CN109900256B - Self-adaptive ocean mobile acoustic tomography system and method - Google Patents
Self-adaptive ocean mobile acoustic tomography system and method Download PDFInfo
- Publication number
- CN109900256B CN109900256B CN201910238943.0A CN201910238943A CN109900256B CN 109900256 B CN109900256 B CN 109900256B CN 201910238943 A CN201910238943 A CN 201910238943A CN 109900256 B CN109900256 B CN 109900256B
- Authority
- CN
- China
- Prior art keywords
- water surface
- observation
- acoustic
- frequency
- low
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Landscapes
- Measurement Of Velocity Or Position Using Acoustic Or Ultrasonic Waves (AREA)
Abstract
A self-adaptive ocean mobile acoustic tomography system and method, the system includes a plurality of water surface mobile platforms distributed outside the observation sea area, three hydrophones are installed at the bottom of each water surface mobile platform for accurate positioning of underwater acoustic transducer, each water surface mobile platform is connected with underwater low-frequency and high-frequency acoustic transducer through dragging cable, the water surface mobile platform has positioning and time service device, realizing self positioning of the water surface mobile platform and making the acoustic transducer synchronously send out detection sound wave, the high-frequency acoustic transducer sends out high-frequency acoustic signal to cooperate with three hydrophones to realize accurate positioning of high-frequency and low-frequency acoustic transducer, the low-frequency acoustic transducer sends out middle-low frequency acoustic signal to mutually receive the emitted signal with the low-frequency acoustic transducer of other water surface mobile platform, realizing acoustic tomography observation of ocean hydrologic information. By using the self-adaptive ocean mobile acoustic chromatography system and the method, the accuracy of ocean hydrologic information measurement can be remarkably improved.
Description
Technical Field
The invention relates to the technical field of ocean monitoring, in particular to a self-adaptive ocean mobile acoustic tomography system and method.
Background
Marine acoustic tomography is an important technical means for measuring marine hydrologic information, and utilizes the change of the propagation speed of sound waves in the ocean to invert marine environment parameters including ocean currents, ocean temperatures and the like. Marine acoustic tomography has the following advantages in acquiring marine environmental information:
1) Because the sound wave propagates in the sea water and has the advantage of small loss, the large-range marine environment information can be obtained.
2) Based on the acoustic propagation multi-path effect, the three-dimensional structure of the marine environment field can be obtained by using limited acoustic chromatography nodes
3) Acoustic chromatography is a non-contact measurement method that can avoid the impact on the marine environment due to instrumentation.
In the existing marine acoustic chromatographic scheme, a plurality of acoustic chromatographic devices are fixed on a buoy or a shore base (such as a marine acoustic measurement buoy system, publication number is CN 106965905A), the positions of measurement nodes are fixed, hydrologic information can be measured only at the fixed positions, and compared with the traditional CT, the method has the advantages that the number of the nodes is small, the number of sound rays is limited, and measurement errors are uncontrollable; when inverting the hydrological information in the observation range, accumulated errors are easy to generate; the platform carrying the measuring device is fixed, self-adaptive sampling cannot be performed, and the minimization of the observation error cannot be achieved by adjusting the position of the sampling point; when the buoy is affected by ocean phenomena such as waves, an internal wave field and the like to incline, the position of the observation device cannot be accurately determined, and the measurement accuracy is affected.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provides a self-adaptive ocean mobile acoustic chromatography system and a self-adaptive ocean mobile acoustic chromatography method for improving the accuracy of ocean hydrologic information measurement.
In order to achieve the above purpose, the present invention adopts the following technical scheme:
the utility model provides a self-adaptation ocean mobile acoustic chromatography system, includes a plurality of surface of water moving platform that survey sea periphery was put, and three hydrophone is equipped with to every surface of water moving platform bottom for underwater acoustic transducer's accurate positioning, every surface of water moving platform is connected with underwater low frequency acoustic transducer and high frequency acoustic transducer through dragging the cable, the surface of water moving platform has location and time service device for realize the self location of surface of water moving platform and make the acoustic transducer send out the sound wave of surveying in step, high frequency acoustic transducer be used for sending high frequency acoustic signal with three hydrophone cooperation, in order to realize high frequency acoustic transducer with the accurate positioning of low frequency acoustic transducer, low frequency acoustic transducer is used for sending well low frequency acoustic signal, and the corresponding low frequency acoustic transducer of other surface of water moving platform receives the signal of transmission each other, in order to realize the acoustic chromatography observation of ocean hydrologic information.
Further:
the water surface mobile platform is a wave glider, an AUV, an unmanned ship or an artificial ship.
The positioning and time service device is a GPS or Beidou positioning system.
The water surface moving platform is a wave glider, the wave glider comprises a water surface ship, an underwater traction body and an umbilical cable connected with the water surface ship and the underwater traction body, the water surface ship moves through dragging of the underwater traction body, and the high-frequency underwater acoustic transducer and the low-frequency underwater acoustic transducer are installed on the underwater traction body.
The bottom of the water surface ship is fixedly provided with the hydrophone bracket, the hydrophone bracket is provided with three connecting rods which form an included angle of 120 degrees, and the three hydrophones are respectively arranged on the three connecting rods.
The system comprises a water surface ship, a positioning and timing device, a gyroscope, a system and a three-point positioning method, wherein the gyroscope is arranged on the water surface ship and is used for measuring the gesture of the water surface ship so as to obtain the position of each hydrophone relative to the water surface ship, the gyroscope is matched with the position information of the water surface moving platform determined by the positioning and timing device to obtain the actual positions of the three hydrophones, and the system determines the exact positions of the high-frequency underwater acoustic transducer and the low-frequency underwater acoustic transducer which are arranged on the underwater traction body according to the actual positions of the three hydrophones.
