CN115860714B - Power equipment safe operation management system and method based on industrial Internet - Google Patents

Power equipment safe operation management system and method based on industrial Internet Download PDF

Info

Publication number
CN115860714B
CN115860714B CN202211498512.6A CN202211498512A CN115860714B CN 115860714 B CN115860714 B CN 115860714B CN 202211498512 A CN202211498512 A CN 202211498512A CN 115860714 B CN115860714 B CN 115860714B
Authority
CN
China
Prior art keywords
current
immersed transformer
sound signal
fault sound
sample
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.)
Active
Application number
CN202211498512.6A
Other languages
Chinese (zh)
Other versions
CN115860714A (en
Inventor
舒蔚蔚
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.)
Zhuhai Dereis Technology Co ltd
Original Assignee
Zhuhai Dereis Technology Co ltd
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 Zhuhai Dereis Technology Co ltd filed Critical Zhuhai Dereis Technology Co ltd
Priority to CN202211498512.6A priority Critical patent/CN115860714B/en
Publication of CN115860714A publication Critical patent/CN115860714A/en
Application granted granted Critical
Publication of CN115860714B publication Critical patent/CN115860714B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
    • Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
    • Y04S10/00—Systems supporting electrical power generation, transmission or distribution
    • Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications

Landscapes

  • Testing Relating To Insulation (AREA)

Abstract

The invention discloses a power equipment safe operation management system and method based on an industrial Internet, and relates to the technical field of power equipment safe operation management. The system comprises a data acquisition module, a data processing module, an influence function construction analysis module, a classification model construction analysis module, a similarity judgment analysis module and a feedback reminding module; the output end of the data acquisition module is connected with the input end of the data processing module; the output end of the data processing module is connected with the input end of the influence function construction analysis module; the output end of the influence function construction analysis module is connected with the input end of the classification model construction analysis module; the output end of the classification model construction analysis module is connected with the input end of the similarity judgment analysis module; the output end of the similarity judging and analyzing module is connected with the input end of the feedback reminding module.

