Sound source localization method and system based on improved adaptive beam forming

Abstract

The invention relates to the technical field of signal processing, in particular to a sound source localization method and system based on improved adaptive beam forming. A microphone array is used for collecting sound source signals, and a signal receiving model is established; designing an improved adaptive filtering beam forming algorithm, and optimizing a cost function; solving an optimal weighting vector through a minimization cost function, and introducing an adaptive updating mechanism to dynamically adjust the weighting vector; selecting a reference microphone, calculating time difference of arrival and distance difference of other microphones, and establishing a distance difference equation from the sound source to the receiver; and solving a sound source coordinate and calculating the distance between the center of the microphone array and the sound source. According to the invention, through innovative algorithm design and a self-adaptive mechanism, the precision and stability of sound source localization are significantly improved, and especially, the method shows strong anti-interference capability and adaptability in a complex noise environment. The method can effectively improve the equipment maintenance efficiency in the substation equipment noise monitoring, reduces the fault risk, and provides reliable guarantee for the safe operation of a power grid.

Description

Technical Field

[0001] The invention relates to the technical field of signal processing, and in particular to a sound source localization method and system based on improved adaptive beamforming. Background Art

[0002] At present, beamforming algorithms are widely used in fields such as wireless communications, radar systems, and sonar to enhance the reception of target signals and suppress interference and noise. However, with the dynamic changes in the environment and the increase in signal complexity, traditional beamforming algorithms have certain limitations in dealing with rapidly changing channel conditions. Especially in noise detection of power grid equipment, accurate sound source positioning is of great significance for equipment fault diagnosis and preventive maintenance. However, it is difficult for traditional algorithms to accurately locate noise sources in complex noise environments, which affects the safe operation of power grid equipment. In order to solve these problems, the present invention proposes an improved adaptive beamforming algorithm that can achieve more accurate noise source positioning in complex environments. At present, commonly used beamforming methods include minimum mean square error (MMSE), minimum variance distortionless response (MVDR), and maximum signal-to-noise ratio (SNR) methods. These methods can achieve good performance in ideal environments, but their performance may not be satisfactory in dynamic environments or complex interference conditions. In particular, when environmental noise and interference signals change over time, the performance of traditional algorithms will significantly decrease, resulting in a decrease in beam directivity and signal quality.

[0003] To address this problem, the present invention proposes an adaptive beamforming algorithm that combines channel characteristics and environmental noise factors. By introducing adjustment factors related to channel characteristics, the algorithm can adapt to environmental changes more flexibly and enhance the ability to capture target signals. In addition, an adjustment coefficient related to noise power is introduced into the cost function, so that the algorithm has higher robustness in complex noise environments. In order to further improve the adaptability of the algorithm, the present invention designs a dynamically adjusted adaptive update mechanism. By introducing dynamic step factors and adjustment factors, the mechanism can update the weighted vector in real time according to channel conditions and environmental changes, thereby maintaining the optimal performance of beamforming. Accurate noise source positioning is of great significance in fault detection and status monitoring of power grid equipment. Through the improved algorithm of the present invention, the noise source in the power grid equipment can be more accurately identified and located, potential faults can be discovered in advance, the risk of power outages can be reduced, and the stability and safety of the power grid can be improved. By combining environmental characteristics, dynamic adjustment mechanisms and advanced optimization technologies, the improved algorithm of the present invention shows higher robustness and adaptability under complex and changeable channel conditions, providing a more effective solution for noise detection and fault positioning of power grid equipment. Summary of the invention

[0004] In view of the problems existing in the prior art, the present invention is proposed.

[0005] Therefore, the problem to be solved by the present invention is that with the dynamic changes in the environment and the increase in signal complexity, traditional beamforming algorithms have certain limitations in dealing with rapidly changing channel conditions. Especially in the noise detection of power grid equipment, accurate sound source positioning is of great significance for equipment fault diagnosis and preventive maintenance. Traditional algorithms are difficult to accurately locate noise sources in complex noise environments, affecting the safe operation of power grid equipment.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, an embodiment of the present invention provides a sound source localization method based on improved adaptive beamforming, which includes using a microphone array to collect sound source signals, establishing a received signal model, and designing an improved adaptive filtering beamforming algorithm to optimize a cost function;

[0008] The optimal weight vector is solved by minimizing the cost function, and an adaptive update mechanism is introduced to dynamically adjust the weight vector;

[0009] Select a reference microphone, calculate the arrival time difference and distance difference of other microphones, and establish the distance difference equation from the sound source to the receiver to solve the coordinates of the sound source;

[0010] Calculate the distance between the center of the microphone array and the sound source to achieve accurate sound source positioning.