The initial sampling points of the water surface moving platforms are uniformly distributed on the periphery circumference of the region to be observed, and the water surface moving platforms randomly move along the circumference to perform multiple sampling; the plurality of water surface moving platforms are preferably five wave gliders.
After the low-frequency underwater acoustic transducer of each water surface mobile platform completes acoustic wave emission, converting into a receiving mode; each water surface mobile platform receives sound waves sent by low-frequency underwater sound transducers carried by other water surface mobile platforms, and calculates the arrival time delay of each sound wave; when each water surface mobile platform receives low-frequency sound waves sent by all other water surface mobile platforms, one-time measurement is completed; after the measurement is completed at the initial sampling point, relevant data including the position of the low-frequency underwater acoustic transducer, the sequence of arrival and the propagation time of the sound waves are transmitted to a shore base through a communication system carried by a water surface mobile platform for analysis, and marine hydrologic information at the corresponding moment of an observation area is calculated in an inversion mode; preferably, the entire observation area has a total of 10 sound ray coverage per measurement.
An adaptive ocean mobile acoustic tomography method, using the adaptive ocean mobile acoustic tomography system to perform acoustic tomography observation of ocean hydrologic information, preferably, wherein the determining process of the actual positions of the three hydrophones includes:
1) Selecting the positioning position of the water surface moving platform as the origin of a world coordinate system, wherein the X axis is in the northwest direction, and the Y axis is in the northwest direction to form a right-hand coordinate system OXYZ;
2) Selecting the center of three hydrophones as the origin O ', the X ' axis points to the starboard of the water surface moving platform, and the Y ' axis points to the head direction of the water surface moving platform, so as to form a right-hand coordinate system O ' X ' Y ' Z ', wherein the origin of the hydrophone coordinate system in the world coordinate system is P 0 ＝(x o ，y o ，z o ) T ；
3) According to the coordinates p of three hydrophones in the coordinate system O ' X ' Y ' Z i ＝(x i ，y i ，z i ) T I=1, 2,3, determining the transformation of the hydrophone coordinate system into the world coordinate system is:
R＝R x R y R z ，
wherein roll, pitch, and yaw are angles of rotation about three coordinate axes of XYZ, respectively, from the world coordinate system to the hydrophone coordinate system in the order of rotation about X axis, rotation about Y axis, and rotation about Z axis, respectively, thereby from the hydrophone coordinate system to the world coordinate systemWhen the system is in the system, the conversion sequence is that the system rotates around the Z axis, then rotates around the Y axis and finally rotates around the X axis; r is R x ，R y ，R z The transformation matrix is respectively a transformation matrix when rotating around three coordinate axes, and R is a transformation matrix from a hydrophone coordinate system to a world coordinate system;
4) Determining the position of three hydrophones in a world coordinate system as P i ＝(X i ，Y i ，Z i ) T ，i＝1，2，3
P i ＝Rp i +P 0 。
Further, searching the optimal placement position of the mobile node through a neural network algorithm: after the water surface mobile platform completes observation at an initial sampling point, transmitting observation data to a shore base for analysis through a communication device of the water surface mobile platform, and inverting hydrological information in an observed sea area through measurement data to obtain an estimated error; the neural network algorithm takes observed measurement data as a training sample, predicts the optimal point setting target position, and because the ocean power environment changes with time, the optimal position of the water surface mobile platform also changes with time, and the obtained optimal point setting position is the optimal point at the current moment; the water surface mobile platform carries out position adjustment according to the optimal target position obtained by shore-based analysis, then carries out repeated observation, and after the measurement is completed, transmits data back to shore for analysis again, so that the data is reciprocated;
the method comprises the steps of taking errors of inversion hydrologic information as input of a neural network, inverting errors to be errors of different points of the whole observation area, wherein the data quantity is related to grid division, and the number of neurons of an input layer is equal to the number of errors; preferably, the normalization is performed prior to use of the data; preferably, a small neural network is used, and the set network comprises one to two hidden layers;
preferably, the positions of the five observation points are used as the output of the neural network, the output layer of the neural network is set to be 10 neurons, and the 10 neurons respectively represent one component in the two-dimensional positions of the observation points and correspond to the five observation positions; defining an objective function:
L＝L δ (y,f(e))+L s (e)，
wherein L is δ (y, f (e)) is the Huber loss function as follows:
wherein e is input error information, y represents an actual observation position, f (e) is a predicted optimal observation point position, and delta is a converging radius; the Huber function is used as an objective function, each component is guaranteed to be near an actual observation position, and the sensitivity degree to abnormal points in data is reduced by adjusting the super parameter delta, so that the optimal point is guaranteed not to be missed when gradient descent is about to end; preferably, an auxiliary loss function L is additionally provided s (e) Ce, where e is the hydrologic error of the corresponding observation point, C is the penalty coefficient, and the auxiliary objective function is used to ensure that the obtained observation point error is relatively small;
evaluating the network while adjusting network parameters, and randomly selecting samples with set proportions as verification samples; performing multiple rounds of training, recording model parameters and verification scores, and finally selecting the model parameter with the highest verification score to predict the optimal point distribution position; the positions of the five wave gliders are adjusted to the optimal observation points, and repeated observation is carried out; after the target point is observed, the position information of the observation point and the hydrologic information obtained by reproduction are added into a training sample, and as the position of the optimal observation point is changed along with time, the failure time of the sample is set to be T, the T is selected according to the hydrologic environment of the observation area, the sample with the existence time exceeding T in the training sample set is deleted, the neural network is retrained, the position of the new optimal observation point is predicted, and the operation is repeated.