Description

Power equipment safe operation management system and method based on industrial Internet
Technical Field
The invention relates to the technical field of power equipment safe operation management, in particular to a power equipment safe operation management system and method based on an industrial Internet.
Background
The industrial internet is a novel infrastructure, an application mode and industrial ecology which are deeply fused with new generation information communication technology and industrial economy, and a brand new manufacturing and service system which covers a full industrial chain and a full value chain is constructed by comprehensively connecting people, machines, objects, systems and the like, so that an implementation way is provided for the development of industrialization and even industrialization digitization, networking and intellectualization, and the industrial internet is an important foundation stone of the fourth industrial revolution.
The power equipment mainly comprises two major types of power generation equipment and power supply equipment, wherein the power generation equipment mainly comprises a power station boiler, a steam turbine, a gas turbine, a water turbine, a generator, a transformer and the like, and the power supply equipment mainly comprises power transmission lines, transformers, contactors and the like with various voltage levels. The method mainly relies on experienced operation staff to check fault points which are possibly problematic one by one in the prior art, and finally determines the fault point which is caused by faults.
Disclosure of Invention
The invention aims to provide a power equipment safe operation management system and method based on the industrial Internet, so as to solve the problems in the background technology.
In order to solve the technical problems, the invention provides the following technical scheme:
an electric power equipment safe operation management method based on industrial internet, the method comprises the following steps:
step S1: acquiring historical data of power equipment by utilizing industrial Internet big data, wherein the power equipment comprises an oil immersed transformer, the historical data comprises historical fault sound signals and historical operation fault points of the oil immersed transformer, and performing spectrum analysis on the historical fault sound signals to obtain a historical fault sound spectrum characteristic set;
step S2: constructing a machine learning classification model, combining a historical fault sound signal set and a historical operation fault point set of the oil immersed transformer to generate a training sample data set, and calculating the Euclidean distance between any two training samples;
step S3: acquiring a current fault sound signal and a current operating environment of the oil-immersed transformer, constructing an environment influence function, and calculating the probability of the current operating environment of the oil-immersed transformer influencing the current fault sound signal based on the current operating environment of the oil-immersed transformer;
Step S4: setting a probability threshold value of influence of the current running environment of the oil-immersed transformer on the current fault sound signal, carrying out spectrum analysis on the current fault sound signal of the oil-immersed transformer to obtain a current fault sound spectrum feature set when the probability of influence of the current running environment of the oil-immersed transformer on the current fault sound signal does not exceed the threshold value, substituting the current fault sound signal of the oil-immersed transformer as a current test sample, substituting the current fault sound spectrum feature set as a sample feature set of the current test sample into a machine learning classification model, and calculating the Euclidean distance between the current test sample and a training sample;
step S5: setting a Euclidean distance threshold value of a current test sample and a training sample, judging the training sample which is most similar to the current test sample when the Euclidean distance between the current test sample and the training sample does not exceed the threshold value, acquiring a sample output result of the training sample, taking the sample output result as a sample output result of the current test sample, namely obtaining a current operation fault point under a current fault sound signal, and feeding back the current operation fault point to an operation and maintenance personnel;
Step S6: when the probability that the current running environment of the oil immersed transformer affects the current fault sound signal exceeds a threshold value, the system sends out early warning reminding to a manager and stops feeding back the current running fault point under the current fault sound signal.
Further, in step S1,
acquiring a historical fault sound signal set A= { x of the oil immersed transformer W 1 (n)，x 2 (n)，...，x m (n) }; wherein x is 1 (n)，x 2 (n)，...，x m (n) represents the 1 st, 2 nd, and the third and fourth historical fault sound signals of the oil immersed transformer W; historical operation fault point location set c= { y of corresponding oil immersed transformer W 1 ，y 2 ，...，y m -a }; wherein y is 1 ，y 2 ，...，y m Respectively representing 1 st and 2 nd and the third and fourth historical operation fault points of the oil immersed transformer W; to the instituteThe historical fault sound signal is subjected to spectrum analysis to obtain a historical fault sound spectrum feature setWherein (1)>The i-th historical fault sound signal includes 1,2, and a-th fault sound spectrum characteristic values, respectively;
performing a spectral analysis on the historical fault sound signal includes:
performing spectrum analysis on the historical fault sound signal by using discrete Fourier transform;
according to the formula:
wherein k=0, 1,2,; x (K) represents the frequency spectrum of the ith historical fault sound signal after discrete Fourier transform; x is x i (n) represents an i-th history trouble sound signal.
Further, in step S2, the building a machine learning classification model includes:
combining the historical fault sound signal set and the historical operation fault point position set of the oil immersed transformer W to generate a training sample data set D= { (x) 1 (n)，y 1 )，(x 2 (n)，y 2 )，......，(x m (n)，y m ) -a }; wherein x is 1 (n)，x 2 (n)，...，x m (n) represents the 1 st, 2 nd, and/or the third, m th training samples of the machine learning classification model, respectively; y is 1 ，y 2 ，...，y m Each of the samples represents a 1 st, 2 nd and a.i.m. output result;
any training sample x i (n) including a sample feature set
According to the formula, the Euclidean distance between any two training samples is calculated as follows:
wherein L is 2 (x i (n)，x r (n)) represents training samples x i (n) and training sample x r (n) Euclidean distance between; l represents training sample x i (n) and training sample x r Serial number of sample feature of (n).
In the technical scheme, the discrete Fourier transform is the most basic method for signal analysis, the Fourier transform is the core of the Fourier analysis, and the signal is transformed from a time domain to a frequency domain through the discrete Fourier transform, so that the frequency spectrum structure and the change rule of the signal are researched; in the method, different historical fault sound frequency spectrum characteristics can be obtained quickly and clearly by performing discrete Fourier transform on different historical fault sound signals, and a basis is provided for constructing a machine learning classification model.
Further, in step S3, a current fault sound signal and a current operating environment of the oil immersed transformer W are obtained;
the current running environment of the oil immersed transformer E comprises time T for acquiring the current fault sound signal of the oil immersed transformer W w And the industrial noise intensity Q of the position where the oil immersed transformer W is w ；
The time T for acquiring the current fault sound signal of the oil immersed transformer W w ∈[0，24]；
Industrial noise intensity Q of position of oil immersed transformer W w ∈[Q min ，Q max ]The method comprises the steps of carrying out a first treatment on the surface of the Wherein Q is min A minimum value representing the industrial noise intensity of the position where the oil immersed transformer W is located; q (Q) max A maximum value of the industrial noise intensity indicating the position of the oil immersed transformer W;
for the time T for acquiring the current fault sound signal of the oil immersed transformer W w And the industrial noise intensity Q of the position where the oil immersed transformer W is located w Normalization processing is carried out to obtain a normalized value T of the time for obtaining the current fault sound signal of the oil immersed transformer W v And normalized value Q of industrial noise intensity of position where oil immersed transformer W is v ；
Setting an influence coefficient of time for acquiring a current fault sound signal of the oil immersed transformer W, and marking the influence coefficient as alpha;
setting an influence coefficient of industrial noise intensity of the position where the oil immersed transformer W is positioned, and recording the influence coefficient as beta;
Building an environmental impact function:
p v ＝α*T v +β*Q v
wherein p is v The probability that the current running environment of the oil immersed transformer W affects the current fault sound signal is represented; alpha represents an influence coefficient of time for acquiring a current fault sound signal of the oil immersed transformer W; t (T) v A normalized value representing the time of acquiring the current fault sound signal of the oil immersed transformer W; beta represents an influence coefficient of industrial noise intensity of the position where the oil immersed transformer W is located; q (Q) v A normalized value representing the intensity of industrial noise at the location of the oil immersed transformer W.
In the above technical solution, the application places of the oil-immersed transformers are different, which may cause different operation environments of the oil-immersed transformers, and in the present application, considering that different operation environments may have different effects on the obtained fault sound signals of the oil-immersed transformers, for example, the effect of the fault sound signals of the oil-immersed transformers obtained in daytime on the working day may be greater than that in nighttime on the working day due to the operation environments; the influence of the operation environment on the fault sound signal of the oil-immersed transformer obtained in the factory building with high industrial noise intensity is larger than that of the oil-immersed transformer obtained outside the open room with low industrial noise intensity, so that the larger the influence of the operation environment on the obtained fault sound signal of the oil-immersed transformer is, the larger the judgment error of the system on the fault point is, and the influence function model is constructed to preprocess the operation environment, so that the system precision can be optimized.