[0011] As a preferred solution of the sound source localization method based on improved adaptive beamforming described in the present invention, wherein: the use of a microphone array to collect sound source signals includes using a microphone array to measure independent sound sources, the number of microphone arrays is m, numbered 1, 2, 3...m, the distance between the sound source and the center of the microphone array is R, each microphone unit receives a signal from space, and the received signal model is established as:

[0012] r=Mt+j+u

[0013] Where r is the received signal vector, M is the gain matrix of the array, which represents the gain and phase offset of the target signal when it passes through the antenna array, t is the target signal, j is the interference signal, and u is the noise.

[0014] As a preferred solution of the sound source localization method based on improved adaptive beamforming of the present invention, wherein: the improved adaptive filtering beamforming algorithm is designed, and the optimized cost function includes that the improved adaptive filtering beamforming algorithm continuously adjusts the weighting vector p to minimize the loss function, and the cost function is in the form of:

[0015] D(p)=E[p H rt 2 ]+λp 2

[0016] Among them, E[〃] represents the expected value, p H is the weighted output signal, and λ is the regularization parameter used to prevent overfitting.

[0017] As a preferred solution of the sound source localization method based on improved adaptive beamforming of the present invention, wherein: solving the optimal weighted vector by minimizing the cost function includes:

[0018] To minimize the cost function, we differentiate p and set the derivative to zero to find the optimal solution:

[0019]

[0020] The optimal weight vector is:

[0021] p opt =(C r +λI) -1 C tr γ

[0022] Among them, C r =E[rr H ] is the covariance matrix of the received signal, C tr =E[tr H ] is the cross-covariance matrix between the target signal and the received signal, I is the unit matrix, and γ is the adjustment coefficient.

[0023] As a preferred solution of the sound source localization method based on improved adaptive beamforming of the present invention, the introduction of the adaptive update mechanism and the dynamic adjustment of the weighting vector include the following steps:

[0024] In a dynamic environment, an adaptive update mechanism is introduced to continuously adjust the weight vector and introduce a new variable δ to adjust the step size factor α(n):

[0025] p(n+1)=p(n)+δ·α(n)·r(n)·(γt(n)-p H (n)r(n))

[0026] Among them, δ can be dynamically adjusted according to the change of the channel or the change of the environmental noise.

[0027] As a preferred solution of the sound source localization method based on improved adaptive beamforming of the present invention, wherein: the selection of the reference microphone and the calculation of the arrival time difference and distance difference of other microphones include:

[0028] Select microphone 1 as the reference receiver and obtain the weight vector p opt Calculate the arrival time difference Δt of other receivers relative to the reference receiveri =t i -t1, for each microphone receiver calculate the distance difference Δd using the sound speed c ij =c·(Δt i -Δt j ).

[0029] As a preferred solution of the sound source localization method based on improved adaptive beamforming of the present invention, wherein: the distance between the center of the microphone array and the sound source is calculated to achieve accurate sound source localization, including:

[0030] For each pair of microphone receivers (i, j), an equation is created to represent the difference in distance from the sound source to the receiver:

[0031]

[0032] By solving the sound source coordinates (x, y), we can calculate the distance R between the center of the microphone array and the sound source.

[0033] In a second aspect, an embodiment of the present invention provides a sound source localization system based on improved adaptive beamforming, which includes a collection module, which uses a microphone array to collect sound source signals, establishes a receiving signal model, and designs an improved adaptive filtering beamforming algorithm to optimize a cost function;

[0034] The adjustment module solves the optimal weight vector by minimizing the cost function, introduces an adaptive update mechanism, and dynamically adjusts the weight vector;

[0035] The solution module selects the reference microphone, calculates the arrival time difference and distance difference of other microphones, and establishes the distance difference equation from the sound source to the receiver to solve the coordinates of the sound source;

[0036] The positioning module calculates the distance between the center of the microphone array and the sound source to achieve accurate sound source positioning.

