CN114186184A - Power distribution terminal fault diagnosis method, system, device and storage medium - Google Patents

Power distribution terminal fault diagnosis method, system, device and storage medium Download PDF

Info

Publication number
CN114186184A
CN114186184A CN202111463158.9A CN202111463158A CN114186184A CN 114186184 A CN114186184 A CN 114186184A CN 202111463158 A CN202111463158 A CN 202111463158A CN 114186184 A CN114186184 A CN 114186184A
Authority
CN
China
Prior art keywords
state
fault
distribution terminal
power distribution
measurement data
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202111463158.9A
Other languages
Chinese (zh)
Inventor
赖奎
林浩泉
黄景云
赵国荣
苏博波
戴雄杰
李志娟
施明
李岳锋
苏海林
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.)
Guangdong Power Grid Co Ltd
Jiangmen Power Supply Bureau of Guangdong Power Grid Co Ltd
Original Assignee
Guangdong Power Grid Co Ltd
Jiangmen Power Supply Bureau of Guangdong Power Grid 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 Guangdong Power Grid Co Ltd, Jiangmen Power Supply Bureau of Guangdong Power Grid Co Ltd filed Critical Guangdong Power Grid Co Ltd
Priority to CN202111463158.9A priority Critical patent/CN114186184A/en
Publication of CN114186184A publication Critical patent/CN114186184A/en
Priority to PCT/CN2022/135595 priority patent/WO2023098753A1/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/14—Fourier, Walsh or analogous domain transformations, e.g. Laplace, Hilbert, Karhunen-Loeve, transforms
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R23/00—Arrangements for measuring frequencies; Arrangements for analysing frequency spectra
    • G01R23/16—Spectrum analysis; Fourier analysis
    • 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
    • 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
    • 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
    • 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
    • Y04S10/52—Outage or fault management, e.g. fault detection or location

Abstract

The invention relates to the technical field of power failure diagnosis and discloses a power distribution terminal failure diagnosis method, a power distribution terminal failure diagnosis system, a power distribution terminal failure diagnosis device and a storage medium. The method comprises the steps of collecting the running state of the power distribution terminal and generating a state sequence, establishing a measurement data model corresponding to each fault type according to the state sequence, processing the multiple measurement data models through short-time Fourier transform to convert time domain data types into time-frequency graphs, inputting the processed measurement data models into a one-dimensional neural network for training to obtain a trained fault diagnosis model, and identifying the fault of the power distribution terminal through the fault diagnosis model; the method and the device can realize the rapid diagnosis of the running state of the power distribution terminal.