Further, in steps S4-S6,
setting oil immersedThe probability threshold value of the influence of the current running environment of the transformer W on the current fault sound signal is recorded as p 0 ；
When the probability of the current running environment of the oil immersed transformer W affecting the current fault sound signal does not exceed a threshold value, performing spectrum analysis on the current fault sound signal of the oil immersed transformer W to obtain a current fault sound spectrum feature set;
taking the current fault sound signal of the oil immersed transformer W as a current test sample, and marking the current fault sound signal as x u (n) taking the current fault sound spectrum feature set as a sample feature set of the current test sample
Calculating a current test sample x u (n) and training sample x i Euclidean distance between (n):
wherein L is 2 (x i (n)，x u (n)) represents the current test sample x u (n) and training sample x i (n) Euclidean distance between;representing training samples x i The first sample feature of (n); />Representing the current test sample x u The first sample feature of (n);
setting a current test sample x u (n) and training sample x i The Euclidean distance threshold between (n) is recorded as
When (when)When it is judged that the current test sample x is identical to the current test sample x u (n) training sample x which is most similar i (n)；
Acquisition of training samples x i Sample output result y of (n) i The method comprises the steps of carrying out a first treatment on the surface of the Outputting the sample to result y i Sample output result y as current test sample u Obtaining the current operation fault point position under the current fault sound signal;
when the probability that the current running environment of the oil immersed transformer W affects the current fault sound signal exceeds a threshold value, the system sends out early warning reminding to a manager and stops feeding back the current running fault point under the current fault sound signal.
In the above technical solution, the machine learning classification model specifically refers to a KNN learning algorithm, namely a K nearest neighbor learning algorithm, and its working principle is to input a new sample (test sample) with unknown class, compare each feature of the new sample with the feature corresponding to the sample in the training sample set, and calculate the euclidean distance between the test sample and each sample in the training sample set; sorting the obtained Euclidean distances from small to large, wherein smaller values represent more similar values; the K nearest neighbor learning algorithm has the advantages of simple and understandable model, higher precision, insensitivity to abnormal values, no need of training and no need of parameter estimation. In the application, the historical fault sound signal of the oil immersed transformer W is used as a training sample of a machine learning classification model, the current fault sound signal of the oil immersed transformer E is used as a current test sample, the Euclidean distance between the two samples is calculated, so that the training sample most similar to the current test sample is obtained, and different training samples correspond to different sample output results, so that the current running fault point under the current fault sound signal can be obtained.
The system comprises a data acquisition module, a data processing module, an influence function construction analysis module, a classification model construction analysis module, a similarity judgment analysis module and a feedback reminding module;
the data acquisition module is used for acquiring historical data of power equipment, current fault sound signals of the oil immersed transformer and current operation environments by utilizing industrial Internet big data, wherein the power equipment comprises the oil immersed transformer, and the historical data comprises historical fault sound signals and historical operation fault points of the oil immersed transformer; the data processing module is used for carrying out spectrum analysis on the historical fault sound signals and the current fault sound signals of the oil immersed transformer to obtain a historical fault sound spectrum characteristic set and a current fault sound spectrum characteristic set; the influence function construction analysis module is used for constructing an environment influence function, calculating the probability of influence of the current operation environment of the oil-immersed transformer on the current fault sound signal based on the current operation environment of the oil-immersed transformer, setting a probability threshold value of influence of the current operation environment of the oil-immersed transformer on the current fault sound signal, and judging whether the probability of influence of the current operation environment of the oil-immersed transformer on the current fault sound signal exceeds the threshold value; the classification model construction analysis module is used for constructing a machine learning classification model, combining a historical fault sound signal set and a historical operation fault point set of the oil immersed transformer to generate a training sample data set, calculating the Euclidean distance between any two training samples, taking the current fault sound signal of the oil immersed transformer as a current test sample, substituting the current fault sound spectrum characteristic set as a sample characteristic set of the current test sample into the machine learning classification model, and calculating the Euclidean distance between the current test sample and the training sample; the similarity judging and analyzing module is used for setting the Euclidean distance threshold value of the current test sample and the training sample, judging the training sample which is most similar to the current test sample when the Euclidean distance between the current test sample and the training sample does not exceed the threshold value, acquiring a sample output result of the training sample, taking the sample output result as the sample output result of the current test sample, namely obtaining a current operation fault point under the current fault sound signal, and feeding back the current operation fault point to an operation and maintenance personnel; the feedback reminding module is used for sending out early warning reminding to a manager when the probability of the current running environment of the oil immersed transformer affecting the current fault sound signal exceeds a threshold value, and stopping feeding back the current running fault point under the current fault sound signal;
The output end of the data acquisition module is connected with the input end of the data processing module; the output end of the data processing module is connected with the input end of the influence function construction analysis module; the output end of the influence function construction analysis module is connected with the input end of the classification model construction analysis module; the output end of the classification model construction analysis module is connected with the input end of the similarity judgment analysis module; the output end of the similarity judging and analyzing module is connected with the input end of the feedback reminding module.
Further, the data acquisition module comprises a historical data acquisition unit and a current data acquisition unit;
the historical data acquisition unit is used for acquiring historical data of the power equipment by utilizing industrial Internet big data, the power equipment comprises an oil immersed transformer, and the historical data comprises historical fault sound signals and historical operation fault points of the oil immersed transformer;
the current data acquisition unit is used for acquiring a current fault sound signal and a current running environment of the oil immersed transformer;
the current running environment comprises time T for acquiring current fault sound signals of the oil immersed transformer w And the industrial noise intensity Q of the position of the oil immersed transformer w ；
The output end of the historical data acquisition unit is connected with the input end of the current data acquisition unit; the output end of the current data acquisition unit is connected with the input end of the data processing module;
the data processing module comprises a spectrum analysis unit and a spectrum characteristic acquisition unit;
the spectrum analysis unit is used for carrying out spectrum analysis on the historical fault sound signals and the current fault sound signals of the oil immersed transformer;
the frequency spectrum characteristic acquisition unit is used for acquiring a historical fault sound frequency spectrum characteristic set and a current fault sound frequency spectrum characteristic set;
the output end of the spectrum analysis unit is connected with the input end of the spectrum characteristic acquisition unit; the output end of the frequency spectrum characteristic acquisition unit is connected with the input end of the influence function construction analysis module.
Further, the influence function construction analysis module comprises an influence function construction unit and an analysis judgment unit;
the influence function construction unit is used for constructing an environment influence function and calculating the probability of influence of the current operation environment of the oil-immersed transformer on the current fault sound signal based on the current operation environment of the oil-immersed transformer;
The analysis judging unit is used for setting a probability threshold value of the influence of the current running environment of the oil-immersed transformer on the current fault sound signal and judging whether the probability of the influence of the current running environment of the oil-immersed transformer on the current fault sound signal exceeds the threshold value;
the output end of the influence function construction unit is connected with the input end of the analysis judging unit; the output end of the analysis judging unit is connected with the input end of the classification model construction analysis module.
Further, the classification model construction and analysis module comprises a machine learning classification model construction unit and a machine learning classification model analysis unit;
