A method and device for extracting leakage noise of a water supply network

Abstract

The embodiment of the present specification discloses a leakage noise extraction method and device for a water supply network. The method comprises acquiring a collected sound signal of the water supply network. The method further comprises determining an expected signal of the leakage noise based on the collected sound signal of the water supply network and a main frequency band thereof, wherein the main frequency band is determined by a main distribution range of cumulative energy calculated according to a power spectral density of the collected sound signal. In addition, the method further comprises extracting the leakage noise signal of the water supply network from the collected sound signal by using an adaptive filtering algorithm with the expected signal as a guide signal. The embodiment of the present specification calculates the main distribution range of the cumulative energy by using the power spectral density, so as to determine the expected signal conforming to the characteristics of the leakage noise of the water supply network. Then, the adaptive filtering algorithm is used to filter out the environmental noise in the collected sound signal with the relatively accurate expected signal as the guide signal, so as to extract the relatively accurate leakage noise, thereby improving the accuracy of subsequent leakage identification and positioning.

Description

TECHNICAL FIELD

[0001] One or more embodiments of the present specification relate to the technical field of leak noise filtering of water supply pipe network, in particular to a leak noise extraction method and device of water supply pipe network. BACKGROUND

[0002] Urban water supply pipeline is the infrastructure to maintain the normal operation of the city, and its safety and reliability is directly related to the life of residents, industrial production and social public safety. However, with the development of urbanization, factors such as pipeline aging, poor pipe material, construction quality problems and external environmental erosion have led to increasingly serious pipeline leakage problems. Usually, noise is collected on the ground, pipeline wall and pipeline water using leak detection instruments, noise recorders, correlation detectors and other equipment.

[0003] Before identifying and locating the leak noise, it is necessary to remove as much environmental noise as possible and retain the leak noise. However, since the pipeline leak noise is a wideband random signal, the existing filtering scheme cannot effectively filter out the environmental noise to obtain a relatively accurate leak noise. SUMMARY

[0004] One or more embodiments of the present specification describe a leak noise extraction method and device of water supply pipe network.

[0005] In a first aspect of the present specification, a leak noise extraction method of water supply pipe network is provided. The method includes obtaining a collected sound signal of the water supply pipe network. The method further includes determining an expected signal of the leak noise based on the collected sound signal of the water supply pipe network and a main frequency band thereof, the main frequency band being determined from a main distribution range of cumulative energy calculated according to a power spectral density of the collected sound signal. In addition, the method further includes extracting a leak noise signal of the water supply pipe network from the collected sound signal by an adaptive filtering algorithm with the expected signal as a guide signal.

[0006] In a second aspect of the present specification, a leak noise extraction device of water supply pipe network is provided. The device includes a collected signal obtaining module configured to obtain a collected sound signal of the water supply pipe network. The device further includes an expected signal determining module configured to determine an expected signal of the leak noise based on the collected sound signal of the water supply pipe network and a main frequency band thereof, the main frequency band being determined from a main distribution range of cumulative energy calculated according to a power spectral density of the collected sound signal. In addition, the device further includes a leak noise extracting module configured to extract a leak noise signal of the water supply pipe network from the collected sound signal by an adaptive filtering algorithm with the expected signal as a guide signal.

[0007] In a third aspect of the present specification, an electronic device is provided. The electronic device includes a processor and a memory connected to the processor. The memory is configured to store executable program code. The processor is configured to execute a program corresponding to the executable program code by reading the executable program code stored in the memory, to perform the method described above.

[0008] In a fourth aspect of the present specification, a computer readable storage medium is provided, and the computer readable storage medium stores a computer program. The computer program is executed by a processor to implement the method described above.

[0009] It should be understood that the content described in the summary section is not intended to limit or define key or important features of the embodiments of the present disclosure, nor is it used to limit the scope of the present disclosure. Other features of the present disclosure will become apparent from the following description. BRIEF DESCRIPTION OF DRAWINGS

[0010] In order to more clearly illustrate the technical solutions in the embodiments of the present specification, the drawings required to be used in the embodiments will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the present specification, and other drawings can also be obtained by those skilled in the art without any creative effort based on these drawings.