Description

Power distribution terminal fault diagnosis method, system, device and storage medium
Technical Field
The invention relates to the technical field of power fault diagnosis, in particular to a power distribution terminal fault diagnosis method, system, device and storage medium.
Background
Power distribution terminals are important devices to address the problem of power distribution reliability, and may monitor related devices, signals, and power in a power distribution network. Power distribution terminals are susceptible to installation environments and to failure during actual operation. For a distribution automation system, a maintainer can only realize the fault of a distribution terminal under the condition that the distribution terminal is completely disconnected or refuses to operate, and cannot quickly diagnose the hidden trouble of the fault. In addition, troubleshooting of the power distribution terminal is often performed offline under the condition of complete forbidding, and the passive and delayed troubleshooting mode makes it difficult for the power distribution terminal to perform troubleshooting at an early stage, so that operation and maintenance expenses are greatly increased.
Disclosure of Invention
The invention provides a method, a system, a device and a storage medium for diagnosing faults of a power distribution terminal, and solves the technical problem of rapid identification of the faults of the power distribution terminal.
The invention provides a fault diagnosis method for a power distribution terminal, which comprises the following steps:
acquiring the running state of a power distribution terminal and generating a state sequence;
establishing a measurement data model corresponding to each fault type according to the state sequence, wherein the measurement data model corresponding to each fault type comprises a measurement data model in a normal state, a measurement data model corresponding to a precision distortion type fault, a measurement data model corresponding to an attenuation oscillation type fault and a measurement data model corresponding to a harmonic interference type fault;
processing the multiple measurement data models through short-time Fourier transform to convert the time domain data types in the measurement data models into time-frequency graphs;
inputting the processed measurement data model into a one-dimensional neural network for training to obtain a trained fault diagnosis model;
and inputting a state sequence to be tested, diagnosing the fault type by using the fault diagnosis model and outputting a diagnosis result.
According to an implementation manner of the first aspect of the present invention, the acquiring the operation state of the power distribution terminal and generating the state sequence includes:
setting corresponding monitoring points according to the fault types of the functional modules of the power distribution terminal;
collecting the running state of each monitoring point, and judging the running state of each monitoring point by a self-diagnosis method to obtain a corresponding running state judgment vector;
and generating a state sequence according to the running state evaluation vector of each monitoring point.
According to a mode that can be realized by the first aspect of the present invention, each functional module of the power distribution terminal includes a central processing unit, an operation control loop, a communication module, a power module and an acquisition module, and the setting of the corresponding monitoring point according to the fault type of each functional module of the power distribution terminal includes:
setting monitoring points of the central processing unit to comprise a task running state, a system parameter setting state and a global positioning system time setting state;
setting monitoring points of the operation control loop to comprise a telemetering state;
setting monitoring points of the communication module to comprise an uplink communication state and a downlink communication state;
setting monitoring points of the power supply module to comprise a power supply state and a battery state;
the monitoring points of the acquisition module are set to comprise an acquisition unit state and a sensor probe state.
According to a possible implementation manner of the first aspect of the present invention, the inputting the processed measurement data model into a one-dimensional neural network for training includes:
constructing a loss function by using the improved cross entropy, and training a model of the one-dimensional neural network by using the loss function;
wherein the improved cross entropy is calculated as:
Figure BDA0003389395530000021
in the formula (I), the compound is shown in the specification,
Figure BDA0003389395530000022
interval of the function of degree of matching for the fault model, yiMatching value for distribution terminal operation state class, alpha is balance factor, alpha belongs to (0,1), gamma is modulation factor, and
Figure BDA0003389395530000023
the weight factors are constructed to reduce the loss function value to a single sample that is easy to classify.
A second aspect of the present invention provides a power distribution terminal fault diagnosis system, including:
the acquisition module is used for acquiring the running state of the power distribution terminal and generating a state sequence;
the measurement data model building module is used for building a measurement data model corresponding to each fault type according to the state sequence, wherein the measurement data model corresponding to each fault type comprises a measurement data model in a normal state, a measurement data model corresponding to a precision distortion fault, a measurement data model corresponding to an attenuation oscillation fault and a measurement data model corresponding to a harmonic interference fault;
the time-frequency transformation module is used for processing the multiple measurement data models through short-time Fourier transformation so as to convert the time-domain data types in the measurement data models into time-frequency graphs;
the training module is used for inputting the processed measurement data model into a one-dimensional neural network for training to obtain a trained fault diagnosis model;
and the fault diagnosis module is used for inputting a state sequence to be tested, diagnosing the fault type by using the fault diagnosis model and outputting a diagnosis result.
According to an implementable manner of the second aspect of the present invention, the acquisition module comprises:
the monitoring point setting unit is used for setting corresponding monitoring points according to the fault types of the functional modules of the power distribution terminal;
the running state judgment vector acquisition unit is used for acquiring the running state of each monitoring point and judging the running state of each monitoring point through the self-diagnosis system to obtain a corresponding running state judgment vector;
and the state sequence generating unit is used for generating a state sequence according to the running state evaluation vector of each monitoring point.
According to a mode that can be realized in the second aspect of the present invention, each functional module of the power distribution terminal includes a central processing unit, an operation control loop, a communication module, a power module, and an acquisition module, and the monitoring point setting unit includes:
the first setting subunit is used for setting monitoring points of the central processing unit to comprise a task running state, a system parameter setting state and a global positioning system time setting state;
the second setting subunit is used for setting monitoring points of the operation control loop to comprise a telemetering state;
the third setting subunit is used for setting monitoring points of the communication module to comprise an uplink communication state and a downlink communication state;
the fourth setting subunit is used for setting monitoring points of the power supply module to comprise a power supply state and a battery state;
and the fifth setting subunit is used for setting the monitoring point of the acquisition module to comprise an acquisition unit state and a sensor probe state.
According to an implementable manner of the second aspect of the present invention, the training module is specifically configured to:
constructing a loss function by using the improved cross entropy, and training a model of the one-dimensional neural network by using the loss function;
wherein the improved cross entropy is calculated as:
Figure BDA0003389395530000041
in the formula (I), the compound is shown in the specification,
Figure BDA0003389395530000042
interval of the function of degree of matching for the fault model, yiMatching value for distribution terminal operation state class, alpha is balance factor, alpha belongs to (0,1), gamma is modulation factor, and
Figure BDA0003389395530000043
the weight factors are constructed to reduce the loss function value to a single sample that is easy to classify.
A third aspect of the present invention provides a power distribution terminal fault diagnosis apparatus, including:
a memory to store instructions; the instruction is an instruction which can realize the fault diagnosis method of the power distribution terminal in any one of the realizable modes;
a processor to execute the instructions in the memory.
A fourth aspect of the present invention is a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements a power distribution terminal fault diagnosis method as described in any one of the above-implementable manners.
According to the technical scheme, the invention has the following advantages:
the method comprises the steps of collecting the running state of the power distribution terminal and generating a state sequence, establishing a measurement data model corresponding to each fault type according to the state sequence, processing the multiple measurement data models through short-time Fourier transform to convert time domain data types into time-frequency graphs, inputting the processed measurement data models into a one-dimensional neural network for training to obtain a trained fault diagnosis model, and identifying the fault of the power distribution terminal through the fault diagnosis model; the method and the device can realize the rapid diagnosis of the running state of the power distribution terminal, solve the problems of outstanding quality problem, high operation and maintenance difficulty and high operation and maintenance cost of the power distribution terminal at present, are beneficial to improving the intelligent level and the working efficiency of the operation and maintenance work of the power distribution terminal, and have higher practicability and applicability to the actual engineering problem.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to these drawings without inventive exercise.
Fig. 1 is a flow chart of a method for diagnosing faults of a power distribution terminal according to an alternative embodiment of the present invention;
fig. 2 is a block diagram of a power distribution terminal fault diagnosis system according to an alternative embodiment of the present invention.
Description of the drawings:
1-an acquisition module; 2-a measurement data model building module; 3-a time-frequency transform module; 4-a training module; 5-fault diagnosis module.
Detailed Description
The embodiment of the invention provides a method, a system, a device and a storage medium for diagnosing faults of a power distribution terminal, which are used for solving the technical problem of rapid identification of the faults of the power distribution terminal.
In order to make the objects, features and advantages of the present invention more obvious and understandable, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is obvious that the embodiments described below are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1, fig. 1 is a flowchart illustrating a method for diagnosing a fault of a power distribution terminal according to an embodiment of the present invention.
The power distribution terminal fault diagnosis method provided by the embodiment of the invention comprises the steps of S1-S5.
And step S1, acquiring the operation state of the power distribution terminal and generating a state sequence.
In one implementation, the acquiring the operating state of the power distribution terminal and generating the state sequence includes:
setting corresponding monitoring points according to the fault types of the functional modules of the power distribution terminal;
collecting the running state of each monitoring point, and judging the running state of each monitoring point by a self-diagnosis method to obtain a corresponding running state judgment vector;
and generating a state sequence according to the running state evaluation vector of each monitoring point.
In order to realize the standardization of the fault diagnosis process of the power distribution terminal, all functional modules of the power distribution terminal are divided into a central processing unit, an operation control loop, a communication module, a power supply module and an acquisition module.
