Method for ensuring operation of hydroacoustic station with flexible long towed array when tug's course changes

Abstract

FIELD: hydroacoustics.SUBSTANCE: invention relates namely to hydroacoustic stations with a flexible long towed array (HAS with FLTA), designed to detect noisy objects. In the claimed method, using the spatial directivity of combined receivers (CR), the direction to the tug is continuously determined by each CR, which makes it possible to calculate the current coordinates of each CR relative to the tug and thereby determine the correct phasing coefficients for the formation of direction-finding relief.EFFECT: ensuring the correct operation of the HAS with the FLTA when the tug's course changes.1 cl, 6 dwg

Description

[0001] The invention relates to the field of hydroacoustics, namely to hydroacoustic stations with a flexible extended towed antenna (GAS with FLTA), designed to detect noisy objects.

[0002] GAS with GPBA are an important component of the hydroacoustic armament of submarines (SS) and surface ships (SC) [1, 2]. It is planned to equip autonomous unmanned underwater vehicles with them.

[0003] The design of the GPBA is shown in Fig. 1, where the numbers indicate:

[0004] 1 - a rotating current junction that ensures electrical contact between the cables located in the towing vehicle and the cables located in the towing vehicle body during rotation of the towing vehicle around its own axis;

[0005] 2 - a cable-rope designed to remove the active part of the GPBA several hundred meters from the towing vehicle, which ensures the isolation of receivers located in the active part of the GPBA from the noise of the towing vehicle;

[0006] 3 - vibration section, eliminating jerks and vibrations of the active part of the gas turbine engine that occur during towing;

[0007] 4 - the active part, in which pressure receivers are located equidistantly with a step equal to half the wavelength of the signal at the upper frequency of the operating frequency range;

[0008] 5 - end body that creates mechanical resistance to towing and thereby ensures the rectilinear shape of the gas turbine.

[0009] The difference between the GAS with GPBA and the rest of the hydroacoustic weapons of submarines and surface ships lies in the use of a linear hydroacoustic antenna with equidistantly located pressure receivers [1, 2] or combined receivers (hereinafter referred to as CR) [3-6], the features of which are:

[0010] - omnidirectionality of the GPBA in the horizontal and vertical planes, which ensures the detection of both underwater and air targets;

[0011] - the location of the GPBA behind the stern of the submarine and NK, which allows monitoring the space in the stern sector of the course angles, shaded for the bow and side antennas of the submarine and NK;

[0012] - the active part of the GPBA is located several hundred meters from the tug, which ensures isolation from the noise it generates.

[0013] At the same time, the sonar with the gas-propelled baffle has a number of significant drawbacks, one of which is the disruption of the sonar's performance when the tug's course changes, caused by a violation of the linearity (bending) of the gas-propelled baffle.

[0014] This drawback can be eliminated by developing a method for determining the coordinates of each GPBA receiver when the tug's course changes. A significant number of publications have been devoted to the development of such methods, for example [7-27]. These publications discuss the following methods for reconstructing the GPBA shape:

[0015] 1) Using sensors. Directional and depth sensors are placed at several points along the antenna to provide local information about the antenna's lateral displacement. For example, the depth sensor and compass outputs located on each receiver are used to estimate its position.

[0016] 2) Taking into account acoustic data. The antenna shape is modeled as a parabolic curve or a circular arc, with the parabolic model being suitable for slow turns.

[0017] 3) ​​Using Kalman filters. Based on the Paidossis equation, which describes the dynamic behavior of a thin flexible antenna in water, it is discretized in space and time to obtain a Kalman filter representation that describes the physics of motion propagation along the antenna.

[0018] 4) Combining sensor and acoustic data. For example, to estimate the antenna shape, directional sensors along the antenna and acoustic data from hydrophones are used, combining these estimates using adaptive weights.

[0019] 5) In the GPBA based on fiber-optic sensors, the antenna shape is determined by mathematically processing reflections from micro-inhomogeneities of the fiber.

