obtained by receiving the radio wave from the satellite at the electronic reference point whose position is known in advance by the antenna 2. The error correction information is received at a frequency of 5 Hz to 10 Hz. As a result, the positioning accuracy becomes high. The GNSS receiver 3 obtains the three-dimensional position information P1 of the buoy 1 (antenna 2). The three-dimensional position information is represented by (X: longitude, Y: latitude, Z: above sea level or altitude).

The three-dimensional position information P1 from the GNSS receiver 3 is supplied to the wireless communication unit 4. The wireless communication unit 4 has an antenna 5 and performs wireless communication with a website on the Internet or a center on the ground. For example, communication is performed using a mobile phone network. Communication may be performed using a satellite telephone network. Further, the buoy 1 is provided with an interface 6 for receiving wave height data from a wave height meter 8 laid on the seabed. The buoy 1 is provided with a power supply unit 7 that supplies power to the GNSS receiver 3, the wireless communication unit 4, and the interface 6. The power supply unit 7 is controlled to supply power from the battery on a regular basis (for example, every hour). Further, when the wave number required for the wave height calculation, for example, the amplitude (three-dimensional position information) corresponding to 100 waves is collected, the power to each part is controlled to be turned off. The power supply unit 7 is composed of, for example, a secondary battery and a solar power generation device.

The wave height meter 8 measures the wave height received by the buoy 1 by detecting, for example, a change in water pressure. In addition to the method of detecting changes in water pressure, a wave height meter that uses ultrasonic waves can also be used. The wave height meter 8 defines the water depth that can be laid. The measured wave height data R1 of the wave height meter 8 is supplied to the wireless communication unit 4 via the interface 6 by wire or wirelessly. Alternatively, the measured peak height data R1 is collected via a recording medium such as a memory card. The wireless communication unit 4 transmits three-dimensional position information and wave height data. The three-dimensional position information received by the GNSS receiver 3 may be subjected to processing such as noise removal.

As shown in FIG. 2, the radio wave transmitted from the communication unit 4 is received by the antenna 11 and supplied to the wireless communication unit 12. Reception processing is performed by the wireless communication unit 12, and the three-dimensional position information P1 and the measured peak height data R1 are taken out from the reception unit 12 and supplied to the analysis computer 13. The analysis computer 13 performs machine learning with a teacher using the three-dimensional position information P1 and the measured wave height data R1 as training data, and obtains a wave height estimation model M1 representing the relationship between the three-dimensional position information P1 and the wave height. The wave height estimation model M1 is an algorithm and a function.

For example, three-dimensional information is given as an input of an estimation model (neural network), and a wave height (estimation) is obtained as an output. Repeated calculations and parameter adjustments are performed so that the estimated wave height approaches the correct answer data (wave height measured by the wave height meter 8 installed on the seafloor) (this is machine learning), and the wave height value is estimated. By this machine learning, the fluctuation pattern of the sea surface is learned from the three-dimensional position information, and abnormal values such as noise are discriminated and automatically removed to estimate an accurate peak value.

Using the wave height estimation model M1 thus obtained, a sea state information measurement buoy (hereinafter referred to as a test buoy) 21 is configured as shown in FIG. The verification buoy 21 has the same shape as the buoy 1 described above, and includes an antenna 22, a GNSS receiver 23, a wireless communication unit 24, an antenna 25, a computer 26, and a power supply unit 27. The power supply unit 7 is controlled to supply power from the battery on a regular basis (for example, every hour). Further, when the sample data necessary for the wave height calculation is collected, the power to each part is controlled to be turned off. The power supply unit 7 is composed of, for example, a secondary battery and a solar power generation device. The GNSS receiver 23 receives radio waves from satellites and error correction information, and outputs three-dimensional position information P1. The wave height estimation model M1 is installed as software on the computer 26.

In the verification buoy 21, the three-dimensional position information P1 acquired by the GNSS receiver 23 is supplied to the wireless communication unit 24 and the computer 26. The computer 26 obtains the wave height data R1 using the wave height estimation model M1 and supplies the obtained wave height data R1 to the wireless communication unit 24. The wireless communication unit 24 transmits the three-dimensional position information P1 and the wave height data R1 to a predetermined website or ground center on the Internet.

A sea state information measuring device using the verification buoy 21 will be described with reference to FIG. For example, the test buoy 21 and the mooring buoy 31 as the target buoy are moored to the common buoy 41 at a short distance. The buoy 41 is connected to the sinker 42 on the seabed via a cable 43. Therefore, since the distance between the test buoy 21 and the mooring buoy 31 is relatively small, they are floating in almost the same sea state and are affected by the same sea state such as waves. The mooring buoy 31 may be connected to the cable 43 without providing the buoy 41.

As shown in FIG. 5, the mooring buoy 31 includes an antenna 32, a GNSS receiver 33, a wireless communication buoy 34, an antenna 35, and a power supply unit 37. These components are the same as those of the test buoy 21. The GNSS receiver 33 outputs the three-dimensional position information P2 of the verification buoy 31 itself by receiving and processing the radio wave from the satellite. The three-dimensional position information P2 is transmitted to the website or the center on the ground by the wireless communication unit 34 and the antenna 35.