The central processing unit is a core component of the power distribution terminal and is mainly responsible for data acquisition, error analysis, remote control command transmission, communication processing of the power distribution terminal and the like.
In order to set corresponding monitoring points, the faults of each module need to be analyzed to obtain fault description information. The fault description information of each module is as follows:
(1) a power supply module:
the fault phenomena comprise PT burnout and explosion, frequent starting of a backup power supply and abnormal voltage of the backup power supply; the fault types caused by the fault phenomenon comprise a power supply PT (potential transformer) fault, a power supply CT (current transformer) fault and a backup power supply fault;
(2) an acquisition module:
the fault phenomenon comprises the fact that collected data are far from the mean value and the collected data are not collected, and the corresponding fault types comprise measuring equipment aging, mutual inductor faults and filter circuit faults;
(3) a communication module:
the fault phenomenon comprises weak communication signals, no response, repeated link establishment or frequent initialization, and the corresponding fault types comprise communication interface faults and wireless communication faults;
(4) operating the control loop:
the fault phenomena comprise switch failure and switch misoperation, and the corresponding fault types comprise control loop faults;
(5) a central processing unit:
the fault phenomena comprise that the terminal is in an off-line state, the terminal software is restarted and the switch is refused to operate, and the corresponding fault types comprise circuit board aging faults and terminal software crash.
Based on the above-mentioned fault description information, as a preferred embodiment, the corresponding monitoring points may be set as follows:
setting monitoring points of the central processing unit to comprise a task running state, a system parameter setting state and a global positioning system time setting state;
setting monitoring points of the operation control loop to comprise a telemetering state;
setting monitoring points of the communication module to comprise an uplink communication state and a downlink communication state;
setting monitoring points of the power supply module to comprise a power supply state and a battery state;
the monitoring points of the acquisition module are set to comprise an acquisition unit state and a sensor probe state.
The method for judging the running state of each monitoring point by a self-diagnosis method to obtain a corresponding running state judgment vector comprises the following steps:
(1) a central processing unit:
self-diagnosis of the task running state: judging whether the software task is normally executed or not through the software execution monitoring program; the corresponding serial number of the operation state evaluation vector in the state sequence is M1;
self-diagnosis of parameters: the parameter self-diagnosis mainly comprises self-diagnosis of power distribution terminal system parameters and limit setting parameters, wherein the power distribution terminal system parameters comprise IP addresses and equipment addresses, and the limit setting parameters comprise anti-shake time; the corresponding serial number of the operation state evaluation vector in the state sequence is M2;
and (3) time setting self-diagnosis: by designing a self-diagnosis synchronous interruption program, comparing a time synchronization result with a program after interruption recovery, and if the comparison result is consistent, indicating that the global positioning system has no fault when synchronizing time; the corresponding serial number of the operation state evaluation vector in the state sequence is M3;
(2) operating the control loop:
the remote measuring state is realized by intelligently comparing remote control commands sent by the central processing unit and the acquisition module and determining the causal relationship of two times of hits, and the number of the operation state judgment vector in the state sequence is M4;
(3) self-diagnosis of the communication module:
the self-diagnosis of the power distribution terminal communication module comprises an uplink communication part and a downlink communication part. The self-diagnosis method of the communication module fault is realized by confirming the return of the instruction; the self-diagnosis module determines whether the module has a fault by sending a query command to the master station or the upstream and downstream intelligent equipment and checking whether the module can receive a response; the numbers of the operation state evaluation vectors in the state sequence are M5 and M6.
(4) Self-diagnosis of the power module:
self-diagnosis of the power module includes detecting whether the power is off and whether a connected battery is missing or causes under-voltage; the corresponding numbers of the operation state evaluation vectors in the state sequence are M7, M8 and M9 respectively;
(5) self-diagnosis of the acquisition module:
the data collected for the power distribution terminal includes telemetry and external signal data, with external measurements including voltage and current values; in order to conveniently describe the self-diagnosis model of the acquisition module, the telemetering data is ξ, the threshold value is ε, the average value is θ, the difference value is set to be larger than δ, and the self-diagnosis judgment principle of the acquisition module is as follows:
if the | xi-theta | is larger than 0, the data is far away from the mean value, and the fault of the acquisition unit is judged;
if xi is larger than delta, indicating that the data is larger than a threshold value, judging that the acquisition unit has a fault;
if delta is greater than 0, indicating that the data is 0, judging that the sensor probe has a failure fault;
the numbers of the operation state evaluation vectors corresponding to the three fault types in the state sequence are M10, M11 and M12 respectively.
In the self-diagnosis described above, the value of the operation state evaluation vector is 1 or 0. The definition rule of the value of each operation state judgment vector can be set according to the actual situation. For example, when M [10] is 0, it indicates that the acquisition unit is not faulty, and when M [10] is 1, it indicates that the acquisition unit is faulty. For another example, when M [1] is equal to 1, it indicates that the software task is normally executed, and when M [1] is equal to 0, it indicates that the software task is abnormally executed. The values of the operating state evaluation vectors of all monitoring points form a state sequence M at a given time, where M ═ M [1], M [2], …, M [ n ]), and M [ k ] is the value of the operating state evaluation vector of monitoring point k at the current time.