[0020] The disadvantage of all of these methods is the low accuracy of GPBA shape reconstruction and the difficulty of their practical implementation. As a result, none of these methods are used in practice in Russia. No data on their practical application abroad are available from publications.

[0021] The problem can be solved relatively simply and inexpensively if the active part of the GPBA contains equidistantly located combined (vector-scalar) receivers (hereinafter referred to as CR) [28-30]. We will demonstrate this using Fig. 2, which shows:

[0022] - B - GPBA tug;

[0023] - GPBA - flexible extended towed antenna;

[0024] - C - local remote source of noise signal emission (target);

[0025] - KP i - i-th CP located in the active part of the GPBA;

[0026] - - the target course angle, i.e. the angle between the direction to the target and the course line of the tug;

[0027] - - the angle with the vertex at the location of the i-th control point between the direction to the tug and the tangent to the GPBA at the location of the i-th control point with the “+” sign if the direction to the tug is to the right of the tangent and with the “-” sign otherwise;

[0028] - - the length of the part of the GPBA from the stern of the tug to the i-th control point;

[0029] - - the straight line distance between the stern of the tug and the i-th control point;

[0030] - - the path length of a plane wave of a radiation source signal located at the course angle tug, from a straight line parallel to the signal front line and passing through the stern of the tug, to the i-th control point;

[0031] - - the angle with its vertex at the stern of the tug between the straight line connecting the stern of the tug with the i-th control point and the line parallel to the plane front of the sound wave passing through the stern of the tug.

[0032] When the tug's course changes, the GPBA begins to bend, acquiring a shape close to a circle [7]. If this fact is not taken into account when forming the bearing relief, the situational picture will be distorted. This will be demonstrated below.

[0033] The claimed invention considers a GPBA, the active part of which contains equidistantly located combined (vector-scalar) receivers [28-30]. The advantage of such a GPBA is higher noise immunity (the concentration coefficient increases by 3 times

[29] ) and the ability to unambiguously determine the direction (bearing or course angle) of the detected target [3-6].

[0034] As a prototype method, we will choose the method described in [5] for forming the direction-finding relief at the output of the GPBA, consisting of the CP.

[0035] When embedding the CP into the GPBA, the axis of one vector channel, for example, , is rigidly aligned with the axis of the GPBA, and the axes of the other two vector channels And are oriented perpendicular to each other and perpendicular to the axis of the GPBA. With this arrangement of the CP, its spatial orientation relative to the GPBA is determined by the zero angle between the axis of the GPBA (at the point of attachment of the CP) and the axis of the vector channel , as well as an angle deviation of the vector channel axis from the zenith (we will conventionally call it the bank angle), which is assumed to be known.

[0036] Detection and direction finding of targets using the prototype method is carried out based on the assumption that the GPBA is a straight line continuing the course line of the tug, by performing the following sequence of actions:

[0037] 1) Synchronously with the sampling frequency equal to the upper frequency of the operating frequency range increased by 2.4 - 2.5 times, the recording is carried out at moments in time countdowns signal at the output of each -th of the 4 receiving channels of each -th of CPs that are part of the active part of the GPBA. Using the taken readings, realizations are accumulated from signal readings at the output of each channel of each control panel over a time interval equal to the inverse value of the specified resolution of the spectral analysis.

[0038] 2) The vector is calculated dimensions complex spectra of the output signal combined receivers when they are phased in the direction . Each element of the vector calculated using the formula:

[0039] (1)

[0040] where

[0041] - frequency;

[0042] - horizontal direction;

[0043] - vertical direction;

[0044] - the number of samples in the signal implementation;

[0045] - directional characteristic -th channel of the KP ( - vector channels, - scalar pressure channel):

[0046] , (2)

[0047] Formulas (2) correspond to the case when the vector channels of the control system are receivers of oscillatory velocity

[28] . In the case when they are receivers of oscillatory acceleration or pressure gradient, it must be taken into account that the signals at their output are phase-shifted by 90° relative to the signals at the output of the pressure receiver, for which we multiply per imaginary unit.