[0011] Figure 1 A schematic diagram showing an example environment in which a plurality of embodiments of the present specification can be implemented is shown;

[0012] Figure 2 A flowchart of a leakage noise extraction method of a water supply network according to some embodiments of the present specification is shown;

[0013] Figure 3 A principle block diagram of a leakage noise extraction process of a water supply network according to some embodiments of the present specification is shown;

[0014] Figure 4a A time-domain image of a collected acoustic signal under a water supply network simulation test according to some embodiments of the present specification is shown;

[0015] Figure 4b A frequency spectrum image of a collected acoustic signal under a water supply network simulation test according to some embodiments of the present specification is shown;

[0016] Figure 4c A time-frequency image of a collected acoustic signal under a water supply network simulation test according to some embodiments of the present specification is shown;

[0017] Figure 4d A distribution image of a power spectral density of a collected acoustic signal under a water supply network simulation test according to some embodiments of the present specification is shown;

[0018] Figure 4e A distribution image of the accumulated energy of the collected acoustic signals at different frequencies under the water supply network simulation test of some embodiments of the present specification is shown;

[0019] Figure 4f A time-frequency image of the expected signal under the water supply network simulation test of some embodiments of the present specification is shown;

[0020] Figure 4g A time-frequency image of the leakage noise signal under the water supply network simulation test of some embodiments of the present specification is shown;

[0021] Figure 5 An example block diagram of a leakage noise extraction device of a water supply network of some embodiments of the present specification is shown;

[0022] Figure 6 A structural schematic diagram of an electronic device of some embodiments of the present specification is shown. DETAILED DESCRIPTION

[0023] The technical solutions in the embodiments of the present specification will be described clearly and completely below in combination with the drawings in the embodiments of the present specification.

[0024] The terms "first", "second", "third" and the like in the specification and claims of the present specification and the above drawings are used to distinguish different objects, and are not used to describe a specific order. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but can optionally include steps or units that are not listed, or can optionally include other steps or units inherent to the process, method, product or device.

[0025] As described above, human activities can produce more environmental noise, which seriously interferes with the accurate judgment of leakage noise. Therefore, before noise recognition and positioning, it is necessary to remove as much environmental noise as possible and retain leakage noise.

[0026] Pipeline leakage noise is a wideband random signal, whose energy and frequency vary irregularly within a certain range. In addition to the basic band-pass filter, researchers also use wavelet denoising (WD), empirical mode decomposition (EMD), singular value decomposition (SVD) and other noise suppression methods. Although WD, EMD and SVD can process random noise, they all need to select appropriate parameters to achieve better filtering effect. Among them, WD needs to select appropriate wavelet basis function and threshold, and EMD and SVD need prior information of noise IMF (Intrinsic Mode Function), so the application in different engineering faces difficulties.

[0027] Adaptive filter is an important modern signal processing method, which can automatically adjust the filter coefficients according to the input signal, and better solve the problems of parameter selection and prior information faced by other signal processing methods. The least mean squares (LMS) algorithm is an adaptive filtering algorithm based on the least mean square error criterion, but the LMS algorithm needs to calculate the error signal through the expected signal, and it is usually difficult to directly obtain the expected signal of pipeline leakage noise in actual engineering. This also leads to the LMS algorithm, although it has been applied in time delay estimation of leakage point positioning, but it is currently difficult to be applied in extraction of pipeline leakage noise.

[0028] Therefore, an embodiment of the present specification proposes a leakage noise extraction scheme for a water supply network. In the embodiment of the present specification, the power spectral density (PSD) is used to determine the main distribution range (i.e. the main frequency band referred to herein) of the cumulative energy in the sound signal collected from the water supply network (i.e. the collected sound signal referred to herein), and according to the main distribution range of the cumulative energy, the expected signal of the leakage noise is determined. Then, using the adaptive filtering algorithm, the relatively accurate expected signal is used as the guide signal, and the relatively accurate leakage noise signal can be extracted from the collected sound signal.

