CN116699400A - Generator rotor short-circuit fault monitoring system, method and readable storage medium - Google Patents
Generator rotor short-circuit fault monitoring system, method and readable storage medium Download PDFInfo
- Publication number
- CN116699400A CN116699400A CN202310708701.XA CN202310708701A CN116699400A CN 116699400 A CN116699400 A CN 116699400A CN 202310708701 A CN202310708701 A CN 202310708701A CN 116699400 A CN116699400 A CN 116699400A
- Authority
- CN
- China
- Prior art keywords
- generator
- fault
- data
- short
- database
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 54
- 238000012544 monitoring process Methods 0.000 title claims abstract description 54
- 238000003745 diagnosis Methods 0.000 claims abstract description 73
- 239000013598 vector Substances 0.000 claims abstract description 46
- 238000000605 extraction Methods 0.000 claims abstract description 36
- 230000008569 process Effects 0.000 claims description 24
- 238000001228 spectrum Methods 0.000 claims description 20
- 238000012423 maintenance Methods 0.000 claims description 15
- 238000005457 optimization Methods 0.000 claims description 10
- 238000003062 neural network model Methods 0.000 claims description 8
- 238000012549 training Methods 0.000 claims description 8
- 238000003066 decision tree Methods 0.000 claims description 7
- 238000010801 machine learning Methods 0.000 claims description 7
- 238000004088 simulation Methods 0.000 claims description 7
- 230000001502 supplementing effect Effects 0.000 claims description 7
- 238000012360 testing method Methods 0.000 claims description 7
- 238000011084 recovery Methods 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 3
- 230000009466 transformation Effects 0.000 claims description 3
- 238000001514 detection method Methods 0.000 description 8
- 238000012706 support-vector machine Methods 0.000 description 7
- 230000002159 abnormal effect Effects 0.000 description 3
- 239000000284 extract Substances 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 230000000295 complement effect Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000009434 installation Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000019771 cognition Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000009413 insulation Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000002194 synthesizing effect Effects 0.000 description 1
- 238000001931 thermography Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/34—Testing dynamo-electric machines
- G01R31/346—Testing of armature or field windings
-
- 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
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Testing And Monitoring For Control Systems (AREA)
Abstract
The application belongs to the technical field of generator fault monitoring, and discloses a system and a method for monitoring short-circuit faults of a generator rotor and a readable storage medium, wherein the system comprises the following components: the data acquisition module is used for acquiring target data when the generator operates, wherein the target data comprises electric parameters, vibration signals, noise signals, temperature parameters, image information and electromagnetic parameters of the generator; the feature extraction module is used for extracting features of the target data to obtain corresponding feature vectors; the fault diagnosis module is used for comparing the feature vectors based on the fault diagnosis model, the normal operation data of the generator normal state database and the fault condition data of the generator fault mode database, and determining whether the generator is faulty or not and the fault type and the fault position of the generator. The application can realize accurate positioning and classified diagnosis of the short-circuit fault of the generator rotor.
Description
Technical Field
The application relates to the technical field of generator fault monitoring, in particular to a system and a method for monitoring short-circuit faults of a generator rotor and a readable storage medium.
Background
Existing generator rotor short circuit on-line monitoring techniques typically focus on a particular detection method, such as acoustic emission monitoring, vibration analysis, thermal imaging detection, and eddy current detection, which are advantageous but limited. On-line monitoring of the short circuit of the generator rotor is carried out based on a single detection principle, the state monitoring range of equipment is limited, the running condition of the equipment cannot be comprehensively reflected, the judgment accuracy is low, misjudgment or missed judgment is easy to occur, the type and the position of the fault cannot be accurately judged, and intelligent precise diagnosis is realized.
Disclosure of Invention
The application provides a system and a method for monitoring short-circuit faults of a generator rotor and a readable storage medium, which can realize accurate positioning and classified diagnosis of the short-circuit faults of the generator rotor.
In a first aspect, an embodiment of the present application provides a system for monitoring a short-circuit fault of a generator rotor, the system comprising:
the data acquisition module is used for acquiring target data when the generator operates, wherein the target data comprises electric parameters, vibration signals, noise signals, temperature parameters, image information and electromagnetic parameters of the generator;
the feature extraction module is used for extracting features of the target data to obtain corresponding feature vectors;
the fault diagnosis module is used for comparing the feature vectors based on the fault diagnosis model, the normal operation data of the generator normal state database and the fault condition data of the generator fault mode database, and determining whether the generator is faulty or not and the fault type and the fault position of the generator.
Further, the generator normal state database is constructed based on generator target data collected under a generator normal running state; the generator fault mode database is constructed based on the generator state data monitored under different short circuit types in the generator rotor short circuit simulation test.