the machine learning classification model construction unit is used for constructing a machine learning classification model, combining a historical fault sound signal set and a historical operation fault point set of the oil immersed transformer to generate a training sample data set, and calculating Euclidean distance between any two training samples;
the machine learning classification model analysis unit is used for taking a current fault sound signal of the oil immersed transformer as a current test sample, substituting the current fault sound spectrum characteristic set as a sample characteristic set of the current test sample into the machine learning classification model, and calculating the Euclidean distance between the current test sample and the training sample;
The output end of the machine learning classification model construction unit is connected with the input end of the machine learning classification model analysis unit; the output end of the machine learning classification model analysis unit is connected with the input end of the similarity judgment analysis module;
further, the similarity judging and analyzing module comprises a threshold setting unit and a judging and outputting unit;
the threshold setting unit is used for setting a Euclidean distance threshold between a current test sample and a training sample;
the judging and outputting unit is used for judging a training sample which is most similar to the current test sample when the Euclidean distance between the current test sample and the training sample does not exceed a threshold value, obtaining a sample output result of the training sample, taking the sample output result as the sample output result of the current test sample, namely obtaining a current operation fault point under a current fault sound signal, and feeding back the current operation fault point to an operation and maintenance personnel;
the output end of the threshold setting unit is connected with the input end of the judging output unit; and the output end of the judging output unit is connected with the input end of the feedback reminding module.
Compared with the prior art, the invention has the following beneficial effects: according to the invention, the historical fault sound signals of the oil immersed transformer can be subjected to spectrum analysis to obtain a historical fault sound spectrum feature set, and the Euclidean distance between the current test sample and the training sample is calculated by using the machine learning classification model so as to rapidly determine the historical fault sound signal which is the most similar to the current test sample, namely the current fault sound signal, thereby determining the historical operation fault point; the method and the device can preprocess the operating environment of the oil immersed transformer, quickly judge the probability of influence of the operating environment on the current fault signal, improve the accuracy of analyzing and obtaining the current fault point by the system, further simplify the complicated step that operation and maintenance personnel need to check each fault point which is likely to have problems one by one, save the time of the operation and maintenance personnel for maintaining the power equipment, improve the maintenance efficiency of the operation and maintenance personnel, and reduce the influence on the life of residents caused by power failure.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the invention. In the drawings:
fig. 1 is a schematic structural diagram of an industrial internet-based power equipment safe operation management system of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Referring to fig. 1, the present invention provides the following technical solutions:
an electric power equipment safe operation management method based on industrial internet, the method comprises the following steps:
step S1: acquiring historical data of power equipment by utilizing industrial Internet big data, wherein the power equipment comprises an oil immersed transformer, the historical data comprises historical fault sound signals and historical operation fault points of the oil immersed transformer, and performing spectrum analysis on the historical fault sound signals to obtain a historical fault sound spectrum characteristic set;
Step S2: constructing a machine learning classification model, combining a historical fault sound signal set and a historical operation fault point set of the oil immersed transformer to generate a training sample data set, and calculating the Euclidean distance between any two training samples;
step S3: acquiring a current fault sound signal and a current operating environment of the oil-immersed transformer, constructing an environment influence function, and calculating the probability of the current operating environment of the oil-immersed transformer influencing the current fault sound signal based on the current operating environment of the oil-immersed transformer;
step S4: setting a probability threshold value of influence of the current running environment of the oil-immersed transformer on the current fault sound signal, carrying out spectrum analysis on the current fault sound signal of the oil-immersed transformer to obtain a current fault sound spectrum feature set when the probability of influence of the current running environment of the oil-immersed transformer on the current fault sound signal does not exceed the threshold value, substituting the current fault sound signal of the oil-immersed transformer as a current test sample, substituting the current fault sound spectrum feature set as a sample feature set of the current test sample into a machine learning classification model, and calculating the Euclidean distance between the current test sample and a training sample;
Step S5: setting a Euclidean distance threshold value of a current test sample and a training sample, judging the training sample which is most similar to the current test sample when the Euclidean distance between the current test sample and the training sample does not exceed the threshold value, acquiring a sample output result of the training sample, taking the sample output result as a sample output result of the current test sample, namely obtaining a current operation fault point under a current fault sound signal, and feeding back the current operation fault point to an operation and maintenance personnel;
step S6: when the probability that the current running environment of the oil immersed transformer affects the current fault sound signal exceeds a threshold value, the system sends out early warning reminding to a manager and stops feeding back the current running fault point under the current fault sound signal.
Further, in step S1,
acquiring a historical fault sound signal set A= { x of the oil immersed transformer W 1 (n)，x 2 (n)，...，x m (n) }; wherein x is 1 (n)，x 2 (n)，...，x m (n) represents the 1 st, 2 nd, and the third and fourth historical fault sound signals of the oil immersed transformer E; historical operation fault point location set c= { y of corresponding oil immersed transformer W 1 ，y 2 ，...，y m -a }; wherein y is 1 ，y 2 ，...，y m Respectively representing 1 st and 2 nd and the third and fourth historical operation fault points of the oil immersed transformer W; performing spectrum analysis on the historical fault sound signals to obtain a historical fault sound spectrum feature set Wherein (1)>The i-th historical fault sound signal includes 1,2, and a-th fault sound spectrum characteristic values, respectively;
performing a spectral analysis on the historical fault sound signal includes:
performing spectrum analysis on the historical fault sound signal by using discrete Fourier transform;
according to the formula:
wherein k=0, 1,2,; x (K) represents the frequency spectrum of the ith historical fault sound signal after discrete Fourier transform; x is x i (n) represents an i-th history trouble sound signal.
Further, in step S2, the building a machine learning classification model includes:
combining the historical fault sound signal set and the historical operation fault point position set of the oil immersed transformer W to generate a training sample data set D= { (x) 1 (n)，y 1 )，(x 2 (n)，y 2 )，......，(x m (n)，y m ) -a }; wherein x is 1 (n)，x 2 (n)，...，x m (n) represents the 1 st, 2 nd, and/or the third, m th training samples of the machine learning classification model, respectively; y is 1 ，y 2 ，...，y m Each of the samples represents a 1 st, 2 nd and a.i.m. output result;
any training sample x i (n) including a sample feature set
According to the formula, the Euclidean distance between any two training samples is calculated as follows:
wherein L is 2 (x i (n)，x r (n)) represents training samples x i (n) and training sample x r (n) Euclidean distance between; l represents training sample x i (n) and training sample x r Serial number of sample feature of (n).
Further, in step S3, a current fault sound signal and a current operating environment of the oil immersed transformer W are obtained;
the current running environment of the oil immersed transformer W comprises time T for acquiring the current fault sound signal of the oil immersed transformer E w And the industrial noise intensity Q of the position where the oil immersed transformer W is w ；
The time T for acquiring the current fault sound signal of the oil immersed transformer W w ∈[0，24]；
Industrial noise intensity Q of position of oil immersed transformer W w ∈[Q min ，Q max ]The method comprises the steps of carrying out a first treatment on the surface of the Wherein Q is min A minimum value representing the industrial noise intensity of the position where the oil immersed transformer W is located; q (Q) max A maximum value of the industrial noise intensity indicating the position of the oil immersed transformer W;
for the time T for acquiring the current fault sound signal of the oil immersed transformer W w And the industrial noise intensity Q of the position where the oil immersed transformer W is located w Normalization processing is carried out to obtain a normalized value T of the time for obtaining the current fault sound signal of the oil immersed transformer W v And normalized value Q of industrial noise intensity of position where oil immersed transformer E is v ；
Setting an influence coefficient of time for acquiring a current fault sound signal of the oil immersed transformer W, and marking the influence coefficient as alpha;
setting an influence coefficient of industrial noise intensity of the position where the oil immersed transformer W is positioned, and recording the influence coefficient as beta;
Building an environmental impact function:
p v ＝α*T v +β*Q v