[0029] In this way, by calculating the main distribution range of accumulated energy using power spectral density, the main frequency range of leakage noise energy distribution can be determined. Using this main frequency range, a desired signal close to the actual leakage noise can be obtained from the acquired acoustic signal, solving the problem of directly obtaining the desired signal of leakage noise due to its randomness. Furthermore, an adaptive filtering algorithm addresses the problems faced by other signal processing methods in parameter selection and prior information. In particular, by using a relatively accurate desired signal of leakage noise as a guiding signal, the adaptive filtering algorithm can filter out environmental noise from the acquired acoustic signal, extracting a more accurate leakage noise and thus improving the accuracy of subsequent leakage identification and localization.

[0030] Figure 1 Example environment 100 in which several embodiments of this specification may be implemented is shown. For example... Figure 1 As shown, environment 100 includes a water supply network located below ground 104 (taking water pipe 106 as an example). A sound signal acquisition device 108 is typically installed against the wall on the outside of water pipe 106; this can be an acoustic sensor (such as a detection probe from a leak detector or noise recorder). Alternatively, the sound signal acquisition device 108 can also be installed on the ground or in the water pipes to collect sound signals. Figure 1 As shown, the environment 100 also includes a computing device 102, including desktop computers, laptops, tablets, servers, and other devices with computing capabilities. The sound signal acquisition device 108 sends the detected sound signal to the computing device 102, which performs filtering processing on the acquired sound signal 110 to extract the leakage noise signal 120.

[0031] like Figure 1As shown, the computing device 102 determines the dominant frequency band 112 of the leakage noise energy distribution from the acquired acoustic signal 110. In some examples, the power spectral density of the accumulated energy is calculated based on the power spectral density of the acquired acoustic signal, and the main distribution range of the accumulated energy, i.e., the main frequency range, is determined based on the power spectral density of the accumulated energy, and this frequency range is determined as the dominant frequency band 112. Next, the bandpass filter 114 is configured according to the determined dominant frequency band 112, including the bandwidth, center frequency, and upper and lower limit frequencies of the passband. The upper and lower limit frequencies of the passband are determined based on the upper and lower limits of the dominant frequency band 112, the center frequency is half the sum of the upper and lower limit frequencies of the passband, and the bandwidth is the difference between the upper and lower limit frequencies of the passband. In addition, other parameters of the bandpass filter 114 are configured according to filtering needs (including signal amplification or attenuation, filter selectivity, etc.). The configured bandpass filter 114 can extract the desired signal 116 of leakage noise located in the dominant frequency range from the input acquired acoustic signal 110. Subsequently, the adaptive filter 118 uses the extracted desired signal 116 as a guide signal to extract the input acquired acoustic signal 110, outputting a leakage noise signal 120. During the extraction process, the difference between the desired signal 116 and the leakage noise signal 120 is calculated, i.e., the error signal 122. The parameters of the adaptive filter are iteratively adjusted with the goal of minimizing the error signal 122 until convergence. The final output leakage noise signal 120 is the leakage noise signal after filtering out environmental noise from the acquired acoustic signal.

[0032] It should be understood that the architecture and functionality in example environment 100 are described for illustrative purposes only and do not imply any limitation on the scope of this disclosure. Embodiments of this disclosure can also be applied to other environments with different structures and / or functionalities.

[0033] Figure 2 A flowchart of a method 200 for extracting leakage noise from a water supply network, according to some embodiments of this specification, is shown. This method 200 can, for example, be derived from... Figure 1 The computing device 102 in the middle performs the operation. For example... Figure 2 As shown in box 202, method 200 can acquire the collected acoustic signal from the water supply network. In one example, from... Figure 1 The acoustic signal is acquired in the acoustic signal acquisition device 102. In another example, the acoustic signal acquired in the acoustic signal acquisition device 102 is amplified, filtered, and processed to obtain the acoustic signal of the water supply network.

