Configuration design and channel weighting fused high-performance circular array and arraying method

Abstract

The invention discloses a high-performance circular array integrating configuration design and channel weighting and an arraying method, and belongs to the technical field of underwater acoustic detection.The method comprises the following steps that S1, a configuration optimization efficiency function is constructed, and an optimal array configuration is screened from a configuration master mask; s2, constructing a weighted efficiency function, processing the optimal array configuration, and determining optimal weights of a plurality of channels; and S3, generating a circular array based on the optimal array configuration and the optimal weights of the plurality of channels. According to the invention, the hardware cost and the development investment are greatly reduced, and the system power consumption and the load of rear-end signal processing are obviously reduced, so that the scale of the sonar system is greatly simplified and the overall cost-effectiveness ratio is improved on the premise of ensuring high directivity.

Description

Technical Field

[0001] This invention belongs to the field of underwater acoustic detection technology, specifically relating to a high-performance circular array and its arrangement method that integrates configuration design and channel weighting. Background Technology

[0002] In modern sonar systems, the circular receiver transducer array is the core sensing unit, and its configuration design directly determines the key performance indicators of the sonar system, such as resolution and anti-interference capability.

[0003] For a long time, in order to achieve high spatial resolution (i.e., to form a narrow receiving beam), the design of circular sonar arrays has generally followed the principle of increasing the physical size (aperture) of the array. However, increasing the physical aperture brings a series of technical challenges: First, in order to avoid high-intensity spurious peaks (i.e., grating lobes) in the beam direction due to spatial undersampling, the number of array elements must be increased to follow the spatial Nyquist sampling theorem and form a densely arranged array with a large number of array elements. This design paradigm is usually called a fully filled array design.

[0004] While this design theoretically ensures lobe-free detection, in engineering practice it falls into the inherent contradiction of high performance accompanied by high cost. The surge in the number of array elements has also directly led to an exponential increase in the number of hardware channels, system power consumption, signal reception and processing, and development costs of the sonar system. This keeps the cost of high-performance sonar systems high, greatly limiting their application and widespread use in cost-sensitive applications.

[0005] Secondly, even with a fully filled array design, circular arrays still have a high sidelobe, a theoretical ceiling that is difficult to overcome. This means that due to the inherent periodicity and symmetry of the array element arrangement, its sidelobe suppression capability has an inherent upper limit. For example, for a typical equidistant circular array, its first sidelobe is only 8 dB lower than the main lobe. In applications where strong targets are used to detect nearby weak targets, such as detecting weak leaks near subsea pipelines or small scour pits near bridge piers, the high sidelobe of the strong target will cause acoustic energy leakage, forming a high-energy sidelobe contamination zone. Similar to the glare effect in optics, the reflection from the weak target on the main lobe is completely overwhelmed by the reflection from the strong target on the sidelobe, ultimately leading to detection "blindness" or "distortion."

[0006] To break this deadlock, existing technologies have attempted to use random or simple rule-based reduction of array elements. However, due to the lack of systematic optimization theory guidance, beam performance often suffers severe degradation and becomes uncontrollable. In particular, it is prone to generating messy high sidelobes at large angles outside the sound beam, or causing the array performance to fluctuate drastically with changes in the scanning direction. Summary of the Invention

[0007] To address the inherent problems of high performance accompanied by high system complexity, high cost, and high sidelobe level inherent in the traditional circular array fully-filled array paradigm, this invention proposes a high-performance circular array and arraying method that integrates configuration design and channel weighting.

[0008] The technical solution of this invention is: a high-performance circular array method integrating configuration design and channel weighting, comprising the following steps:

[0009] S1. Construct a configuration optimization performance function to select the optimal array configuration from the configuration master.

[0010] S2. Construct a weighted efficiency function to process the optimal array configuration and determine the optimal weights for several channels;

[0011] S3. Generate a circular array based on the optimal array configuration and the optimal weights of several channels.

[0012] Furthermore, S1 includes the following sub-steps:

[0013] S11. Establish design goals and physical constraints;

[0014] S12. Based on the design objectives and physical constraints, construct the configuration optimization efficiency function;

[0015] S13. Construct the configuration gene encoding and generate the initial configuration population;

[0016] S14. With minimizing or maximizing the configuration optimization efficiency function as the optimization objective, iteratively search the initial configuration population until the first preset convergence condition is met.