The invention has the following beneficial effects:
the invention provides a self-adaptive ocean mobile acoustic chromatography system, which aims at the defects that the measuring range is relatively fixed and the measuring error is uncontrollable when the existing acoustic chromatography technology observes ocean environment information, and comprises a plurality of water surface mobile platforms distributed outside an observation sea area, wherein each water surface mobile platform is provided with three hydrophones, low-frequency and high-frequency underwater acoustic transducers and a positioning and time service device, the positioning and time service device realizes the self-positioning of the water surface mobile platform and enables the underwater acoustic transducers to synchronously send out detection sound waves, the high-frequency acoustic transducers send high-frequency acoustic signals to be matched with the three hydrophones to realize the accurate positioning of the high-frequency and low-frequency underwater acoustic transducers, and the low-frequency underwater acoustic transducers send middle-low-frequency acoustic signals and the low-frequency underwater acoustic transducers corresponding to other water surface mobile platforms mutually receive the transmitted signals so as to realize the acoustic chromatography observation of ocean hydrologic information. According to the mobile acoustic chromatography scheme for measuring by using the water surface mobile platform (such as a wave glider, an underwater autonomous vehicle, an unmanned ship and the like) to carry the acoustic observation device, the defect that the position of a measurement node is relatively fixed in the traditional marine acoustic chromatography can be overcome, the self-adaptive sampling of the observation node can be realized, and the measurement precision of marine hydrologic information is remarkably improved; based on three hydrophones of each water surface moving platform, the underwater acoustic transducer of each water surface moving platform can be accurately positioned, so that the accurate positioning of the observation device is realized; further, the mobile acoustic chromatography system and the method provided by the invention fully utilize the mobility of the water surface mobile platform, realize the optimal position arrangement and adjustment of the observation equipment by using the neural network algorithm, and realize the self-adaptive sampling of ocean environment fields with different structures by using the method of searching the optimal arrangement point by using the neural network algorithm, thereby overcoming the influence of various ocean phenomena on the measurement precision and realizing the minimization of the measurement error.
Drawings
Fig. 1 is a schematic view of a surface mobile platform according to an embodiment of the present invention (a wave glider is taken as an example of the surface mobile platform), and fig. 1a is a bottom view of a surface ship of the surface mobile platform.
FIG. 2 is a three-point positioning hydrophone mounting arrangement for one embodiment of the invention.
Fig. 3 is a schematic diagram illustrating horizontal propagation of sound rays between mobile nodes according to an embodiment of the present invention.
FIG. 4 is a schematic diagram of optimal placement adjustment according to an embodiment of the present invention.
FIG. 5 is a general flow chart of an observation scheme according to one embodiment of the present invention.
Detailed Description
The following describes embodiments of the present invention in detail. It should be emphasized that the following description is merely exemplary in nature and is in no way intended to limit the scope of the invention or its applications.
Referring to fig. 1 to 5, in one embodiment, an adaptive ocean mobile acoustic tomography system includes a plurality of water surface mobile platforms distributed around the observation sea, wherein three hydrophones 9, 10, 11 are installed at the bottom of each water surface mobile platform for accurate positioning of underwater acoustic transducers, each water surface mobile platform is connected with underwater low-frequency underwater acoustic transducers 5 and high-frequency underwater acoustic transducers 6 through drag cables, the water surface mobile platform is provided with positioning and timing devices for realizing self positioning of the water surface mobile platform and enabling the underwater acoustic transducers to synchronously emit detection sound waves, the high-frequency underwater acoustic transducers 6 are used for emitting high-frequency acoustic signals and are matched with the three hydrophones 9, 10, 11 to realize accurate positioning of the high-frequency underwater acoustic transducers 6 and the low-frequency underwater acoustic transducers 5, and the low-frequency underwater acoustic transducers 5 are used for emitting medium-low frequency acoustic signals, and the low-frequency underwater acoustic transducers 5 corresponding to other water surface mobile platforms mutually receive the emitted signals to realize acoustic tomography observation of ocean hydrologic information.
In a specific embodiment, the surface mobile platform may be a wave glider, an AUV, an unmanned ship, or an artificial ship.
In a specific embodiment, the positioning and timing device may be a GPS or a beidou positioning system.
In a preferred embodiment, the water surface moving platform is a wave glider, the wave glider comprises a water surface ship 1, an underwater traction body 2 and an umbilical cable 3 connecting the water surface ship 1 and the underwater traction body 2, the water surface ship 1 moves through the dragging of the underwater traction body 2, and the high-frequency underwater sound transducer 6 and the low-frequency underwater sound transducer 5 are mounted on the underwater traction body 2.
In a more preferred embodiment, the hydrophone carrier 4 is fixed to the bottom of the surface vessel 1, the hydrophone carrier 4 has three connecting rods forming an angle of 120 degrees between each other, and the three hydrophones 9, 10, 11 are mounted on the three connecting rods, respectively.
In a more preferred embodiment, the surface vessel 1 is provided with a gyroscope 7 for measuring the attitude of the surface vessel 1 to obtain the position of each hydrophone relative to the surface vessel 1, and the actual positions of the three hydrophones 9, 10, 11 are obtained in cooperation with the position information of the surface mobile platform determined by the positioning and timing device, and the system determines the exact positions of the high frequency underwater sound transducer 6 and the low frequency underwater sound transducer 5 mounted on the underwater traction body by a three-point positioning method according to the actual positions of the three hydrophones 9, 10, 11.