[0034] In block 204, method 200 can determine the desired signal of leakage noise based on the collected acoustic signal and its dominant frequency band from the water supply network. The dominant frequency band is determined by the main distribution range of accumulated energy calculated based on the power spectral density of the collected acoustic signal. In some implementations, the power spectral density of the collected acoustic signal is determined, thus allowing the distribution of random noise frequency energy to be measured. Next, the power spectral density is normalized, and the cumulative energy distribution is calculated, i.e., the cumulative power spectral density is determined. Then, based on the cumulative power spectral density and preset upper and lower energy limits, the dominant frequency band of the collected acoustic signal is determined. The frequency response of a bandpass filter is designed according to the dominant frequency band to obtain the desired signal of leakage noise.

[0035] In block 206, method 200 can use the desired signal as a guide signal to extract the leakage noise signal of the water supply network from the acquired acoustic signal using an adaptive filtering algorithm. In some implementations, using the desired signal as a guide signal and aiming to minimize the error signal determined by the difference between the desired signal and the leakage noise signal, the adaptive filter outputs the leakage noise signal by adjusting the adaptive weight vector according to the acquired acoustic signal.

[0036] By employing the above method, the frequency range of leakage noise can be captured through power spectral density, and this range reflects the stability of the leakage noise in the time-frequency domain. This allows for a relatively accurate reflection of the characteristics of pipeline network leakage noise, and further, based on the frequency range of the main energy distribution, a desired signal close to the actual pipeline network leakage noise can be obtained. Thus, the embodiments in this specification can guide adaptive filtering by extracting a high-quality desired signal from the input signal (i.e., the acquired acoustic signal mentioned above), effectively suppressing environmental noise and enhancing the leakage noise component. Furthermore, based on a more accurate leakage noise signal, leakage can be identified and located more accurately.

[0037] The foregoing has described specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than that shown in the embodiments and may still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require the specific or sequential order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are possible or may be advantageous.

[0038] Figure 3 A schematic block diagram illustrating the principle of a leakage noise extraction process 300 for a water supply network according to some embodiments of this specification is shown. Figure 3 As shown, the Discrete Fourier Transform (DFT) is performed on the acquired acoustic signal x(n) in box 302 to obtain... In some implementations, assuming the acquired sound signal x(n) collected by the noise recorder consists of leakage noise and ambient noise, then the leakage noise... It can be approximated as the superposition of m sinusoidal components:

[0039] (1)

[0040] In equation (1), m is the number of harmonic components; It is the amplitude of the i-th component, which is related to the leak flow rate and pressure; It is the angular frequency of the i-th component; It is the initial phase.

[0041] In other implementations, the leakage noise is determined based on the characteristics of the leakage noise and prior information. Represented as modulation band-limited noise:

[0042] (2)

[0043] In equation (2), Indicates the amplitude of a slowly varying random variable. As the reference amplitude, For fluctuation intensity, random variable ; Indicates Gaussian white noise; This represents the impulse response of a bandpass filter.

[0044] Environmental noise can generally be considered as a type of Gaussian white noise containing various interferences. satisfy: .in, This indicates a normal distribution, with 0 representing the mean. Indicates variance.

[0045] Leakage noise determined based on any one of the above embodiments. as well as The collected acoustic signal x(n) can be determined.

[0046] Then, a discrete Fourier transform is performed on the acquired acoustic signal x(n) to obtain:

[0047] (3)

[0048] In box 304, based on The power spectral density P(k) of the acquired acoustic signal x(n) is obtained. In some implementations, the power spectral density is calculated using the periodogram method. :

[0049] (4)

[0050] In equation (4), This represents the power of the signal at the k-th frequency point. This represents the absolute value of the complex amplitude of the signal at the k-th frequency point, and N represents the total number of sampling points of the signal. express The real part, express The imaginary part.

[0051] In some implementations, the power spectral density is estimated using the Welch method:

[0052] (5)

[0053] In equation (5), M is the number of segments. This represents the estimated power spectral density at the k-th frequency point. Let represent the power of the i-th signal segment at the k-th frequency point. The core idea of ​​the Welch method is to divide the long signal into several shorter, equal-length segments (usually overlapping), calculate the periodogram for each segment, and then average these periodograms to obtain an estimate of the power spectrum. This method can effectively reduce the estimation variance caused by the randomness and noise of the signal.