[0048] 3) The spectral correlation matrix is ​​calculated and accumulated over a specified time interval. dimensions × :

[0049] (3)

[0050] (the horizontal line in (3) symbolizes accumulation over a time interval).

[0051] 4) The energy spectrum of the signal in the direction is calculated

[0052] (4)

[0053] where - frequency-dependent control vector of dimension phasing coefficients, providing phasing of the antenna from non-directional receivers to the direction , calculated as:

[0054] (5)

[0055] - the distance between adjacent checkpoints in the GPBA;

[0056] - the speed of sound at the towing depth.

[0057] 5) The distribution of the input signal energy in vertical and horizontal directions, called the direction-finding relief, is calculated

[31]

[0058] (6)

[0059] where - lower and upper limit frequencies of the operating frequency range.

[0060] 6) In the direction-finding relief, the method of two-sided contrast is used to detect maxima, the source of which are target signals, and determine the abscissas of the maxima, which are the directions to the targets.

[0061] The disadvantage of the prototype, as noted above, is that the operability of the sonar with the gas-propelled baffle is disrupted when the tug's course changes due to the curvature of the gas-propelled baffle, which does not allow monitoring the current position of each control point.

[0062] Technical problem to be solved - improvement of the gas control system with gas-powered aircraft.

[0063] The technical result achieved is to ensure the correct operation of the sonar with the gas-powered navigation system when the tugboat's course changes.

[0064] The stated goal is achieved by the fact that, using the spatial directionality of the control point, each control point continuously determines the direction to the tug, which makes it possible to calculate the current coordinates of each control point relative to the tug and thereby determine the correct phasing coefficients for the formation of the direction-finding relief.

[0065] The essence of the claimed method is that, unlike the prototype, when the course of the tug changes, according to the data of each combined receiver, the angles are determined between the tangent to the antenna directed towards the tug at the place of attachment of the combined receiver and the straight line connecting the combined receiver with the stern of the tug, using these angles, taking into account the distances of the combined receivers from the stern of the tug, determined by the length of the emitted cable-rope, phasing coefficients are determined, used to form the direction-finding relief, this order of determining the phasing coefficients is maintained until the angles between the tangent to the antenna directed towards the tug at the place of attachment of the combined receiver and the straight line connecting the combined receiver with the stern of the tug become equal to zero, taking into account the errors in their determination.

[0066] Let's demonstrate this. Let's assume that the GPBA takes the shape of a circle when the tug's course changes [7]. Then, from the figure in Fig. 2 it follows:

[0067] (7)

[0068] Thus, having measured with -th KP angle (in radians) and determined Based on the length of the elongated cable, the distance can be calculated .

[0069] Fig. 3 shows the direction-finding relief at the exit -th CP, corresponding to the situation shown in Fig. 2. The angles are plotted along the abscissa axis in degrees between the tangent to the GPBA at the attachment point -th CP and directions to the sources. The ordinate axis shows the levels of the radiation sources. The highest radiation level corresponds to the direction =-10° to the stern of the tug, where the propeller is located, which is the main source of the tug's noise emissions. Considering that the active part of the GPBA is located no further than a few hundred meters from the stern of the tug, the signal level at the input of each -th CP will be large, which allows for a more accurate determination of the direction to the stern of the tug to use one of the super-resolution algorithms, for example, the Capon algorithm

[31] , as shown in Fig. 3.

[0070] Note that the angles , plotted along the abscissa axis in Fig. 3, correspond to the course angles of the tug according to the formula

[0071] (8)

[0072] To phase a curved GPBA in a certain direction of observation, it is necessary to determine for this direction the path length of a plane wave from this direction from some fixed position (for example, from a straight line parallel to the wave front line and passing through the stern of the tug) to each control point ( in Fig. 2).

[0073] Fig. 2 shows the length of the path plane wave from the direction corresponding to the course angle of the tug , to -th CP. As can be seen from Fig. 2, this path length is determined by the expression:

[0074] (9)

[0075] As a result, to calculate the phasing coefficients of a curved GPBA, instead of formula (5), you need to use the formula

[0076] , (10)

[0077] in which is determined by formula (9) when substituted into it instead of .