[0017] S15. Select the configuration gene encoding that maximizes the configuration optimization efficiency function from the iterative search results, and decode it to obtain the optimal array configuration.

[0018] Furthermore, in S12, the configuration optimization efficiency function The expression is:

[0019] ;

[0020] in, The peak sidelobe level under uniform excitation, The main lobe width under uniform excitation. To activate the array elements, This is the first performance conversion function; This is the second performance conversion function. This is the third performance conversion function. As the first weighting factor, As the second weighting factor, It is the third weighting factor.

[0021] Furthermore, in S13, binary strings are used for gene encoding. Specifically, for several candidate positions on the configuration master, each bit represents a candidate position. 1 indicates that the position element is activated, and 0 indicates that the position element is in a dormant state.

[0022] Furthermore, in S13, an initial configuration population containing the encoding of several configuration genes is randomly generated.

[0023] Furthermore, in S14, a discrete hybrid simulated annealing particle swarm optimization algorithm is used to iteratively search the initial configuration population.

[0024] Furthermore, S2 includes the following sub-steps:

[0025] S21. Encode the channel excitation weights of several active array elements in the optimal array configuration into real number vectors and generate an initial weight population.

[0026] S22. Construct a weighted efficiency function;

[0027] S23. With minimizing or maximizing the weighted efficiency function as the optimization objective, iteratively search the initial weight population until the second preset convergence condition is met.

[0028] S24. Determine the optimal weights for several channels based on the weights corresponding to the iteration results that satisfy the second preset convergence condition.

[0029] Furthermore, in S22, the weighted efficiency function The expression is:

[0030] ;

[0031] in, The peak sidelobe level when using a set of non-uniform channel weights to be optimized. The first preset pitch scan angle within a specific azimuth plane. The second preset pitch scan angle within a specific azimuth plane. The first in a specific azimuth plane A preset pitch scanning angle To obtain the maximum value,

[0032] Furthermore, in S23, a particle swarm optimization algorithm is used for iterative search.

[0033] Based on the above methods, this invention also proposes a high-performance circular array that integrates configuration design and channel weighting, including a physical array and a signal acquisition and processing system;

[0034] The physical array is used to install several equally spaced transducer units to form a pool of physical candidate array elements;

[0035] The signal acquisition and processing system is used to determine the optimal array configuration and the optimal weights of several channels to generate a circular array.

[0036] The beneficial effects of this invention are:

[0037] (1) This invention effectively addresses high system complexity and high cost, and achieves system scale reduction: By integrating configuration design and channel weighting top-level design, it breaks the inherent contradiction that high performance of traditional circular arrays is inevitably accompanied by high system complexity and high cost; by algorithm optimization, it effectively reduces the number of active array elements on the basis of the equidistant circular array configuration template, directly responding to and solving the problem mentioned in the background technology of the exponential growth of the number of hardware channels, system power consumption, signal reception and processing, and development cost of sonar system caused by the surge in array elements; it not only significantly reduces hardware cost and development investment, but also significantly reduces system power consumption and back-end signal processing load, thereby achieving a significant reduction in the scale of sonar system and an improvement in overall cost-effectiveness while ensuring high directivity.

[0038] (2) This invention breaks through the bottleneck of sidelobe suppression of traditional arrays and improves the detection capability of weak targets: The core of this invention is to generate an asymmetric and non-periodic optimized topology configuration. This structure fundamentally breaks the performance limitation of the 8dB main-sidelobe level upper limit difference determined by the periodicity of the arrangement of traditional circular arrays. It "disperses" the concentrated and high-intensity sidelobe energy in traditional arrays and transforms it into a diffuse and extremely low-level noise-like form, just like replacing the dazzling "optical glare" with a uniform dark background, thereby significantly reducing the sidelobe pollution area caused by strong targets. This qualitative change in sidelobe suppression capability is of decisive significance for application scenarios of detecting nearby weak targets in the background of strong targets, effectively avoiding detection "blindness" or "distortion", and greatly improving the detection resolution capability in complex environments.