[0037] In a third aspect, an embodiment of the present invention provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program instructions are executed by the processor, the steps of the sound source localization method based on improved adaptive beamforming as described in the first aspect of the present invention are implemented.

[0038] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program instructions are executed by a processor, the steps of the sound source localization method based on improved adaptive beamforming as described in the first aspect of the present invention are implemented.

[0039] The beneficial effects of the present invention are as follows: the present invention proposes an improved beamforming algorithm, which optimizes the cost function by introducing factors related to channel characteristics and noise power adjustment coefficients. The algorithm optimizes the weighting vector by minimizing the improved cost function, and combined with an adaptive update mechanism, can adjust the weighting vector in real time in a dynamic environment. Ultimately, the obtained optimal weighting vector provides a higher precision detection capability for the target signal, while effectively suppressing environmental noise and interference. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0041] Figure 1 A flow chart of a sound source localization method based on improved adaptive beamforming;

[0042] Figure 2 A diagram of a computer device for a sound source localization method based on improved adaptive beamforming. DETAILED DESCRIPTION

[0043] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0044] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0045] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor is it a separate or selective embodiment that is mutually exclusive with other embodiments.

[0046] Example 1

[0047] Reference Figure 1-2 , which is the first embodiment of the present invention, and provides a sound source localization method based on improved adaptive beamforming, comprising:

[0048] S100: Use a microphone array to collect sound source signals, establish a receiving signal model, and design an improved adaptive filtering beamforming algorithm to optimize the cost function;

[0049] S101: Using a microphone array to collect sound source signals includes using a microphone array to measure independent sound sources. The number of microphone arrays is m, numbered 1, 2, 3...m, the distance between the sound source and the center of the microphone array is R, and each microphone unit receives a signal from space. Due to the limited speed of sound wave propagation, there will be a time delay between the signals received by different microphones. This delay depends on the relative position relationship between the microphone and the sound source. The signal propagates in space through sound waves and is transmitted to each microphone unit. For the i-th microphone unit, the received signal model is established as follows:

[0050] r=Mt+j+u

[0051] Where r is the received signal vector, M is the gain matrix of the array, which represents the gain and phase offset of the target signal when it passes through the antenna array, t is the target signal, j is the interference signal, and u is the noise.

[0052] Preferably, a microphone array is used to measure an independent sound source. The number of microphone arrays is 128, numbered 1, 2, 3...128. The distances between the sound source and the center of the microphone array are 1m, 2m, and 5m for independent testing. Each microphone unit receives a signal from the space, and each microphone unit can receive a signal from the space. In order to test sound sources at different distances, we set the sound source at 1m, 2m, and 5m from the center of the microphone array for independent measurements. This multiple measurement method can help evaluate the beamforming performance and accuracy of the microphone array at different distances. During each measurement, all microphone units receive signals from the sound source and record these data for subsequent processing.

[0053] S102: And design an improved adaptive filtering beamforming algorithm, the optimization cost function includes, the improved adaptive filtering beamforming algorithm minimizes the loss function by continuously adjusting the weighting vector p. The core idea of ​​this algorithm is to dynamically adjust the weight of the filter according to the error between the received signal and the expected output to adapt to the complex and changing signal environment. In practical applications, the signal environment is usually affected by factors such as multipath propagation, interference sources and noise. Therefore, by adaptively adjusting the weighting vector, interference and noise can be effectively suppressed, thereby improving the quality of the signal. The form of the cost function is:

[0054] D(p)=E[p H rt 2 ]+λp 2

[0055] Among them, E[〃] represents the expected value, p H is the weighted output signal, and λ is the regularization parameter used to prevent overfitting.

[0056] Preferably, the weighting vector p is continuously adjusted by an improved adaptive filtering beamforming algorithm to minimize the loss function. The regularization parameter λ is determined to be 0.81. After acquiring the raw data of the microphone array, the signals are then processed by an improved adaptive filtering beamforming algorithm. The core of the algorithm is to minimize the loss function in the process of dynamically adjusting the weighting vector p, thereby optimizing the directivity and performance of the beam. To this end, the regularization parameter is determined to be 0.81 to control the complexity of the algorithm and avoid overfitting. Through this step, the algorithm can ensure that the adjustment of the weighting vector is more robust and more adaptable, while reducing the impact of noise and interference on signal processing.