Step S2, establishing a measured data model corresponding to each fault type according to the state sequence, wherein the measured data model corresponding to each fault type comprises a measured data model in a normal state, a measured data model corresponding to a precision distortion fault, a measured data model corresponding to an attenuation oscillation fault and a measured data model corresponding to a harmonic interference fault.
Because measuring devices such as a mutual inductor, an analog-to-digital converter and the like in a power distribution terminal are often interfered by various factors in actual operation, certain random noise is added to actually acquired alternating current voltage or current, and the noise can be generally considered to be subjected to Gaussian distribution. Therefore, as a preferred embodiment, the measurement data model in the normal state is set as:
Figure BDA0003389395530000081
in the formula, N (0, delta)2) Is a mean value of 0 and a variance of δ2A is the amplitude of the measurement voltage or current, ω is the angular frequency of the measurement voltage or current,
Figure BDA0003389395530000082
the phase angle of the voltage or current is measured.
When signal processing modules such as transformers and the like and a signal transmission unit and the like in a power distribution terminal acquisition system break down, the precision of a measured value is distorted, and the measured value has the characteristic of large noise, which is also called as high-noise fault data. When a precision distortion fault occurs, the average value of voltage and current measurement is unchanged, the measurement variance is changed, and the measurement data model corresponding to the precision distortion fault at the moment is as follows:
Figure BDA0003389395530000083
in the formula, Nf(0,δf 2) Is a mean value of 0 and a variance of δf 2The measurement noise of the gaussian distribution of (a);
when the measuring equipment in the power distribution terminal ages or performance is attenuated, the measured data often has certain oscillation attenuation characteristics, the measured data model at the moment is the measured data model corresponding to the oscillation attenuation type fault, and the measured data model is set as:
Figure BDA0003389395530000091
in the formula, b is an oscillation attenuation constant, and t is oscillation measurement time;
when a measurement component in a power distribution terminal is subjected to environmental interference or component failure, the measured ac power deviates from the fundamental frequency. For example, ferromagnetic resonance occurs when the excitation characteristic of the electromagnetic transformer is degraded, so that the measurement contains higher harmonic components, the measurement data model at the moment is the measurement data model corresponding to the harmonic interference type fault, and the measurement data model is set as:
Figure BDA0003389395530000092
in the formula, A2、
Figure BDA0003389395530000093
Respectively representing the amplitude and phase angle, A, of the 2 nd harmonic componentz、
Figure BDA0003389395530000094
Respectively representing the magnitude and phase angle of the z-th harmonic component.
And step S3, processing the multiple measured data models through short-time Fourier transform to convert the time domain data types in the measured data models into time-frequency graphs.
The established measurement data model of the power distribution terminal is a time domain model, and Short-time Fourier Transform (STFT) is introduced in the step for enhancing the class characteristics of different data types, so that the digital data sample is converted into a time-frequency map containing time domain and frequency domain information.
The STFT is a joint time-frequency analysis method aiming at time-varying and non-stationary signals, can convert one-dimensional fault signal data into a characteristic spectrum containing time domain and frequency domain information, and the conversion formula is shown as the following formula:
Figure BDA0003389395530000095
wherein f (t) is a measured data model to be transformed, and g (t-mu) is a window function centered at mu time;
in this embodiment, a commonly used hamming window function is selected, and its expression is shown in the following formula:
Figure BDA0003389395530000096
in the formula, T is the measurement time length of the window function, and T is time.
Step S4, inputting the processed measurement data model into a one-dimensional neural network for training to obtain a trained fault diagnosis model;
and step S5, inputting the state sequence to be tested, utilizing the fault diagnosis model to diagnose the fault type and outputting the diagnosis result.
In one implementation, the inputting the processed measurement data model into a one-dimensional neural network for training includes:
constructing a loss function by using the improved cross entropy, and training a model of the one-dimensional neural network by using the loss function;
wherein the improved cross entropy is calculated as:
Figure BDA0003389395530000101
in the formula (I), the compound is shown in the specification,
Figure BDA0003389395530000102
interval of the function of degree of matching for the fault model, yiMatching value for distribution terminal operation state class, alpha is balance factor, alpha belongs to (0,1), gamma is modulation factor, and
Figure BDA0003389395530000103
the weight factors being constructed to reduce the value of the loss function to a value which is easy to implementA single sample of the classification.
Although the conventional one-dimensional neural network can process sequence data, the following problems still exist: 1) as the depth of the neural network increases, the amount of parameters increases sharply, scattering of the gradient or explosion of the gradient becomes more severe and the model is difficult to form and optimize; 2) the size of a simple convolution kernel is difficult to capture time information of different time granularities, and the performance of evaluation is influenced. In view of the above, there is a need to improve the structure of conventional one-dimensional neural networks.