The radio wave including the three-dimensional position information P1 and the wave height data R1 from the verification buoy 21 and the three-dimensional position information P2 from the mooring buoy 31 is received by the antenna 51 on the receiving side as shown in FIG. These data are supplied to the receiving server 53. The analysis computer 54 is connected to the receiving server 53.

A modified wave height estimation model M2 for obtaining the wave height from the three-dimensional position information P2 obtained by the GNSS receiver 33 of the mooring buoy 31 by the analysis of the analysis computer 54 is created. In the modified wave height estimation model M2, the relationship between the three-dimensional position information of the test buoy 21 and the wave height does not match the relationship between the three-dimensional position information of the mooring buoy 31 of different sizes, shapes, and weights and the wave height. The parameters of the wave height estimation model M1 are modified.

Then, by mounting this modified wave height estimation model M2 on the mooring buoy, the mooring buoy has a function as a wave height meter. The obtained wave height data is transmitted to the website (cloud server) or the center on the ground by wireless communication, and the wave height information of the sea area where the mooring buoy exists can be obtained in real time.

A preset user, for example, a fisherman can know the wave height information in a form processed into a graph or the like by a personal computer, a mobile terminal (for example, a smartphone) or the like. In addition, each mooring buoy transmits only 3D position information, the website (cloud server) or the center on the ground receives the 3D position information from each mooring buoy, and the corrected wave height estimation model M2 is used. The wave height at the position of each mooring buoy may be measured.

Not limited to the mooring buoy, as shown in FIG. 7A, the drifting buoy 61 and the verification buoy 21 may be connected by a cable or the like to obtain a corrected wave height estimation model corresponding to the drifting buoy 61. Once a modified crest height estimation model is available, a website (cloud server) or ground center that receives 3D position information from a large number of drifting buoys by mounting the modified crest height estimation model on a large number of drifting buoys. By using the modified wave height estimation model in, the wave height at the position of each drifting buoy can be measured in real time.

Further, as shown in FIG. 7B, the verification buoy 21 may be connected to a relatively small vessel 71 such as a fishing vessel loaded with the GNSS receiver 72. In this case, a modified wave height estimation model corresponding to the ship 71 can be obtained. Similar to the case of the mooring buoy and the drifting buoy described above, by loading the GNSS receiver on the ship, the ship can have a role as a wave height measuring device.

Although the measurement of wave height has been explained, the flow velocity estimation model is obtained by machine learning, and the modified flow velocity estimation model with the parameters modified according to the target floating body is obtained, so that the flow velocity is measured at the same time as the wave height. can do.

Further, the flow velocity can be measured more accurately by using the configuration shown in FIG. 8 during machine learning. The mooring buoy 81 is connected to the sinker 88 on the seabed by a cable 89. The mooring buoy 81 is provided with an antenna 82, a GNSS receiver 83, a wireless communication unit 84, an antenna 85, a power supply unit 86, and a load cell 87. The configuration other than the load cell 87 is the same as that in FIG. 1. The GNSS receiver 83 obtains the three-dimensional position information of the mooring buoy 81, and the three-dimensional position information and the tension data measured by the load cell 87 are the wireless communication unit 84. Sent by.

The load cell 87 is provided near the joint between the mooring buoy 81 and the cable 89, and measures the tension applied to the cable 89. This tension reflects the strength of the flow, and the flow velocity is estimated from the tension. In this way, training data consisting of tension (flow velocity) and three-dimensional position information is obtained, and machine learning is performed using this training data.

Although the embodiments of the present invention have been specifically described above, the present invention is not limited to the above-described embodiments, and various modifications based on the technical idea of the present invention are possible. For example, the flow direction may be estimated from the drifting position of the buoy with the position when there is no flow (tide stop) as the center (reference point). In addition, the wave height information obtained by the present invention can be used to correct the sway of the ship, and accurate water depth measurement becomes possible. The configurations, methods, processes, shapes, materials, numerical values, etc. mentioned in the above-described embodiments are merely examples, and different configurations, methods, processes, shapes, materials, numerical values, etc. may be used as necessary. ..

1 ... buoy, 3 ... GNSS receiver, 4 ... wireless communication unit, 8 ... wave height meter,
13 ... Analysis computer, 21 ... Certification buoy, 31 ... Mooring buoy,
P1, P2 ... 3D position information, R1 ... wave height data, M1 ... wave height estimation model,
M2 ・ ・ ・ Corrected wave height estimation model

Claims (1)

Is a structure floating on the sea surface, is used in a state affected by almost the same sea conditions and other floating body, a sea condition information measuring buoys have application to enable other floating body as Oceanographic information measuring device,
With the antenna and GNSS receiver,
A processing device that calculates sea state information from the three-dimensional position information obtained by the GNSS receiver and the sea state information estimation model previously obtained by machine learning.
It is equipped with a communication device that transmits the sea state information obtained by the processing device.
The sea state information estimation model floats a buoy having almost the same shape as the sea state information measurement buoy, and is provided with a sea state information measuring device for measuring the sea state information received by the buoy.
By supplying the three-dimensional position information received by the GNSS receiver of the buoy and the measured sea condition information measured by the sea condition information measuring device to a computer, and performing machine learning using the three-dimensional position information and the measured sea condition information as training data. The required sea condition information measurement buoy.

Images