In a preferred embodiment, the initial sampling points of the water surface moving platforms are uniformly distributed on the periphery circumference of the region to be observed, and the water surface moving platforms randomly move along the circumference to perform multiple sampling; the plurality of water surface moving platforms are preferably five wave gliders.
In the preferred embodiment, the low-frequency underwater acoustic transducer 5 of each water surface mobile platform is converted into a receiving mode after completing the acoustic wave emission; each water surface mobile platform receives sound waves sent by a low-frequency underwater sound transducer 5 carried by other water surface mobile platforms, and calculates the arrival time delay of each sound wave; when each water surface mobile platform receives low-frequency sound waves sent by all other water surface mobile platforms, one-time measurement is completed; after the measurement is completed at the initial sampling point, relevant data including the position of the low-frequency underwater acoustic transducer 5, the sequence of arrival and the propagation time of the sound waves are transmitted to a shore base through a communication system carried on a water surface mobile platform for analysis, and marine hydrologic information at the corresponding moment of an observation area is calculated in an inversion mode; preferably, the entire observation area has a total of 10 sound ray coverage per measurement.
In another embodiment, an adaptive ocean mobile acoustic tomography method uses the adaptive ocean mobile acoustic tomography system to perform acoustic tomography of ocean hydrologic information.
In some embodiments, the marine mobile acoustic chromatography observation system is realized by arranging a plurality of water surface mobile platforms outside an observation sea area, wherein the water surface mobile platforms can be autonomous observation devices such as wave gliders, AUVs, unmanned ships and the like, and can also be realized by artificial ships. The water surface mobile platform realizes self positioning and time service through GPS or Beidou, and the purpose of time service is to ensure that a plurality of observation devices synchronously send detection signals. The bottom of the water surface moving platform is provided with three hydrophones 9, 10 and 11 for accurately positioning underwater acoustic transducers. The water surface moving platform is connected with two underwater acoustic transducers through a dragging cable, one underwater acoustic transducer transmits high-frequency acoustic signals to be matched with three hydrophones 9, 10 and 11 arranged at the bottom of the moving platform, and accurate positioning of the underwater acoustic transducers is achieved through a three-point positioning method; the other underwater acoustic transducer transmits medium-low frequency acoustic signals, and the underwater acoustic transducers corresponding to other mobile platforms mutually receive the transmitted signals, so that acoustic chromatographic observation of marine hydrologic information is realized.
The general flow of the observation method of the system is shown in fig. 5. The following water surface moving platform uses five wave gliders as an example to describe the self-adaptive ocean moving sound chromatography method.
First, the underwater traction body 2 drags the surface ship 1 of the wave glider to initial sampling points through the umbilical cable 3, as shown in fig. 3, the initial sampling points are uniformly distributed on the periphery circumference of the area to be observed a, and the surface ship 1 of the five wave gliders randomly moves along the circumference to perform multiple sampling. After the five wave gliders reach the initial sampling point, the time service of the system is finished through the GPS8 mounted in the cabin of the water surface ship of each wave glider, so that all underwater sound transducers can emit detection sound waves simultaneously. The wave glider is provided with a low-frequency underwater acoustic transducer 5 and a high-frequency underwater acoustic transducer 6, and simultaneously sends out a low-frequency acoustic signal and a high-frequency acoustic signal respectively, and the wave glider records the time of sound wave emission. As shown in fig. 2, three hydrophones, each of which forms an angle of 120 degrees, are mounted on a hydrophone frame 4, and the hydrophone frame 4 is fixed at the bottom of the surface ship 1 through bolts. Because the lengths of the three connecting rods are known, after the attitude of the surface ship 1 is measured through the gyroscope 7 arranged in the cabin of the surface ship 1, the position of each hydrophone relative to the surface ship 1 can be obtained, and then the position of each hydrophone is matched with the position of the surface ship 1 determined by the GPS8, so that the actual positions of the three hydrophones can be obtained. The position determining method comprises the following steps:
1. the position of the GPS8 carried by the wave glider is selected as the origin O of a world coordinate system, the X axis is in the direction of the positive east, and the Y axis is in the direction of the positive north, so that a right-hand coordinate system OXYZ is formed.
2. The center of a hydrophone bracket is selected as an origin O ', an X ' axis points to the starboard of the water surface ship 1, a Y ' axis points to the direction of the bow of the ship, a right-hand coordinate system O ' X ' Y ' Z ' is formed, and the origin of the hydrophone coordinate system in a world coordinate system has a coordinate P 0 ＝(x o ，y o ，z o ) T 。
3. Because the included angle between the rod length and the hydrophone is known, the coordinate p of the three hydrophones under the coordinate system O 'X' Y 'Z' can be obtained i ＝(x i ，y i ，z i ) T I=1, 2,3, the transformation from the hydrophone coordinate system to the world coordinate system can be expressed as:
R＝R x R y R z ，
the roll, pitch and yaw are angles rotating around three coordinate axes of XYZ respectively, and the rotation from the world coordinate system to the hydrophone coordinate system is performed according to the sequence of rotating around the X axis, rotating around the Y axis and finally rotating around the Z axis, so that when the hydrophone coordinate system to the world coordinate system is performed, the transformation sequence is performed by rotating around the Z axis, then rotating around the Y axis and finally rotating around the X axis. R is R x ，R y ，R z The transformation matrix rotates around three coordinate axes, R is from the hydrophone coordinate system to the world coordinate systemAnd transforming the matrix.