[0054] Subsequently, the power spectral density obtained from frame 304 was... Normalization is performed:

[0055] (6)

[0056] In equation (6), This represents the sum of the power spectral densities at all frequency points from k=0 to k=N / 2−1. This represents the normalized power spectral density at the k-th frequency point, typically between 0 and 1.

[0057] In box 306, the cumulative energy distribution is calculated based on the normalized power spectral density, yielding:

[0058] (7)

[0059] Equation (7) calculates the cumulative power spectral density based on the normalized power spectral density, and the normalized cumulative energy distribution. (i.e., the cumulative power spectral density) is monotonically increasing and satisfies:

[0060] (8)

[0061] Equation (8) indicates that the sum of the cumulative normalized power spectral densities is equal to 1.

[0062] Subsequently, based on the cumulative power spectral density obtained from box 306, the main frequency band 308 is determined. From the signal frequency perspective, the frequency range of leakage noise exhibits certain patterns under different conditions. Generally, the frequency is higher when the pipe diameter is small, the flow velocity is high, the water pressure is high, and the distance to the leak is close, typically with a leakage noise frequency range greater than 1200 Hz; the frequency is lower when the pipe diameter is large, the flow velocity is slow, the water pressure is low, and the distance to the leak is very far, typically with a leakage noise frequency range less than 600 Hz. Based on prior information about leakage noise, the frequency range of leakage noise in most pipelines is between 0 and 4000 Hz. Based on the time-frequency characteristics of leakage noise, an energy threshold range [L,H] is set, such as [0.05,0.95] or [0.1,0.9]. Therefore, the main frequency band... Defined as:

[0063] (9-1)

[0064] (9-2)

[0065] In equations (9-1) and (9-2) Sampling rate, and Here, N represents the total number of sampling points in the signal, and the upper and lower limits of the main frequency band can be determined from the cumulative power spectral density based on the upper and lower limits of the energy threshold range.

[0066] In box 308, according to the main frequency band Design the frequency response of a bandpass filter. The desired signal d(n) is obtained:

[0067] (10)

[0068] In equation (9), IFFT stands for Inverse Discrete Fourier Transform, which is used to convert a frequency domain signal back to a time domain signal. This indicates the frequency response of a bandpass filter. A bandpass filter allows signals within a specific frequency range (i.e., the main frequency band) to pass through, while suppressing signals of other frequencies.

[0069] The desired signal d(n) calculated in block 308 is fed into the adaptive filter, which performs adaptive filtering and outputs the signal.

[0070] (11)

[0071] In equation (11), Let x(n) represent the adaptive weight vector, x(n) represent the input signal vector, i.e. the acquired acoustic signal, and y(n) represent the output signal, i.e. the leakage noise signal (reference box 312).

[0072] The error signal is See box 314:

[0073] (12)

[0074] In some implementations, based on a fixed step size Perform adaptive weight updates:

[0075] (13)

[0076] With the goal of minimizing the error signal, an adaptive filtering algorithm is used for iterative updates, and the final output signal is the leakage noise signal after filtering out environmental noise.

[0077] In other implementations, based on the changing step size Perform adaptive weight updates. In each iteration, update the step size once, and then update the adaptive weights using equation (13) based on the updated step size. In some examples, based on the error signal... The ratio of the error power to the input power of the acquired acoustic signal x(n) is used to determine the step size for the adaptive weight vector update using an exponential decay function:

[0078] (14)

[0079] In equation (14), Indicates the upper limit of the learning rate. The error power of the error signal. To acquire the input power of the acoustic signal, The sensitivity factor represents the rate at which the control step size decays with respect to the power ratio. It is a small positive number to prevent the denominator from being 0.

[0080] The error power and input power are updated as follows:

[0081] (15-1)

[0082] (15-2)

[0083] The error power and input power at the first time step are set to preset values, such as... Set to 0, Set it to 0. Determine the error power and input power at the next time step according to equations (15-1) and (15-2). For example, based on the error power at the previous time step and the error signal at the current time step. The square of the input power at the previous time step is used to update the error power at the current time step, and the square of the leakage noise signal x(n) at the current time step is used to update the input power at the current time step. Then, the step size is updated according to equation (14) based on the determined error power and input power.