[0039] (3) This invention constructs a new design paradigm that is systematic and quantifiable: This invention provides not only a specific array configuration, but a complete and quantifiable methodology for circular array layout. Designers can customize the array configuration and weight scheme that best suits the task by flexibly adjusting the evolutionary efficiency function in the two-stage optimization process according to different task requirements (such as pursuing the lowest sidelobe or balancing resolution and cost). This capability provides a reliable technical approach for array optimization design in different application scenarios, making the array design process more efficient, accurate and controllable. Attached Figure Description

[0040] Figure 1 A flowchart illustrating the overall process of a high-performance circular array method that integrates configuration design and channel weighting;

[0041] Figure 2 This is a diagram of the master template structure of the fully filled circular array configuration used as the basis for screening in an embodiment of the present invention;

[0042] Figure 3 This is a schematic diagram of the optimal array configuration generated by the first stage of the optimization method in an embodiment of the present invention;

[0043] Figure 4(a) shows the beam pattern of a conventional fully filled array under uniform excitation in an embodiment of the present invention;

[0044] Figure 4(b) shows the beam pattern of the optimal array configuration generated in the embodiment of the present invention under uniform excitation;

[0045] Figure 4(c) shows the beam pattern of the optimal array configuration generated in the embodiment of the present invention after loading the optimal channel weights;

[0046] Figure 4(d) shows the sidelobe suppression performance comparison curves of the three schemes (fully filled array, unweighted optimal configuration array, and weighted optimal configuration array) in the same coordinate system;

[0047] Figure 5 This is a breakdown diagram illustrating the comparative verification process of the detection and resolution performance of equal-spacing arrays and the optimized array of the present invention for strong and weak targets in the "high brightness-dark contrast acoustic imaging" scenario (the left image shows the detection results of the traditional array, and the right image shows the detection results of the method of the present invention). Detailed Implementation

[0048] The embodiments of the present invention will be further described below with reference to the accompanying drawings.

[0049] like Figure 1 As shown, this invention provides a high-performance circular array method that integrates configuration design and channel weighting, comprising the following steps:

[0050] S1. Construct a configuration optimization performance function to select the optimal array configuration from the configuration master.

[0051] S2. Construct a weighted efficiency function to process the optimal array configuration and determine the optimal weights for several channels;

[0052] S3. Generate a circular array based on the optimal array configuration and the optimal weights of several channels.

[0053] This invention aims to solve the following core technical challenges:

[0054] (1) Decoupling hardware scale and detection performance: Break the strong coupling relationship between hardware scale and detection performance, significantly reduce the number of array elements while ensuring high directivity, and achieve a significant reduction in the scale of the circular array sonar system under the same directivity.

[0055] (2) Break through the bottleneck of traditional array sidelobe suppression technology: Break the 8dB main sidelobe level upper limit difference determined by the periodicity and symmetry of array element arrangement. By optimizing the array topology design, the concentrated sidelobe energy is converted into diffuse, low-level noise-like sidelobe energy, which significantly reduces the negative impact of sidelobe contamination area on weak target detection.

[0056] (3) Construct a quantifiable and evaluable array design paradigm: break away from the traditional array design model that relies on expert experience and a lot of trial and error, and form a systematic and quantifiable array design method to provide a technical approach for customized array configuration optimization design for different application scenarios.

[0057] In a first aspect, this invention provides a high-performance circular array arrangement method that integrates configuration design and channel weighting. The core idea is to decouple the array design problem into a two-stage sequential optimization mathematical problem: First, through intelligent optimization, an array topology configuration with a small number of array elements and excellent beam performance potential is found in a high-dimensional discrete space; then, a set of channel excitation weights that can further improve performance are fine-tuned for this configuration in a continuous weight space. The entire process focuses on improving the overall performance of the array at a specific azimuth angle. The method specifically includes the following steps:

[0058] Phase 1: Configuration topology optimization based on intelligent optimization.

[0059] The goal of this stage is to select an array configuration from a complete master configuration that has a small number of array elements and can achieve robust elevation electronic scanning performance within a specific azimuth plane (e.g., 0° azimuth) solely through physical layout. This lays a solid structural foundation for subsequent in-depth performance optimization. This stage first establishes clear design objectives and physical constraints based on system application requirements, such as expected beamwidth, operating frequency, and target resolution. These objectives and constraints will directly guide the subsequent definition and optimization of the master configuration.

[0060] Second stage: Channel weighted optimization based on the second intelligent optimization algorithm.

[0061] This stage aims to equip the physical configuration solidified in the first stage with a set of non-uniform channel excitation weights, thereby further optimizing its pitch scanning performance in a specific azimuth plane through post-processing on the basis of the inherently superior physical layout.