4. The position of three hydrophones in the world coordinate system can be expressed as P i ＝(X i ，Y i ，Z i ) T ，i＝1，2，3
P i ＝Rp i +P 0
Three hydrophones on the hydrophone frame 4 for receiving signals from the high frequency hydrophone transducers 6, the distance x from each hydrophone transducer 6 to each hydrophone being determined by calculating the delay of arrival of sound waves i I=1, 2,3. According to the calculated exact positions of the three hydrophones, the exact position of the high-frequency underwater acoustic transducer 6 can be determined by a three-point positioning method, namely, the exact position of the low-frequency underwater acoustic transducer 5 is obtained.
After the low-frequency underwater acoustic transducer 5 completes the acoustic wave transmission, it is converted into a receiving mode. The wave glider receives sound waves emitted by the low-frequency underwater sound transducer 5 carried by other wave gliders, and calculates the time delay of arrival of each sound wave. The entire observation area is covered with 10 sound rays S per measurement. When each wave glider receives the low-frequency sound waves emitted by all other wave gliders, one measurement is completed. After the measurement is completed at the initial sampling point, relevant data (such as the position of the low-frequency underwater acoustic transducer 5, the sequence of arrival and propagation time of the sound wave and the like) are transmitted to a shore base through a communication system carried by the wave glider for analysis, and hydrological information such as temperature, flow speed and the like at the corresponding moment of an observation area and hydrological information errors are calculated in an inversion mode.
In addition, mobility of the mobile platform is fully utilized, and in a preferred embodiment, the optimal placement position of the mobile node is found through a neural network algorithm. After the mobile platform completes observation at the initial sampling point, the communication device of the water surface part of the mobile platform transmits the observation data to a shore base for analysis, and the hydrological information in the observed sea area is inverted through the measurement data to obtain an estimated error. The neural network algorithm takes observed measurement data as a training sample, predicts the optimal point setting target position, and because the ocean power environment changes with time, the optimal position of the mobile platform also changes with time, and the obtained optimal point setting position is the optimal point at the current moment. And the mobile platform adjusts the position according to the optimal target position obtained by the shore-based analysis, repeatedly observes, and transmits the data back to the shore-based for analysis after the measurement is completed, so that the operation is repeated.
Taking the error of inversion hydrologic information as the input of a neural network, inverting the error to be the error of different points of the whole observation area, wherein the data volume is related to grid division, and the number of neurons of an input layer is equal to the number of errors; because the difference of the value ranges of the features is large, normalization processing is carried out before data use, and meanwhile, the number of samples is small, in order to avoid the overfitting phenomenon, a small-sized neural network is adopted, and the network is arranged to comprise one to two hidden layers. The positions of the five observation points are used as the output of the neural network, the output layer of the neural network is set to be 10 neurons, and the 10 neurons respectively represent one component in the two-dimensional positions of the observation points and correspond to the five observation positions. Defining an objective function:
L＝L δ (y,f(e))+L s (e)，
wherein L is δ (y, f (e)) is the Huber loss function as follows:
where e is input error information, y represents an actual observation position, f (e) is a predicted optimal observation point position, and δ is a converging radius. The Huber function is used as an objective function, each component is guaranteed to be near an actual observation position, and the sensitivity degree to abnormal points in data can be reduced and the optimal point is guaranteed not to be missed when gradient descent is about to end by adjusting the super parameter delta. Additionally set up an auxiliary loss function L s (e) Ce, where e is the hydrologic error of the corresponding observation point, C is the penalty coefficient, the auxiliary objective function ensures that the obtained observation point error is relatively small, and in practical application, the coordination error and the positional relationship, i.e., L, are considered s (e) And L δ (y,f(e))。
To evaluate the network while adjusting network parameters, a certain proportion of samples are randomly selected as validation samples (e.g., 20% of samples are validated). And performing multiple rounds of training, recording model parameters and verification scores, and finally selecting the model parameter with the highest verification score to predict the optimal point distribution position. Fig. 4 is an optimal distribution point adjustment schematic diagram, and positions of five wave gliders are adjusted to an optimal observation point for repeated observation. After the target point is observed, the position information of the observation point and the hydrologic information obtained by reproduction are added into a training sample, and the optimal observation point position is changed along with time, so that the failure time of the sample is set to be T, the T is selected according to the hydrologic environment of the observation area, the sample with the existence time exceeding T in the training sample set is deleted, the neural network is retrained, and the new optimal observation point position is predicted, and the operation is repeated.
The foregoing is a further detailed description of the invention in connection with specific/preferred embodiments, and it is not intended that the invention be limited to such description. It will be apparent to those skilled in the art that several alternatives or modifications can be made to the described embodiments without departing from the spirit of the invention, and these alternatives or modifications should be considered to be within the scope of the invention.
Claims (10)
1. The self-adaptive ocean mobile acoustic chromatographic system is characterized by comprising a plurality of water surface mobile platforms distributed outside an observation sea area, wherein three hydrophones are arranged at the bottom of each water surface mobile platform and used for accurately positioning an underwater low-frequency underwater acoustic transducer and a high-frequency underwater acoustic transducer, each water surface mobile platform is connected with the underwater low-frequency underwater acoustic transducer and the underwater high-frequency underwater acoustic transducer through a dragging cable, the water surface mobile platform is provided with a positioning and time service device and is used for realizing self positioning of the water surface mobile platform and enabling the underwater acoustic transducer to synchronously emit detection sound waves, the high-frequency underwater acoustic transducer is used for transmitting high-frequency acoustic signals and is matched with the three hydrophones so as to realize accurate positioning of the high-frequency underwater acoustic transducer and the low-frequency underwater acoustic transducer, and the low-frequency underwater acoustic transducer is used for transmitting medium-low frequency acoustic signals and the low-frequency underwater acoustic transducer corresponding to other mobile platforms mutually receive the transmitted signals so as to realize acoustic chromatographic observation of ocean hydrologic information; the water surface moving platform is a wave glider, the wave glider comprises a water surface ship, an underwater traction body and an umbilical cable for connecting the water surface ship and the underwater traction body, the water surface ship moves through the dragging of the underwater traction body, and the high-frequency underwater acoustic transducer and the low-frequency underwater acoustic transducer are arranged on the underwater traction body; the hydrophone bracket is fixed at the bottom of the water surface ship and is provided with three connecting rods forming an included angle of 120 degrees between every two hydrophone brackets, and the three hydrophones are respectively arranged on the three connecting rods; the system comprises a water surface ship, a positioning and timing device, a gyroscope, a system and a three-point positioning method, wherein the gyroscope is arranged on the water surface ship and is used for measuring the gesture of the water surface ship so as to obtain the position of each hydrophone relative to the water surface ship, the gyroscope is matched with the position information of the water surface moving platform determined by the positioning and timing device to obtain the actual positions of the three hydrophones, and the system determines the exact positions of the high-frequency underwater acoustic transducer and the low-frequency underwater acoustic transducer which are arranged on the underwater traction body according to the actual positions of the three hydrophones.