[0084] like Figure 3 As shown in box 316, process 300 iteratively updates the adaptive filter using the VSS-LMS (Variable Step Size Least Mean Squares) algorithm described in the example above, converging rapidly until the error signal is minimized. The adaptive filter outputs the optimal filtered leakage noise signal.

[0085] Figures 4a to 4g Images from water supply network simulation tests of some embodiments of this specification are shown to illustrate the leakage noise extraction method for water supply networks according to embodiments of this specification. First, based on the characteristics of leakage noise, an input signal x(n) with a main frequency band of [500, 800] Hz is constructed using a bandpass filter, with a sampling rate of 8192 Hz and a signal-to-noise ratio of 10 dB. Figure 4a A time-domain image 400A of the acquired acoustic signal under a water supply network simulation test according to some embodiments of this specification is shown. Image 400A normalizes the amplitude of the acquired acoustic signal and shows the amplitude variation of the acquired acoustic signal within 1 second. Based on the method of the embodiments of this specification, the power density is calculated using the acquired acoustic signal from the water supply network. Figure 4d The diagram 400D illustrates the power spectral density distribution of acoustic signals collected during a water supply network simulation experiment according to some embodiments of this specification. Distribution diagram 400D shows the power spectral density distribution within the range of 0–4000 Hz. According to the method described in the embodiments of this specification, the main distribution range of accumulated energy is determined based on the power spectral density, thereby determining the dominant frequency band. Figure 4e The diagram 400E shows the distribution of accumulated energy of acoustic signals collected under a water supply network simulation test of some embodiments of this specification at different frequencies. Distribution diagram 400E shows the accumulated energy in the range of 0~4000Hz. According to the set energy threshold range [L,H], such as [0.1,0.9], the range [520,840]Hz between the two dashed lines in the diagram can be determined as the main frequency band.

[0086] Figure 4bThe spectral image 400B of the acquired acoustic signal under a water supply network simulation test according to some embodiments of this specification is shown. This spectral image 400B is obtained by performing a Fast Fourier Transform (FFT) based on the time-domain image 400A, and shows the amplitude variation of the acquired acoustic signal in the range of 0~4000Hz. Figure 4c The present specification shows a time-frequency image 400C of the acoustic signal collected under a water supply network simulation test according to some embodiments of this specification. This time-frequency image 400C is obtained by performing a short-time Fourier transform (STFT) based on the time-domain image 400A, showing the frequency change of the collected acoustic signal within 1 second. Figure 4b and Figure 4c All of them show that the main frequency range is basically in the range of 500~800Hz. Figure 4b and Figure 4c Based on Figure 4a By performing a Fourier transform on a pre-constructed input signal, it can be proven that... Figure 4d The main frequency band can be determined more accurately by calculating the main distribution range of the accumulated energy based on the power spectral density of the collected acoustic signal.

[0087] The method described in this specification is based on a determined and relatively accurate dominant frequency band, and extracts the desired signal from the acquired acoustic signal. Figure 4f The present invention illustrates a time-frequency image 400F of the desired signal under a water supply network simulation test of some embodiments of this specification. This time-frequency image 400F shows the frequency variation of the desired signal within 1 second.

[0088] The method described in this specification uses the extracted desired signal as a guide signal and employs an adaptive filtering algorithm to extract the leakage noise signal from the input signal. Figure 4g The following is a 400G time-frequency image of leakage noise signal under a water supply network simulation test, illustrating some embodiments of this specification. Assuming the adaptive filter order is 32, the initial step size of VSS-LMS is... The VSS-LMS step size update forgetting factor is The step size range is The 400G time-frequency image shows the changes in leakage noise signal within 1 second. (Comparison) Figure 4f and Figure 4g Because the main frequency band matching is relatively accurate, the expected signal and the leakage noise signal are highly similar, which greatly suppresses the environmental noise while preserving the leakage noise energy.