[0057] S200: solving the optimal weight vector by minimizing the cost function, introducing an adaptive update mechanism, and dynamically adjusting the weight vector;

[0058] S201: Solving the optimal weighted vector by minimizing the cost function includes:

[0059] To minimize the cost function, we differentiate p and set the derivative to zero to find the optimal solution:

[0060]

[0061] The optimal weight vector is:

[0062] p opt =(C r +λI) -1 C tr γ

[0063] Among them, C r =E[rr H ] is the covariance matrix of the received signal, C tr =E[tr H ] is the cross-covariance matrix between the target signal and the received signal, I is the unit matrix, and γ is the adjustment coefficient.

[0064] Preferably, the cost function is minimized by taking the derivative of p and setting the derivative to zero to obtain the optimal solution p opt , where the adjustment coefficient γ is 3, which is used to control the adjustment amplitude of the weights during the optimization process. Through this step, the system can find the optimal weight vector more quickly and accurately, thereby improving the convergence of the algorithm.

[0065] S202: Introducing an adaptive update mechanism to dynamically adjust the weight vector includes the following steps:

[0066] In a dynamic environment, an adaptive update mechanism is introduced to continuously adjust the weight vector and introduce a new variable δ to adjust the step size factor α(n):

[0067] p(n+1)=p(n)+δ·α(n)·r(n)·(γt(n)-p H (n)r(n))

[0068] Among them, δ can be dynamically adjusted according to the change of the channel or the change of the environmental noise.

[0069] Preferably, in order to maintain high performance in a dynamic environment, the system introduces an adaptive update mechanism to continuously adjust the weight vector. This enables the beamforming algorithm to adapt to changes in the environment in real time, such as the movement of the sound source position or changes in external interference. To this end, a new variable is introduced to adjust the step size factor, the value range of which is [0.01,1], and the weight vector is continuously optimized through 1500 iterations. In each iteration, the step size factor affects the adjustment rate of the weight. A larger one will make the weight update faster, but may bring instability, while a smaller one will make the weight update smoother but slower. Through these iterations, the system finally obtains an optimal weight vector suitable for the current environment. In a dynamic environment, the weight vector is continuously adjusted by introducing an adaptive update mechanism. In a dynamic environment, the weight vector is continuously adjusted by introducing an adaptive update mechanism. A new variable δ is introduced to adjust the step size factor α(n). The variable δ range is [0.01,1], and 1500 iterations are performed to finally obtain the weight vector p op t .

[0070] S300: Select a reference microphone, calculate the arrival time difference and distance difference of other microphones, and establish a distance difference equation from the sound source to the receiver to solve the sound source coordinates;

[0071] S301: Select a reference microphone and calculate the arrival time difference and distance difference of other microphones, including:

[0072] Select microphone 1 as the reference receiver and obtain the weight vector p opt Calculate the arrival time difference Δt of other receivers relative to the reference receiver i =t i - t1, for each microphone receiver, calculate the distance difference Δd using the sound speed c = 340 m / s ij =c·(Δt i -Δt j ).

[0073] S400: Calculates the distance between the center of the microphone array and the sound source to achieve accurate sound source positioning.

[0074] S401: Calculate the distance between the center of the microphone array and the sound source to achieve accurate sound source positioning, including:

[0075] For each pair of microphone receivers (i, j), an equation is established to represent the distance difference from the sound source to the receiver. This equation is used to describe the distance difference relationship between the sound source and the receiver. By modeling the distance difference between multiple microphone pairs, a system equation group can be formed:

[0076]

[0077] By solving the sound source coordinates (x, y), we can calculate the distance R between the center of the microphone array and the sound source.

[0078] The R calculated for different sound source distances is shown in the following table. Each different distance is measured 10 times.