[0062] The present invention aims to systematically generate a circular array with a small number of active array elements and high performance when performing pitch electronic scanning in a specific azimuth plane from a configuration master containing M equally spaced candidate array element positions through a two-stage sequential optimization process.

[0063] In this embodiment of the invention, S1 includes the following sub-steps:

[0064] S11. Establish design goals and physical constraints;

[0065] S12. Based on the design objectives and physical constraints, construct the configuration optimization efficiency function;

[0066] S13. Construct the configuration gene encoding and generate the initial configuration population;

[0067] S14. With minimizing or maximizing the configuration optimization efficiency function as the optimization objective, iteratively search the initial configuration population until the first preset convergence condition is met.

[0068] S15. Select the configuration gene encoding that maximizes the configuration optimization efficiency function from the iterative search results, and decode it to obtain the optimal array configuration.

[0069] Configuration topology optimization aims to use intelligent optimization algorithms to select an N-element optimized configuration (N0) from the configuration master template that has a small number of active array elements and excellent pitch electronic scan robustness in physical structure. <M)。

[0070] Based on the system application requirements, such as expected beamwidth, operating frequency, target resolution, and other indicators, establish clear design goals and physical constraints.

[0071] In this embodiment of the invention, in S12, according to the design goals and physical constraints, a configuration optimization efficiency function is constructed to quantitatively evaluate the beam performance of any part of the activation configuration when electronically scanning in a specific azimuth plane under uniform excitation conditions. The configuration optimization efficiency function is a function of one or more beam performance indicators, which can be selected from, but are not limited to, peak sidelobe level (PSL), -3dB main lobe width (HPBW), and the number of activation elements N used.

[0072] The first intelligent optimization algorithm is adopted to perform iterative search and evaluation of the configuration population with the goal of minimizing or maximizing the configuration optimization efficiency function. This includes configuration decoding and efficiency evaluation, elite selection and reproduction decision-making, gene recombination and mutation, and population renewal, until the preset convergence condition is met.

[0073] Configuration optimization efficiency function The expression is:

[0074] ;

[0075] in, The peak sidelobe level under uniform excitation, The main lobe width under uniform excitation. To activate the array elements, This is the first performance conversion function; This is the second performance conversion function. This is the third performance conversion function. As the first weighting factor, As the second weighting factor, It is the third weighting factor.

[0076] The one or more beam performance metrics are evaluated using a "minimize worst-case" strategy; the configuration optimization performance function is defined as the worst-case peak sidelobe level that occurs when the main beam scans a series of preset pitch angles in the specific azimuth plane under uniform excitation.

[0077] In this embodiment of the invention, in S13, binary strings are used for gene encoding. Specifically, for several candidate positions on the configuration master plate, each bit represents a candidate position. 1 indicates that the position element is activated, and 0 indicates that the position element is in a dormant state.

[0078] In an embodiment of the present invention, in S13, an initial configuration population containing several configuration gene codes is randomly generated.

[0079] In this embodiment of the invention, in S14, the discrete hybrid simulated annealing particle swarm optimization algorithm is used to iteratively search the initial configuration population.

[0080] In this embodiment of the invention, S2 includes the following sub-steps:

[0081] S21. Encode the channel excitation weights of several active array elements in the optimal array configuration into real number vectors and generate an initial weight population.

[0082] S22. Construct a weighted efficiency function;

[0083] S23. With minimizing or maximizing the weighted efficiency function as the optimization objective, iteratively search the initial weight population until the second preset convergence condition is met.

[0084] S24. Determine the optimal weights for several channels based on the weights corresponding to the iteration results that satisfy the second preset convergence condition.

[0085] Using a set of randomly generated weight vectors as the initial population, an intelligent optimization algorithm is employed for iterative processing. In each generation, for each weight particle, the following operations are performed:

[0086] 1. Performance Evaluation: Using the current weights, simulate and calculate the beam pattern of this configuration in a specific azimuth plane at a series of preset pitch scanning angles.

[0087] 2. Performance Calculation: Based on the weighted performance function, calculate the performance evaluation value of the current weight (i.e., the worst-case PSL).

[0088] 3. Particle Update: According to the algorithm rules, update the weighted particles to make them evolve towards the individual optimal and global optimal directions.

[0089] Repeat this process until convergence, and finally obtain a set of globally optimal channel weights.