wherein p is v The probability that the current running environment of the oil immersed transformer W affects the current fault sound signal is represented; alpha represents an influence coefficient of time for acquiring a current fault sound signal of the oil immersed transformer W; t (T) v A normalized value representing the time of acquiring the current fault sound signal of the oil immersed transformer W; beta represents an influence coefficient of industrial noise intensity of the position where the oil immersed transformer W is located; q (Q) v A normalized value representing the intensity of industrial noise at the location of the oil immersed transformer W.
Further, in steps S4-S6,
setting a probability threshold value of the influence of the current running environment of the oil immersed transformer W on the current fault sound signal, and recording the probability threshold value as p 0 ；
When the probability of the current running environment of the oil immersed transformer W affecting the current fault sound signal does not exceed a threshold value, performing spectrum analysis on the current fault sound signal of the oil immersed transformer W to obtain a current fault sound spectrum feature set;
taking the current fault sound signal of the oil immersed transformer W as a current test sample, and marking the current fault sound signal as x u (n) taking the current fault sound spectrum feature set as a sample feature set of the current test sample
Calculating a current test sample x u (n) and training sample x i Euclidean distance between (n):
wherein L is 2 (x i (n)，x u (n)) represents the current test sample x u (n) and training sample x i (n) Euclidean distance between;representing training samples x i The first sample feature of (n); />Representing the current test sample x u The first sample feature of (n);
setting a current test sample x u (n) and training sample x i The Euclidean distance threshold between (n) is recorded as
When (when)When it is judged that the current test sample x is identical to the current test sample x u (n) training sample x which is most similar i (n)；
Acquisition of training samples x i Sample output result y of (n) i The method comprises the steps of carrying out a first treatment on the surface of the Outputting the sample to result y i Sample output result y as current test sample u Obtaining the current operation fault point position under the current fault sound signal;
when the probability that the current running environment of the oil immersed transformer W affects the current fault sound signal exceeds a threshold value, the system sends out early warning reminding to a manager and stops feeding back the current running fault point under the current fault sound signal.
The system comprises a data acquisition module, a data processing module, an influence function construction analysis module, a classification model construction analysis module, a similarity judgment analysis module and a feedback reminding module;
The data acquisition module is used for acquiring historical data of power equipment, current fault sound signals of the oil immersed transformer and current operation environments by utilizing industrial Internet big data, wherein the power equipment comprises the oil immersed transformer, and the historical data comprises historical fault sound signals and historical operation fault points of the oil immersed transformer; the data processing module is used for carrying out spectrum analysis on the historical fault sound signals and the current fault sound signals of the oil immersed transformer to obtain a historical fault sound spectrum characteristic set and a current fault sound spectrum characteristic set; the influence function construction analysis module is used for constructing an environment influence function, calculating the probability of influence of the current operation environment of the oil-immersed transformer on the current fault sound signal based on the current operation environment of the oil-immersed transformer, setting a probability threshold value of influence of the current operation environment of the oil-immersed transformer on the current fault sound signal, and judging whether the probability of influence of the current operation environment of the oil-immersed transformer on the current fault sound signal exceeds the threshold value; the classification model construction analysis module is used for constructing a machine learning classification model, combining a historical fault sound signal set and a historical operation fault point set of the oil immersed transformer to generate a training sample data set, calculating the Euclidean distance between any two training samples, taking the current fault sound signal of the oil immersed transformer as a current test sample, substituting the current fault sound spectrum characteristic set as a sample characteristic set of the current test sample into the machine learning classification model, and calculating the Euclidean distance between the current test sample and the training sample; the similarity judging and analyzing module is used for setting the Euclidean distance threshold value of the current test sample and the training sample, judging the training sample which is most similar to the current test sample when the Euclidean distance between the current test sample and the training sample does not exceed the threshold value, acquiring a sample output result of the training sample, taking the sample output result as the sample output result of the current test sample, namely obtaining a current operation fault point under the current fault sound signal, and feeding back the current operation fault point to an operation and maintenance personnel; the feedback reminding module is used for sending out early warning reminding to a manager when the probability of the current running environment of the oil immersed transformer affecting the current fault sound signal exceeds a threshold value, and stopping feeding back the current running fault point under the current fault sound signal;
The output end of the data acquisition module is connected with the input end of the data processing module; the output end of the data processing module is connected with the input end of the influence function construction analysis module; the output end of the influence function construction analysis module is connected with the input end of the classification model construction analysis module; the output end of the classification model construction analysis module is connected with the input end of the similarity judgment analysis module; the output end of the similarity judging and analyzing module is connected with the input end of the feedback reminding module.
Further, the data acquisition module comprises a historical data acquisition unit and a current data acquisition unit;
the historical data acquisition unit is used for acquiring historical data of the power equipment by utilizing industrial Internet big data, the power equipment comprises an oil immersed transformer, and the historical data comprises historical fault sound signals and historical operation fault points of the oil immersed transformer;
the current data acquisition unit is used for acquiring a current fault sound signal and a current running environment of the oil immersed transformer;
the current running environment comprises time T for acquiring current fault sound signals of the oil immersed transformer w And the industrial noise intensity Q of the position of the oil immersed transformer w ；
The output end of the historical data acquisition unit is connected with the input end of the current data acquisition unit; the output end of the current data acquisition unit is connected with the input end of the data processing module;
the data processing module comprises a spectrum analysis unit and a spectrum characteristic acquisition unit;
the spectrum analysis unit is used for carrying out spectrum analysis on the historical fault sound signals and the current fault sound signals of the oil immersed transformer;
the frequency spectrum characteristic acquisition unit is used for acquiring a historical fault sound frequency spectrum characteristic set and a current fault sound frequency spectrum characteristic set;
the output end of the spectrum analysis unit is connected with the input end of the spectrum characteristic acquisition unit; the output end of the frequency spectrum characteristic acquisition unit is connected with the input end of the influence function construction analysis module.
Further, the influence function construction analysis module comprises an influence function construction unit and an analysis judgment unit;
the influence function construction unit is used for constructing an environment influence function and calculating the probability of influence of the current operation environment of the oil-immersed transformer on the current fault sound signal based on the current operation environment of the oil-immersed transformer;
The analysis judging unit is used for setting a probability threshold value of the influence of the current running environment of the oil-immersed transformer on the current fault sound signal and judging whether the probability of the influence of the current running environment of the oil-immersed transformer on the current fault sound signal exceeds the threshold value;
the output end of the influence function construction unit is connected with the input end of the analysis judging unit; the output end of the analysis judging unit is connected with the input end of the classification model construction analysis module.
Further, the classification model construction and analysis module comprises a machine learning classification model construction unit and a machine learning classification model analysis unit;
the machine learning classification model construction unit is used for constructing a machine learning classification model, combining a historical fault sound signal set and a historical operation fault point set of the oil immersed transformer to generate a training sample data set, and calculating Euclidean distance between any two training samples;
the machine learning classification model analysis unit is used for taking a current fault sound signal of the oil immersed transformer as a current test sample, substituting the current fault sound spectrum characteristic set as a sample characteristic set of the current test sample into the machine learning classification model, and calculating the Euclidean distance between the current test sample and the training sample;
The output end of the machine learning classification model construction unit is connected with the input end of the machine learning classification model analysis unit; the output end of the machine learning classification model analysis unit is connected with the input end of the similarity judgment analysis module;
further, the similarity judging and analyzing module comprises a threshold setting unit and a judging and outputting unit;
the threshold setting unit is used for setting a Euclidean distance threshold between a current test sample and a training sample;