[0079] Table 1 Results of sound source localization distance at different distances

[0080]

[0081]

[0082] Furthermore, this embodiment also provides a sound source localization system based on improved adaptive beamforming, comprising:

[0083] The acquisition module uses a microphone array to collect sound source signals, establishes a receiving signal model, and designs an improved adaptive filtering beamforming algorithm to optimize the cost function;

[0084] The adjustment module solves the optimal weight vector by minimizing the cost function, introduces an adaptive update mechanism, and dynamically adjusts the weight vector;

[0085] The solution module selects the reference microphone, calculates the arrival time difference and distance difference of other microphones, and establishes the distance difference equation from the sound source to the receiver to solve the coordinates of the sound source;

[0086] The positioning module calculates the distance between the center of the microphone array and the sound source to achieve accurate sound source positioning.

[0087] This embodiment also provides a computer device, which is applicable to the case of a sound source localization method based on improved adaptive beamforming, and includes a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the sound source localization method based on improved adaptive beamforming as proposed in the above embodiment.

[0088] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0089] This embodiment further provides a storage medium on which a computer program is stored. When the program is executed by a processor, the sound source localization method based on improved adaptive beamforming as proposed in the above embodiment is implemented.

[0090] In summary, by using a microphone array to collect sound source signals and establish a receiving signal model, we can accurately capture the complex sound field environment. This step not only takes into account the target signal, but also includes interference signals and noise, so that subsequent processing can more accurately separate the target sound source, thereby improving the ability to locate the noise source of equipment in a complex power grid environment.

[0091] By designing an improved adaptive filtering beamforming algorithm and optimizing the cost function, the traditional beamforming algorithm is innovated. The regularization parameter λ is introduced to effectively prevent the overfitting problem, making the algorithm more robust in the face of different noise environments, which is crucial for the accuracy of noise monitoring of power grid equipment.

[0092] The optimal weighted vector is solved by minimizing the cost function, and an adaptive update mechanism is introduced to achieve dynamic adjustment of the weighted vector. This innovation enables the system to adapt to changing channel conditions and environmental noise in real time, greatly improving the accuracy of locating sound sources in complex and dynamically changing power grid environments.

[0093] By selecting a reference microphone, calculating the arrival time difference and distance difference of other microphones, and establishing the distance difference equation from the sound source to the receiver, the sound source can be accurately located. This method not only improves the positioning accuracy, but also reduces the computational complexity, allowing the system to respond quickly in actual power grid equipment monitoring.

[0094] By calculating the distance between the center of the microphone array and the sound source, accurate sound source positioning is finally achieved. This step combines the results of all previous processing to provide an intuitive and accurate sound source location information, providing a reliable basis for fault diagnosis and preventive maintenance of power grid equipment.

[0095] In summary, this solution effectively solves the problem of performance degradation of traditional beamforming algorithms in dynamic environments through a series of innovative steps. It not only improves the accuracy of sound source localization, but also enhances the system's adaptability in complex noise environments. This method is particularly suitable for noise monitoring of power grid equipment, which can detect potential equipment failures in a timely manner and improve the safety and reliability of the power grid. In addition, the adaptive characteristics of this method enable it to maintain high performance under different environmental conditions, which is an advantage that traditional methods do not have.

[0096] This sound source localization method based on improved adaptive beamforming is not only innovative in technology, but also shows significant economic benefits in practical applications. By improving the accuracy and timeliness of equipment fault diagnosis, the cost of power grid maintenance can be greatly reduced, and the risk of power outages caused by equipment failures can be reduced, thereby improving the overall power grid operation efficiency. This unexpected technical effect makes the present invention have important application prospects in the field of power system management and maintenance.

[0097] Example 2

[0098] Reference Figure 1 - Figure 2 , which is the second embodiment of the present invention, and this embodiment provides a sound source localization method based on improved adaptive beamforming. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments.

[0099] To verify the effectiveness of the method of the present invention, a field test was conducted in a 500kV substation. A running main transformer was selected as the noise source, and a circular array consisting of 64 microphones was used for sound collection. The microphone array had a diameter of 2 meters and was placed 10 meters away from the transformer. The tests were conducted under three different environmental conditions: normal operation (ambient noise of about 50dB), high load operation (ambient noise of about 65dB), and simulated fault state (ambient noise of about 80dB).

[0100] First, the microphone array was calibrated to ensure that the sensitivity of each microphone was consistent. The sampling rate was set to 48kHz to capture the full spectrum characteristics of the transformer noise. Five minutes of continuous recording were performed for each condition, and the data was split into 10-second segments for processing.