[0078] Figure 4 shows the direction-finding relief corresponding to the situation in Figure 2, with the GPBA phasing in accordance with formula (9). Two maxima are observed in the direction-finding relief: the highest maximum corresponds to the noise of the tug, the second maximum corresponds to the noise of the target located at a course angle of -50° relative to the tug.

[0079] Since the towing noise does not carry useful information at this stage of processing, it can be coherently suppressed by one of the methods described in

[31] , for example, by the compensation method. Fig. 5 differs from Fig. 4 only in that the towing noise is suppressed by 28 dB using the compensation algorithm and is observed as a low-level maximum.

[0080] For comparison, Fig. 6 shows the direction-finding relief corresponding to the situation in Fig. 2, obtained by phasing the GPBA using formula (5), i.e., without taking into account the curvature of the GPBA. From an examination of Fig. 6 it follows that the direction-finding relief depicted therein does not correspond to the actual situation.

[0081] Detection and bearing of targets using the claimed method begins from the moment the tug changes course and is carried out by performing the following sequence of actions:

[0082] 1) Synchronously with the sampling frequency equal to the upper frequency of the operating frequency range increased by 2.4 - 2.5 times, synchronous sampling is carried out at moments in time countdowns signal at the output of each -th of the 4 receiving channels of each -th of CPs that are part of the active part of the GPBA. Using the taken readings, realizations are accumulated from signal readings at the output of each channel of each control unit over a time interval equal to the reciprocal of the specified spectral analysis resolution, lying in the range of 0.01 - 0.1 from the upper frequency of the frequency range used.

[0083] 2) Using each implementation from complex spectra of signals are calculated using the fast Fourier transform (FFT) from signal samples at the exit of each -th of 4 channels each -th of KP. Calculated spectra are combined into a vector

[0084] . (11)

[0085] 3) Calculated and accumulated by implementations from counts ( =5…10) correlation (cross-spectral) matrix (CM) of 4×4 dimensions of signals at the output of 4 channels of each -th of KP:

[0086] (12)

[0087] 4) Using the Capon algorithm

[31] , complex spectra are calculated at the exit of each -th of CP when they are phased in the direction , Where

[0088] - horizontal angle in the range [-180°; +180°] relative to the direction towards the tug tangent to the GPBA line at the attachment point -th CP, selected with a step equal to one tenth of the width of the directivity characteristic of the GPBA at the geometric mean frequency of the frequency range used.

[0089] - vertical angle, for which it is sufficient to use one value =0°, corresponding to the horizontal direction:

[0090] , (13)

[0091] where - inverted correlation matrix .

[0092] 5) The bearing reliefs at the output of each -th CP in coordinates

[31] :

[0093] (14)

[0094] 6) In the bearing relief for everyone -th control point detects the largest maximum (caused by the noise of the tug) and determines the corresponding angle value .

[0095] 7) According to formula (9), when substituting into it for everyone -th CP the lengths of the paths are calculated , used in formula (10) for phasing the signals of individual CPs.

[0096] 8) The complex spectrum at the output of the entire GPBA, phased in each direction, is calculated :

[0097] , (15)

[0098] where according to formula (8) .

[0099] Directions are moved in the range of course angles [-180°; +180°] with a step equal to one tenth of the width of the directional characteristic of the GPBA in the horizontal plane at the geometric mean frequency of the frequency range used. As a direction it is advisable to use only one (horizontal) direction =0°.

[0100] 9) The direction-finding relief

[31] is calculated at the output of the entire GPBA

[0101] (16)

[0102] 10) In the direction-finding relief, the method of two-sided contrast is used to detect maxima, the source of which are target signals, and their abscissas, which are the directions to the targets, are determined.