Since the number of faults is less than that in the normal operation state, the state data of the stable operation of the power distribution terminal is more than that of the fault of the power distribution terminal, which causes the model to pay more attention to the stable sample, thereby causing misjudgment of the unstable sample.
In order to solve the problems, the embodiment of the invention improves the cross entropy, and utilizes the improved cross entropy to construct the loss function to train the model of the one-dimensional neural network, so that the precision of training the fault diagnosis model can be effectively improved.
The invention also provides a power distribution terminal fault diagnosis system.
Referring to fig. 2, fig. 2 is a block diagram illustrating a power distribution terminal fault diagnosis system according to an embodiment of the present invention.
The power distribution terminal fault diagnosis system provided by the embodiment of the invention comprises:
the acquisition module 1 is used for acquiring the running state of the power distribution terminal and generating a state sequence;
the measurement data model building module 2 is used for building a measurement data model corresponding to each fault type according to the state sequence, wherein the measurement data model corresponding to each fault type comprises a measurement data model in a normal state, a measurement data model corresponding to a precision distortion fault, a measurement data model corresponding to an attenuation oscillation fault and a measurement data model corresponding to a harmonic interference fault;
the time-frequency transformation module 3 is used for processing the multiple measurement data models through short-time Fourier transformation so as to convert the time-domain data types in the measurement data models into time-frequency graphs;
the training module 4 is used for inputting the processed measurement data model into a one-dimensional neural network for training to obtain a trained fault diagnosis model;
and the fault diagnosis module 5 is used for inputting a state sequence to be tested, diagnosing the fault type by using the fault diagnosis model and outputting a diagnosis result.
In an implementable manner, the acquisition module 1 comprises:
the monitoring point setting unit is used for setting corresponding monitoring points according to the fault types of the functional modules of the power distribution terminal;
the running state judgment vector acquisition unit is used for acquiring the running state of each monitoring point and judging the running state of each monitoring point through the self-diagnosis system to obtain a corresponding running state judgment vector;
and the state sequence generating unit is used for generating a state sequence according to the running state evaluation vector of each monitoring point.
In an implementation manner, each functional module of the power distribution terminal includes a central processing unit, an operation control loop, a communication module, a power module and an acquisition module 1, and the monitoring point setting unit includes:
the first setting subunit is used for setting monitoring points of the central processing unit to comprise a task running state, a system parameter setting state and a global positioning system time setting state;
the second setting subunit is used for setting monitoring points of the operation control loop to comprise a telemetering state;
the third setting subunit is used for setting monitoring points of the communication module to comprise an uplink communication state and a downlink communication state;
the fourth setting subunit is used for setting monitoring points of the power supply module to comprise a power supply state and a battery state;
and the fifth setting subunit is used for setting the monitoring point of the acquisition module 1 to comprise an acquisition unit state and a sensor probe state.
In an implementation manner, the training module 4 is specifically configured to:
constructing a loss function by using the improved cross entropy, and training a model of the one-dimensional neural network by using the loss function;
wherein the improved cross entropy is calculated as:
Figure BDA0003389395530000121
in the formula (I), the compound is shown in the specification,
Figure BDA0003389395530000122
interval of the function of degree of matching for the fault model, yiMatching value for distribution terminal operation state class, alpha is balance factor, alpha belongs to (0,1), gamma is modulation factor, and
Figure BDA0003389395530000123
the weight factors are constructed to reduce the loss function value to a single sample that is easy to classify.
The invention also provides a power distribution terminal fault diagnosis device, which comprises:
a memory to store instructions; the instruction is an instruction which can implement the power distribution terminal fault diagnosis method according to any one of the above embodiments;
a processor to execute the instructions in the memory.
The present invention also provides a computer-readable storage medium, on which a computer program is stored, which, when executed by a processor, implements the power distribution terminal fault diagnosis method according to any one of the above embodiments.
According to the embodiment of the invention, on the basis of analyzing the field operation faults of various power distribution terminals, typical faults of the power distribution terminals are analyzed for each module, prior knowledge of the power distribution terminals is utilized to the maximum extent, and the success rate of fault diagnosis and identification can be greatly improved. The method, the system and the device can be applied to solving the problem of the fault of the terminal equipment of the complex power distribution network, and can quickly obtain a comprehensive fault diagnosis result, so that the load recovery rate is better improved.
It can be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working processes of the system, the apparatus and the module described above may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
In the several embodiments provided in the present application, it should be understood that the disclosed system, apparatus and method may be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the units is only one logical division, and other divisions may be realized in practice, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.
The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
The above-mentioned embodiments are only used for illustrating the technical solutions of the present invention, and not for limiting the same; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.