[0089] Figure 5 An example block diagram of a leakage noise extraction device 500 for a water supply network, representing some embodiments of this specification, is shown. Figure 5As shown, the device 500 includes a signal acquisition module 502, configured to acquire acoustic signals from a water supply network. The device 500 also includes a desired signal determination module 504, configured to determine a desired leakage noise signal based on the acoustic signals from the water supply network and their dominant frequency band. The dominant frequency band is determined by the main distribution range of accumulated energy calculated based on the power spectral density of the acoustic signals. Finally, the device 500 includes a leakage noise extraction module 506, configured to extract the leakage noise signal from the acoustic signals from the water supply network using an adaptive filtering algorithm, guided by the desired signal.

[0090] In some implementations, the device 500 further includes a main frequency band determination module, configured to determine the power spectral density of the acquired acoustic signal based on the acquired acoustic signal; determine the cumulative power spectral density based on the normalized power spectral density; and determine the main frequency band of the acquired acoustic signal based on the cumulative power spectral density and preset upper and lower energy limits.

[0091] In some implementations, the desired signal determination module 504 includes a configuration unit configured to configure the main frequency band range of the bandpass filter based on the main frequency band of the water supply network. The desired signal determination module 504 also includes a determination unit configured to determine the desired signal of leakage noise output by the bandpass filter based on the collected acoustic signal from the water supply network.

[0092] In some implementations, the leakage noise extraction module 506 includes a first iteration unit configured to minimize the error signal determined by the difference between the desired signal and the leakage noise signal, wherein an adaptive filter outputs the leakage noise signal by adjusting an adaptive weight vector based on the acquired acoustic signal. The leakage noise extraction module 506 also includes a second iteration unit configured to determine the step size for updating the adaptive weight vector based on the ratio of the error power of the error signal to the input power of the acquired acoustic signal, wherein both the error power and the input power are updated recursively; and to determine the updated adaptive weight vector based on the step size, thereby adjusting the updated adaptive weight vector according to the acquired acoustic signal and outputting the updated leakage noise signal.

[0093] In this way, the apparatus of this specification embodiment can capture the frequency range of leakage noise through power spectral density, and then obtain a desired signal that is close to the actual pipeline network leakage noise based on the frequency range of the main energy distribution. Thus, this specification embodiment can guide adaptive filtering by extracting a high-quality desired signal from the acquired acoustic signal (i.e., the acquired acoustic signal referred to above), effectively suppressing environmental noise and enhancing the leakage noise component. Furthermore, based on a more accurate leakage noise signal, leakage can be identified and located more accurately.

[0094] Figure 6A block diagram of an electronic device 600 that can implement various embodiments of the present disclosure is shown. For example... Figure 6 As shown, device 600 includes a processor 601, which can perform various appropriate actions and processes based on computer program instructions loaded into random access memory (RAM) 603 according to computer program instructions stored in read-only memory (ROM) 602. RAM 603 may also store various programs and data required for the operation of device 600. The processor 601, ROM 602, and RAM 603 are interconnected via bus 604. Input / output (I / O) interface 605 is also connected to bus 604.

[0095] The various processes and procedures described above, such as method 200, can be executed by processor 601. For example, in some embodiments, method 200 may be implemented as a software program tangibly contained in a machine-readable medium. In some embodiments, part or all of the software program may be loaded and / or installed on device 600 via ROM 602. When the software program is loaded into RAM 603 and executed by processor 601, one or more actions of method 200 described above may be performed.

[0096] This specification also provides a computer-readable storage medium storing instructions that, when executed on a computer or processor, cause the computer or processor to perform one or more steps in the above-described method embodiments. If the constituent modules of the above-described electronic device are implemented as software functional units and sold or used as independent products, they can be stored in the computer-readable storage medium.

[0097] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this specification are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in or transmitted through a computer-readable storage medium. The computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, Digital Subscriber Line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium accessible to a computer or a data storage device such as a server or data center that integrates one or more available media. The available media may be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., Digital Versatile Discs (DVDs)), or semiconductor media (e.g., Solid State Disks (SSDs)).

[0098] The embodiments described above are merely preferred embodiments of this specification and are not intended to limit the scope of this specification. Any modifications and improvements made by those skilled in the art to the technical solutions of this specification without departing from the spirit of this specification should fall within the protection scope defined by the claims of this specification.