the judging and outputting unit is used for judging a training sample which is most similar to the current test sample when the Euclidean distance between the current test sample and the training sample does not exceed a threshold value, obtaining a sample output result of the training sample, taking the sample output result as the sample output result of the current test sample, namely obtaining a current operation fault point under a current fault sound signal, and feeding back the current operation fault point to an operation and maintenance personnel;
the output end of the threshold setting unit is connected with the input end of the judging output unit; and the output end of the judging output unit is connected with the input end of the feedback reminding module.
Example 1:
The oil immersed transformer should be a uniform "buzzing" sound during normal operation, because the alternating magnetic flux changes cause the vibration of the iron core to produce a loud sound when alternating current passes through the windings of the transformer, but the transformer is faulty if it produces the following abnormal sounds: (1) The strong and uneven noise occurs and the amplitude is increased, because the core penetrating screw of the iron core is not firmly fixed, the iron core is loosened, and the silicon steel sheets vibrate; (2) The discharging sound of the 'squeak' is generated in the transformer, and is caused by the fact that the winding or the outgoing line flashover and discharge of the shell or the broken line of the grounding wire of the iron core causes the high voltage generated by the induction of the iron core to the shell (ground); (3) The inside of the transformer is provided with a 'jingle', which is that the individual parts on the transformer are not firmly fixed to generate sound; (4) The 'camping' sound of a matrix is arranged in the transformer, and is caused by the vibration of certain silicon steel sheet ends which are separated from the laminated layers under the condition of light load or empty load; (5) Where there is breakdown inside the transformer, a "squeak" or "hum" sound is generated, which is a pilot trial that suddenly gets thicker and suddenly becomes thinner;
acquiring a historical fault sound signal set A= { x of the oil immersed transformer W 1 (n)，x 2 (n)，...，x 5 (n) }; wherein x is 1 (n)，x 2 (n)，...，x 5 (n) represents the 1 st, 2 nd, and the.i. and 5 th historical fault sound signals of the oil immersed transformer W, respectively; historical operation fault point location set of corresponding oil immersed transformer WC＝{y 1 ，y 2 ，...，y 5 -a }; wherein y is 1 ，y 2 ，...，y 5 The 1 st, 2 nd, and the.i. 5 th historical operating fault points of the oil-immersed transformer W are respectively represented; performing spectrum analysis on the historical fault sound signals to obtain a historical fault sound spectrum feature setWherein (1)>The i-th historical fault sound signal has the 1 st, 2 nd, the 4 th fault sound spectrum characteristic values;
therefore, the 1 st historical fault sound signal x 1 (n) historical Fault Sound Spectrum feature set
2 nd historical fault sound Signal x 2 (n) historical Fault Sound Spectrum feature set
3 rd historical fault sound Signal x 3 (n) historical Fault Sound Spectrum feature set
4 th historical Fault Sound Signal x 4 (n) historical Fault Sound Spectrum feature set
5 th historical fault sound Signal x 5 (n) historical Fault Sound Spectrum feature set
Combining the historical fault sound signal set and the historical operation fault point position set of the oil immersed transformer W to generate a training sample data set D= { (x) 1 (n)，y 1 )，(x 2 (n)，y 2 )，......，(x 5 (n)，y 5 ) -a }; wherein x is 1 (n)，x 2 (n)，...，x 5 (n) represents the 1 st, 2 nd, and the.once-every, 5 th training samples of the machine learning classification model, respectively; y is 1 ，y 2 ，...，y 5 Each of the samples represents a 1 st, 2 nd and 5 th output result; any training sample x i (n) including a sample feature set
Acquiring a current fault sound signal and a current running environment of the oil immersed transformer W;
the current running environment of the oil immersed transformer W comprises time T for acquiring the current fault sound signal of the oil immersed transformer W w And the industrial noise intensity Q of the position where the oil immersed transformer W is w ；
The time T for acquiring the current fault sound signal of the oil immersed transformer W w ∈[0，24]；
Industrial noise intensity Q of position of oil immersed transformer W w ∈[20，100]；
Time T for acquiring current fault sound signal of oil immersed transformer W w ＝10；
Acquiring industrial noise intensity Q of position of oil immersed transformer W w ＝70；
For the time T for acquiring the current fault sound signal of the oil immersed transformer W w And the industrial noise intensity Q of the position where the oil immersed transformer W is located w Normalization processing is carried out to obtain a normalized value T of the time for obtaining the current fault sound signal of the oil immersed transformer W v Normalized value Q of=0.5 and industrial noise intensity at the location of oil immersed transformer W v ＝0.6；
Setting an influence coefficient alpha=0.3 of the time for acquiring the current fault sound signal of the oil immersed transformer W;
setting an influence coefficient beta=0.7 of the industrial noise intensity of the position where the oil immersed transformer W is located;
Probability of the current operating environment of the oil immersed transformer W affecting the current fault sound signal:
p v ＝α*T v +β*Q v ＝0.3*0.5+0.6*0.7＝0.57
wherein p is v The probability that the current running environment of the oil immersed transformer W affects the current fault sound signal is represented; alpha represents an influence coefficient of time for acquiring a current fault sound signal of the oil immersed transformer W; t (T) v A normalized value representing the time of acquiring the current fault sound signal of the oil immersed transformer W; beta represents an influence coefficient of industrial noise intensity of the position where the oil immersed transformer W is located; q (Q) v A normalized value representing the intensity of industrial noise at the location of the oil immersed transformer W.
Setting a probability threshold p of influence of the current running environment of the oil immersed transformer W on the current fault sound signal 0 ＝0.6；
Because p is v ＜p 0 Therefore, the current fault sound signal of the oil immersed transformer W is subjected to spectrum analysis to obtain a current fault sound spectrum characteristic set;
taking the current fault sound signal of the oil immersed transformer W as a current test sample, and marking the current fault sound signal as x u (n) taking the current fault sound spectrum feature set as a sample feature set of the current test sample
Calculating a current test sample x u (n) and training sample x i Euclidean distance between (n):
l can be obtained 2 (x 1 (n)，x u (n))＝2.87；L 2 (x 2 (n)，x u (n))＝0.7；L 2 (x 3 (n)，x u (n))＝5.12；L 2 (x 4 (n)，x u (n))＝6.18；L 2 (x 5 (n)，x u (n))＝3.04；
Setting a current test sample x u (n) and training sample x i (n) Euclidean distance threshold between
Because L is 2 (x 2 (n)，x u (n))=0.7 < 1, and the current test sample x is determined u (n) training sample x which is most similar 2 (n)；
Acquisition of training samples x 2 Sample output result y of (n) 2 The method comprises the steps of carrying out a first treatment on the surface of the Outputting the sample to result y 2 Sample output result y as current test sample u Obtaining the current operation fault point position of y under the current fault sound signal 2 。
Example 2:
acquiring a current fault sound signal and a current running environment of the oil immersed transformer W;
the current running environment of the oil immersed transformer W comprises time T for acquiring the current fault sound signal of the oil immersed transformer W w And the industrial noise intensity Q of the position where the oil immersed transformer W is w ；
The time T for acquiring the current fault sound signal of the oil immersed transformer W w ∈[0，24]；
Industrial noise intensity Q of position of oil immersed transformer W w ∈[20，100]；
Time T for acquiring current fault sound signal of oil immersed transformer W w ＝11；
Acquiring industrial noise intensity Q of position of oil immersed transformer W w ＝80；
For the time T for acquiring the current fault sound signal of the oil immersed transformer W w And the industrial noise intensity Q of the position where the oil immersed transformer W is located w Normalization processing is carried out to obtain a normalized value T of the time for obtaining the current fault sound signal of the oil immersed transformer W v Normalized value Q of=0.6 and industrial noise intensity at the location of oil immersed transformer W v ＝0.7；
Setting an influence coefficient alpha=0.3 of the time for acquiring the current fault sound signal of the oil immersed transformer W;
setting an influence coefficient beta=0.7 of the industrial noise intensity of the position where the oil immersed transformer W is located;
probability of the current operating environment of the oil immersed transformer W affecting the current fault sound signal:
p v ＝α*T v +β*Q v ＝0.3*0.6+0.7*0.7＝0.67
wherein p is v The probability that the current running environment of the oil immersed transformer W affects the current fault sound signal is represented; alpha represents an influence coefficient of time for acquiring a current fault sound signal of the oil immersed transformer W; t (T) v A normalized value representing the time of acquiring the current fault sound signal of the oil immersed transformer W; beta represents an influence coefficient of industrial noise intensity of the position where the oil immersed transformer W is located; q (Q) v A normalized value representing the industrial noise intensity at the location of the oil immersed transformer W;
setting a probability threshold p of influence of the current running environment of the oil immersed transformer W on the current fault sound signal 0 ＝0.6；
Because p is v ＞p 0 Therefore, the system sends out early warning reminding to the manager and stops feeding back the current operation fault point under the current fault sound signal.
It is noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus.
Finally, it should be noted that: the foregoing description of the preferred embodiments of the present invention is not intended to be limiting, but rather, although the present invention has been described in detail with reference to the foregoing embodiments, it will be apparent to those skilled in the art that modifications may be made to the embodiments described above, or equivalents may be substituted for elements thereof. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims (7)

