Dry-type reactor insulation condition intelligent diagnosis and early warning method based on acoustic imaging

Abstract

The invention discloses an intelligent diagnosis and early warning method for the insulation condition of a dry-type air-core reactor based on acoustic imaging. The method comprises the steps that S1, a non-intrusive online insulation monitoring simulation platform for the dry-type air-core reactor is built; s2, a microphone array is used for receiving sound signals in the space and processing the sound signals; s3, an array signal mathematical model for collecting the direction of arrival of the sound signals is constructed, and the fault direction is judged; and S4, according to an acoustic imaging principle, drawing an acoustic image, completing identification of a fault type, and positioning the position of a sound source. The method has the advantages of simple principle, simplicity and convenience in implementation, high diagnosis precision and the like.

Description

Technical Field

[0001] This invention mainly relates to the field of online detection technology for power equipment, specifically a method for intelligent diagnosis and early warning of insulation status of dry-type reactors based on acoustic imaging. Background Technology

[0002] High-voltage dry-type air-core reactors, characterized by being oil-free, coreless, and explosion-proof, have been widely used in power grids at all levels. However, the main operational fault of dry-type air-core reactors is inter-turn insulation breakdown. Causes of this fault include line dampness or weak points in the insulation, partial discharge arcing, localized overheating leading to insulation breakdown, overvoltage, external forces such as small animals bridging short circuits, and mechanical stress from short-circuit current causing coil deformation and damage. During an inter-turn short circuit, the reactor's phase current changes little, but the large circulating current between strands causes a localized temperature rise in the coil, accelerating the aging of the short-turn insulation and potentially developing into a multi-turn short circuit. Ultimately, this can lead to the reactor catching fire. Even if the protection system disconnects the reactor from the grid, it will be severely burned and beyond repair. Therefore, for reactors, an inter-turn short circuit directly affects overvoltage limiting, filtering, and compensation at the line's end, causing equipment damage and impacting other power system equipment, resulting in significant economic losses to the power grid.

[0003] Furthermore, with the continuous improvement of reactor capacity and withstand voltage levels, the problems of large reactor vibration and severe noise pollution have become increasingly prominent, seriously affecting on-site production order and the daily lives of surrounding residents. Distinguishing between electromagnetic vibration, mechanical vibration, core saturation-induced noise, and noise caused by inter-turn short circuits within the normal operating range of reactors has gradually attracted the attention of relevant researchers. Timely analysis and classification of noise during reactor operation, and intelligent diagnosis of the insulation condition of dry-type reactors, are of great significance for preventing faults and ensuring the normal operation of the power system.

[0004] In existing technologies, research directions for inter-turn short-circuit fault diagnosis in dry-type air-core reactors mainly include: the detection coil method, also known as the magnetic field detection method, but it cannot provide timely warnings of early discharge faults caused by reactor defects; the temperature detection method, including smoke detection using smoke generated by the decomposition of internal materials in the dry-type air-core reactor when the temperature rises, but heat accumulation is a long and slow process, making it impossible to achieve early warning of faults; and other methods include inter-turn overvoltage experiments and electrical quantity detection. However, each of these existing technologies has its advantages and disadvantages. Some scholars have combined these methods with intelligent algorithms, optimizing existing methods and achieving good progress.

[0005] With the development of electrical technology, fault monitoring and early warning systems for dry-type air-core reactors will evolve towards intelligence. This requires in-depth analysis of the fault occurrence and development process and fault mechanism of reactors, full consideration of various factors, and improvement of the accuracy and reliability of insulation monitoring and early warning to achieve unmanned operation and ensure the stable operation of dry-type air-core reactors. To this end, it is necessary to consider the influence of multiple parameters and the environmental factors of reactor operation, reduce the impact of external interference, and adopt relevant technologies to filter noise and obtain effective signals.