Claims

1. A method for extracting a leakage noise of a water supply network, characterized by, The method comprises the following steps: acquiring a collected sound signal of the water supply network; determining an expected signal of the leakage noise of the water supply network based on the collected sound signal and a main frequency band of the water supply network, including: configuring a main frequency band range of a band-pass filter based on the main frequency band of the water supply network; and outputting the expected signal of the leakage noise of the water supply network by the band-pass filter based on the collected sound signal; the main frequency band is determined by a main distribution range of cumulative energy calculated according to a power spectral density of the collected sound signal; the determination process of the main frequency band comprises: determining the power spectral density of the collected sound signal based on the collected sound signal; determining a cumulative power spectral density based on the normalized power spectral density; and determining the main frequency band of the collected sound signal based on the cumulative power spectral density and preset upper and lower energy limit values; and extracting a leakage noise signal of the water supply network from the collected sound signal by an adaptive filtering algorithm with the expected signal as a guide signal.

2. The method of claim 1, wherein, extracting a leakage noise signal of the water supply network from the collected sound signal by an adaptive filtering algorithm with the expected signal as a guide signal comprises: an adaptive filter outputs the leakage noise signal by adjusting an adaptive weight vector according to the collected sound signal, aiming at minimizing an error signal determined by a difference between the expected signal and the leakage noise signal.

3. The method of claim 2, wherein, The method further comprises: determining a step length of adaptive weight vector update based on a ratio of an error power of the error signal to an input power of the collected sound signal, both the error power and the input power being updated by a recursive method; and determining an updated adaptive weight vector based on the step length, so as to output an updated leakage noise signal by adjusting the updated adaptive weight vector according to the collected sound signal.

4. The method of claim 3, wherein, determining a step length of adaptive weight vector update based on a ratio of an error power of the error signal to an input power of the collected sound signal comprises: determining a step length of adaptive weight vector update based on a ratio of an error power of the error signal to an input power of the collected sound signal by exponential function attenuation.

5. The method of claim 3, wherein, both the error power and the input power being updated by a recursive method comprises: updating the error power at a current time based on the error power at a previous time and a square value of the error signal at the current time, and updating the input power at the current time based on the input power at the previous time and a square value of the leakage noise signal at the current time, wherein the error power and the input power at a first time are preset values.

6. A leakage noise extraction device for a water supply network, characterized by The method comprises the following steps: an acquisition signal acquisition module is configured to acquire a collected sound signal of the water supply network; The expected signal determination module is configured to determine an expected signal of the leakage noise based on the collected sound signal of the water supply network and a main frequency band thereof, the main frequency band being determined according to a main distribution range of cumulative energy calculated according to a power spectral density of the collected sound signal; the expected signal determination module comprises a configuration unit configured to configure a main frequency band range of a band-pass filter based on the main frequency band of the water supply network, and a determination unit configured to output the expected signal of the leakage noise from the band-pass filter based on the collected sound signal of the water supply network; The main frequency band determination module is configured to determine the power spectral density of the collected sound signal based on the collected sound signal. The cumulative power spectral density is determined based on the normalized power spectral density, and the main frequency band of the collected sound signal is determined based on the cumulative power spectral density and preset upper and lower limit values of energy. The leakage noise extraction module is configured to extract the leakage noise signal of the water supply network from the collected sound signal by an adaptive filtering algorithm with the expected signal as a guide signal. The leakage noise extraction module comprises:

7. The apparatus of claim 6, wherein, The first iteration unit is configured to minimize an error signal determined by a difference between the expected signal and the leakage noise signal, and to output the leakage noise signal from the adaptive filter based on the collected sound signal by adjusting an adaptive weight vector. The second iteration unit is configured to determine a step size of adaptive weight vector updating based on a ratio of an error power of the error signal to an input power of the collected sound signal, the error power and the input power being updated by a recursive method, and to determine an updated adaptive weight vector based on the step size, so as to output the updated leakage noise signal of the water supply network by adjusting the updated adaptive weight vector based on the collected sound signal. ​

Images