1. The power equipment safe operation management method based on the industrial Internet is characterized by comprising the following steps of:
step S1: acquiring historical data of power equipment by utilizing industrial Internet big data, wherein the power equipment comprises an oil immersed transformer, the historical data comprises historical fault sound signals and historical operation fault points of the oil immersed transformer, and performing spectrum analysis on the historical fault sound signals to obtain a historical fault sound spectrum characteristic set;
step S2: constructing a machine learning classification model, combining a historical fault sound signal set and a historical operation fault point set of the oil immersed transformer to generate a training sample data set, and calculating the Euclidean distance between any two training samples;
Step S3: acquiring a current fault sound signal and a current operating environment of the oil-immersed transformer, constructing an environment influence function, and calculating the probability of the current operating environment of the oil-immersed transformer influencing the current fault sound signal based on the current operating environment of the oil-immersed transformer;
step S4: setting a probability threshold value of influence of the current running environment of the oil-immersed transformer on the current fault sound signal, carrying out spectrum analysis on the current fault sound signal of the oil-immersed transformer to obtain a current fault sound spectrum feature set when the probability of influence of the current running environment of the oil-immersed transformer on the current fault sound signal does not exceed the threshold value, substituting the current fault sound signal of the oil-immersed transformer as a current test sample, substituting the current fault sound spectrum feature set as a sample feature set of the current test sample into a machine learning classification model, and calculating the Euclidean distance between the current test sample and a training sample;
step S5: setting a Euclidean distance threshold value of a current test sample and a training sample, judging the training sample which is most similar to the current test sample when the Euclidean distance between the current test sample and the training sample does not exceed the threshold value, acquiring a sample output result of the training sample, taking the sample output result as a sample output result of the current test sample, namely obtaining a current operation fault point under a current fault sound signal, and feeding back the current operation fault point to an operation and maintenance personnel;
Step S6: when the probability that the current running environment of the oil immersed transformer affects the current fault sound signal exceeds a threshold value, the system sends out early warning reminding to a manager and stops feeding back the current running fault point under the current fault sound signal;
acquiring a historical fault sound signal set A= { x of the oil immersed transformer W 1 (n),x 2 (n),...,x m (n) }; wherein x is 1 (n),x 2 (n),...,x m (n) represents the 1 st, 2 nd, and the third and fourth historical fault sound signals of the oil immersed transformer W; historical operation fault point location set c= { y of corresponding oil immersed transformer W 1 ,y 2 ,...,y m -a }; wherein y is 1 ,y 2 ,...,y m Respectively representing 1 st and 2 nd and the third and fourth historical operation fault points of the oil immersed transformer W; performing spectrum analysis on the historical fault sound signals to obtain a historical fault sound spectrum feature setWherein (1)>The i-th historical fault sound signal includes 1,2, and a-th fault sound spectrum characteristic values, respectively;
performing a spectral analysis on the historical fault sound signal includes:
performing spectrum analysis on the historical fault sound signal by using discrete Fourier transform;
according to the formula:
wherein k=0, 1,2,; x (K) represents the frequency spectrum of the ith historical fault sound signal after discrete Fourier transform; x is x i (n) represents an i-th historical fault sound signal;
the building of the machine learning classification model includes:
combining the historical fault sound signal set and the historical operation fault point position set of the oil immersed transformer W to generate a training sample data set D= { (x) 1 (n),y 1 ),(x 2 (n),y 2 ),......,(x m (n),y m ) -a }; wherein x is 1 (n),x 2 (n),...,x m (n) represents the 1 st, 2 nd, and/or the third, m th training samples of the machine learning classification model, respectively; y is 1 ,y 2 ,...,y m Each of the samples represents a 1 st, 2 nd and a.i.m. output result;
any training sample x i (n) including a sample feature set
According to the formula, the Euclidean distance between any two training samples is calculated as follows:
wherein L is 2 (x i (n),x r (n)) represents training samples x i (n) and training sample x r (n) Euclidean distance between; l represents training sample x i (n) and training sample x r A serial number of the sample feature of (n);
the current running environment of the oil immersed transformer W comprises time T for acquiring the current fault sound signal of the oil immersed transformer W w And the industrial noise intensity Q of the position where the oil immersed transformer W is w ；
The time T for acquiring the current fault sound signal of the oil immersed transformer W w ∈[0,24]；
Industrial noise intensity Q of position of oil immersed transformer W w ∈[Q min ,Q max ]The method comprises the steps of carrying out a first treatment on the surface of the Wherein Q is min A minimum value representing the industrial noise intensity of the position where the oil immersed transformer W is located; q (Q) max A maximum value of the industrial noise intensity indicating the position of the oil immersed transformer W;
for the time T for acquiring the current fault sound signal of the oil immersed transformer W w And the industrial noise intensity Q of the position where the oil immersed transformer W is located w Normalization processing is carried out to obtain a normalized value T of the time for obtaining the current fault sound signal of the oil immersed transformer W v And normalized value Q of industrial noise intensity of position where oil immersed transformer W is v ；
Setting an influence coefficient of time for acquiring a current fault sound signal of the oil immersed transformer W, and marking the influence coefficient as alpha;
setting an influence coefficient of industrial noise intensity of the position where the oil immersed transformer W is positioned, and recording the influence coefficient as beta;
building an environmental impact function:
p v ＝α*T v +β*Q v
wherein p is v The probability that the current running environment of the oil immersed transformer W affects the current fault sound signal is represented; alpha represents an influence coefficient of time for acquiring a current fault sound signal of the oil immersed transformer W; t (T) v Normalized value representing time for acquiring current fault sound signal of oil immersed transformer WThe method comprises the steps of carrying out a first treatment on the surface of the Beta represents an influence coefficient of industrial noise intensity of the position where the oil immersed transformer W is located; q (Q) v A normalized value representing the intensity of industrial noise at the location of the oil immersed transformer W.
2. The industrial internet-based power equipment safe operation management method according to claim 1, wherein: in the steps S4-S6,
setting a probability threshold value of the influence of the current running environment of the oil immersed transformer W on the current fault sound signal, and recording the probability threshold value as p 0 ；
When the probability of the current running environment of the oil immersed transformer W affecting the current fault sound signal does not exceed a threshold value, performing spectrum analysis on the current fault sound signal of the oil immersed transformer W to obtain a current fault sound spectrum feature set;
taking the current fault sound signal of the oil immersed transformer W as a current test sample, and marking the current fault sound signal as x u (n) taking the current fault sound spectrum feature set as a sample feature set of the current test sample
Calculating a current test sample x u (n) and training sample x i Euclidean distance between (n):
wherein L is 2 (x i (n),x u (n)) represents the current test sample x u (n) and training sample x i (n) Euclidean distance between;representing training samples x i The first sample feature of (n); />Representing the current test sample x u The first sample feature of (n);
setting a current test sample x u (n) and training sample x i The Euclidean distance threshold between (n) is recorded as
When (when)When it is judged that the current test sample x is identical to the current test sample x u (n) training sample x which is most similar i (n)；
Acquisition of training samples x i Sample output result y of (n) i The method comprises the steps of carrying out a first treatment on the surface of the Outputting the sample to result y i Sample output result y as current test sample u Obtaining the current operation fault point position under the current fault sound signal;
when the probability that the current running environment of the oil immersed transformer W affects the current fault sound signal exceeds a threshold value, the system sends out early warning reminding to a manager and stops feeding back the current running fault point under the current fault sound signal.
3. An industrial internet-based power equipment safe operation management system applying the industrial internet-based power equipment safe operation management method of any one of claims 1 to 2, characterized in that: the system comprises a data acquisition module, a data processing module, an influence function construction analysis module, a classification model construction analysis module, a similarity judgment analysis module and a feedback reminding module;
the data acquisition module is used for acquiring historical data of power equipment, current fault sound signals of the oil immersed transformer and current operation environments by utilizing industrial Internet big data, wherein the power equipment comprises the oil immersed transformer, and the historical data comprises historical fault sound signals and historical operation fault points of the oil immersed transformer; the data processing module is used for carrying out spectrum analysis on the historical fault sound signals and the current fault sound signals of the oil immersed transformer to obtain a historical fault sound spectrum characteristic set and a current fault sound spectrum characteristic set; the influence function construction analysis module is used for constructing an environment influence function, calculating the probability of influence of the current operation environment of the oil-immersed transformer on the current fault sound signal based on the current operation environment of the oil-immersed transformer, setting a probability threshold value of influence of the current operation environment of the oil-immersed transformer on the current fault sound signal, and judging whether the probability of influence of the current operation environment of the oil-immersed transformer on the current fault sound signal exceeds the threshold value; the classification model construction analysis module is used for constructing a machine learning classification model, combining a historical fault sound signal set and a historical operation fault point set of the oil immersed transformer to generate a training sample data set, calculating the Euclidean distance between any two training samples, taking the current fault sound signal of the oil immersed transformer as a current test sample, substituting the current fault sound spectrum characteristic set as a sample characteristic set of the current test sample into the machine learning classification model, and calculating the Euclidean distance between the current test sample and the training sample; the similarity judging and analyzing module is used for setting a Euclidean distance threshold value of a current test sample and a training sample, judging the training sample which is most similar to the current test sample when the Euclidean distance between the current test sample and the training sample does not exceed the threshold value, acquiring a sample output result of the training sample, taking the sample output result as a sample output result of the current test sample, namely obtaining a current operation fault point under a current fault sound signal, and feeding back the current operation fault point to an operation and maintenance person; the feedback reminding module is used for sending out early warning reminding to a manager when the probability of the current running environment of the oil immersed transformer affecting the current fault sound signal exceeds a threshold value, and stopping feeding back the current running fault point under the current fault sound signal;
The output end of the data acquisition module is connected with the input end of the data processing module; the output end of the data processing module is connected with the input end of the influence function construction analysis module; the output end of the influence function construction analysis module is connected with the input end of the classification model construction analysis module; the output end of the classification model construction analysis module is connected with the input end of the similarity judgment analysis module; the output end of the similarity judging and analyzing module is connected with the input end of the feedback reminding module.
4. The industrial internet-based power equipment safe operation management system according to claim 3, wherein: the data acquisition module comprises a historical data acquisition unit and a current data acquisition unit;
the historical data acquisition unit is used for acquiring historical data of the power equipment by utilizing industrial Internet big data, the power equipment comprises an oil immersed transformer, and the historical data comprises historical fault sound signals and historical operation fault points of the oil immersed transformer;
the current data acquisition unit is used for acquiring a current fault sound signal and a current running environment of the oil immersed transformer;
The current running environment comprises time T for acquiring current fault sound signals of the oil immersed transformer w And the industrial noise intensity Q of the position of the oil immersed transformer w ；
The output end of the historical data acquisition unit is connected with the input end of the current data acquisition unit; the output end of the current data acquisition unit is connected with the input end of the data processing module;
the data processing module comprises a spectrum analysis unit and a spectrum characteristic acquisition unit;
the spectrum analysis unit is used for carrying out spectrum analysis on the historical fault sound signals and the current fault sound signals of the oil immersed transformer;
the frequency spectrum characteristic acquisition unit is used for acquiring a historical fault sound frequency spectrum characteristic set and a current fault sound frequency spectrum characteristic set;
the output end of the spectrum analysis unit is connected with the input end of the spectrum characteristic acquisition unit; the output end of the frequency spectrum characteristic acquisition unit is connected with the input end of the influence function construction analysis module.
5. The industrial internet-based power equipment safe operation management system according to claim 4, wherein: the influence function construction analysis module comprises an influence function construction unit and an analysis judgment unit;
The influence function construction unit is used for constructing an environment influence function and calculating the probability of influence of the current operation environment of the oil-immersed transformer on the current fault sound signal based on the current operation environment of the oil-immersed transformer;
the analysis judging unit is used for setting a probability threshold value of the influence of the current running environment of the oil-immersed transformer on the current fault sound signal and judging whether the probability of the influence of the current running environment of the oil-immersed transformer on the current fault sound signal exceeds the threshold value;
the output end of the influence function construction unit is connected with the input end of the analysis judging unit; the output end of the analysis judging unit is connected with the input end of the classification model construction analysis module.
6. The industrial internet-based power equipment safe operation management system according to claim 5, wherein: the classification model construction and analysis module comprises a machine learning classification model construction unit and a machine learning classification model analysis unit;
the machine learning classification model construction unit is used for constructing a machine learning classification model, combining a historical fault sound signal set and a historical operation fault point set of the oil immersed transformer to generate a training sample data set, and calculating Euclidean distance between any two training samples;
The machine learning classification model analysis unit is used for taking a current fault sound signal of the oil immersed transformer as a current test sample, substituting the current fault sound spectrum characteristic set as a sample characteristic set of the current test sample into the machine learning classification model, and calculating the Euclidean distance between the current test sample and the training sample;
the output end of the machine learning classification model construction unit is connected with the input end of the machine learning classification model analysis unit; the output end of the machine learning classification model analysis unit is connected with the input end of the similarity judgment analysis module.
7. The industrial internet-based power equipment safe operation management system according to claim 6, wherein: the similarity judging and analyzing module comprises a threshold setting unit and a judging and outputting unit;
the threshold setting unit is used for setting a Euclidean distance threshold between a current test sample and a training sample;
the judging and outputting unit is used for judging a training sample which is most similar to the current test sample when the Euclidean distance between the current test sample and the training sample does not exceed a threshold value, obtaining a sample output result of the training sample, taking the sample output result as the sample output result of the current test sample, namely obtaining a current operation fault point under a current fault sound signal, and feeding back the current operation fault point to an operation and maintenance personnel;
The output end of the threshold setting unit is connected with the input end of the judging output unit; and the output end of the judging output unit is connected with the input end of the feedback reminding module.
CN202211498512.6A 2022-11-28 2022-11-28 Power equipment safe operation management system and method based on industrial Internet Active CN115860714B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202211498512.6A CN115860714B (en) 2022-11-28 2022-11-28 Power equipment safe operation management system and method based on industrial Internet

