CN112883628A - Method and system for positioning abnormal sound source of transformer substation equipment - Google Patents

Method and system for positioning abnormal sound source of transformer substation equipment Download PDF

Info

Publication number
CN112883628A
CN112883628A CN202110308028.1A CN202110308028A CN112883628A CN 112883628 A CN112883628 A CN 112883628A CN 202110308028 A CN202110308028 A CN 202110308028A CN 112883628 A CN112883628 A CN 112883628A
Authority
CN
China
Prior art keywords
sound signal
particle
transformer substation
sound source
value
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202110308028.1A
Other languages
Chinese (zh)
Inventor
张瑶
罗林根
盛戈皞
王辉
宋辉
钱勇
江秀臣
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shanghai Jiaotong University
Original Assignee
Shanghai Jiaotong University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Shanghai Jiaotong University filed Critical Shanghai Jiaotong University
Priority to CN202110308028.1A priority Critical patent/CN112883628A/en
Publication of CN112883628A publication Critical patent/CN112883628A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G06F30/25—Design optimisation, verification or simulation using particle-based methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/004—Artificial life, i.e. computing arrangements simulating life
    • G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2111/00—Details relating to CAD techniques
    • G06F2111/08—Probabilistic or stochastic CAD
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
    • G06F2119/10—Noise analysis or noise optimisation

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Evolutionary Computation (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Health & Medical Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Data Mining & Analysis (AREA)
  • Biophysics (AREA)
  • Biomedical Technology (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • Computational Linguistics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Health & Medical Sciences (AREA)
  • Computer Hardware Design (AREA)
  • Geometry (AREA)
  • Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)

Abstract

The invention discloses a method for positioning an abnormal sound source of transformer substation equipment, which comprises the following steps: (1) collecting sound signal sample Y of transformer substation ═ Y1,y2,…,yn](ii) a (2) Establishing a likelihood function of a transformer substation sound signal sample; (3) and solving the maximum value of the likelihood function of the sound signal sample of the transformer substation by adopting an improved particle swarm algorithm, and obtaining the estimated value of the direction angle D corresponding to the maximum value based on the maximum value so as to position the abnormal sound source of the transformer substation equipment. In addition, the invention also discloses a system for positioning the abnormal sound source of the transformer substation equipment, which comprises: the sound sensor array is used for collecting a sound signal sample of the transformer substation; processing and positioning module for carrying out the inventionAnd (3) and (2) in the method for positioning the abnormal sound source of the transformer substation equipment, so as to position the abnormal sound source of the transformer substation equipment.