[0090] When the iteration process terminates, the optimal weighted particle obtained is the optimal weighting scheme found by this invention. This weighting scheme, combined with the optimal array configuration, constitutes the final design result of this invention, which achieves low sidelobes and high-performance elevation scanning at this specific azimuth angle.

[0091] The present invention aims to equip the first-stage solidified N-element optimal array configuration with a set of non-uniform channel excitation weights that can further optimize its pitch scanning performance in the specific azimuth plane.

[0092] In this embodiment of the invention, in S22, a weighted efficiency function is constructed to quantify the beam performance of any weighting scheme of N active array elements when performing pitch electronic scanning in the specific azimuth plane after weighting. The weighted efficiency function is a function of one or more beam performance indicators, which can be selected from, but are not limited to, the peak sidelobe level (PSL).

[0093] The channel excitation weights of the N active array elements selected in the first stage are encoded into a real vector of length N, serving as the "particles" for optimization in this stage. A weighted performance function is constructed to evaluate the final performance of any weighting scheme in the elevation scan within a specific azimuth plane. Similar to the configuration optimization performance function, the weighted performance function is also a function of one or more beam performance indices, but it evaluates the weighted radiation pattern. To achieve performance consistency across the scan range, this weighted performance function also adopts the form of "minimizing the worst value".

[0094] Weighted efficiency function The expression is:

[0095] ;

[0096] in, The peak sidelobe level when using a set of non-uniform channel weights to be optimized. The first preset pitch scan angle within a specific azimuth plane. The second preset pitch scan angle within a specific azimuth plane. The first in a specific azimuth plane A preset pitch scanning angle To obtain the maximum value.

[0097] This means using the current set of non-uniform weights and pointing the main beam towards the first... Each pitch scan angle Peak sidelobe level at that time. The operation is the same: find the PSL value with the worst performance among all pitch scan angles.

[0098] The goal of this performance function is to find a set of optimal weights that can maximize the performance across the entire pitch scan domain by finely adjusting the excitation amplitude of each channel, based on the structural robustness already guaranteed in the first stage.

[0099] The beam performance indices in the configuration optimization efficiency function and the weighted efficiency function are further selected from at least one of the following:

[0100] Peak sidelobe level (PSL) in a specific azimuth plane under one or more preset pitch scanning angles.

[0101] -3dB main lobe width (HPBW) within a specific azimuth plane under one or more preset pitch scanning angles.

[0102] Integrated sidelobe level (ISL) in a specific azimuth plane under one or more preset pitch scanning angles.

[0103] The directivity index (DI) of the array;

[0104] The number of active array elements used is N.

[0105] In this embodiment of the invention, in S23, a particle swarm optimization algorithm is used for iterative search.

[0106] The first intelligent optimization algorithm of S1 and the second intelligent optimization algorithm of S2 are independently selected from at least one of the following: genetic algorithm (GA), particle swarm optimization algorithm (PSO), simulated annealing algorithm (SA), ant colony optimization algorithm (ACO), differential evolution algorithm (DE), or any combination or improvement of the above algorithms.

[0107] Based on the above methods, this invention also proposes a high-performance circular array that integrates configuration design and channel weighting, including a physical array and a signal acquisition and processing system;

[0108] The physical array is used to install several equally spaced transducer units to form a pool of physical candidate array elements;

[0109] The signal acquisition and processing system is used to determine the optimal array configuration and the optimal weights of several channels to generate a circular array.

[0110] A signal acquisition and processing system with N channels, where N is less than M;

[0111] The hardware wiring and signal connection topology of the N-channel signal acquisition and processing system is uniquely determined by the optimal array configuration generated in the first stage. This optimal array configuration contains N active array elements and exhibits an asymmetric and aperiodic topology.

[0112] The remaining MN transducer units in the physical array are in an electrically "dormant" or "inactive" state, and do not occupy any electronic system resources of the signal acquisition and processing system, thereby effectively reducing the system hardware cost, power consumption and complexity.

[0113] The processing system integrates a digital signal processing module. Before beamforming the N signals, the module multiplies each signal by a set of digital weights defined by the optimal weighting scheme generated in the second stage and stored in memory, to ensure that the array has excellent sidelobe suppression performance when electronically scanning in a preset specific azimuth plane.

[0114] The physical array has M=52, and the signal acquisition and processing system has N=28 channels.