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202211498512.6A CN115860714B (en) 2022-11-28 2022-11-28 Power equipment safe operation management system and method based on industrial Internet

Publications (2)

Publication Number Publication Date
CN115860714A CN115860714A (en) 2023-03-28
CN115860714B true CN115860714B (en) 2023-08-08

Family

ID=85666995

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202211498512.6A Active CN115860714B (en) 2022-11-28 2022-11-28 Power equipment safe operation management system and method based on industrial Internet

Country Status (1)

Country Link
CN (1) CN115860714B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116339267B (en) * 2023-05-25 2023-08-08 深圳市星火数控技术有限公司 Automatic production line control system based on Internet of things

Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110018389A (en) * 2019-02-21 2019-07-16 国网山东省电力公司临沂供电公司 A kind of transmission line of electricity on-line fault monitoring method and system
CN112257988A (en) * 2020-09-29 2021-01-22 中广核工程有限公司 Complex accident feature identification and risk early warning system and method for nuclear power plant
CN112383052A (en) * 2020-11-16 2021-02-19 国网电子商务有限公司 Power grid fault repairing method and device based on power internet of things
CN112395959A (en) * 2020-10-30 2021-02-23 天合云能源互联网技术(杭州)有限公司 Power transformer fault prediction and diagnosis method and system based on audio features
CN113011656A (en) * 2021-03-22 2021-06-22 内蒙古电力(集团)有限责任公司内蒙古电力科学研究院分公司 Power station auxiliary machine fault early warning method and system
CN113466616A (en) * 2021-06-22 2021-10-01 海南电网有限责任公司乐东供电局 Method and device for quickly positioning cable fault point
CN113985194A (en) * 2021-09-28 2022-01-28 广西电网有限责任公司电力科学研究院 Power distribution network fault positioning method based on stack self-encoder
CN114036998A (en) * 2021-09-24 2022-02-11 浪潮集团有限公司 Method and system for fault detection of industrial hardware based on machine learning
CN115081584A (en) * 2022-05-16 2022-09-20 合肥科大智能机器人技术有限公司 Power equipment health management method, system and medium based on machine learning
CN115372752A (en) * 2021-05-18 2022-11-22 中移(上海)信息通信科技有限公司 Fault detection method, device, electronic equipment and storage medium

Patent Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110018389A (en) * 2019-02-21 2019-07-16 国网山东省电力公司临沂供电公司 A kind of transmission line of electricity on-line fault monitoring method and system
CN112257988A (en) * 2020-09-29 2021-01-22 中广核工程有限公司 Complex accident feature identification and risk early warning system and method for nuclear power plant
CN112395959A (en) * 2020-10-30 2021-02-23 天合云能源互联网技术(杭州)有限公司 Power transformer fault prediction and diagnosis method and system based on audio features
CN112383052A (en) * 2020-11-16 2021-02-19 国网电子商务有限公司 Power grid fault repairing method and device based on power internet of things
CN113011656A (en) * 2021-03-22 2021-06-22 内蒙古电力(集团)有限责任公司内蒙古电力科学研究院分公司 Power station auxiliary machine fault early warning method and system
CN115372752A (en) * 2021-05-18 2022-11-22 中移(上海)信息通信科技有限公司 Fault detection method, device, electronic equipment and storage medium
CN113466616A (en) * 2021-06-22 2021-10-01 海南电网有限责任公司乐东供电局 Method and device for quickly positioning cable fault point
CN114036998A (en) * 2021-09-24 2022-02-11 浪潮集团有限公司 Method and system for fault detection of industrial hardware based on machine learning
CN113985194A (en) * 2021-09-28 2022-01-28 广西电网有限责任公司电力科学研究院 Power distribution network fault positioning method based on stack self-encoder
CN115081584A (en) * 2022-05-16 2022-09-20 合肥科大智能机器人技术有限公司 Power equipment health management method, system and medium based on machine learning

Also Published As

Publication number Publication date
CN115860714A (en) 2023-03-28

Similar Documents

Publication Publication Date Title
US6785637B1 (en) Method for monitoring wind power plants
Leite et al. Detection of localized bearing faults in induction machines by spectral kurtosis and envelope analysis of stator current
KR20200014129A (en) Diagnosis method of electric transformer using Deep Learning
JP2013013075A (en) Electrical substation fault monitoring and diagnostics
CN115860714B (en) Power equipment safe operation management system and method based on industrial Internet
CN109507510A (en) A kind of transformer fault diagnosis system
Coelho et al. Performance comparison of classifiers in the detection of short circuit incipient fault in a three-phase induction motor
CN107798283A (en) A kind of neural network failure multi classifier based on the acyclic figure of decision-directed
Chen et al. Fault anomaly detection of synchronous machine winding based on isolation forest and impulse frequency response analysis
CN110910897B (en) Feature extraction method for motor abnormal sound recognition
Leonidovich et al. The development and use of diagnostic systems and estimation of residual life in industrial electrical equipment
CN114157023B (en) Distribution transformer early warning information acquisition method
Cao et al. Remaining useful life prediction of wind turbine generator bearing based on EMD with an indicator
CN117153193B (en) Power equipment fault voiceprint recognition method integrating physical characteristics and data diagnosis
CN117849495A (en) Transformer operation performance evaluation method and system
CN114485379A (en) Transformer winding on-line monitoring method
CN202281934U (en) Power transformer on-line state monitoring device based on audio frequency identification technique
Sun et al. Research progress on oil-immersed transformer mechanical condition identification based on vibration signals
Secic et al. Using deep neural networks for on-load tap changer audio-based diagnostics
Ha et al. Degradation trend estimation and prognostics for low speed gear lifetime
Chang et al. Fuzzy theory-based partial discharge technique for operating state diagnosis of high-voltage motor
CN112001550A (en) Multipoint power quality monitoring system for transformer substation
CN116884432A (en) VMD-JS divergence-based power transformer fault voiceprint diagnosis method
Refaat et al. Smart grid condition assessment: concepts, benefits, and developments
Gopinath et al. Insulation condition assessment of high‐voltage rotating machines using hybrid techniques

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
GR01 Patent grant
GR01 Patent grant