[0006] In summary, existing detection methods focus more on whether electrical testing is used to detect faults, have long response times, and cannot provide guidance on location, early warning status, etc. Summary of the Invention

[0007] The technical problem to be solved by this invention is: in view of the technical problems existing in the prior art, this invention provides a simple, easy-to-implement, and highly accurate intelligent diagnosis and early warning method for the insulation condition of dry reactors based on acoustic imaging.

[0008] To solve the above-mentioned technical problems, the present invention adopts the following technical solution: A method for intelligent diagnosis and early warning of insulation condition of dry-type reactors based on acoustic imaging, comprising: Step S1: Build a non-invasive online insulation monitoring simulation platform for dry-type air-core reactors; Step S2: Use a microphone array to receive and process the sound signals in the space; Step S3: Construct a mathematical model of the array signal to collect the direction of arrival of the sound signal, and determine the direction of the fault; Step S4: Based on the principle of acoustic imaging, draw an acoustic image to complete the identification of the fault type and locate the sound source.

[0009] As a further improvement to the method of the present invention: in step S1, the fault types simulated in the online insulation monitoring simulation platform include inter-turn discharge, surface discharge, loose fasteners and loose epoxy resin binding; wherein inter-turn discharge and surface discharge are discharge faults during operation, and loose fasteners and loose epoxy resin binding are abnormal operating states.

[0010] As a further improvement to the method of the present invention: in step S2, a microphone array is used to sample the signal in the sound field space, and the radiation data of the sound source signal after being discrete in space is collected by utilizing the different spatial characteristics of the signal.

[0011] As a further improvement to the method of the present invention: In step S3, an array signal mathematical model for collecting the direction of arrival of the sound signal is constructed. The array covariance matrix in the mathematical model is decomposed by the music algorithm. Then, the eigenvalues ​​of the matrix are sorted from smallest to largest. The larger eigenvalues ​​correspond to the signal, and the rest correspond to noise, thereby determining the fault direction.

[0012] As a further improvement to the method of the present invention: in step S3, an array signal mathematical model is constructed to acquire the direction of arrival of the sound signal, wherein the number of signal sources to be measured is D, each signal source to be measured has the same polarization and is uncorrelated with each other, and each signal source has the same center frequency. A microphone array is a linear array of M elements with equal spacing, where M > D. Each element is identical and isotropic. The element spacing is d, and the element spacing is no greater than half the wavelength of the highest frequency signal.

[0013] As a further improvement to the method of the present invention: in step S3, the estimated value of the direction of arrival is found by using the Music algorithm to determine the fault direction of the dry-type air-core reactor. The process includes: The estimated value of the covariance matrix is ​​obtained from N received signal vectors, and eigenvalue decomposition is performed on the obtained covariance matrix. The array signal is regarded as the superposition of several spatial harmonics of noise interference, thus linking the direction-of-arrival estimation problem with spectral estimation. Among them, there are only D eigenvalues ​​related to the signal, which are equal to the matrix AR. S A H The eigenvalues ​​and The sum of the remaining MD eigenvalues , It is the smallest eigenvalue of R; To obtain the noise matrix, we take the eigenvalues ​​that are equal to the number of signals D and their corresponding eigenvectors as the signal space, and take the remaining MD eigenvalues ​​and eigenvectors as the noise space to obtain the noise matrix. Calculate the spectral function; construct a noise matrix En using each noise eigenvector as a column; define the spatial spectrum. The change in arrival angle is estimated by finding the wave crest, and the wave arrival direction is estimated by finding the peak value.

[0014] As a further improvement to the method of the present invention: in step S4, fault location imaging is performed by a delay summation beamforming algorithm.

[0015] As a further improvement to the method of the present invention: In step S4, according to the principle of acoustic imaging, the far-field sound source, i.e. the fault sound signal, is transmitted over a long distance and is collected by the microphone array in the form of a plane wave. Different sound power is calculated for different focal points and displayed in the form of an image, i.e., drawn into an acoustic image, to complete the identification of the fault type and thus locate the position of the sound source.