2. The adaptive marine mobile acoustic tomography system of claim 1 wherein the positioning and timing device is a GPS or beidou positioning system.
3. The adaptive marine mobile acoustic tomography system of any of claims 1 to 2, wherein the plurality of surface mobile platforms have initial sampling points uniformly distributed around the periphery of the region to be observed, and wherein the plurality of surface mobile platforms move randomly around the periphery for multiple samples.
4. The adaptive marine mobile acoustic chromatography system of claim 3, wherein the plurality of surface mobile platforms are five wave gliders.
5. The adaptive marine mobile acoustic tomography system of any one of claims 1 to 2 wherein the low frequency underwater acoustic transducer of each surface mobile platform converts to a receive mode after completion of acoustic wave emission; each water surface mobile platform receives sound waves sent by low-frequency underwater sound transducers carried by other water surface mobile platforms, and calculates the arrival time delay of each sound wave; when each water surface mobile platform receives low-frequency sound waves sent by all other water surface mobile platforms, one-time measurement is completed; after the measurement is completed at the initial sampling point, relevant data including the position of the low-frequency underwater acoustic transducer, the sequence of arrival and the propagation time of the sound waves are transmitted to a shore base through a communication system carried by the water surface mobile platform for analysis, and the marine hydrologic information of the observation area at the corresponding moment is calculated in an inversion mode.
6. The adaptive marine mobile acoustic tomography system of claim 5, wherein the total observation area has a total of 10 acoustic line coverage per measurement.
7. A self-adaptive ocean mobile acoustic chromatography method based on a self-adaptive ocean mobile acoustic chromatography system is characterized in that,
the self-adaptive ocean mobile acoustic chromatographic system comprises a plurality of water surface mobile platforms distributed on the periphery of an observation sea, wherein three hydrophones are arranged at the bottom of each water surface mobile platform and used for accurately positioning underwater low-frequency underwater acoustic transducers and high-frequency underwater acoustic transducers, each water surface mobile platform is connected with the underwater low-frequency underwater acoustic transducers and the underwater high-frequency underwater acoustic transducers through a dragging cable, the water surface mobile platform is provided with a positioning and time service device and is used for realizing self positioning of the water surface mobile platform and enabling the underwater acoustic transducers to synchronously emit detection sound waves, the high-frequency underwater acoustic transducers are used for emitting high-frequency acoustic signals and are matched with the three hydrophones so as to realize accurate positioning of the high-frequency underwater acoustic transducers and the low-frequency underwater acoustic transducers, and the low-frequency underwater acoustic transducers corresponding to other water surface mobile platforms are mutually used for receiving emitted signals so as to realize acoustic chromatographic observation of ocean hydrologic information; the water surface moving platform measures the gesture of the water surface ship through a gyroscope so as to obtain the position of each hydrophone relative to the water surface ship, and the gesture is matched with the position information of the water surface moving platform determined by the positioning and time service device so as to obtain the actual positions of the three hydrophones, and the system determines the exact positions of the high-frequency underwater sound transducer and the low-frequency underwater sound transducer through a three-point positioning method according to the actual positions of the three hydrophones;
the method uses the adaptive marine mobile acoustic tomography system to perform acoustic tomography of marine hydrologic information.