[0101] Next, the collected signal is subjected to a short-time Fourier transform (STFT) using a Hanning window, a frame length of 1024 points, and a frame shift of 256 points. Then, the improved adaptive beamforming algorithm of the present invention is applied. When optimizing the cost function, the regularization parameter λ is set to 0.01 to balance noise suppression and signal distortion.

[0102] When solving the optimal weighted vector, the adjustment coefficient α is initially set to 0.5 and is dynamically adjusted according to the change of the signal-to-noise ratio during the processing. The step factor μ in the adaptive update mechanism is initially set to 0.1, and the variable β is set in the range of [0.05, 0.95] and updated every 50 frames.

[0103] In order to improve computing efficiency, GPU acceleration technology was adopted, and NVIDIA Tesla V100 was used for parallel computing. The average delay of the entire processing process was controlled within 10ms, meeting the requirements of real-time processing.

[0104] Finally, the precise location of the sound source is obtained by calculating the distance between the center of the microphone array and the sound source. In order to compare the performance of this method, the traditional delay and sum beamforming (DS) method and the minimum variance distortion-free response (MVDR) method were also used for testing. Each method was tested 10 times under each environmental condition, and the results are shown in the following table:

[0105]

[0106]

[0107] By analyzing the table data, we can get:

[0108] Positioning accuracy: This method shows significant advantages under all environmental conditions. Under normal operating conditions, the average positioning error of this method is only 12.4cm, which is 72.6% higher than the traditional DS method and 56.6% higher than the MVDR method. This high-precision positioning capability means that noise sources can be identified more accurately in practical applications, resulting in more effective equipment maintenance and fault diagnosis.

[0109] Anti-interference ability: As the environmental noise increases, the positioning errors of all methods increase, but the performance of this method decreases the least. From normal to fault state, the error of the DS method increases by 96.7%, the MVDR increases by 104.2%, and the error of this method increases by only 79.8%. This shows that this method has stronger robustness in complex noise environments, which is particularly important for complex noise environments such as substations.

[0110] Stability: This method exhibits the smallest standard deviation under various conditions, indicating that its results are more stable and reliable. For example, under fault conditions, the standard deviation of this method is 4.2cm, while the DS and MVDR methods are 18.7cm and 12.3cm, respectively. This stability is critical for long-term monitoring and trend analysis, which can better predict potential problems with equipment.

[0111] Real-time performance: Although the algorithm complexity of this method is higher than that of the DS method, its processing time is still maintained at about 10ms through GPU acceleration and optimization, which fully meets the requirements of real-time processing. In contrast, although the MVDR method performs better than DS, its processing time is significantly higher than this method, which may cause delays in real-time applications.

[0112] Adaptability: The performance of this method varies minimally under different noise environments, thanks to its adaptive update mechanism. By dynamically adjusting the step size factor and regularization parameters, this method can quickly adapt to environmental changes and maintain high performance. This adaptability is particularly important in scenarios such as substations where environmental conditions change frequently.

[0113] Cost-effectiveness: Although not directly reflected in the table, considering the high accuracy and stability of this method, it can significantly reduce false positives and false negatives, thereby reducing unnecessary equipment inspection and maintenance costs. At the same time, its high-precision positioning capability can find the problem faster, reduce fault diagnosis time, and improve maintenance efficiency.

[0114] In summary, the method of this technical solution is significantly superior to traditional methods in terms of positioning accuracy, anti-interference ability, stability and adaptability. It not only overcomes the problem of performance degradation of existing technologies in complex noise environments, this performance improvement is not only reflected in the data, but more importantly, it can bring significant economic benefits and safety improvements in practical applications. By more accurately and stably locating noise sources, the maintenance efficiency of substation equipment can be greatly improved, the risk of failure can be reduced, and the life of equipment can be extended, thereby bringing huge economic benefits to power grid operations. This innovative method that combines deep learning, signal processing and adaptive algorithms has opened up a new path for the intelligent monitoring and maintenance of power systems and has broad application prospects.