[0016] As a further improvement to the method of the present invention: in step S4, the D sound sources are respectively A plane wave is incident in a certain direction and received by the array. The positions of the M microphones are as follows: Then the output of the delay summation beamformer is:

[0017] In the above formula, The output of the far-field delay summation beamformer, It is a unit quantity, and its direction is opposite to the propagation direction of the plane wave. m P is the weighting factor for the m-th microphone. m is the sound pressure value received by the m-th microphone, and c is the speed of sound wave propagation; the signal from the direction of the sound source is enhanced by delay summation, and different sound power is calculated for different focal points. The different sound power is displayed in the form of an image and drawn into a sound image.

[0018] Compared with the prior art, the advantages of the present invention are as follows: 1. The present invention provides an intelligent diagnosis and early warning method for the insulation condition of dry-type reactors based on acoustic imaging. It is simple in principle, easy to implement, and has high diagnostic accuracy. It comprehensively considers the discharge faults and abnormal operating states of dry-type air-core reactors, aiming to detect the early signs of latent faults in dry-type air-core reactors in a timely manner, and realizes accurate judgment of the severity of faults and future evolution trends.

[0019] 2. The present invention provides an intelligent diagnosis and early warning method for the insulation condition of dry-type reactors based on acoustic imaging. Since many power devices are inherently dangerous during operation, it is difficult to collect equipment information at close range using sensors for certain signals. This invention utilizes acoustic imaging technology to ensure the safety of personnel. Furthermore, acoustic imaging provides a more intuitive way to identify faults and accurately pinpoints the location of the fault in the dry-type reactor.

[0020] 3. The intelligent diagnosis and early warning method for insulation status of dry-type reactors based on acoustic imaging of the present invention adopts the MUSIC algorithm. As a classic spatial spectrum estimation method based on feature structure analysis, the MUSIC algorithm has the ability to simultaneously determine the direction of multiple signals and judge whether the dry-type reactor is in a fault state or working normally. It can accurately estimate the direction of arrival of the signal, and the judgment is more accurate and faster than other judgment methods.

[0021] 4. The present invention provides an intelligent diagnostic and early warning method for the insulation condition of dry-type reactors based on acoustic imaging. It introduces acoustic methods into the condition detection of dry-type air-core reactors, uses the proposed diagnostic method to achieve condition assessment of dry-type air-core reactors, and then uses a music algorithm based on array signal processing to achieve high-resolution direction estimation of signal sources in different spatial directions, thereby achieving accurate judgment of fault severity and future evolution trend. It can be seen that the present invention provides a new method for monitoring the inter-turn working state and fault judgment of dry-type reactors. Attached Figure Description

[0022] Figure 1 This is a flowchart illustrating the method of the present invention.

[0023] Figure 2 This is a schematic diagram illustrating the principle of the microphone equidistant linear array and far-field signal in a specific application example of the present invention.

[0024] Figure 3 This is a schematic diagram illustrating the principle of acoustic imaging using the far-field delay summation beamforming algorithm in a specific application example of the present invention. Detailed Implementation

[0025] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0026] In the description of this invention, it should be understood that the terms "length", "width", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0027] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of the technical features being referred to. Thus, a feature identified as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, "plurality" means two or more, unless otherwise specifically defined.

[0028] In this invention, unless otherwise explicitly specified and limited, the terms "assembly," "connection," "linking," and "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0029] like Figures 1-3 As shown, the present invention provides an intelligent diagnosis and early warning method for the insulation condition of dry reactors based on acoustic imaging. This method is implemented using an equally spaced linear array composed of M microphone elements, and its steps include: Step S1: Build a non-invasive online insulation monitoring simulation platform for dry-type air-core reactors; simulate various normal operation and fault conditions of dry-type reactors; Step S2: Use a microphone array to receive and process sound signals in the space; Step S3: Construct a mathematical model of the array signal to collect the direction of arrival of the sound signal, and determine the direction of the fault; Step S4: Based on the principle of acoustic imaging, draw an acoustic image to complete the identification of the fault type and locate the sound source.