Claims (10)

1. A power distribution terminal fault diagnosis method is characterized by comprising the following steps:
acquiring the running state of a power distribution terminal and generating a state sequence;
establishing a measurement data model corresponding to each fault type according to the state sequence, wherein the measurement data model corresponding to each fault type comprises a measurement data model in a normal state, a measurement data model corresponding to a precision distortion type fault, a measurement data model corresponding to an attenuation oscillation type fault and a measurement data model corresponding to a harmonic interference type fault;
processing the multiple measurement data models through short-time Fourier transform to convert the time domain data types in the measurement data models into time-frequency graphs;
inputting the processed measurement data model into a one-dimensional neural network for training to obtain a trained fault diagnosis model;
and inputting a state sequence to be tested, diagnosing the fault type by using the fault diagnosis model and outputting a diagnosis result.
2. The power distribution terminal fault diagnosis method according to claim 1, wherein the collecting the operation state of the power distribution terminal and generating the state sequence comprises:
setting corresponding monitoring points according to the fault types of the functional modules of the power distribution terminal;
collecting the running state of each monitoring point, and judging the running state of each monitoring point by a self-diagnosis method to obtain a corresponding running state judgment vector;
and generating a state sequence according to the running state evaluation vector of each monitoring point.
3. The method for diagnosing the faults of the power distribution terminal according to claim 2, wherein each functional module of the power distribution terminal comprises a central processing unit, an operation control loop, a communication module, a power supply module and an acquisition module, and the setting of the corresponding monitoring point according to the fault type of each functional module of the power distribution terminal comprises the following steps:
setting monitoring points of the central processing unit to comprise a task running state, a system parameter setting state and a global positioning system time setting state;
setting monitoring points of the operation control loop to comprise a telemetering state;
setting monitoring points of the communication module to comprise an uplink communication state and a downlink communication state;
setting monitoring points of the power supply module to comprise a power supply state and a battery state;
the monitoring points of the acquisition module are set to comprise an acquisition unit state and a sensor probe state.
4. The method according to claim 1, wherein the inputting the processed measurement data model into a one-dimensional neural network for training comprises:
constructing a loss function by using the improved cross entropy, and training a model of the one-dimensional neural network by using the loss function;
wherein the improved cross entropy is calculated as:
Figure FDA0003389395520000021
in the formula (I), the compound is shown in the specification,
Figure FDA0003389395520000022
interval of the function of degree of matching for the fault model, yiMatching value for distribution terminal operation state class, alpha is balance factor, alpha belongs to (0,1), gamma is modulation factor, and
Figure FDA0003389395520000023
the weight factors are constructed to reduce the loss function value to a single sample that is easy to classify.
5. A power distribution terminal fault diagnostic system, comprising:
the acquisition module is used for acquiring the running state of the power distribution terminal and generating a state sequence;
the measurement data model building module is used for building a measurement data model corresponding to each fault type according to the state sequence, wherein the measurement data model corresponding to each fault type comprises a measurement data model in a normal state, a measurement data model corresponding to a precision distortion fault, a measurement data model corresponding to an attenuation oscillation fault and a measurement data model corresponding to a harmonic interference fault;
the time-frequency transformation module is used for processing the multiple measurement data models through short-time Fourier transformation so as to convert the time-domain data types in the measurement data models into time-frequency graphs;
the training module is used for inputting the processed measurement data model into a one-dimensional neural network for training to obtain a trained fault diagnosis model;
and the fault diagnosis module is used for inputting a state sequence to be tested, diagnosing the fault type by using the fault diagnosis model and outputting a diagnosis result.
6. The power distribution terminal fault diagnostic system of claim 5, wherein the acquisition module comprises:
the monitoring point setting unit is used for setting corresponding monitoring points according to the fault types of the functional modules of the power distribution terminal;
the running state judgment vector acquisition unit is used for acquiring the running state of each monitoring point and judging the running state of each monitoring point through the self-diagnosis system to obtain a corresponding running state judgment vector;
and the state sequence generating unit is used for generating a state sequence according to the running state evaluation vector of each monitoring point.
7. The system of claim 6, wherein each functional module of the power distribution terminal comprises a central processing unit, an operation control loop, a communication module, a power module and an acquisition module, and the monitoring point setting unit comprises:
the first setting subunit is used for setting monitoring points of the central processing unit to comprise a task running state, a system parameter setting state and a global positioning system time setting state;
the second setting subunit is used for setting monitoring points of the operation control loop to comprise a telemetering state;
the third setting subunit is used for setting monitoring points of the communication module to comprise an uplink communication state and a downlink communication state;
the fourth setting subunit is used for setting monitoring points of the power supply module to comprise a power supply state and a battery state;
and the fifth setting subunit is used for setting the monitoring point of the acquisition module to comprise an acquisition unit state and a sensor probe state.
8. The power distribution terminal fault diagnostic system of claim 5, wherein the training module is specifically configured to:
constructing a loss function by using the improved cross entropy, and training a model of the one-dimensional neural network by using the loss function;
wherein the improved cross entropy is calculated as:
Figure FDA0003389395520000031
in the formula (I), the compound is shown in the specification,
Figure FDA0003389395520000032
interval of the function of degree of matching for the fault model, yiMatching value for distribution terminal operation state class, alpha is balance factor, alpha belongs to (0,1), gamma is modulation factor, and
Figure FDA0003389395520000033
the weight factors are constructed to reduce the loss function value to a single sample that is easy to classify.
9. A power distribution terminal fault diagnostic apparatus, comprising:
a memory to store instructions; the instruction is an instruction which can realize the fault diagnosis method of the power distribution terminal according to any one of claims 1 to 4;
a processor to execute the instructions in the memory.
10. A computer-readable storage medium, wherein a computer program is stored on the computer-readable storage medium, and when executed by a processor, implements the power distribution terminal fault diagnosis method according to any one of claims 1 to 4.
CN202111463158.9A 2021-12-02 2021-12-02 Power distribution terminal fault diagnosis method, system, device and storage medium Pending CN114186184A (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
CN202111463158.9A CN114186184A (en) 2021-12-02 2021-12-02 Power distribution terminal fault diagnosis method, system, device and storage medium
PCT/CN2022/135595 WO2023098753A1 (en) 2021-12-02 2022-11-30 Power distribution terminal fault diagnosis method, system and apparatus, and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111463158.9A CN114186184A (en) 2021-12-02 2021-12-02 Power distribution terminal fault diagnosis method, system, device and storage medium