8. The method of adaptive marine mobile acoustic tomography of claim 7 wherein the determining of the actual position of the three hydrophones comprises:
1) Selecting the positioning position of the water surface moving platform as the origin O of a world coordinate system, wherein the X axis is in the northwest direction, and the Y axis is in the northwest direction to form a right-hand coordinate system OXYZ;
2) Selecting the center of three hydrophones as the origin O ', the X ' axis points to the starboard of the water surface moving platform, and the Y ' axis points to the head direction of the water surface moving platform, so as to form a right-hand coordinate system O ' X ' Y ' Z ', wherein the origin of the hydrophone coordinate system in the world coordinate system is P 0 ＝(x o ，y o ，z o ) T ；
3) According to the coordinates p of three hydrophones in the coordinate system O ' X ' Y ' Z i ＝(x i ，y i ，z i ) T I=1, 2,3, determining the transformation of the hydrophone coordinate system into the world coordinate system is:
R＝R x R y R z ，
the roll, the pitch and the yaw are angles rotating around three coordinate axes of XYZ respectively, and the rotation from the world coordinate system to the hydrophone coordinate system is firstly performed in the order of rotating around the X axis, rotating around the Y axis and finally rotating around the Z axis, so that when the hydrophone coordinate system to the world coordinate system is performed, the transformation order is that the hydrophone coordinate system firstly rotates around the Z axis, then rotates around the Y axis and finally rotates around the X axis; r is R x ，R y ，R z The transformation matrix is respectively a transformation matrix when rotating around three coordinate axes, and R is a transformation matrix from a hydrophone coordinate system to a world coordinate system;
4) Determining the position of three hydrophones in a world coordinate system as P i ＝(X i ，Y i ，Z i ) T ，i＝1，2，3
P i ＝Rp i +P 0 。
9. An adaptive ocean mobile acoustic tomography method according to claim 7 or 8 wherein,
searching the optimal placement position of the mobile node through a neural network algorithm: after the water surface mobile platform completes observation at an initial sampling point, transmitting observation data to a shore base for analysis through a communication device of the water surface mobile platform, and inverting hydrological information in an observed sea area through measurement data to obtain an estimated error; the neural network algorithm takes observed measurement data as a training sample, predicts the optimal point setting target position, and because the ocean power environment changes with time, the optimal position of the water surface mobile platform also changes with time, and the obtained optimal point setting position is the optimal point at the current moment; the water surface mobile platform carries out position adjustment according to the optimal target position obtained by shore-based analysis, then carries out repeated observation, and after the measurement is completed, transmits data back to shore for analysis again, so that the data is reciprocated;
the method comprises the steps of taking the error of inversion hydrologic information as the input of a neural network, inverting the error of different points of the whole observation area, and making the data quantity related to grid division, wherein the number of neurons of the input layer is equal to the number of errors.
10. The method of adaptive marine mobile acoustic tomography of claim 9,
taking the positions of the five observation points as the output of the neural network, setting the neural network output layer as 10 neurons, and respectively representing one component in the two-dimensional positions of the observation points, wherein the two components correspond to the five observation positions; defining an objective function:
L＝L δ (y,f(e))+L s (e)，
wherein L is δ (y, f (e)) is the Huber loss function as follows:
wherein e is input error information, y represents an actual observation position, f (e) is a predicted optimal observation point position, and delta is a converging radius; the Huber function is used as an objective function, each component is guaranteed to be near an actual observation position, and the sensitivity degree to abnormal points in data is reduced by adjusting the super parameter delta, so that the optimal point is guaranteed not to be missed when gradient descent is about to end; additionally set up an auxiliary loss function L s (e) Ce, where e is the hydrologic error of the corresponding observation point, C is the penalty coefficient, and the auxiliary objective function is used to ensure that the obtained observation point error is relatively small;
evaluating the network while adjusting network parameters, and randomly selecting samples with set proportions as verification samples; performing multiple rounds of training, recording model parameters and verification scores, and finally selecting the model parameter with the highest verification score to predict the optimal point distribution position; the positions of the five wave gliders are adjusted to the optimal observation points, and repeated observation is carried out; after the target point is observed, the position information of the observation point and the hydrologic information obtained by reproduction are added into a training sample, and as the position of the optimal observation point is changed along with time, the failure time of the sample is set to be T, the T is selected according to the hydrologic environment of the observation area, the sample with the existence time exceeding T in the training sample set is deleted, the neural network is retrained, the position of the new optimal observation point is predicted, and the operation is repeated.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910238943.0A CN109900256B (en) | 2019-03-27 | 2019-03-27 | Self-adaptive ocean mobile acoustic tomography system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910238943.0A CN109900256B (en) | 2019-03-27 | 2019-03-27 | Self-adaptive ocean mobile acoustic tomography system and method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109900256A CN109900256A (en) | 2019-06-18 |
| CN109900256B true CN109900256B (en) | 2023-10-20 |
Family
ID=66953032
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910238943.0A Active CN109900256B (en) | 2019-03-27 | 2019-03-27 | Self-adaptive ocean mobile acoustic tomography system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109900256B (en) |
Families Citing this family (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110319811B (en) * | 2019-06-27 | 2021-07-20 | 华南理工大学 | Underwater single-beam high-precision detection system and method adaptive to wave effect |
| CN111308474B (en) * | 2020-03-11 | 2021-12-21 | 自然资源部第二海洋研究所 | Towed deep sea seabed shallow structure acoustic detection system and method |