[0030] In a specific application example, in step S1, the fault types simulated in the online insulation monitoring simulation platform include four types: inter-turn discharge, surface discharge, loose fasteners, and loose epoxy resin binding. Among them, inter-turn discharge and surface discharge are discharge faults during operation, while loose fasteners and loose epoxy resin binding are abnormal operating states.

[0031] In a specific application example, in step S2, a microphone array is used to sample the signal in the sound field space, and the radiation data of the sound source signal after spatial discretization is collected by utilizing the different spatial characteristics of the signal.

[0032] In a specific application example, in step S3, an array signal mathematical model for collecting the direction of arrival of the sound signal is constructed. The array covariance matrix in the mathematical model is decomposed using the music algorithm. Then, the eigenvalues ​​of the matrix are sorted from smallest to largest. The larger eigenvalues ​​correspond to the signal, and the rest correspond to noise, thereby determining the fault direction.

[0033] In a preferred embodiment, the detailed process of step S3 includes: Step S301: Construct a mathematical model of the array signal for acquiring the direction of arrival of the sound signal; The number of signal sources to be tested is D. Each signal source has the same polarization and is uncorrelated with the others, and each signal source has the same center frequency. .

[0034] The microphone array is a linear array of M (M>D) elements with equal spacing. Each element is identical and isotropic. The element spacing is d and is no greater than half the wavelength of the highest frequency signal.

[0035] Step S302: Use the Music algorithm to find the estimated value of the direction of arrival and determine the fault direction of the dry air-core reactor.

[0036] In a preferred embodiment, the detailed process in step S302 above may include: Step S3021: Obtain the estimated value of the covariance matrix based on the N received signal vectors, and perform eigenvalue decomposition on the obtained covariance matrix: wavefront signal is , It is a narrowband signal. It can be represented in the following form: (1) In the above formula, It is the wavefront signal radiated onto the microphone array by the k-th (k = 1, 2, ..., D) signal source, where the subscript k indicates the corresponding signal source. The signal at time t The complex envelope, It is a signal angular frequency.

[0037] The electromagnetic wave passing through the antenna array size over time t1 is: (2) Taking the first element as the reference point, the induced signal of the m-th (m=1,2,...M) element in the equally spaced linear array at time t to the k-th signal source is: (3) In the above formula, a k Let m be the influence of the m-th array element on the th signal source. Assume that each array element is non-directional, and take a=1. Let be the azimuth angle of the i-th signal source. This represents the signal phase difference caused by the path difference between the m-th array element and the 1-th array element.

[0038] Calculate the measurement noise and incoming signals from all sources. The output signal of the m-th array element is: (4) In the above formula, k indicates that the quantity belongs to the k-th signal source, and m indicates that the quantity belongs to the m-th array element. It measures noise, among which Let be the response function of the m-th array element to the k-th signal source. It is the signal strength of the k-th signal source on the array element.

[0039] Convert the expression into matrix form. (5) In the above formula,

[0040]

[0041]

[0042]

[0043] right By performing N-point sampling, the problem becomes estimating the direction of arrival (DOA) of the signal source using the samples {x(i), i=1,2,...,M) of the output signal x). By treating the array signal as a superposition of several spatial harmonics of noise interference, the direction-of-arrival estimation problem is linked to spectral estimation.

[0044] Perform correlation processing on the array output X to obtain its covariance matrix Rx. (6) Substituting equation (5) into it, we get: (7) In the above formula, The correlation matrix of the signal, It is the correlation matrix of the noise, o 2 I is the noise power, and I is the MxM identity matrix.

[0045] The covariance matrix is ​​estimated by performing maximum likelihood estimation on Rx: (8) In the above formula, N is the number of samples. When the number of samples N is infinite, Rx and They are consistent.