[0115] The following description is based on specific embodiments.

[0116] Example 1

[0117] This implementation provides a complete flowchart of a high-performance circular array method that integrates configuration design and channel weighting. This method aims to overcome the sidelobe suppression bottleneck of traditional circular arrays by significantly reducing system hardware size and cost through two-stage sequential optimization. Specifically, it includes the following steps:

[0118] Step 101: Define the configuration master and construct the configuration optimization performance function.

[0119] Reference Figure 2 This paper demonstrates a pool of candidate element positions for topology optimization. First, a master configuration template for array configuration optimization is defined. In this embodiment, based on the required detection accuracy, the template is determined to be an equally spaced circular array containing M=52 candidate element positions, with a physical radius R set to 16.5 cm. These 52 positions collectively constitute the candidate element pool for the algorithm.

[0120] Subsequently, a configuration optimization efficiency function is constructed to evaluate the robustness of arbitrary partial element activated array configurations when performing elevation scans in a specific azimuth plane (0° azimuth in this example). The performance function is specifically set as follows:

[0121] ;

[0122] in, This refers to scanning the electron beam to the elevation angle under uniform excitation. The peak sidelobe level at that time. The evaluation value of this performance function is determined by the worst PSL value that appears after scanning a series of key pitch angles from -60° to +60°. This "minimize worst value" strategy forces the optimization engine to find robust configurations that do not significantly deteriorate in sidelobe performance even at the extreme scanning angles.

[0123] Step 102: Configuration gene encoding and initial population generation.

[0124] To facilitate computer processing, this step employs a configuration gene encoding mechanism. For a master pattern with M=52, the array configuration of any randomly activated array element is encoded into a binary string of length 52, where '1' represents that the array element at that position is activated and '0' represents that it is inactive.

[0125] Subsequently, an initial population containing P=50 different binary codes is randomly generated, with each code representing a random initial array configuration with activated partial array elements, serving as the iterative starting point for the intelligent evolutionary optimization engine.

[0126] Step 103: Two-stage joint optimization

[0127] This step is the core of the method, achieving joint optimization of configuration and weights through two stages of sequential optimization.

[0128] Step 103a: First stage – Sparse configuration optimization based on discrete hybrid simulated annealing particle swarm optimization algorithm

[0129] In this embodiment, a discrete hybrid simulated annealing particle swarm optimization algorithm is used as the optimization engine to find an array configuration with the fewest active elements and the strongest sidelobe suppression performance on a template of M=52. Based on the performance function defined in step 101, the algorithm simulates the electronic scanning process at multiple pitch angles for each configuration and takes the worst PSL as the evaluation value. After this stage, a fixed set of 28-element physical layouts with good pitch scan robustness is obtained, which is the optimal array configuration.

[0130] Step 103b: Second stage – Channel-weighted optimization based on particle swarm optimization (PSO) algorithm.

[0131] Based on the 28-element physical layout determined in the first phase, the second phase of optimization is initiated. The channel excitation amplitudes of these 28 elements are used as optimization variables (a 28-dimensional real vector). Particle swarm optimization is employed for iterative optimization.

[0132] The weighted performance function in this stage is set as the worst (highest) peak sidelobe level produced when scanning a series of key pitch angles. The goal of the PSO algorithm is to find a set of channel weights that minimizes this worst PSL, thereby maximizing performance across the entire pitch scan range.

[0133] Step 104: Solidify the optimal array configuration and weighting scheme.

[0134] After both iterations terminate, the optimal individual obtained in the first stage is the optimal array configuration, defining the physical position of the array elements; the optimal particle obtained in the second stage is the optimal weight scheme, defining the excitation amplitude of the corresponding channel. Together, they constitute the final design result of this invention.

[0135] Implementation Method 2

[0136] This embodiment demonstrates the performance advantages of the N=28-element optimized configuration designed using the method described in Embodiment 1, compared to the traditional M=52-element fully filled array baseline scheme. Simulation parameters are set as follows:

[0137] Center frequency ( 40kHz;

[0138] Speed ​​of sound: 1500 m / s;

[0139] Array radius: 16.5cm;

[0140] Evolutionary performance function: The primary objective is to minimize the peak sidelobe level of the pitch scan within a specific azimuth plane.