[0103] The distinctive features of the claimed GPBA are:

[0104] - determination by the length of the ejected cable for each -th CP distance values in a straight line from the stern of the tug to each -th KP;

[0105] - definition for everyone -th CP angle values between the tangent to the GPBA directed towards the tug at the attachment point of each -th CP and direction to the stern of the tug;

[0106] - Using Arrays of Values And to calculate phasing coefficients that ensure the correct construction of direction-finding reliefs when bending the GPBA;

[0107] - calculation of the direction-finding relief using the phasing coefficients calculated in this way.

[0108] The possibility of obtaining the claimed technical result by implementing the method was verified by means of modeling, the results of which are shown in Fig. 3-6.

[0109] Thus, it can be stated that the declared technical result - ensuring the correct operation of the GAS with the GPBA when the tug's course changes - has been achieved.

[0110] Sources of information:

[0111] 1. Koryakin Yu.A., Smirnov S.A., Yakovlev G.V. Shipborne sonar technology. State of the art and current problems / / St. Petersburg: Science. 2004.

[0112] 2. Urick R.J. Fundamentals of hydroacoustics / / L.: Shipbuilding. 1978.

[0113] 3. Karishnev N.S., Polkanov K.I., Belova N.I., Kuznetsov G.N. Unambiguous direction finding of targets by an extended antenna with scalar and vector-scalar modules / / Hydroacoustics. 2012. Issue 15. P. 42-47.

[0114] 4. Kuznetsov G.N., Mikhnyuk A.N., Smirnov N.M., Polkanov K.I. Using a towed vector-scalar module and matched filtering for unambiguous estimation of the coordinates of a broadband source in passive mode / / Hydroacoustics. 2015. Issue 24. P. 36-51.

[0115] 5. US 7599253 B1. Hybrid pressure and vector sensor towed array. 2009.

[0116] 6 Poulsen AJ Robust vector sensor array processing and performance analysis. PhD thesis. Massachusetts Institute of Technology. 2009. 186 p.

[0117] 7 Learn С.А. Picture of snow and ice cream ГПБА / / Дисс. к.т.н. СПб ГЭТУ «ЛЭТИ». 2018.

[0118] 8 Ballegooijen EC, Wednesday GWM, Schooneveld C, Zalm PPM, Parsons AT, Field AT Measurement of towed array position. IEEE Journal of Oceanic Engineering, vol. 14(4), pp. 101-1 375–383,

[0119] 9 Park. A simplified subspace fitting method for estimating shape of a towed array. IEEE Journal of Oceanic Engineering, vol. 33(2), pp. 101-1 215–223,

[0120] 1 Bucker HP Beamforming Towed Line-Array of Unknown Shape. J. Acoust. Soc. Am. 63(5), pp. 101-1 1451-1454,

[0121] 11. Ferguson B.G. Remedying the effects of array shape distortion on the spatial filtering of acoustic data from a line array of hydrophones. IEEE Journal of Oceanic Engineering, vol. 18(4), pp. 565-571, 1993.

[0122] 12. Gray D.A. Models for The Application of Kalman Filtering to the Estimation of the Shape of a Towed Array. NATO ASI Series, Underwater Acoustic Data Processing, pp. 491-495, 1989.

[0123] 13. Smith J.J., Leung Y.H., Cantoni A. Broadband eugenvector methods for towed array shape estimation with a single source. IEEE International Conference on Acoustics, Speech, and Signal Processing, 1996.

[0124] 14. Wahl D.E. Towed array shape estimation using frequency-wavenumber data. IEEE Journal of Oceanic Engineering, vol. 18(4), pp. 582-590, 1993.

[0125] 15. Dosso S.E. Estimation of towed array curvature. Defence Research Establishment Atlantic, 1999.

[0126] 16. Varadarajan V., Krolik J. Array shape estimation and tracking using active sonar reverberation. IEEE Transactions on Aerospace and Electronic Systems, vol. 40(3), pp. 1073-1086, 2004.

[0127] 17. Rogers J., Krolik J. An online method for time-varying spatial spectrum estimation using a towed acoustic array, IEEE Forty Fourth Asilomar Conference on Signals, Systems and Computers, 2010.