[0115] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A sound source localization method based on improved adaptive beamforming, characterized in that: include, Use microphone array to collect sound source signals, establish receiving signal model, design improved adaptive filtering beamforming algorithm, and optimize cost function; The optimal weight vector is solved by minimizing the cost function, and an adaptive update mechanism is introduced to dynamically adjust the weight vector; Select a reference microphone, calculate the arrival time difference and distance difference of other microphones, and establish the distance difference equation from the sound source to the receiver to solve the coordinates of the sound source; Calculate the distance between the center of the microphone array and the sound source to achieve accurate sound source positioning.

2. The sound source localization method based on improved adaptive beamforming according to claim 1, characterized in that: The method of collecting the sound source signal using the microphone array includes measuring the independent sound source using the microphone array, the number of the microphone array is m, numbered 1, 2, 3 ... m, the distance between the sound source and the center of the microphone array is R, each microphone unit receives a signal from the space, and the received signal model is established as: r=Mt+j+u Where r is the received signal vector, M is the gain matrix of the array, which represents the gain and phase offset of the target signal when it passes through the antenna array, t is the target signal, j is the interference signal, and u is the noise.

3. The sound source localization method based on improved adaptive beamforming according to claim 2, characterized in that: The improved adaptive filtering beamforming algorithm is designed and the cost function is optimized, including that the improved adaptive filtering beamforming algorithm continuously adjusts the weight vector p to minimize the loss function, and the cost function is in the form of: D(p)=E[p H r-t 2 ]+λp 2 Among them, E[·] represents the expected value, p H is the weighted output signal, and λ is the regularization parameter used to prevent overfitting.

4. The sound source localization method based on improved adaptive beamforming according to claim 3, characterized in that: The method of solving the optimal weight vector by minimizing the cost function includes: To minimize the cost function, we differentiate p and set the derivative to zero to find the optimal solution: The optimal weight vector is: p opt =(C r +λI) -1 C tr γ Among them, C r =E[rr H ] is the covariance matrix of the received signal, C tr =E[tr H ] is the cross-covariance matrix between the target signal and the received signal, I is the unit matrix, and γ is the adjustment coefficient.

5. The sound source localization method based on improved adaptive beamforming according to claim 4, characterized in that: The method of introducing an adaptive update mechanism and dynamically adjusting the weight vector comprises the following steps: In a dynamic environment, an adaptive update mechanism is introduced to continuously adjust the weight vector and introduce a new variable δ to adjust the step size factor α(n): p(n+1)=p(n)+δ·α(n)·r(n)·(γt(n)-p H (n)r(n)) Among them, δ can be dynamically adjusted according to the change of the channel or the change of the environmental noise.

6. The sound source localization method based on improved adaptive beamforming according to claim 5, characterized in that: The selecting of the reference microphone and calculating the arrival time difference and distance difference of other microphones include: Select microphone 1 as the reference receiver and obtain the weight vector p opt Calculate the arrival time difference Δt of other receivers relative to the reference receiver i =t i -t1, for each microphone receiver calculate the distance difference Δd using the sound speed c ij =c·(Δt i -Δt j ).

7. The sound source localization method based on improved adaptive beamforming according to claim 6, characterized in that: The calculating the distance between the center of the microphone array and the sound source to achieve accurate sound source positioning includes: For each pair of microphone receivers (i, j), an equation is created to represent the difference in distance from the sound source to the receiver: By solving the sound source coordinates (x, y), we can calculate the distance R between the center of the microphone array and the sound source.

8. A sound source localization system based on improved adaptive beamforming, based on the sound source localization method based on improved adaptive beamforming according to any one of claims 1 to 7, characterized in that: Also includes, The acquisition module uses a microphone array to collect sound source signals, establishes a receiving signal model, and designs an improved adaptive filtering beamforming algorithm to optimize the cost function; The adjustment module solves the optimal weight vector by minimizing the cost function, introduces an adaptive update mechanism, and dynamically adjusts the weight vector; The solution module selects the reference microphone, calculates the arrival time difference and distance difference of other microphones, and establishes the distance difference equation from the sound source to the receiver to solve the coordinates of the sound source; The positioning module calculates the distance between the center of the microphone array and the sound source to achieve accurate sound source positioning.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the sound source localization method based on improved adaptive beamforming according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the sound source localization method based on improved adaptive beamforming according to any one of claims 1 to 7 are implemented.

Images