Publications (1)

Publication Number Publication Date
CN114186184A true CN114186184A (en) 2022-03-15

Family

ID=80603318

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111463158.9A Pending CN114186184A (en) 2021-12-02 2021-12-02 Power distribution terminal fault diagnosis method, system, device and storage medium

Country Status (2)

Country Link
CN (1) CN114186184A (en)
WO (1) WO2023098753A1 (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
TWI800365B (en) * 2022-04-29 2023-04-21 華碩電腦股份有限公司 Power supply health check system and method thereof
WO2023098753A1 (en) * 2021-12-02 2023-06-08 广东电网有限责任公司江门供电局 Power distribution terminal fault diagnosis method, system and apparatus, and storage medium

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116540006B (en) * 2023-07-06 2023-10-17 国网天津市电力公司滨海供电分公司 Fault detection device and method for power distribution automation terminal
CN117290803B (en) * 2023-11-27 2024-03-26 深圳鹏城新能科技有限公司 Energy storage inverter remote fault diagnosis method, system and medium

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2011056144A1 (en) * 2009-11-06 2011-05-12 Nanyang Polytechnic Fault signature-based case library system and method for power system trouble-shooting and recovery
CN107102633B (en) * 2017-04-26 2019-08-02 上海金智晟东电力科技有限公司 A kind of distribution terminal fault self-diagnosis method and system
KR20200014118A (en) * 2018-07-31 2020-02-10 주식회사 비엠티 Power Distribution Board with Fault Fiagnosis Function
CN110516742A (en) * 2019-08-28 2019-11-29 广东电网有限责任公司 A kind of distribution terminal fault distinguishing method and system based on Combination neural network model
CN113111591B (en) * 2021-04-29 2022-06-21 南方电网电力科技股份有限公司 Automatic diagnosis method, device and equipment based on internal fault of modular power distribution terminal
CN114186184A (en) * 2021-12-02 2022-03-15 广东电网有限责任公司江门供电局 Power distribution terminal fault diagnosis method, system, device and storage medium

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2023098753A1 (en) * 2021-12-02 2023-06-08 广东电网有限责任公司江门供电局 Power distribution terminal fault diagnosis method, system and apparatus, and storage medium
TWI800365B (en) * 2022-04-29 2023-04-21 華碩電腦股份有限公司 Power supply health check system and method thereof

Also Published As

Publication number Publication date
WO2023098753A1 (en) 2023-06-08

Similar Documents

Publication Publication Date Title
CN114186184A (en) Power distribution terminal fault diagnosis method, system, device and storage medium
CN111474510B (en) Error evaluation method and system for voltage transformer with non-stable output
EP3157111B1 (en) Power grid protection control system test device, test method, and program
CN111812451B (en) Phase current transient fault component-based distributed line selection method for power distribution network
CN109828168A (en) Converter method for diagnosing faults based on Density Estimator
CN111123009A (en) Intelligent substation avalanche testing method and system based on SCD file and terminal equipment
CN117233687B (en) CVT initial error assessment method, medium and terminal based on historical data
CN106646106B (en) Electric network fault detection method based on outlier's detection technology
CN112414520A (en) Fault detection system of flowmeter
CN111664933A (en) Method and system for filtering vibration and sound detection signal by utilizing static vector optimization
CN112240992B (en) Protection polarity verification method, device and equipment based on line empty charge transient current
US11836036B2 (en) Method of detecting faults in intelligent electronic devices
CN110261716B (en) Transformer operation state vibration sound detection method based on Dirac mixing mechanism
CN109038683B (en) Method, device and equipment for evaluating accuracy of primary frequency modulation frequency signal source
CN113283316A (en) Switch mechanical fault diagnosis method, device and equipment based on sound signals
CN117849691B (en) Multi-dimensional collaborative operation monitoring and early warning system and method for capacitive voltage transformer
CN111649819A (en) Transformer state vibration and sound detection signal filtering method and system using iteration soft threshold
CN110646691A (en) Transformer vibration sound signal filtering method and system by utilizing stretching transformation
CN110286291A (en) A kind of running state of transformer vibration sound detection method and system using principal component
CN113607992B (en) Detection wave generation method of direct current power distribution protection device and related device
CN113341251B (en) Power transmission and transformation field construction monitoring system
CN112254808B (en) Method and system for detecting vibration and sound of running state of transformer by utilizing gradient change
CN110702215B (en) Transformer running state vibration and sound detection method and system using regression tree
CN115421093B (en) Method and system for self-recovering measured data of all-fiber current transformer
Eisenbarth et al. A data-driven method for estimating parameter uncertainty in PMU-based Power Plant Model Validation

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