[0141] After the optimization process described in Implementation Method 1, an optimal array configuration consisting of N=28 array elements is obtained, and its element layout is as follows: Figure 3 As shown, it illustrates the asymmetric, aperiodic topology of the N active elements ultimately selected on an M-element configuration master. Figure 3 In the diagram, 52 hollow circles represent candidate positions for the master configuration, while 28 solid circles represent the ultimately activated array elements. It can be seen that this layout exhibits an asymmetric, aperiodic topological structure.

[0142] Referring to Figures 4(a)-(d), a comparison is shown between the beam patterns of the 28-element optimal weighted configuration after two optimization stages and the traditional 52-element fully filled array at the center frequency and 0° azimuth. It can be clearly seen from the figures that:

[0143] (1) Sidelobe suppression performance: In the 0° pitch direction, the peak sidelobe level (PSL) of the reference scheme is about -7.95dB, which is the typical performance upper limit of traditional equally spaced circular arrays. However, the 28-element optimized configuration obtained in this invention can reduce its PSL to -11.91dB after the first stage of configuration optimization; after the second stage of robust weight optimization, its PSL in the 0° direction can be further suppressed to -13.6dB, and it can maintain an excellent performance of about -13dB throughout the entire ±60° pitch scanning range, successfully breaking through the bottleneck of the -8dB main-sidelobe upper limit difference of traditional circular arrays.

[0144] (2) Streamlined hardware: The present invention requires only 28 effective processing channels, compared with the baseline solution of 52 channels, which greatly reduces the hardware of the electronic system and effectively addresses the high cost problem.

[0145] (3) Performance trade-off: As a performance trade-off, the main lobe width of the present invention is slightly increased from 4.50° to 5.6°. For many application scenarios, this slight loss in resolution is completely acceptable and extremely valuable in exchange for a huge improvement in sidelobe suppression capability and a significant reduction in system cost.

[0146] Implementation Method 3

[0147] This embodiment provides a sonar receiving array device for implementing the above method. The device mainly includes:

[0148] (1) A physical array with M=52 transducer units mounted on its base constitutes a complete physical candidate pool;

[0149] (2) A multi-channel signal acquisition and processing system, which is equipped with only N=28 effective hardware processing channels.

[0150] The core technical feature of this embodiment is that the signal input terminal of the 28-channel signal acquisition and processing system is only connected to the physical array consisting of... Figure 3 The 28 transducer units determined by the optimal array configuration shown are physically connected and processed for signals. The remaining MN=24 transducer units on the physical array are in an electrically "dormant" or "inactive" state, not occupying any electronic system resources. Before entering the beamformer, the signals of these N=28 channels are each multiplied by a set of optimal channel weights generated by the second stage of the method of this invention.

[0151] The sonar device of this invention uses a single 28-channel electronic system to drive a physically complete 52-element base, achieving detection performance that surpasses that of traditional 52-channel fully configured arrays, thereby achieving significant reductions in cost, power consumption, and complexity at the electronic system level.

[0152] Implementation Method 4

[0153] This embodiment aims to visually verify the beneficial effects of the present invention in practical applications through a typical "high contrast acoustic imaging" scenario. This scenario simulates the detection of a weakly scattering target (such as a scour pit, with an echo intensity of -16dB) next to a strongly reflective target (such as a stone pier, with an echo intensity of 0dB).

[0154] Reference Figure 5 This figure illustrates the received signal level after a simulated scan using a process breakdown approach. In the figure, the green dashed line represents the "main lobe weak target reflection" (signal) from a weak target at -20°, with a peak energy of -16dB; the red area represents the "side lobe strong target reflection" (interference) from a strong target at +10°; and the blue solid line represents the final composite received signal after the two are superimposed.

[0155] The left subplot shows the detection results of a conventional equidistant array. At a scanning direction of -20°, the sidelobe reflection energy (interference) from strong targets reaches as high as -15.0 dB, stronger than the main lobe reflection energy (signal) from weak targets, which is only -16.0 dB. Therefore, the final synthesized signal (blue curve) fails to form any identifiable features at -20°; the weak target is completely overwhelmed by the interference, resulting in detection failure.

[0156] The right-hand subplot illustrates the detection results of the array after configuration optimization and channel weighting using the method of this invention. Through joint optimization of configuration and weights, the sidelobe reflection energy from strong targets was successfully suppressed to -19.3 dB in the -20° scanning direction. This interference energy is lower than the signal energy of -16.0 dB, at which point the signal energy forms a clearly discernible local signal peak of -14 dB on the interference background (the bulge of the blue curve at -20°). This peak feature can be effectively captured by subsequent signal processing algorithms, thus achieving successful differentiation between strong and weak targets.