Description

Method and system for positioning abnormal sound source of transformer substation equipment
Technical Field
The invention relates to a positioning method and a positioning system, in particular to a method and a system for positioning a transformer substation.
Background
As is well known, the sound signals of the power equipment of the transformer substation have abundant running state information, and in the prior art, the sound information during the running of the power equipment can be provided by utilizing an acoustic monitoring technology, so that the running state of the equipment is judged. This is a method for monitoring and evaluating the operating state of the electrical equipment, which is often used in the prior art.
In the prior art, a common sound source orientation method, such as a multiple signal classification (MUSIC) method, is low in positioning accuracy in a low signal-to-noise ratio environment, and the effect of positioning an abnormal sound source in a transformer substation environment by using the MUSIC method is poor, so that a satisfactory effect is difficult to achieve.
Accordingly, the Maximum Likelihood Estimation (MLE) method has higher orientation accuracy and is less computationally expensive than the MUSIC method in the prior art. However, when the Maximum Likelihood Estimation (MLE) method is used to locate an abnormal sound source, spectral peak search is still required, and the smaller the search step size is, the larger the calculation amount is.
Based on the above, aiming at the defects and shortcomings in the prior art, the invention is expected to obtain a method and a system for positioning the abnormal sound source of the transformer substation equipment, and the method and the system for positioning the abnormal sound source of the transformer substation equipment can effectively solve the problem of large calculated amount when a Maximum Likelihood Estimation (MLE) method is adopted by introducing an improved particle swarm optimization (MPSO) of a variation idea, so that the abnormal sound source of the transformer substation equipment is quickly and accurately positioned, and have high positioning accuracy, wide applicability, good popularization prospect and application value.
Disclosure of Invention
One of the purposes of the invention is to provide a method for positioning an abnormal sound source of transformer substation equipment, which can effectively solve the problem of large calculation amount when a Maximum Likelihood Estimation (MLE) method is adopted by introducing an improved particle swarm optimization (MPSO) method of variation thought, so as to quickly and accurately position the abnormal sound source of the transformer substation equipment, and has high positioning accuracy, wide applicability, good popularization prospect and application value.
Based on the purpose, the invention provides a method for positioning an abnormal sound source of transformer substation equipment, which comprises the following steps:
(1) collecting sound signal sample Y of transformer substation ═ Y1,y2,…,yn]；
(2) Establishing a likelihood function of a transformer substation sound signal sample;
(3) and solving the maximum value of the likelihood function of the sound signal sample of the transformer substation by adopting an improved particle swarm algorithm, and obtaining the estimated value of the direction angle D corresponding to the maximum value based on the maximum value so as to position the abnormal sound source of the transformer substation equipment.
Further, in the method for locating an abnormal sound source of substation equipment according to the present invention, for a sound signal sample Y ═ Y1,y2,…,yn]The likelihood function of the substation sound signal sample is configured as:
Figure BDA0002988355230000021
Figure BDA0002988355230000022
where theta represents the pitch angle of the sound signal,
Figure BDA0002988355230000023
representing the azimuth angle, σ, of the acoustic signal2Represents the variance of the noise signal, tr (-) represents the trace of the matrix; r represents the covariance matrix of the sound signal, yi(t) represents a sound signal, and,
Figure BDA0002988355230000024
is a direction vector of the sound signal,
Figure BDA0002988355230000025
is a direction vector
Figure BDA0002988355230000026
Projection operator for generating a space, n representing the number of sound signals in a sound signal sample, yi H(t) denotes the sound signal yi(t) conjugate transpose of matrix.
Further, in the method for locating an abnormal sound source of substation equipment according to the present invention, in step (3), an estimated value of the sound signal direction angle D is obtained according to the following formula:
Figure BDA0002988355230000027
further, in the method for locating an abnormal sound source of a substation device according to the present invention, the step (3) of obtaining a maximum value of a likelihood function of a sound signal sample of the substation by using an improved particle swarm algorithm includes the steps of:
taking the likelihood function as a fitness function of the particle swarm algorithm to calculate the value of the fitness function; wherein the search space of the particle swarm optimization is set as a pitch angle theta and an azimuth angle of the sound signal respectively representing the sound signal
Figure BDA0002988355230000031
The particles in the particle swarm algorithm correspond to different pitch angles theta and azimuth angles
Figure BDA0002988355230000032
Searching individual optimal solution and group optimal solution of the particles based on the following calculation formula of the speed and the position of the particles:
vid(t+1)＝ωvid(t)+c1r1(pid(t)-xid(t))+c2r2(pgd(t)-xid(t))
xid(t+1)＝xid(t)+vid(t+1)
where t represents the current number of searches, xidIndicating the d-dimensional position, v, of the ith particleidRepresents the d-dimensional velocity of the ith particle, d is 1, 2; i is 1,2, … …, m; m represents the number of particles in the particle group; c. C1And c2Acceleration constants are respectively cognitive parameters and social parameters; r is1And r2Represents the interval [0,1]A random number of (c); p is a radical ofidAnd pgdRespectively representing an individual optimal solution and a group optimal solution of the particle swarm in historical search; omega is a random weight;
and continuously updating the particle speed and the particle position based on the particle individual optimal solution and the group optimal solution, endowing the particle individual optimal position with 1 random trust value P after the optimal particle position is updated every time, performing variation when the random trust value P is smaller than a variation trust threshold, calculating a fitness function value of the new particle individual position, and replacing the previous particle individual position with the varied particle individual position if the fitness function value is superior to the position before the variation operation, otherwise, keeping the particle individual position unchanged.
Further, in the method for positioning an abnormal sound source of substation equipment according to the present invention, the expression of the random weight ω is:
Figure BDA0002988355230000033
in the formula, N (0,1) is a random number following a normal distribution; mu.smaxAnd muminThe maximum value and the minimum value of the inertia weight mu are respectively; σ is a random weight variance; r is uniformly distributed between 0 and 1The random number of (2).
Furthermore, in the method for positioning the abnormal sound source of the transformer substation equipment, the variant is implemented by adopting a cuckoo algorithm.
Accordingly, another object of the present invention is to provide a system for locating an abnormal sound source of substation equipment, by which the abnormal sound source of substation equipment can be accurately located in a low signal-to-noise ratio environment (e.g., in a substation field environment).
Based on the above purpose, the present invention further provides a system for positioning an abnormal sound source of a substation device, including:
the sound sensor array is used for collecting a sound signal sample of the transformer substation;
a processing and positioning module that performs the steps of:
(1) based on sound signal sample Y of transformer substation ═ Y1,y2,…,yn]Establishing a likelihood function of a transformer substation sound signal sample;
(2) and solving the maximum value of the likelihood function of the sound signal sample of the transformer substation by adopting an improved particle swarm algorithm, and obtaining the estimated value of the direction angle D corresponding to the maximum value based on the maximum value so as to position the abnormal sound source of the transformer substation equipment.
Further, in the system for locating an abnormal sound source of substation equipment according to the present invention, for a sound signal sample Y ═ Y1,y2,…,yn]The likelihood function of the substation sound signal sample is configured as:
Figure BDA0002988355230000041
Figure BDA0002988355230000042
where theta represents the pitch angle of the sound signal,
Figure BDA0002988355230000043
representing sound signalsAzimuth angle, σ2Represents the variance of the noise signal, tr (-) represents the trace of the matrix; r represents the covariance matrix of the sound signal, yi(t) represents a sound signal, and,
Figure BDA0002988355230000044
is a direction vector of the sound signal,
Figure BDA0002988355230000045
is a direction vector
Figure BDA0002988355230000046
Projection operator for generating a space, n representing the number of sound signals in a sound signal sample, yi H(t) denotes the sound signal yi(t) conjugate transpose of matrix.
Further, in the system for locating an abnormal sound source of substation equipment according to the present invention, in step (2), the estimated value of the sound signal direction angle D is obtained according to the following formula:
Figure BDA0002988355230000047
further, in the system for locating an abnormal sound source of a transformer substation device according to the present invention, the step (2) of obtaining a maximum value of a likelihood function of a transformer substation sound signal sample by using an improved particle swarm algorithm includes the steps of:
taking the likelihood function as a fitness function of the particle swarm algorithm to calculate the value of the fitness function; wherein the search space of the particle swarm optimization is set as a pitch angle theta and an azimuth angle of the sound signal respectively representing the sound signal
Figure BDA0002988355230000048
The particles in the particle swarm algorithm correspond to different pitch angles theta and azimuth angles
Figure BDA0002988355230000049
Searching individual optimal solution and group optimal solution of the particles based on the following calculation formula of the speed and the position of the particles:
vid(t+1)＝ωvid(t)+c1r1(pid(t)-xid(t))+c2r2(pgd(t)-xid(t))
xid(t+1)＝xid(t)+vid(t+1)
where t represents the current number of searches, xidIndicating the d-dimensional position, v, of the ith particleidRepresents the d-dimensional velocity of the ith particle, d is 1, 2; i is 1,2, … …, m; m represents the number of particles in the particle group; c. C1And c2Acceleration constants are respectively cognitive parameters and social parameters; r is1And r2Represents the interval [0,1]A random number of (c); p is a radical ofidAnd pgdRespectively representing an individual optimal solution and a group optimal solution of the particle swarm in historical search; omega is a random weight;
and continuously updating the particle speed and the particle position based on the particle individual optimal solution and the group optimal solution, endowing the particle individual optimal position with 1 random trust value P after the optimal particle position is updated every time, performing variation when the random trust value P is smaller than a variation trust threshold, calculating a fitness function value of the new particle individual position, and replacing the previous particle individual position with the varied particle individual position if the fitness function value is superior to the position before the variation operation, otherwise, keeping the particle individual position unchanged.
The method and the system for positioning the abnormal sound source of the transformer substation equipment have the following advantages and beneficial effects:
according to the method for positioning the abnormal sound source of the transformer substation equipment, the problem of large calculation amount when a Maximum Likelihood Estimation (MLE) method is adopted can be effectively solved by introducing an improved particle swarm optimization (MPSO) method of a variation idea, so that the abnormal sound source of the transformer substation equipment can be quickly and accurately positioned, the positioning accuracy is high, the applicability is very wide, and the method has good popularization prospect and application value.
Different from the existing power equipment sound detection technology, the transformer substation equipment abnormal sound source positioning method does not need to install a sound sensor or a signal conditioning unit on the equipment, does not need to consider the arrangement of a field power line and a signal line, but utilizes a non-contact sound sensor array to develop a portable device applied to sound detection and positioning of the whole transformer substation equipment, and realizes the active orientation of the whole transformer substation equipment abnormal sound source.
Correspondingly, the transformer substation equipment abnormal sound source positioning system can be used for implementing the transformer substation equipment abnormal sound source positioning method, and has the advantages and beneficial effects.
Drawings
Fig. 1 schematically shows a perspective view of a uniform area array model of an acoustic sensor.
Fig. 2 schematically shows a projection view of a uniform area array model of an acoustic sensor.
Fig. 3 schematically shows a flow chart of an MPSO-MLE algorithm of the method for locating an abnormal sound source of substation equipment according to an embodiment of the present invention.
Fig. 4 schematically shows a system configuration diagram of a substation equipment abnormal sound source localization system according to an embodiment of the present invention.
Fig. 5 schematically shows a time-domain sound signal diagram collected by a sound sensor array of the substation equipment abnormal sound source positioning system according to an embodiment of the invention.
Detailed Description
The method and system for locating an abnormal sound source of substation equipment according to the present invention will be further explained and explained with reference to the drawings and the specific embodiments of the specification, but the explanation and explanation do not unduly limit the technical solution of the present invention.
Fig. 1 schematically shows a perspective view of a uniform area array model of an acoustic sensor.
Fig. 2 schematically shows a projection view of a uniform area array model of an acoustic sensor.
In the invention, the method for positioning the abnormal sound source of the transformer substation equipment comprises the following steps:
(1) miningCollecting sound signal sample Y of transformer substation ═ Y1,y2,…,yn]；
(2) Establishing a likelihood function of a transformer substation sound signal sample;
(3) and solving the maximum value of the likelihood function of the sound signal sample of the transformer substation by adopting an improved particle swarm algorithm, and obtaining the estimated value of the direction angle D corresponding to the maximum value based on the maximum value so as to position the abnormal sound source of the transformer substation equipment.
In the method for positioning the abnormal sound source of the transformer substation equipment, the sound signal samples of the transformer substation can be acquired by adopting a square matrix formed by M × M sound sensors, and the sound sensor array can be formed by the square matrix formed by the M × M sound sensors.
As shown in fig. 1 and fig. 2, in the formed acoustic sensor array, the distance between adjacent array elements may be represented as d, and the acoustic signal received by the acoustic sensor may be represented as y (t), then the following formula (1) may be obtained:
Figure BDA0002988355230000061
in the above-mentioned formula (1),
Figure BDA0002988355230000062
expressed as a direction vector, which is related to the array distribution of the sound sensors; s (t) is a sound signal vector; n (t) is a noise signal.
With further reference to FIG. 1, it can be seen that in the present invention, the acoustic signal received by the acoustic sensor exists in an incident direction W, where
Figure BDA0002988355230000063
Is the incident azimuth angle, theta is the incident pitch angle. Incident azimuth angle
Figure BDA0002988355230000064
Starting from an x-axis positive half shaft, returning to the x-axis positive half shaft anticlockwise, wherein the range of the x-axis positive half shaft can be 0-360 degrees; angle of incidence theta being positive half from z-axisThe axis starts at the positive x-axis and can range from 0 to 90.
Accordingly, in the present invention, the direction vector of the uniform sound sensor array composed of sound sensors can be aligned
Figure BDA0002988355230000065
The following formula (2) can be obtained by considering the x-axis direction and the y-axis direction:
Figure BDA0002988355230000071
in the above formula (2), axAnd ayRespectively representing direction vectors of an x axis and a y axis; j represents an imaginary part; ω 2 pi f 2 pi c/λ, c being the speed of sound, f being the frequency of the incident sound signal, λ being the wavelength of the incident sound signal; dxAnd dyRespectively showing the spacing between two adjacent sound sensors in the x-axis direction and the y-axis direction.
Referring further to fig. 2, it can be seen that in the acoustic transducer array according to the present invention, the direction vector of the sub-array 1 is axAnd the offset in the y-axis direction needs to be considered for the other sub-arrays, and then the direction vectors are respectively:
Figure BDA0002988355230000072
in the above formula (3), ay(1) Representation matrix ayThe 1 st component of (a); a isy(2) Representation matrix ayThe 2 nd component of (a); accordingly, ay(M) the representation matrix ayThe mth component of (2).
In the invention, the method for positioning the abnormal sound source of the transformer substation equipment needs to be assisted by a Maximum Likelihood Estimation (MLE) method so as to establish a likelihood function of a transformer substation sound signal sample. In the Maximum Likelihood Estimation (MLE) method, the likelihood function of a received signal is referred to as a conditional probability density function with unknown parameters, and the purpose of this method is to use known sample results to extrapolate back the parameter values that are most likely to lead to such results.
By utilizing the principle, in the step (1) of the method for positioning the abnormal sound source of the transformer substation equipment, a large number of transformer substation sound signal samples can be acquired by adopting a square matrix consisting of M × M sound sensors, then in the subsequent step (2), the likelihood function L (theta) of the transformer substation sound signal samples is established, and the transformer substation sound signal samples are analyzed to extract the incident azimuth angle corresponding to the maximum value of the likelihood function
Figure BDA0002988355230000073
And an incident pitch angle θ.
It should be noted that, in the probability statistics, the probability density function of the random variable x obeying the normal distribution can be expressed as:
Figure BDA0002988355230000081
in the above equation (4), μ is a mathematical expectation; sigma2Is the variance.
Accordingly, for normal population N (μ, σ)2) If there are n samples [ x ]1,x2,…,xn]The unknown parameter is (mu, sigma)2) Then, the likelihood function can be obtained as:
Figure BDA0002988355230000082
thus, by taking logarithms for both sides of the likelihood function equation (5), the log-likelihood function can be obtained as:
Figure BDA0002988355230000083
in the invention, in order to obtain the log-likelihood function of the sound signal sample of the transformer substation, the mathematical expected value and the variance of the sound signal sample of the transformer substation need to be obtained. As can be seen from the formula (1), the signal received by the sound sensorThe numbers generally include: the sound signal emitted by the sound source and the noise signal in the environment both follow a normal distribution. In the present invention, the expected value of the sound signal is
Figure BDA0002988355230000084
Its variance is 0, and the ambient noise is considered to be white gaussian noise, whose expectation value is 0 and variance is σ2。
It follows that in the present invention, the mathematical expectation of the acoustic sensor acceptance signal y (t) can be expressed as
Figure BDA0002988355230000085
The variance may be σ2. Accordingly, knowing the mathematical expected value and variance of the substation sound signal sample, Y ═ Y for the substation sound signal sample1,y2,…,yn]The log-likelihood function can be found as:
Figure BDA0002988355230000086
in the above equation (7), θ represents the pitch angle of the sound signal,
Figure BDA0002988355230000087
representing the azimuth angle, σ, of the acoustic signal2Representing the variance, y, of the noise signali(t) represents a sound signal, and,
Figure BDA0002988355230000088
is a direction vector, si(t) denotes a sound signal vector, and n denotes the number of sound signals in a sound signal sample.
Accordingly, in order to obtain the incident azimuth angle corresponding to the maximum value of the above formula (7)
Figure BDA0002988355230000089
And angle of incidence θ, the variance σ required for the noise signal2And sound signal vector si(t) performing maximum likelihood estimation and comparing the likelihood functionSimplifying the number, neglecting constant terms without influence on the problem, thereby obtaining the product only containing unknown parameters
Figure BDA00029883552300000810
And θ substation sound signal sample Y ═ Y1,y2,…,yn]The log-likelihood function of (a) is:
Figure BDA0002988355230000091
in the above formula (8), yi H(t) denotes the sound signal yi(t) conjugate transpose of matrix;
Figure BDA0002988355230000092
to represent
Figure BDA0002988355230000093
The conjugate transpose of (1);
Figure BDA0002988355230000094
is a direction vector
Figure BDA0002988355230000095
Generating a projection operator of the space.
Figure BDA0002988355230000096
The expression of (c) can be expressed by the following formula (9):
Figure BDA0002988355230000097
correspondingly, the invention can further utilize the property and the inner and outer product conversion relation of the projection matrix to convert the sound signal sample Y of the transformer substation to [ Y ═ Y%1,y2,…,yn]The likelihood function of (d) is constructed as:
Figure BDA0002988355230000098
Figure BDA0002988355230000099
in the above equation (10), θ represents the pitch angle of the sound signal,
Figure BDA00029883552300000910
representing the azimuth angle, σ, of the acoustic signal2Represents the variance of the noise signal, tr (-) represents the trace of the matrix; r represents the covariance matrix of the sound signal, yi(t) represents a sound signal, and,
Figure BDA00029883552300000911
is a direction vector of the sound signal,
Figure BDA00029883552300000912
is a direction vector
Figure BDA00029883552300000913
Projection operator for generating a space, n representing the number of sound signals in a sound signal sample, yi H(t) denotes the sound signal yi(t) conjugate transpose of matrix.
Fig. 3 schematically shows a flow chart of an MPSO-MLE algorithm of the method for locating an abnormal sound source of substation equipment according to an embodiment of the present invention.
In the invention, the method for positioning the abnormal sound source of the substation equipment solves the problem of large calculation amount of a Maximum Likelihood Estimation (MLE) algorithm by utilizing and introducing an improved particle swarm optimization (MPSO) of a variation idea.
As shown in fig. 3, in the present embodiment, the method for positioning an abnormal sound source of substation equipment according to the present invention employs an MPSO-MLE algorithm, which is an algorithm using an improved particle swarm algorithm (MPSO) and a Maximum Likelihood Estimation (MLE) algorithm.
After the likelihood function of the sound signal sample of the transformer substation is established, the maximum value of the likelihood function of the sound signal sample of the transformer substation can be obtained by adopting an improved particle swarm optimization (MPSO) method, and the estimated value of the direction angle D corresponding to the maximum value is obtained based on the maximum value so as to position the abnormal sound source of the equipment of the transformer substation.
It should be noted that a Particle Swarm Optimization (PSO) is a swarm intelligence optimization algorithm for simulating foraging behavior of a bird swarm, and has the advantages of strong search function, good optimization effect, strong convergence performance, and the like. The search space of the particle swarm optimization is assumed to represent the pitch angle theta of the sound signal and the azimuth angle of the sound signal respectively
Figure BDA0002988355230000101
The particles in the particle swarm algorithm correspond to different pitch angles theta and azimuth angles
Figure BDA0002988355230000102
Assuming that the number of particles in the particle group is m, the d-dimensional position of the ith particle is represented by xidThe velocity is denoted by vidI is 1,2, …, m; d is 1, 2; then the individual optimal solution and the population optimal solution of the particle can be searched based on the following calculation formula of the velocity and the position of the particle:
vid(t+1)＝ωvid(t)+c1r1(pid(t)-xid(t))+c2r2(pgd(t)-xid(t)) (12)
xid(t+1)＝xid(t)+vid(t+1) (13)
where t represents the current number of searches, xidIndicating the d-dimensional position, v, of the ith particleidRepresents the d-dimensional velocity of the ith particle, d is 1, 2; i is 1,2, … …, m; m represents the number of particles in the particle group; c. C1And c2Acceleration constants are respectively cognitive parameters and social parameters; r is1And r2Represents the interval [0,1]A random number of (c); p is a radical ofidAnd pgdRespectively representing an individual optimal solution and a group optimal solution of the particle swarm in historical search; ω is a random weight.
In the invention, in order to avoid the algorithm from falling into local optimum and improve the search efficiency, the invention adopts the random weight omega. The random weight omega is larger, which is beneficial to global search and is suitable for being used in the initial search stage; the random weight omega is smaller, which is beneficial to local search and is suitable for being used in the later stage of search.
The expression of the random weight ω can be as follows:
Figure BDA0002988355230000103
in the above formula (14), N (0,1) is a random number following a normal distribution; mu.smaxAnd muminThe maximum value and the minimum value of the inertia weight mu are respectively; σ is a random weight variance; r is a random number uniformly distributed between 0 and 1.
According to the invention, by using the variation thought in the genetic algorithm, in the process that the particle swarm searches the individual optimal solution and the swarm optimal solution in the historical search, after the optimal position is updated each time, the individual optimal position of the particle can be endowed with 1 random trust value P. When the random trust value P is smaller than the variation trust threshold, when the random value P is smaller than the trust threshold, the variation can be implemented, the fitness value of the new particle position is calculated, if the fitness value is better than the position before the variation operation, the position of the particle individual after the variation is used for replacing the position of the particle individual before the variation operation, otherwise, the position of the particle individual is not changed.
It should be noted that the method adopted by the present invention does not change the current ordered optimization process, but can make the particles jump out and stop, thereby improving the global search capability.
Accordingly, in the present invention, the mutation may be implemented using the cuckoo algorithm. When the random confidence value P is less than the variation confidence threshold value PthWhen, i.e. P < PthAnd then, carrying out variation by adopting a cuckoo algorithm to obtain:
pbest_id(t)＝pid(t)+r(pkd(t)-pld(t)) (15)
in the above formula (15), pbest_idIs a mutated individualA location; p is the individual position of the particle before mutation, and subscripts k and l represent the randomly drawn particle number; d is the particle dimension; pthReflecting the variation degree of the optimal position of the particle, wherein the larger variation degree can lead to the larger search range of the algorithm, and the smaller variation degree means the small variation range of the particle, thus leading the variation operation to lose significance and adjusting the threshold value.
Fig. 4 schematically shows a system configuration diagram of a substation equipment abnormal sound source localization system according to an embodiment of the present invention.
In order to better illustrate the effectiveness of the method for positioning the abnormal sound source of the transformer substation equipment, the method acquires a sound signal sample of a certain transformer substation through field test for further explanation.
In the invention, the abnormal sound source positioning system of the transformer substation equipment is adopted, and can be used for implementing the abnormal sound source positioning method of the transformer substation equipment.
As shown in fig. 4, in the present embodiment, the system for locating an abnormal sound source of substation equipment may include: an acoustic transducer array and a processing and localization module. Wherein the processing and positioning module may comprise: the device comprises a preprocessing unit, a synchronous acquisition and sending device, a computer and a battery management module.
Correspondingly, the sound sensor array is a 4 multiplied by 4 uniform area array with array element spacing of 7cm, and can collect sound signal samples of the transformer substation; the processing and positioning module can establish a likelihood function of the sound signal sample of the transformer substation based on the sound signal sample of the transformer substation, then the maximum value of the likelihood function of the sound signal sample of the transformer substation is obtained by adopting an improved particle swarm algorithm, and an estimated value of a direction angle D corresponding to the maximum value is obtained based on the maximum value so as to position an abnormal sound source of the equipment of the transformer substation.
It should be noted that, in the present embodiment, a tester may discharge electricity by using a handheld electrostatic gun at a distance of 3m from the acoustic sensor array in the system according to the present invention in the field, and a time-domain acoustic signal generated by the discharge of the handheld electrostatic gun is shown in fig. 5.
Fig. 5 schematically shows a time-domain sound signal diagram collected by a sound sensor array of the substation equipment abnormal sound source positioning system according to an embodiment of the invention.
As shown in fig. 5, the time domain sound signals respectively representing the sound signals collected by the 16-way sound sensor in fig. 5 can be respectively represented as CH1-CH 16. The signal-to-noise ratio of the field test discharge sound signal is about 5 dB.
Correspondingly, after the collection of the sound signal samples of the transformer substation is completed, a likelihood function of the sound signal samples of the transformer substation can be established based on a Maximum Likelihood Estimation (MLE) method, and then an improved particle swarm optimization (MPSO) is adopted in a matching mode to position the abnormal sound source of the transformer substation equipment.
In order to prove the superiority of the MPSO-MLE algorithm adopted by the invention, in the invention, the transformer substation sound signal samples collected by the sound sensor array formed by 16 sound sensors can be tested by three completely different methods, namely a PSO-MLE algorithm (a standard PSO algorithm), an MPSO-MLE algorithm (the algorithm of the invention) and a multiple signal classification (MUSIC) algorithm, so as to locate the abnormal sound source of the transformer substation equipment and obtain the corresponding test data result. The results of the relevant test data can be shown in table 1 below.
Table 1 lists the results of on-site sound source localization using the PSO-MLE algorithm, the MPSO-MLE algorithm, and the MUSIC algorithm, respectively.
Table 1.
Figure BDA0002988355230000121
As can be seen from table 1, the mean square error of positioning the on-site sound source by using the MUSIC algorithm is about 1.5 °; the mean square error of positioning a field sound source by adopting a PSO-MLE algorithm is about 1.8 degrees, because the standard PSO method is easy to fall into local optimum and cannot jump out, and the final result error is larger. Correspondingly, the mean square error of the MPSO-MLE algorithm adopted by the invention for positioning the site sound source is about 1.1 degrees, and the convergence speed is higher, so that the MPSO-MLE algorithm adopted by the invention not only can reduce the calculation amount, but also can effectively improve the positioning accuracy.
In conclusion, the abnormal sound source positioning method for the transformer substation equipment can effectively solve the problem of large calculation amount when a Maximum Likelihood Estimation (MLE) method is adopted by introducing a variation thought through an improved particle swarm optimization (MPSO) method, can further quickly and accurately position the abnormal sound source of the transformer substation equipment, and has the advantages of high positioning accuracy, wide applicability, good popularization prospect and high application value.
Correspondingly, the transformer substation equipment abnormal sound source positioning system can be used for implementing the transformer substation equipment abnormal sound source positioning method, and has the advantages and beneficial effects.
It should be noted that the prior art in the protection scope of the present invention is not limited to the examples given in the present application, and all the prior art which is not inconsistent with the technical scheme of the present invention, including but not limited to the prior patent documents, the prior publications and the like, can be included in the protection scope of the present invention.
In addition, the combination of the features in the present application is not limited to the combination described in the claims of the present application or the combination described in the embodiments, and all the features described in the present application may be freely combined or combined in any manner unless contradictory to each other.
It should also be noted that the above-mentioned embodiments are only specific embodiments of the present invention. It is apparent that the present invention is not limited to the above embodiments and similar changes or modifications can be easily made by those skilled in the art from the disclosure of the present invention and shall fall within the scope of the present invention.

Claims (10)

1. A method for positioning abnormal sound source of transformer substation equipment is characterized by comprising the following steps:
(1) collecting sound signal sample Y of transformer substation ═ Y1,y2,…,yn]；
(2) Establishing a likelihood function of a transformer substation sound signal sample:
(3) and solving the maximum value of the likelihood function of the sound signal sample of the transformer substation by adopting an improved particle swarm algorithm, and obtaining the estimated value of the direction angle D corresponding to the maximum value based on the maximum value so as to position the abnormal sound source of the transformer substation equipment.
2. The substation equipment abnormal sound source localization method of claim 1, wherein for a sound signal sample Y ═ Y1,y2,…,yn]The likelihood function of the substation sound signal sample is configured as:
Figure FDA0002988355220000011
Figure FDA0002988355220000012
where theta represents the pitch angle of the sound signal,
Figure FDA0002988355220000013
representing the azimuth angle, σ, of the acoustic signal2Represents the variance of the noise signal, tr (-) represents the trace of the matrix; r represents the covariance matrix of the sound signal, yi(t) represents a sound signal, and,
Figure FDA0002988355220000014
is a direction vector of the sound signal,
Figure FDA0002988355220000015
is a direction vector
Figure FDA0002988355220000016
Projection operator for generating a space, n representing the number of sound signals in a sound signal sample, yi H(t) denotes the sound signal yi(t) conjugate transpose of matrix.
3. The method for locating an abnormal sound source of substation equipment according to claim 2, wherein in step (3), the estimated value of the sound signal direction angle D is obtained according to the following formula:
Figure FDA0002988355220000017
4. the method for locating the abnormal sound source of the substation equipment according to any one of claims 1 to 3, wherein the step (3) of adopting the improved particle swarm algorithm to obtain the maximum value of the likelihood function of the sound signal sample of the substation comprises the steps of:
taking the likelihood function as a fitness function of the particle swarm algorithm to calculate the value of the fitness function; wherein the search space of the particle swarm optimization is set as a pitch angle theta and an azimuth angle of the sound signal respectively representing the sound signal
Figure FDA0002988355220000018
The particles in the particle swarm algorithm correspond to different pitch angles theta and azimuth angles
Figure FDA0002988355220000021
Searching individual optimal solution and group optimal solution of the particles based on the following calculation formula of the speed and the position of the particles:
vid(t+1)＝ωvid(t)+c1r1(pid(t)-xid(t))+c2r2(pgd(t)-xid(t))
xid(t+1)＝xid(t)+vid(t+1)
where t represents the current number of searches, xidIndicating the d-dimensional position, v, of the ith particleidRepresents the d-dimensional velocity of the ith particle, d is 1, 2; i is 1,2, … …, m; m represents the number of particles in the particle group; c. C1And c2Acceleration constants are respectively cognitive parameters and social parameters; r is1And r2Represents the interval [0,1]A random number of (c); p is a radical ofidAnd pgdRespectively representing an individual optimal solution and a group optimal solution of the particle swarm in historical search; omega is a random weight;
and continuously updating the particle speed and the particle position based on the particle individual optimal solution and the group optimal solution, endowing the particle individual optimal position with 1 random trust value P after the optimal particle position is updated every time, performing variation when the random trust value P is smaller than a variation trust threshold, calculating a fitness function value of the new particle individual position, and replacing the previous particle individual position with the varied particle individual position if the fitness function value is superior to the position before the variation operation, otherwise, keeping the particle individual position unchanged.
5. The method for positioning the abnormal sound source of the substation equipment according to claim 4, wherein the expression of the random weight ω is:
Figure FDA0002988355220000022
in the formula, N (0,1) is a random number following a normal distribution; mu.smaxAnd muminThe maximum value and the minimum value of the inertia weight mu are respectively; σ is a random weight variance; r is a random number uniformly distributed between 0 and 1.
6. The method for positioning the abnormal sound source of the substation equipment according to claim 4, wherein the variation is implemented by adopting a cuckoo algorithm.
7. The utility model provides a transformer substation equipment unusual sound source positioning system which characterized in that includes:
the sound sensor array is used for collecting a sound signal sample of the transformer substation;
a processing and positioning module that performs the steps of:
(1) based on sound signal sample Y of transformer substation ═ Y1,y2,…,yn]Establishing a likelihood function of a transformer substation sound signal sample;
(2) and solving the maximum value of the likelihood function of the sound signal sample of the transformer substation by adopting an improved particle swarm algorithm, and obtaining the estimated value of the direction angle D corresponding to the maximum value based on the maximum value so as to position the abnormal sound source of the transformer substation equipment.
8. The substation equipment abnormal sound source localization system of claim 7, wherein for a sound signal sample Y ═ Y1,y2,…,yn]The likelihood function of the substation sound signal sample is configured as:
Figure FDA0002988355220000031
Figure FDA0002988355220000032
where theta represents the pitch angle of the sound signal,
Figure FDA0002988355220000033
representing the azimuth angle, σ, of the acoustic signal2Represents the variance of the noise signal, tr (-) represents the trace of the matrix; r represents the covariance matrix of the sound signal, yi(t) represents a sound signal, and,
Figure FDA0002988355220000034
is a direction vector of the sound signal,
Figure FDA0002988355220000035
is a direction vector
Figure FDA0002988355220000036
Projection operator for generating a space, n representing the number of sound signals in a sound signal sample, yi H(t) denotes the sound signal yi(t) conjugate transpose of matrix.
9. The substation equipment abnormal sound source localization system according to claim 7, wherein in step (2), the estimated value of the sound signal direction angle D is obtained according to the following formula:
Figure FDA0002988355220000037
10. the substation equipment abnormal sound source positioning system according to any one of claims 7 to 9, wherein the step (2) of adopting the improved particle swarm optimization to find the maximum value of the likelihood function of the substation sound signal sample comprises the steps of:
taking the likelihood function as a fitness function of the particle swarm algorithm to calculate the value of the fitness function; wherein the search space of the particle swarm optimization is set as a pitch angle theta and an azimuth angle of the sound signal respectively representing the sound signal
Figure FDA0002988355220000038
The particles in the particle swarm algorithm correspond to different pitch angles theta and azimuth angles
Figure FDA0002988355220000039
Searching individual optimal solution and group optimal solution of the particles based on the following calculation formula of the speed and the position of the particles:
vid(t+1)＝ωvid(t)+c1r1(pid(t)-xid(t))+c2r2(pgd(t)-xid(t))
xid(t+1)＝xid(t)+vid(t+1)
where t represents the current number of searches, xidIndicating the d-dimensional position, v, of the ith particleidRepresenting the ith particleD-dimension speed, d ═ 1, 2; i is 1,2, … …, m; m represents the number of particles in the particle group; c. C1And c2Acceleration constants are respectively cognitive parameters and social parameters; r is1And r2Represents the interval [0,1]A random number of (c); p is a radical ofidAnd pgdRespectively representing an individual optimal solution and a group optimal solution of the particle swarm in historical search; omega is a random weight;
and continuously updating the particle speed and the particle position based on the particle individual optimal solution and the group optimal solution, endowing the particle individual optimal position with 1 random trust value P after the optimal particle position is updated every time, performing variation when the random trust value P is smaller than a variation trust threshold, calculating a fitness function value of the new particle individual position, and replacing the previous particle individual position with the varied particle individual position if the fitness function value is superior to the position before the variation operation, otherwise, keeping the particle individual position unchanged.
CN202110308028.1A 2021-03-23 2021-03-23 Method and system for positioning abnormal sound source of transformer substation equipment Pending CN112883628A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110308028.1A CN112883628A (en) 2021-03-23 2021-03-23 Method and system for positioning abnormal sound source of transformer substation equipment

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110308028.1A CN112883628A (en) 2021-03-23 2021-03-23 Method and system for positioning abnormal sound source of transformer substation equipment

Publications (1)

Publication Number Publication Date
CN112883628A true CN112883628A (en) 2021-06-01

Family

ID=76041159

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110308028.1A Pending CN112883628A (en) 2021-03-23 2021-03-23 Method and system for positioning abnormal sound source of transformer substation equipment

Country Status (1)

Country Link
CN (1) CN112883628A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114783462A (en) * 2022-05-11 2022-07-22 安徽理工大学 Mine hoist fault source positioning analysis method based on CS-MUSIC

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20170249955A1 (en) * 2014-05-06 2017-08-31 State Grid Jiangsu Electric Power Company Nanjing Power Supply Company A transformer noise suppression method
CN109752633A (en) * 2019-01-28 2019-05-14 国网山东省电力公司日照供电公司 The method and system that a kind of pair of transformer station partial discharge signals are positioned

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20170249955A1 (en) * 2014-05-06 2017-08-31 State Grid Jiangsu Electric Power Company Nanjing Power Supply Company A transformer noise suppression method
CN109752633A (en) * 2019-01-28 2019-05-14 国网山东省电力公司日照供电公司 The method and system that a kind of pair of transformer station partial discharge signals are positioned

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
张瑶 等: ""基于MPSO-MLE的变电站设备异常声源定位方法"", 《高电压技术》 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114783462A (en) * 2022-05-11 2022-07-22 安徽理工大学 Mine hoist fault source positioning analysis method based on CS-MUSIC

Similar Documents

Publication Publication Date Title
Sheng et al. Maximum likelihood multiple-source localization using acoustic energy measurements with wireless sensor networks
CN102520389B (en) Two-dimensional spatial-spectrum spectral-peak searching method
CN103529364B (en) Many Partial Discharge Sources localization method in oil based on ultrasonic thinned array sensor
CN102147458B (en) Method and device for estimating direction of arrival (DOA) of broadband sound source
WO2020244359A1 (en) Sound source position estimation method, readable storage medium and computer device
CN102183710A (en) Method and system for identifying numbers of partial discharge sources of transformers
CN108549052A (en) A kind of humorous domain puppet sound intensity sound localization method of circle of time-frequency-spatial domain joint weighting
CN108896962A (en) Iteration localization method based on sound position fingerprint
CN112883628A (en) Method and system for positioning abnormal sound source of transformer substation equipment
CN106529700A (en) Space-time correlation based construction method and system of wind power plant prediction model
CN110095755B (en) Sound source positioning method
Noel et al. Improved maximum likelihood estimation of target position in wireless sensor networks using particle swarm optimization
CN113050037B (en) Transformer substation equipment abnormal sound source positioning method and system
Liu et al. DOA estimation of underwater acoustic signals based on PCA-kNN algorithm
CN106646347A (en) Ecological niche differential evolution-based multi-signal classification spectral peak search method
JP2004144579A (en) Sound source survey method and device
CN114595742B (en) Fuel cell fault data sampling method and system
CN112906579B (en) Sea clutter weak target classification method and system based on K-means clustering and SVM
Feng et al. Optimization algorithm improvement of association rule mining based on particle swarm optimization
CN111505581B (en) Passive target detection method based on distributed sensor nodes
Zhao et al. Large array DOA estimation based on extreme learning machine and random matrix theory
CN110161459B (en) Rapid positioning method for amplitude modulation sound source
CN113702923A (en) Radar clutter simulation method and electronic equipment
Liang et al. A DOA Estimation algorithm for the vertical line array of vector hydrophone based on data fusion method
CN110287514A (en) Hypervelocity impact source intelligent locating method based on vibration signal processing

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
WD01 Invention patent application deemed withdrawn after publication

Application publication date: 20210601

WD01 Invention patent application deemed withdrawn after publication