[0046] Eigenvalue decomposition of equation (5) above shows that Rx has M positive real eigenvalues. Each corresponds to one of the M feature vectors. Rx is a Hermitian matrix, so all eigenvectors are orthogonal.

[0047] Among them, there are only D eigenvalues ​​related to the signal, which are respectively equal to matrix AR. s A H The eigenvalues ​​and The sum of the remaining MD eigenvalues , It is the smallest eigenvalue of R.

[0048] Step S3022: Calculate the noise matrix; In order of magnitude of eigenvalues Treating the eigenvalues ​​and corresponding eigenvectors equal to the number D of signals as the signal space, and the remaining MD eigenvalues ​​and eigenvectors as the noise space, we obtain the noise matrix E. n : (9) (10) In the above formula, the eigenvectors corresponding to the noise eigenvalues ​​are orthogonal to the column vectors of matrix A, and each column of A corresponds to the direction of the signal source, thus the direction of the signal source can be solved using the noise eigenvalues.

[0049] Step S3023: Calculate the spectral function; Construct a noise matrix En using each noise feature vector as its column:

[0050] Define spatial spectrum : (10) In the above formula, the denominator is the inner product of the signal vector and the noise matrix, and the denominator is zero; however, due to the presence of noise, it is actually a minimum value, so P has a spike.

[0051] From the above formula, we can make The change is estimated by finding the peak, and the change is obtained by finding the peak value to obtain the estimated value of the angle of arrival.

[0052] In a specific application example, in step S4, according to the principle of acoustic imaging, the far-field sound source, i.e. the fault sound signal, after being transmitted over a long distance, can be regarded as being collected by the microphone array in the form of a plane wave. Different sound power is calculated for different focal points, and finally the different sound power is displayed in the form of an image, which can be drawn into an acoustic image to complete the identification of the fault type and thus locate the position of the sound source.

[0053] In a specific application example, in step S4, fault location imaging is performed using a time-delay summation beamforming algorithm. Since the music algorithm has already been used to classify the acquired multiple sound signals in step S3, using this acoustic imaging algorithm, time-delay summation beamforming, in step S4 provides a more intuitive judgment of the fault location.

[0054] In step S4, the D sound sources are respectively A plane wave is incident in a certain direction and received by the array. The positions of the M microphones are as follows: Then the output of the delay summation beamformer is: (11) In the above formula, The output of the far-field delay summation beamformer, It is a unit quantity, and its direction is opposite to the propagation direction of the plane wave. m P is the weighting factor for the m-th microphone. m is the sound pressure level received by the m-th microphone, and c is the speed of sound wave propagation.

[0055] The signal from the direction of the sound source is enhanced by the above-mentioned delay summation method. Different sound powers are calculated for different focal points, and finally, the different sound powers are displayed in the form of an image, which can be drawn into an acoustic image.

[0056] The above are merely preferred embodiments of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principles of the present invention should be considered within the scope of protection of the present invention.

Claims

1. A method for intelligent diagnosis and early warning of insulation condition of dry-type reactors based on acoustic imaging, characterized in that, include: Step S1: Build a non-invasive online insulation monitoring simulation platform for dry-type air-core reactors; Step S2: Use a microphone array to receive and process the sound signals in the space; Step S3: Construct a mathematical model of the array signal to collect the direction of arrival of the sound signal, and determine the direction of the fault; Step S4: Based on the principle of acoustic imaging, draw an acoustic image to complete the identification of the fault type and locate the sound source.

2. The intelligent diagnosis and early warning method for insulation condition of dry-type reactors based on acoustic imaging according to claim 1, characterized in that, In step S1, the fault types simulated in the online insulation monitoring simulation platform include inter-turn discharge, surface discharge, loose fasteners, and loose epoxy resin binding. The inter-turn discharge and surface discharge are discharge faults during operation, while the loose fasteners and loose epoxy resin binding are abnormal operating states.