[0157] This implementation method irrefutably demonstrates the great engineering value and practical utility of the present invention in solving the core technical problem of weak target resolution under strong interference environment by effectively suppressing sidelobe interference and significantly improving the signal-to-interference ratio.

[0158] Those skilled in the art will recognize that the embodiments described herein are intended to help the reader understand the principles of the invention, and should be understood that the scope of protection of the invention is not limited to such specific statements and embodiments. Those skilled in the art can make various other specific modifications and combinations based on the technical teachings disclosed in this invention without departing from the spirit of the invention, and these modifications and combinations are still within the scope of protection of this invention.

Claims

1. A high-performance circular array arrangement method integrating configuration design and channel weighting, characterized in that, Includes the following steps: S1. Construct a configuration optimization performance function to select the optimal array configuration from the configuration master. S2. Construct a weighted efficiency function to process the optimal array configuration and determine the optimal weights for several channels; S3. Generate a circular array based on the optimal array configuration and the optimal weights of several channels.

2. The high-performance circular array method integrating configuration design and channel weighting according to claim 1, characterized in that, S1 includes the following sub-steps: S11. Establish design goals and physical constraints; S12. Based on the design objectives and physical constraints, construct the configuration optimization efficiency function; S13. Construct the configuration gene encoding and generate the initial configuration population; S14. With minimizing or maximizing the configuration optimization efficiency function as the optimization objective, iteratively search the initial configuration population until the first preset convergence condition is met. S15. Select the configuration gene encoding that maximizes the configuration optimization efficiency function from the iterative search results, and decode it to obtain the optimal array configuration.

3. The high-performance circular array method integrating configuration design and channel weighting according to claim 2, characterized in that, In S12, the configuration optimization efficiency function The expression is: ； in, The peak sidelobe level under uniform excitation. The main lobe width under uniform excitation. To activate the array elements, This is the first performance conversion function. This is the second performance conversion function. This is the third performance conversion function. As the first weighting factor, As the second weighting factor, It is the third weighting factor.

4. The high-performance circular array method integrating configuration design and channel weighting according to claim 2, characterized in that, In S13, binary strings are used for gene encoding. Specifically, for several candidate positions on the configuration master plate, each bit represents a candidate position. 1 indicates that the position element is activated, and 0 indicates that the position element is in a dormant state.

5. The high-performance circular array method integrating configuration design and channel weighting according to claim 2, characterized in that, In S13, an initial configuration population containing several configuration gene codes is randomly generated.

6. The high-performance circular array method integrating configuration design and channel weighting according to claim 2, characterized in that, In step S14, the discrete hybrid simulated annealing particle swarm optimization algorithm is used to iteratively search the initial configuration population.

7. The high-performance circular array method integrating configuration design and channel weighting according to claim 1, characterized in that, S2 includes the following sub-steps: S21. Encode the channel excitation weights of several active array elements in the optimal array configuration into real number vectors and generate an initial weight population. S22. Construct a weighted efficiency function; S23. With minimizing or maximizing the weighted efficiency function as the optimization objective, iteratively search the initial weight population until the second preset convergence condition is met. S24. Determine the optimal weights for several channels based on the weights corresponding to the iteration results that satisfy the second preset convergence condition.

8. The high-performance circular array method integrating configuration design and channel weighting according to claim 7, characterized in that, In S22, the weighted efficiency function The expression is: ； in, The peak sidelobe level when using a set of non-uniform channel weights to be optimized. The first preset pitch scan angle within a specific azimuth plane. The second preset pitch scan angle within a specific azimuth plane. The first in a specific azimuth plane A preset pitch scanning angle To obtain the maximum value.

9. The high-performance circular array method integrating configuration design and channel weighting according to claim 7, characterized in that, In step S23, an iterative search is performed using a particle swarm optimization algorithm.

10. A high-performance circular array integrating configuration design and channel weighting, characterized in that, This includes physical arrays and signal acquisition and processing systems; The physical array is used to install several equally spaced transducer units to form a physical candidate array element pool; The signal acquisition and processing system is used to determine the optimal array configuration and the optimal weights of several channels to generate a circular array.

Images