[0128] 18. Goldberg J.M. Joint direction of arrival and array shape tracking for multiple moving targets. IEEE Journal of Oceanic Engineering, vol. 23(2), pp. 118-126, 1998.

[0129] 19. Odom J., Krolik J. Passive towed array shape estimation using heading and acoustic data. IEEE Journal of Oceanic Engineering, vol. 40(2), 465-474, pp. 2015.

[0130] 20. Wyeth N.C. Methods of array element localization for a towed underwater acoustic array. IEEE Journal of Oceanic Engineering, vol. 19(1), pp. 128-133, 1994.

[0131] 21. Howard B.E., Syck J.M. Calculation of the shape of a towed underwater acoustic array. IEEE Journal of Oceanic Engineering, vol. 17(2), pp. 193-203, 1992.

[0132] 22. Gray D.A., Anderson B.D.O., Bitmead R.R. Towed array shape estimation using Kalman filters - theoretical models. IEEE Journal of Oceanic Engineering, vol. 18(4), pp. 543-556, 1993.

[0133] 23. Gray D.A., Riley J.L. Towed array shape estimation using Kalman filters -experimental investigations. IEEE Journal of Oceanic Engineering, vol. 18(4), pp. 572-581, 1993.

[0134] 24. Feng Lu, Milios E., Stergiopoulos S., Dhanantwari A. New towed-array shape-estimation scheme for real-time sonar systems. IEEE Journal of Oceanic Engineering, vol. 28(3), pp. 552-563, 2003.

[0135] 25. Nikitakos N.V., Leros A.K., Katsikas S.K. Towed array shape estimation using multimodel partitioning filters. IEEE Journal of Oceanic Engineering, vol. 23(4), pp. 380-384, 1998.

[0136] 26. Newhall BK, Jenkins JW, Dietz JEG Improved estimation of the shape of towed sonar arrays, Proc. of the 21st IEEE Instrumentation and Measurement Technology Conference, 2004.

[0137] 27. Andreev M.Ya., Okhrimenko S.N., Rubanov I.L., Shifman F.N. Positioning system for a flexible extended towed antenna / / Hydroacoustics. 2007. issue 7, pp. 30-32.

[0138] 28. Gordienko V.A., Ilyichev V.I., Zakharov L.N. Vector-phase methods in acoustics. / / M.: Nauka. 1989.

[0139] 29. Skrebnev G.K. Combined hydroacoustic receivers / / St. Petersburg: Elmor. 1997. 200 p.

[0140] 30. Gordienko V.A. Vector-phase methods in acoustics / / M.: Fizmatlit. 2007. 480 p.

[0141] 31. Malyshkin G.S. Optimal and adaptive methods of processing hydroacoustic signals. In 2 volumes. Volume 2. Adaptive methods / / JSC "Concern "Central Research Institute "Elektropribor". 2011. 374 p.

Claims

A method for ensuring the operation of a hydroacoustic station with a flexible extended towed antenna based on combined (vector-scalar) receivers when the course of a tug changes, including the synchronous pickup of signals from the output of each channel of each combined receiver, the calculation of the phasing coefficients of each combined receiver in a given direction, the formation of a direction-finding relief using the calculated phasing coefficients, the detection in it of maxima caused by noise emissions of sound sources using a two-sided contrast method, the determination of course angles corresponding to these maxima, characterized in that when the course of the tug changes, using the data from each combined receiver, the angles are determined between the tangent to the antenna directed towards the tug at the point of attachment of the combined receiver and the straight line connecting the combined receiver with the stern of the tug,using these angles, taking into account the distances of the combined receivers from the stern of the tug, determined by the length of the ejected cable-rope, the phasing coefficients used to form the direction-finding relief are determined; this order of determining the phasing coefficients is maintained until the angles between the tangent to the antenna directed towards the tug at the attachment point of the combined receiver and the straight line connecting the combined receiver with the stern of the tug become equal to zero, taking into account the errors in their determination.

Images