3. The intelligent diagnosis and early warning method for insulation condition of dry-type reactors based on acoustic imaging according to claim 1, characterized in that, In step S2, a microphone array is used to sample the signal in the sound field space, and the radiation data of the sound source signal after being discrete in space is collected by utilizing the different spatial characteristics of the signal.

4. The intelligent diagnosis and early warning method for insulation condition of dry-type reactors based on acoustic imaging according to any one of claims 1-3, characterized in that, In step S3, an array signal mathematical model for collecting sound signal arrival direction is constructed. The array covariance matrix in the mathematical model is decomposed using the music algorithm. The eigenvalues ​​of the matrix are then sorted from smallest to largest. Larger eigenvalues ​​correspond to signals, and the rest correspond to noise, thereby determining the fault direction.

5. The intelligent diagnosis and early warning method for insulation condition of dry-type reactors based on acoustic imaging according to claim 4, characterized in that, In step S3, an array signal mathematical model is constructed to acquire the direction of arrival of the sound signal. The number of signal sources to be measured is D, each signal source has the same polarization and is uncorrelated, and each signal source has the same center frequency. A microphone array is a linear array of M elements with equal spacing, where M > D. Each element is identical and isotropic. The element spacing is d, and the element spacing is no greater than half the wavelength of the highest frequency signal.

6. The intelligent diagnosis and early warning method for insulation condition of dry-type reactors based on acoustic imaging according to claim 4, characterized in that, In step S3, the estimated value of the direction of arrival is found using the Music algorithm to determine the fault direction of the dry-type air-core reactor. The process includes: The estimated value of the covariance matrix is ​​obtained from N received signal vectors, and eigenvalue decomposition is performed on the obtained covariance matrix. The array signal is regarded as the superposition of several spatial harmonics of noise interference, thus linking the direction-of-arrival estimation problem with spectral estimation. Among them, there are only D eigenvalues ​​related to the signal, which are equal to the matrix AR. S A H The eigenvalues ​​and The sum of the remaining MD eigenvalues , It is the smallest eigenvalue of R; To obtain the noise matrix, we take the eigenvalues ​​that are equal to the number of signals D and their corresponding eigenvectors as the signal space, and take the remaining MD eigenvalues ​​and eigenvectors as the noise space to obtain the noise matrix. Calculate the spectral function; construct a noise matrix En using each noise eigenvector as a column; define the spatial spectrum. The change in arrival angle is estimated by finding the wave crest, and the wave arrival direction is estimated by finding the peak value.

7. The intelligent diagnosis and early warning method for insulation condition of dry-type reactors based on acoustic imaging according to any one of claims 1-3, characterized in that, In step S4, fault location imaging is performed using a delay-summing beamforming algorithm.

8. The intelligent diagnosis and early warning method for insulation condition of dry-type reactors based on acoustic imaging according to claim 7, characterized in that, In step S4, according to the principle of acoustic imaging, the far-field sound source, i.e. the fault sound signal, is transmitted over a long distance and is collected by the microphone array in the form of a plane wave. Different sound power is calculated for different focal points and displayed in the form of an image, i.e., drawn into an acoustic image, to complete the identification of the fault type and thus locate the position of the sound source.

9. The intelligent diagnosis and early warning method for insulation condition of dry-type reactors based on acoustic imaging according to claim 8, characterized in that, In step S4, the D sound sources are respectively A plane wave is incident in a certain direction and received by the array. The positions of the M microphones are as follows: Then the output of the delay summation beamformer is: In the above formula, The output of the far-field delay summation beamformer, It is a unit quantity, and its direction is opposite to the propagation direction of the plane wave. m P is the weighting factor for the m-th microphone. m is the sound pressure value received by the m-th microphone, and c is the speed of sound wave propagation; the signal from the direction of the sound source is enhanced by delay summation, and different sound power is calculated for different focal points. The different sound power is displayed in the form of an image and drawn into a sound image.

Images