| CN111856936B (en) * | 2020-07-21 | 2023-06-02 | 天津蓝鳍海洋工程有限公司 | Control method for cabled underwater high-flexibility operation platform |
| CN112849349B (en) * | 2021-01-18 | 2022-02-01 | 清华大学深圳国际研究生院 | Online internal wave early warning system and method |
| CN113377842B (en) * | 2021-06-21 | 2022-07-08 | 山东八五信息技术有限公司 | Emergency management method and system based on industrial internet big data |
| CN113556186B (en) * | 2021-07-26 | 2023-01-13 | 广东海洋大学 | Underwater communication auxiliary tracking device and system and method applying same |
| CN113758446B (en) * | 2021-09-09 | 2022-12-27 | 厦门大学 | Underwater acoustic measurement method for thickness of marine organism attachment |
| CN115248030B (en) * | 2021-12-28 | 2023-09-22 | 浙江海洋大学 | Ocean vortex flow measuring method based on acoustic chromatography |
| CN114070419B (en) * | 2022-01-17 | 2022-05-17 | 北京理工大学 | Underwater arbitrary shape detection array system and method based on high-speed wireless communication |
| CN115407404B (en) * | 2022-11-03 | 2023-05-12 | 自然资源部第二海洋研究所 | Self-adaptive ocean mobile acoustic chromatography device |
| CN117760507A (en) * | 2023-12-27 | 2024-03-26 | 广州远动信息技术有限公司 | River acoustic chromatographic flow monitoring system and method based on double-frequency underwater acoustic base station |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6552336B1 (en) * | 2000-10-11 | 2003-04-22 | The United States Of America As Represented By The Secretary Of The Navy | Non-invasive, opto-acoustic water current measurement system and method |
| EP1691214A1 (en) * | 2005-02-15 | 2006-08-16 | Dcn | System for determining the absolute position of a towed or self-propelled submarine engine |
| CN201016731Y (en) * | 2006-12-08 | 2008-02-06 | 国家海洋局第二海洋研究所 | Multi-frequency sea-bottom acoustic original position testing system |
| CN101762824A (en) * | 2010-01-20 | 2010-06-30 | 上海交通大学 | Method for measuring position of marine seismic streamer based on one-way hydroacoustic ranging |
| WO2014028084A2 (en) * | 2012-05-18 | 2014-02-20 | King Abdullah University Of Science And Technology | Satellite and acoustic tracking device |
| CN106842173A (en) * | 2017-02-28 | 2017-06-13 | 浙江海洋大学 | Dispersion extracting method, apparatus and system |
| CN107356233A (en) * | 2017-06-14 | 2017-11-17 | 中国科学院声学研究所 | A kind of vertical array 1 system suitable for the extremely cold marine site acoustics hydrographic survey of high latitude |
| CN209485353U (en) * | 2019-03-27 | 2019-10-11 | 清华大学深圳研究生院 | A kind of mobile sound tomographic system in adaptive ocean |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2010138624A1 (en) * | 2009-05-27 | 2010-12-02 | Teledyne Rd Instruments | System and method for determining wave characteristics from a moving platform |
| WO2011084483A2 (en) * | 2009-12-16 | 2011-07-14 | Shb Instruments, Inc. | Underwater acoustic navigation systems and methods |
-
2019
- 2019-03-27 CN CN201910238943.0A patent/CN109900256B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6552336B1 (en) * | 2000-10-11 | 2003-04-22 | The United States Of America As Represented By The Secretary Of The Navy | Non-invasive, opto-acoustic water current measurement system and method |
| EP1691214A1 (en) * | 2005-02-15 | 2006-08-16 | Dcn | System for determining the absolute position of a towed or self-propelled submarine engine |
| CN201016731Y (en) * | 2006-12-08 | 2008-02-06 | 国家海洋局第二海洋研究所 | Multi-frequency sea-bottom acoustic original position testing system |
| CN101762824A (en) * | 2010-01-20 | 2010-06-30 | 上海交通大学 | Method for measuring position of marine seismic streamer based on one-way hydroacoustic ranging |
| WO2014028084A2 (en) * | 2012-05-18 | 2014-02-20 | King Abdullah University Of Science And Technology | Satellite and acoustic tracking device |
| CN106842173A (en) * | 2017-02-28 | 2017-06-13 | 浙江海洋大学 | Dispersion extracting method, apparatus and system |
| CN107356233A (en) * | 2017-06-14 | 2017-11-17 | 中国科学院声学研究所 | A kind of vertical array 1 system suitable for the extremely cold marine site acoustics hydrographic survey of high latitude |
| CN209485353U (en) * | 2019-03-27 | 2019-10-11 | 清华大学深圳研究生院 | A kind of mobile sound tomographic system in adaptive ocean |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109900256A (en) | 2019-06-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109900256B (en) | Self-adaptive ocean mobile acoustic tomography system and method | |
| CN110146895B (en) | Acoustic velocity profile inversion method based on inverted multi-beam echometer | |
| CN110208812A (en) | Unmanned vehicles seabed dimensional topography detection device and method partly latent | |
| CN109884730B (en) | Virtual array passive detection method based on deep-sea underwater glider platform | |
| CN109724684B (en) | Direct signal propagation time measuring method based on underwater autonomous vehicle | |
| CN103969639B (en) | The signal processing system of five wave beam fish detectors and signal processing method thereof | |
| CN110294080B (en) | Method for realizing underwater accurate operation by using ultra-short baseline | |
| CN111239829A (en) | Towed controllable source electromagnetic and underwater sound composite underwater target detection system and method | |
| CN104133217B (en) | Method and device for three-dimensional velocity joint determination of underwater moving target and water flow | |
| CN209485353U (en) | A kind of mobile sound tomographic system in adaptive ocean | |
| CN102590804A (en) | Overland testing system of Doppler sonar and testing method thereof | |
| CN108680170A (en) | AUV based on electromagnetic wave attenuation principle returns depressed place navigation system and method | |
| CN109319074B (en) | Multi-orthogonal signal emission unmanned submersible vehicle sound guiding and recycling system | |
| CN108107436A (en) | A kind of submarine target based on reliable acoustic path is actively classified and localization method | |
| CN108919324B (en) | Positioning method of underwater glider | |
| CN105738869A (en) | Deepwater beacon searching and positioning method suitable for single hydrophone | |
| WO1997050007A2 (en) | Method of locating hydrophones | |
| CN111578944A (en) | Underwater glider positioning system and method based on single beacon | |
| CN109632258B (en) | Marine internal wave acoustic detection method based on transmitting-receiving separation of vector sensor | |
| CN110806760B (en) | Target tracking control method of unmanned underwater vehicle | |
| CN106707286B (en) | A kind of experimental system and method for underwater mobile submerged body frequency targets detection | |
| CN111427010A (en) | ROV underwater positioning system and positioning method | |
| CN110471032A (en) | A kind of method of submarine target Passive Positioning | |
| CN113009417B (en) | Submarine acoustic array formation estimation method utilizing sound field interference characteristics | |
| CN112666562B (en) | Synthetic aperture sonar motion compensation and imaging method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |