CN112461919A - System and method for detecting physical and chemical properties of transformer oil by applying multi-frequency ultrasonic technology - Google Patents
System and method for detecting physical and chemical properties of transformer oil by applying multi-frequency ultrasonic technology Download PDFInfo
- Publication number
- CN112461919A CN112461919A CN202011249451.0A CN202011249451A CN112461919A CN 112461919 A CN112461919 A CN 112461919A CN 202011249451 A CN202011249451 A CN 202011249451A CN 112461919 A CN112461919 A CN 112461919A
- Authority
- CN
- China
- Prior art keywords
- wolf
- transformer oil
- ultrasonic
- physical
- parameters
- 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
Links
- 239000000126 substance Substances 0.000 title claims abstract description 42
- 238000000034 method Methods 0.000 title claims abstract description 20
- 238000005516 engineering process Methods 0.000 title claims abstract description 10
- 241000282461 Canis lupus Species 0.000 claims abstract description 74
- 238000001514 detection method Methods 0.000 claims abstract description 67
- 238000013528 artificial neural network Methods 0.000 claims abstract description 18
- 238000012545 processing Methods 0.000 claims description 16
- 230000003321 amplification Effects 0.000 claims description 11
- 238000003199 nucleic acid amplification method Methods 0.000 claims description 11
- 238000006243 chemical reaction Methods 0.000 claims description 9
- 239000000523 sample Substances 0.000 claims description 8
- 238000012549 training Methods 0.000 claims description 8
- 238000001914 filtration Methods 0.000 claims description 6
- 239000011159 matrix material Substances 0.000 claims description 4
- 238000013507 mapping Methods 0.000 claims description 3
- 238000007781 pre-processing Methods 0.000 claims description 3
- 238000012360 testing method Methods 0.000 claims description 3
- 230000003247 decreasing effect Effects 0.000 claims description 2
- 230000032683 aging Effects 0.000 abstract description 4
- 230000015556 catabolic process Effects 0.000 description 7
- 239000002245 particle Substances 0.000 description 5
- 239000002253 acid Substances 0.000 description 4
- 238000006731 degradation reaction Methods 0.000 description 3
- 239000012535 impurity Substances 0.000 description 3
- 238000005259 measurement Methods 0.000 description 3
- 238000010521 absorption reaction Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 2
- 238000001816 cooling Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000010292 electrical insulation Methods 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 238000004817 gas chromatography Methods 0.000 description 2
- 238000009413 insulation Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000004566 IR spectroscopy Methods 0.000 description 1
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008033 biological extinction Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 239000000084 colloidal system Substances 0.000 description 1
- 230000006866 deterioration Effects 0.000 description 1
- 239000003344 environmental pollutant Substances 0.000 description 1
- 239000000835 fiber Substances 0.000 description 1
- 230000017525 heat dissipation Effects 0.000 description 1
- 238000001285 laser absorption spectroscopy Methods 0.000 description 1
- 239000007788 liquid Substances 0.000 description 1
- 238000009659 non-destructive testing Methods 0.000 description 1
- 239000001301 oxygen Substances 0.000 description 1
- 229910052760 oxygen Inorganic materials 0.000 description 1
- 231100000719 pollutant Toxicity 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 230000000644 propagated effect Effects 0.000 description 1
- 230000001902 propagating effect Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 230000035882 stress Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 238000002604 ultrasonography Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N29/00—Investigating or analysing materials by the use of ultrasonic, sonic or infrasonic waves; Visualisation of the interior of objects by transmitting ultrasonic or sonic waves through the object
- G01N29/02—Analysing fluids
- G01N29/032—Analysing fluids by measuring attenuation of acoustic waves
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/12—Testing dielectric strength or breakdown voltage ; Testing or monitoring effectiveness or level of insulation, e.g. of a cable or of an apparatus, for example using partial discharge measurements; Electrostatic testing
- G01R31/1209—Testing dielectric strength or breakdown voltage ; Testing or monitoring effectiveness or level of insulation, e.g. of a cable or of an apparatus, for example using partial discharge measurements; Electrostatic testing using acoustic measurements
-
- 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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computing Systems (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- Computational Linguistics (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- General Engineering & Computer Science (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- Data Mining & Analysis (AREA)
- Artificial Intelligence (AREA)
- Acoustics & Sound (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Investigating Or Analyzing Materials By The Use Of Ultrasonic Waves (AREA)
Abstract
The invention relates to a system and a method for detecting physical and chemical properties of transformer oil by applying a multi-frequency ultrasonic technology, and belongs to the field of transformer oil aging degree detection. The method establishes the correlation between the propagation characteristic parameters of the ultrasonic waves with different frequencies in the transformer oil and the physical and chemical performance parameters of the transformer oil, thereby establishing a detection system of the physical and chemical performance parameters of the transformer oil. The weight and function parameters of the back propagation neural network are optimized by adopting an improved wolf algorithm, and the improved neural network is trained, so that a detection system is established to quickly and accurately acquire the physical and chemical performance parameters of the transformer oil. The invention can realize nondestructive detection and simultaneously avoid the interference of external environment on the detection result to the maximum extent.
Description
Technical Field
The invention belongs to the field of transformer oil aging degree detection, and relates to a method and a system for detecting physical and chemical properties of transformer oil based on a multi-frequency ultrasonic technology.
Background
With the gradual expansion of the scale of the intelligent power grid, the requirements on the power supply quality and the safe and reliable operation of the power grid are higher and higher. The transformer, one of the most core devices in the power system, undertakes the transmission and transformation tasks of the whole power grid, and the stable operation state of the transformer is related to the safe operation of the power system. At present, most of power transformers in a power grid are oil-immersed power transformers, and transformer oil is an indispensable part of the oil-immersed power transformers and mainly plays roles in insulation, heat dissipation, cooling and arc extinction in the running process of the transformers. However, in the daily operation process, the transformer oil is affected by light, heat, oxygen, machinery and various environmental factors to cause gradual degradation of various performance indexes, and in the severe case, serious accidents such as damage or burnout of transformer equipment can be caused, so that the safe and reliable operation of the whole power supply system is damaged. Therefore, the transformer oil is detected so as to find the problems of oil quality degradation, insulation performance degradation and the like in time, and then the transformer is filtered or replaced, so that the transformer oil detection device has great significance for ensuring safe and stable operation of the transformer.
At present, infrared spectroscopy and gas chromatography are used as detection methods for transformer oil in the power industry, and the physicochemical properties of the transformer oil are detected by a certain method, but the gas chromatography is easily interfered by environmental factors such as temperature and the like, and can generate certain influence on the detection result, while the laser absorption spectroscopy has weak light source intensity and low measurement precision.
Recently, the physical and chemical properties of the transformer oil are mainly detected in the power industry by adopting detection equipment for various electrical insulation parameters to separately detect various electrical insulation parameters representing the quality of the transformer oil. Although the accuracy of detection is achieved, each parameter requires different detection equipment, so that the detection process is complicated and the equipment is expensive, thereby increasing the economic cost of detection. In addition, the method needs to take the oil sample out of the transformer in operation and send the oil sample to an external laboratory for detection, so that the problems of easy pollution and the like in sampling and storage exist.
Therefore, a need exists for a transformer oil detection system that is less susceptible to interference and can perform nondestructive testing.
Disclosure of Invention
In view of the above, the present invention provides a method and a system for detecting physical and chemical properties of transformer oil based on a multi-frequency ultrasonic technology, which can realize nondestructive detection while ensuring the accuracy of detection.
In order to achieve the purpose, the invention provides the following technical scheme:
1. a method for detecting physical and chemical properties of transformer oil by applying a multi-frequency ultrasonic technology specifically comprises the following steps:
s1: acquiring ultrasonic wave propagation characteristic parameters in the transformer oil;
s2: taking the obtained actual values of the ultrasonic wave propagation characteristic parameters and the transformer oil physical and chemical property parameters as training samples, training the samples by using an IGWO-BPNN algorithm, and establishing an IGWO-BPNN detection model of the transformer oil physical and chemical property parameters; wherein, the IGWO-BPNN algorithm adopts an improved Grey wolf algorithm (IGWO) to optimize the weight and the function parameters of a Back Propagation Neural Network (BPNN);
s3: and analyzing the data obtained by each detection by using an IGWO-BPNN detection model, and giving a detection result of the physical and chemical performance parameters of the transformer oil.
Further, in step S2, the multi-frequency ultrasonic detection technique principle: when ultrasonic waves propagate in a medium, energy gradually weakens along with the increase of the propagation distance, and when the ultrasonic waves propagate in a non-uniform medium, attenuation is mainly scattering attenuation and absorption attenuation. The total attenuation coefficient alpha of the ultrasonic wave propagating in the inhomogeneous medium can be approximated by
Wherein alpha isηIs the absorption attenuation coefficient; alpha is alphasIs the scattering attenuation coefficient; f is the ultrasonic frequency; c is the ultrasonic propagation speed; ρ is the liquid medium density; eta is viscosity coefficient; approximately regarding the scattering particles as rigid spheres, wherein r is the radius of the spheres; n is the number of scattering particles per unit volume.
The transformer oil is an inhomogeneous medium containing large-size particles, and as can be seen from the above formula, when ultrasonic waves are propagated in the transformer oil, the frequency of the ultrasonic waves, the size and the number of the particles in the transformer oil, the performance of the transformer oil and the like all affect the propagation characteristics of the ultrasonic waves, such as the wave speed, the attenuation coefficient and the like. Therefore, the correlation between the propagation characteristic parameters of the ultrasonic waves with different frequencies in the transformer oil and the physical and chemical performance parameters of the transformer oil can be established, and a detection system of the physical and chemical performance parameters of the transformer oil is established. In order to establish the detection system, the weight and function parameters of the back propagation neural network are optimized by adopting an improved wolf algorithm, and the optimized neural network is trained, so that the detection system is established to quickly and accurately acquire the physical and chemical performance parameters of the transformer oil.
The improved grayish wolf algorithm specifically comprises: the gray wolf algorithm divides the whole wolf group into four levels according to the fitness, the first 3 of the fitness are respectively the optimal wolf alpha, the second best wolf beta and the third best wolf delta, the rest wolfs are omega, the gray wolfs are close to and surround the prey in the process of predation, and correspondingly, the distance between the gray wolf individual and the prey is determined in the Improved Gray Wolf (IGWO) algorithm and the positions of the gray wolfs are updated;
D＝|C·XP(t)-X(t)|
X(t+1)＝XP(t)-A·D
wherein t is the current iteration number, XPIs the position of the prey, X (t) is the current gray wolf position, and X (t +1) is the updated position of the gray wolf; c is a wobble factor, C is 2r1，r1Is [0, 1 ]]Random number of inner；A＝2ar2 2-a，r2Is [0, 1 ]]A is a convergence factor, and a is linearly decreased from 2 to 0 along with the increase of the iteration times;
wherein, tmaxIs the maximum iteration number;
in the hunting process, the optimal wolf alpha carries the leads beta and delta to hunt the prey, and the direction of the prey is as follows:
firstly, calculating the direction of an individual hunting target in the wolf group, and mathematically expressing the following steps:
wherein D isα、Dβ、DδThe distances between alpha, beta and delta of the wolf individual and the prey are respectively; c1、C2、C3Are respectively corresponding swing factors; xα(t)、Xβ(t)、Xδ(t) the current positions of alpha, beta and delta of the wolf individual, Xα(t+1)、Xβ(t+1)、Xδ(t +1) are the updated positions of alpha, beta and delta of the wolf individual respectively; a. the1、A2、A3Respectively corresponding coefficients;
then, the direction of the individual moving to the prey is calculated to obtain the updated position X of the wolf individual omegaω(t +1), the mathematical expression is as follows:
Xω(t+1)＝[Xα(t+1)+Xβ(t+1)+Xδ(t+1)]/3
wherein, Xω(t) and Xω(t +1) are the current position and the updated position of the gray wolf ω, respectively.
Further, in step S2, an improved grayish wolf algorithm is used to optimize the weight and function parameters of the back propagation neural network, which specifically includes the following steps:
1) constructing a back propagation neural network;
2) initializing parameters of a back propagation neural network; determining the scale and the maximum iteration number of the gray wolf population, determining the dimensionality of the gray wolf individual position information and the upper and lower boundaries of the gray wolf dimensionality, and randomly initializing the gray wolf position;
3) selecting a fitness function, calculating the fitness, and selecting an optimal wolf alpha, a suboptimal wolf beta and a third best wolf delta;
4) updating the position information omega of the remaining gray wolfs and updating the parameters A, C and a;
5) judging whether the set maximum iteration number or the set error is reached, otherwise, repeating the step 3) and the step 4) until the condition is met;
6) outputting the position of the optimal wolf alpha, mapping the optimal wolf alpha into a weight matrix which is used as the weight from a BPNN neural network hidden layer to an output layer, and realizing an IGWO-BPNN detection model;
7) and preprocessing the data and inputting the preprocessed data into a trained IGWO-BPNN detection model to obtain a prediction result.
2. A system for testing physical and chemical properties of transformer oil using multi-frequency ultrasound technology, comprising: the device comprises a multi-frequency ultrasonic transmitting and receiving module 1, a data acquisition module 2 and an upper computer 3.
The multi-frequency ultrasonic wave transmitting and receiving module 1 is used for transmitting and receiving ultrasonic echo signals.
The data acquisition module 2 is used for acquiring and processing ultrasonic echo signals.
The upper computer 3 analyzes and calculates the acquired ultrasonic wave propagation characteristic parameters, uses the acquired parameters and the actual values of the transformer oil physical and chemical property parameters as training samples, trains the samples by using an IGWO-BPNN algorithm, and establishes an IGWO-BPNN detection model of the transformer oil physical and chemical property parameters; and finally, analyzing the data obtained by each detection by using a detection model, and giving a detection result of the physical and chemical performance parameters of the transformer oil.
Further, the multi-frequency ultrasonic transmitting and receiving module includes a multi-frequency ultrasonic transceiver circuit 11 and an ultrasonic transducer (probe array) 12; the multi-frequency ultrasonic transceiver circuit 11 comprises an ultrasonic emission driving circuit 101 and an ultrasonic echo processing circuit 102;
the ultrasonic emission driving circuit 101 is used for generating square wave pulse signals with different frequencies and driving the ultrasonic transducer to emit ultrasonic signals; the ultrasonic transducer 12 converts an electric signal and an ultrasonic signal to each other; the ultrasonic echo processing circuit 102 performs primary filtering, amplification and other processing on the ultrasonic waves and then transmits the ultrasonic waves to the data acquisition module 2.
Further, the data acquisition module comprises a signal amplification circuit 21, a filter circuit 22 and a digital-to-analog conversion circuit 23; the signal amplification circuit 21 is used for amplifying ultrasonic echo signals received by the multi-frequency ultrasonic wave transmitting and receiving module; the filter circuit 22 is configured to filter interference signals of various other frequencies in the ultrasonic echo signal, so as to obtain more accurate analog quantities of the multi-frequency ultrasonic signal; the digital-to-analog conversion circuit 23 is used for converting the multi-frequency ultrasonic analog quantity into a digital signal and transmitting the digital signal to the upper computer.
The invention has the beneficial effects that: the multi-frequency ultrasonic detection technology adopted by the invention can complete the emission and the reception of 3 different-frequency ultrasonic waves, complete the acquisition of echo signals, obtain parameters such as sound velocity, attenuation coefficient and the like of the ultrasonic waves through the processing of the echo signals, establish the correlation between the parameters and transformer oil parameters, and further detect the physicochemical properties of the transformer oil through detecting the parameters of the multi-frequency ultrasonic waves passing through the transformer oil. The interference of an external environment to a detection result is avoided to the greatest extent, and compared with the prior art, the experimental time and the experimental cost are greatly saved.
Additional advantages, objects, and features of the invention will be set forth in part in the description which follows and in part will become apparent to those having ordinary skill in the art upon examination of the following or may be learned from practice of the invention. The objectives and other advantages of the invention may be realized and attained by the means of the instrumentalities and combinations particularly pointed out hereinafter.
Drawings
For the purposes of promoting a better understanding of the objects, aspects and advantages of the invention, reference will now be made to the following detailed description taken in conjunction with the accompanying drawings in which:
FIG. 1 is a block diagram of a multi-frequency ultrasonic testing system according to the present invention;
FIG. 2 is a block diagram of a multi-frequency ultrasonic detection system according to the present invention;
FIG. 3 is a flow chart of the IGWO-BPNN algorithm employed in the present invention;
FIG. 4 is a graph showing the detection results of breakdown voltage in example 1;
FIG. 5 is a graph showing the results of detection of dielectric loss tangent in example 1;
FIG. 6 is a graph showing the results of kinematic viscosity measurements in example 1;
FIG. 7 is a graph showing the results of detection of an acid value in example 1;
reference numerals: 1-a multi-frequency ultrasonic transmitting and receiving module, 2-a data acquisition module and 3-an upper computer; 11-a multi-frequency ultrasonic wave receiving and transmitting circuit, 101-an ultrasonic wave transmitting driving circuit and 102-an ultrasonic wave echo processing circuit; 12-ultrasonic transducer, 21-signal amplifying circuit, 22-filter circuit and 23-digital-to-analog conversion circuit.
Detailed Description
The embodiments of the present invention are described below with reference to specific embodiments, and other advantages and effects of the present invention will be easily understood by those skilled in the art from the disclosure of the present specification. The invention is capable of other and different embodiments and of being practiced or of being carried out in various ways, and its several details are capable of modification in various respects, all without departing from the spirit and scope of the present invention. It should be noted that the drawings provided in the following embodiments are only for illustrating the basic idea of the present invention in a schematic way, and the features in the following embodiments and examples may be combined with each other without conflict.
Referring to fig. 1 to 3, the present invention preferably discloses a system for detecting physical and chemical properties of transformer oil by using multi-frequency ultrasonic technology, as shown in fig. 1 and 2, the system includes: the system comprises a multi-frequency ultrasonic transmitting and receiving module 1, a data acquisition module 2 and an upper computer 3;
the multi-frequency ultrasonic transmitting and receiving module 1 is mainly used for transmitting and recovering ultrasonic signals, and comprises a multi-frequency ultrasonic receiving and transmitting circuit 11 and an ultrasonic transducer 12. The multi-frequency ultrasonic transceiver circuit 11 includes an ultrasonic emission driving circuit 101 and an ultrasonic echo processing circuit 102, wherein the ultrasonic emission driving circuit 101 generates square wave pulse signals with different frequencies and drives the ultrasonic transducer 12 to emit ultrasonic signals, the ultrasonic transducer 12 can convert the electric signals and the ultrasonic signals into each other, and the ultrasonic echo processing circuit 102 performs preliminary filtering, amplification and other processing on the ultrasonic waves and then transmits the ultrasonic waves to the data acquisition module 2.
The data acquisition module comprises a signal amplification circuit 21, a filter circuit 22 and a digital-to-analog conversion circuit 23, wherein the signal amplification circuit 21 amplifies the signal transmitted back by the controller, the filter circuit 22 filters interference signals of various other frequencies in echo waves to obtain more accurate multi-frequency ultrasonic signal analog quantity, and the digital-to-analog conversion circuit 23 converts the multi-frequency ultrasonic analog quantity into a digital signal and then transmits the digital signal to upper computer software.
The upper computer 3 analyzes and calculates the acquired ultrasonic wave propagation characteristic parameters, uses the acquired parameters and the actual values of the transformer oil physical and chemical property parameters as training samples, trains the samples by using an IGWO-BPNN algorithm, and establishes an IGWO-BPNN detection model of the transformer oil physical and chemical property parameters; and finally, analyzing the data obtained by each detection by using a detection model, and giving a detection result of the physical and chemical performance parameters of the transformer oil.
In the preferred embodiment, the ultrasonic transducer 12 uses three transmitting/receiving probes of 100K, 200K and 300K.
When the detection system works, firstly, the multi-frequency ultrasonic wave emission driving circuit generates square wave pulse signals with different frequencies, the emission probe array is respectively excited to emit a plurality of ultrasonic wave signals with different frequencies to pass through the transformer oil, then, the multi-frequency ultrasonic wave passes through the transformer oil and then the ultrasonic wave receiving probe array receives echo signals, and the echo processing circuit can carry out primary filtering and amplification processing on the echo signals; a signal amplification circuit in the data acquisition module further amplifies processed data such as echo signals and the like and then transmits the amplified data to a filtering processing circuit, interference signals of other frequencies are filtered, accurate multi-frequency ultrasonic signal analog quantity is transmitted to a digital-to-analog conversion circuit, and the digital-to-analog conversion circuit converts the multi-frequency ultrasonic analog quantity into digital signals and transmits the digital signals to an upper computer for analysis and processing; the upper computer obtains ultrasonic wave propagation characteristic parameters through analysis and calculation, the obtained ultrasonic wave propagation characteristic parameters and actual values of transformer oil physical and chemical performance parameters are used as training samples, the samples are trained through an IGWO-BPNN algorithm, and an IGWO-BPNN detection model of the transformer oil physical and chemical performance parameters is established; and finally, analyzing the data obtained by each detection by using a detection model, and giving a detection result of the physical and chemical performance parameters of the transformer oil.
As shown in fig. 3, the improved grayish wolf algorithm optimized back propagation neural network (IGWO-BPNN) algorithm flow includes the following specific steps:
1) and constructing a Back Propagation Neural Network (BPNN) and determining a back propagation neural network topological structure.
2) And initializing IGWO parameters. Determining the scale and the maximum iteration number of the gray wolf population, determining the dimension of the gray wolf individual position information and the upper and lower boundaries of the gray wolf dimension, and randomly initializing the gray wolf position.
3) Selecting a fitness function, calculating the fitness, and selecting an optimal wolf alpha, a suboptimal wolf beta and a third best wolf delta.
4) The position information ω of the remaining grays is updated, and the parameters A, C and a are updated.
5) And judging whether the set maximum iteration number or the set error is reached, otherwise, repeating the step 3) and the step 4) until the condition is met.
6) And outputting the position of the optimal wolf alpha, mapping the optimal wolf alpha into a weight matrix, and using the weight matrix as the weight from the hidden layer of the BP neural network to the output layer to realize the IGWO-BPNN model.
7) And preprocessing the data, inputting the preprocessed data into the trained IGWO-BPNN model to obtain a prediction result, comparing the prediction result with an actual value, and checking the reliability of the model.
Example 1:
the multi-frequency ultrasonic detection system provided by the invention is used for detecting a certain 220kV three-phase transformer of a national grid power supply company, extracting a transformer oil sample, and analyzing the detected data by using a detection model based on IGWO-BPNN. In order to improve the detection accuracy, the transformer oil is detected for 10 times, all data are imported into a software system to obtain the breakdown voltage, the dielectric loss factor, the kinematic viscosity and the acid value of the transformer oil, and the average value of each parameter is obtained. And after analyzing the generated oil parameter detection result by the software system, finally giving a transformer oil parameter detection report. The detection results of the respective parameters are first analyzed below.
As shown in fig. 4, the average value of 10 detection results of the breakdown voltage of the transformer oil is 54.95kV, and the breakdown voltage of the transformer oil is not less than 30kV according to the GB2536-2011 standard, which meets the standard. The transformer oil has the advantages of strong limited electrical stress tolerance, good insulating property, less water, fiber impurities or polar impurities in the oil, and low pollution degree and aging degree of the transformer oil.
As shown in fig. 5, the average value of 10 detection results of the dielectric loss factor of the transformer oil is 0.021%, and according to the GB2536-2011 standard, the dielectric loss factor of the transformer oil is not more than 0.5%, and meets the standard. The transformer oil is relatively low in polar pollutant, moisture or colloid substance content, and low in oil pollution or oil quality deterioration degree.
As shown in FIG. 6, the average value of 10 measurements of the kinematic viscosity of the transformer oil was 9.58mm2And/s, according to the GB2536-2011 standard, the kinematic viscosity of the transformer oil is not more than 12mm2And/s, meeting the standard. The transformer oil has the advantages that solid particle impurities in the transformer oil are few, the intermolecular friction in the oil is small when the oil flows, and the cooling effect of the transformer oil is good.
As shown in FIG. 7, the average value of 10 detection results of the acid value of the transformer oil is 0.0088mg KOH/g, and according to the GB2536-2011 standard, the acid value of the transformer oil is not more than 0.01mg KOH/g and meets the standard. The transformer oil is low in aging degree and pollution degree and has good performance.
The performance parameter values of the transformer oil all accord with the GB2536-2011 standard, and the current transformer oil can be judged to be in a healthy state within a normal range. Since the transformer is put into operation, faults caused by transformer oil never occur, the transformer is always kept in a normal operation state, and the correctness of the detection result given by the system is verified again.
Finally, the above embodiments are only intended to illustrate the technical solutions of the present invention and not to limit the present invention, and although the present invention has been described in detail with reference to the preferred embodiments, it will be understood by those skilled in the art that modifications or equivalent substitutions may be made on the technical solutions of the present invention without departing from the spirit and scope of the technical solutions, and all of them should be covered by the claims of the present invention.
Claims (8)
1. A system for testing physical and chemical properties of transformer oil by using multi-frequency ultrasonic technique, the system comprising: the device comprises a multi-frequency ultrasonic transmitting and receiving module (1), a data acquisition module (2) and an upper computer (3);
the multi-frequency ultrasonic wave transmitting and receiving module (1) is used for transmitting and receiving ultrasonic echo signals;
the data acquisition module (2) is used for acquiring and processing ultrasonic echo signals;
the upper computer (3) analyzes and calculates the acquired ultrasonic wave propagation characteristic parameters, uses the acquired parameters and the actual values of the transformer oil physical and chemical property parameters as training samples, trains the samples by using an IGWO-BPNN algorithm, and establishes an IGWO-BPNN detection model of the transformer oil physical and chemical property parameters; and finally, analyzing the data obtained by each detection by using a detection model, and giving a detection result of the physical and chemical performance parameters of the transformer oil.
2. The system for detecting the oil physical and chemical properties of the transformer according to claim 1, wherein the multi-frequency ultrasonic transmitting and receiving module comprises a multi-frequency ultrasonic transmitting and receiving circuit (11), an ultrasonic transducer (12); the multi-frequency ultrasonic wave transmitting and receiving circuit (11) comprises an ultrasonic wave transmitting and driving circuit (101) and an ultrasonic wave echo processing circuit (102);
the ultrasonic emission driving circuit (101) is used for generating square wave pulse signals with different frequencies and driving the ultrasonic transducer to emit ultrasonic signals; the ultrasonic transducer (12) converts an electric signal and an ultrasonic signal to each other; the ultrasonic echo processing circuit (102) conducts primary filtering and amplification on ultrasonic waves and then transmits the ultrasonic waves to the data acquisition module (2).
3. The system for detecting the oil physical and chemical properties of the transformer according to claim 2, wherein three transmitting/receiving probes of 100K, 200K or 300K are selected for the ultrasonic transducer (12).
4. The system for detecting the oil physical and chemical properties of a transformer according to claim 3, wherein the data acquisition module comprises a signal amplification circuit (21), a filter circuit (22) and a digital-to-analog conversion circuit (23);
the signal amplification circuit (21) is used for amplifying ultrasonic echo signals received by the multi-frequency ultrasonic wave transmitting and receiving module; the filter circuit (22) is used for filtering interference signals of various other frequencies in the ultrasonic echo signals to obtain accurate multi-frequency ultrasonic signal analog quantity; the digital-to-analog conversion circuit (23) is used for converting the multi-frequency ultrasonic analog quantity into a digital signal and transmitting the digital signal to the upper computer.
5. A method for detecting physical and chemical properties of transformer oil by applying a multi-frequency ultrasonic technology is characterized by comprising the following steps:
s1: acquiring ultrasonic wave propagation characteristic parameters in the transformer oil;
s2: taking the obtained actual values of the ultrasonic wave propagation characteristic parameters and the transformer oil physical and chemical property parameters as training samples, training the samples by using an IGWO-BPNN algorithm, and establishing an IGWO-BPNN detection model of the transformer oil physical and chemical property parameters; wherein, the IGWO-BPNN algorithm adopts an improved wolf algorithm to optimize the weight and the function parameters of the back propagation neural network;
s3: and analyzing the data obtained by each detection by using an IGWO-BPNN detection model, and giving a detection result of the physical and chemical performance parameters of the transformer oil.
6. The method for detecting physical and chemical properties of transformer oil according to claim 5, wherein in step S2, the improved Grey wolf algorithm specifically comprises: the gray wolf algorithm divides the whole wolf group into four levels according to the fitness, the first 3 of the fitness are respectively the optimal wolf alpha, the second best wolf beta and the third best wolf delta, the rest wolfs are omega, the gray wolfs are close to and surround the prey in the process of predation, and correspondingly, the distance between the gray wolf individual and the prey is determined in the improved gray wolf algorithm and the position of the gray wolf is updated;
D＝|C·XP(t)-X(t)|
X(t+1)＝XP(t)-A·D
wherein t is the current iteration number, XPIs the position of the prey, X (t) is the current gray wolf position, and X (t +1) is the updated position of the gray wolf; c is a wobble factor, C is 2r1，r1Is [0, 1 ]]A random number within; a is 2ar2 2-a，r2Is [0, 1 ]]A is a convergence factor, and a is linearly decreased from 2 to 0 along with the increase of the iteration times;
wherein, tmaxIs the maximum number of iterations.
7. The method for detecting physical and chemical properties of transformer oil as claimed in claim 6, wherein during hunting, the best wolf α leads β, δ to hunt the prey, and the orientation of the prey is specifically:
firstly, calculating the direction of an individual hunting target in the wolf group, and mathematically expressing the following steps:
wherein D isα、Dβ、DδThe distances between alpha, beta and delta of the wolf individual and the prey are respectively; c1、C2、C3Are respectively corresponding toThe wobble factor of (d); xα(t)、Xβ(t)、Xδ(t) the current positions of alpha, beta and delta of the wolf individual, Xα(t+1)、Xβ(t+1)、Xδ(t +1) are the updated positions of alpha, beta and delta of the wolf individual respectively; a. the1、A2、A3Respectively corresponding coefficients;
then, the direction of the individual moving to the prey is calculated to obtain the updated position X of the wolf individual omegaω(t +1), the mathematical expression is as follows:
Xω(t+1)＝[Xα(t+1)+Xβ(t+1)+Xδ(t+1)]/3
wherein, Xω(t) and Xω(t +1) are the current position and the updated position of the gray wolf ω, respectively.
8. The method for detecting physical and chemical properties of transformer oil according to claim 7, wherein in step S2, the weight and function parameters of the back propagation neural network are optimized by using the improved graying algorithm, which specifically includes the following steps:
1) constructing a Back Propagation Neural Network (BPNN);
2) initializing parameters of a back propagation neural network; determining the scale and the maximum iteration number of the gray wolf population, determining the dimensionality of the gray wolf individual position information and the upper and lower boundaries of the gray wolf dimensionality, and randomly initializing the gray wolf position;
3) selecting a fitness function, calculating the fitness, and selecting an optimal wolf alpha, a suboptimal wolf beta and a third best wolf delta;
4) updating the position information omega of the remaining gray wolfs and updating the parameters A, C and a;
5) judging whether the set maximum iteration number or the set error is reached, otherwise, repeating the step 3) and the step 4) until the condition is met;
6) outputting the position of the optimal wolf alpha, mapping the optimal wolf alpha into a weight matrix which is used as the weight from a BPNN neural network hidden layer to an output layer, and realizing an IGWO-BPNN detection model;
7) and preprocessing the data and inputting the preprocessed data into a trained IGWO-BPNN detection model to obtain a prediction result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011249451.0A CN112461919A (en) | 2020-11-10 | 2020-11-10 | System and method for detecting physical and chemical properties of transformer oil by applying multi-frequency ultrasonic technology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011249451.0A CN112461919A (en) | 2020-11-10 | 2020-11-10 | System and method for detecting physical and chemical properties of transformer oil by applying multi-frequency ultrasonic technology |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112461919A true CN112461919A (en) | 2021-03-09 |
Family
ID=74825382
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011249451.0A Pending CN112461919A (en) | 2020-11-10 | 2020-11-10 | System and method for detecting physical and chemical properties of transformer oil by applying multi-frequency ultrasonic technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112461919A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113075152A (en) * | 2021-03-26 | 2021-07-06 | 云南电网有限责任公司电力科学研究院 | Infrared light enhancement system for detecting content of dissolved gas in transformer oil |
| CN113917287A (en) * | 2021-11-22 | 2022-01-11 | 国家电网有限公司 | Substation bus joint discharge heating fault monitoring system |
| CN117239917A (en) * | 2023-08-28 | 2023-12-15 | 连云港智源电力设计有限公司 | Substation data acquisition and processing device and method |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105182141A (en) * | 2015-09-30 | 2015-12-23 | 广州供电局有限公司电力试验研究院 | Transformer fault detection method and system |
| CN109063928A (en) * | 2018-08-29 | 2018-12-21 | 广东电网有限责任公司 | A kind of transformer oil Forecasting Model of Density modeling method and device |
| CN109738519A (en) * | 2019-01-04 | 2019-05-10 | 国网四川省电力公司广安供电公司 | A kind of denoising method of transformer high-voltage bushing lead ultrasound detection |
| CN111625816A (en) * | 2020-04-21 | 2020-09-04 | 江西理工大学 | Intrusion detection method and device |
| CN111709524A (en) * | 2020-07-03 | 2020-09-25 | 江苏科技大学 | RBF neural network optimization method based on improved GWO algorithm |
| CN111723945A (en) * | 2020-06-03 | 2020-09-29 | 杭州欣禾圣世科技有限公司 | BP neural network optimization method based on improved wolf algorithm |
| WO2020191801A1 (en) * | 2019-03-27 | 2020-10-01 | 东北大学 | Lithium ion battery remaining life prediction method based on wolf pack optimization lstm network |
-
2020
- 2020-11-10 CN CN202011249451.0A patent/CN112461919A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105182141A (en) * | 2015-09-30 | 2015-12-23 | 广州供电局有限公司电力试验研究院 | Transformer fault detection method and system |
| CN109063928A (en) * | 2018-08-29 | 2018-12-21 | 广东电网有限责任公司 | A kind of transformer oil Forecasting Model of Density modeling method and device |
| CN109738519A (en) * | 2019-01-04 | 2019-05-10 | 国网四川省电力公司广安供电公司 | A kind of denoising method of transformer high-voltage bushing lead ultrasound detection |
| WO2020191801A1 (en) * | 2019-03-27 | 2020-10-01 | 东北大学 | Lithium ion battery remaining life prediction method based on wolf pack optimization lstm network |
| CN111625816A (en) * | 2020-04-21 | 2020-09-04 | 江西理工大学 | Intrusion detection method and device |
| CN111723945A (en) * | 2020-06-03 | 2020-09-29 | 杭州欣禾圣世科技有限公司 | BP neural network optimization method based on improved wolf algorithm |
| CN111709524A (en) * | 2020-07-03 | 2020-09-25 | 江苏科技大学 | RBF neural network optimization method based on improved GWO algorithm |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113075152A (en) * | 2021-03-26 | 2021-07-06 | 云南电网有限责任公司电力科学研究院 | Infrared light enhancement system for detecting content of dissolved gas in transformer oil |
| CN113917287A (en) * | 2021-11-22 | 2022-01-11 | 国家电网有限公司 | Substation bus joint discharge heating fault monitoring system |
| CN117239917A (en) * | 2023-08-28 | 2023-12-15 | 连云港智源电力设计有限公司 | Substation data acquisition and processing device and method |
| CN117239917B (en) * | 2023-08-28 | 2024-02-23 | 连云港智源电力设计有限公司 | Substation data acquisition and processing device and method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112461919A (en) | System and method for detecting physical and chemical properties of transformer oil by applying multi-frequency ultrasonic technology | |
| CN103344934B (en) | The detection check method and system of Partial Discharge in Power Transformer sonac | |
| US20030037615A1 (en) | Non-destructive evaluation of wire insulation and coatings | |
| CN102879309A (en) | Gas particle concentration measurement method and device on basis of broadband linear frequency modulation ultrasound | |
| CN110186546A (en) | Hydrophone sensitivity free field wide band calibration method based on pink noise | |
| CN107192930A (en) | A kind of transformer partial discharge positioning method | |
| CN113916989A (en) | Method and system for detecting internal defects of high-performance epoxy insulating part of power system | |
| CN114035007A (en) | Non-contact transformer bushing partial discharge detection system | |
| CN110530972A (en) | A kind of ultrasound detection evaluation method and device and device application method for the assessment of cable ageing state | |
| Tang et al. | A method based on SVD for detecting the defect using the magnetostrictive guided wave technique | |
| CN113341277A (en) | Insulator fault assessment method based on multi-frequency ultrasonic waves and experimental platform thereof | |
| CN113189208A (en) | Ultrasonic characteristic detection method and detection system for lithium battery | |
| CN111678988A (en) | Nonlinear ultrasonic evaluation device and method for concrete material surface damage | |
| CN114839269A (en) | Nondestructive testing method and device for internal defect stress of GIS solid insulating part | |
| CN106918604B (en) | Inhaul cable defect detection system based on electromagnetic wave transmission line theory and detection method thereof | |
| CN202814843U (en) | Gas particle concentration measurement device based on broadband linear frequency modulation ultrasounds | |
| CN113758617B (en) | Efficient nondestructive testing system and method for stress gradient based on broadband frequency-sweep signal frequency domain calculation | |
| CN115327306A (en) | Particle identification method used in gas insulation environment | |
| CN115792427A (en) | Cable coupling interference effect test system and method | |
| CN105676225B (en) | Ranging system and method in opaque troubled liquor | |
| CN113109638A (en) | Audible noise sound power prediction method for alternating current transmission line in high-altitude area | |
| CN114113951B (en) | Ultrasonic stress detection system and method for insulation defect of high-voltage switch cabinet | |
| CN115577630A (en) | Method and system for predicting moisture content of transformer oil | |
| CN117743981A (en) | Transformer oil breakdown voltage prediction method based on random forest model | |
| CN213337476U (en) | Meat moisture detection device based on audio signal characteristics |
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 |