Further, the feature extraction module further includes:
the time domain feature extraction sub-module is used for extracting the features of the time domain of the target data to obtain the time domain features corresponding to the target data, wherein the time domain features comprise the maximum value, the minimum value, the average value, the variance, the median value, the slope, the peak value and the valley value of the data corresponding to the electrical parameters, the vibration signals, the noise signals and the temperature parameters;
the frequency domain feature extraction submodule is used for calculating frequency spectrums of vibration signals, noise signals and current waveforms based on the fast Fourier transform, and carrying out frequency domain feature extraction on data corresponding to the vibration signals, the noise signals and the current waveforms based on the frequency spectrums to obtain frequency domain features corresponding to target data, wherein the frequency domain features comprise main frequencies, frequency zero crossings, cut-off frequencies, ratios of low-frequency signal power and high-frequency signal power of the frequency spectrums and power in frequency sub-bands;
the spatial domain feature extraction sub-module is used for extracting the spatial domain features of the data corresponding to the image information to obtain the spatial domain features corresponding to the target data, wherein the spatial domain features comprise gray distribution of the image, geometric features in the image and edge and contour information of the image.
Further, the fault diagnosis model is trained based on the normal operation data of the generator normal state database and the fault condition data of the generator fault mode database, and is used for judging the fault type of the generator based on the feature vector corresponding to the generator after the training is completed, wherein the fault diagnosis model is a machine learning model and comprises an SVM, a neural network model and a decision tree.
Further, the fault diagnosis module further includes:
the classifying sub-module is used for inputting the feature vector corresponding to the generator into the fault diagnosis model to classify faults and determining whether the generator has a rotor short circuit fault or not;
the matching sub-module is used for calculating the matching degree of the feature vector corresponding to the generator and fault condition data of the generator fault mode database when the rotor short-circuit fault of the generator is determined, determining the short-circuit fault data with the highest matching degree and determining the fault type of the generator;
and the position estimation sub-module is used for determining the fault position of the preset possibility corresponding to the fault type of the generator based on the short-circuit fault data with the highest matching degree and expert experience data.
Further, the system also comprises an optimizing module for:
acquiring the collected feedback data in a preset period, wherein the feedback data comprises various characteristic parameter data collected in the process of monitoring the running state of the generator, real fault data confirmed and recorded by technicians in the process of overhauling the short-circuit fault of the rotor of the generator and data for recovering the normal running of the generator after overhauling;
supplementing and correcting normal operation data of a normal state database of the generator based on various characteristic parameter data collected in the monitoring process of the running state of the generator and the data of the normal operation recovery of the generator after maintenance to obtain an optimized and updated normal state database of the generator;
verifying and correcting fault condition data of a generator fault mode database based on real fault data confirmed and recorded by technicians in the process of short-circuit fault maintenance of the generator rotor, and obtaining an optimized and updated generator fault mode database;
and adjusting parameters and structures corresponding to the fault diagnosis model based on the feedback data to obtain the fault diagnosis model after optimization and updating.
In a second aspect, the present application also provides a method for monitoring short-circuit faults of a generator rotor, the method comprising:
collecting target data when the generator operates, wherein the target data comprises electric parameters, vibration signals, noise signals, temperature parameters, image information and electromagnetic parameters of the generator;
extracting the characteristics of the target data to obtain corresponding characteristic vectors;
and comparing the feature vectors in a classified mode based on the fault diagnosis model, the normal operation data of the generator normal state database and the fault condition data of the generator fault mode database, and determining whether the generator is faulty or not and the fault type and the fault position of the generator.
Further, the generator normal state database is constructed based on generator target data collected under a generator normal running state; the generator fault mode database is constructed based on the generator state data monitored under different short circuit types in the generator rotor short circuit simulation test.
Further, the step of extracting features of the target data to obtain corresponding feature vectors includes:
performing time domain feature extraction on data corresponding to the electrical parameters, the vibration signals, the noise signals and the temperature parameters to obtain time domain features corresponding to the target data, wherein the time domain features comprise maximum values, minimum values, average values, variances, medians, slopes, peaks and valleys of the corresponding data;
calculating frequency spectrums of data corresponding to vibration signals, noise signals and current waveforms based on fast Fourier transformation, and carrying out frequency domain feature extraction on the data corresponding to the vibration signals, the noise signals and the current waveforms based on the frequency spectrums to obtain frequency domain features corresponding to target data, wherein the frequency domain features comprise main frequencies, frequency zero crossings, cut-off frequencies, ratios of low-frequency signal power and high-frequency signal power of the frequency spectrums and power in frequency sub-bands;
and extracting the spatial domain features of the data corresponding to the image information to obtain the spatial domain features corresponding to the target data, wherein the spatial domain features comprise gray distribution of the image, geometric features in the image and edge and contour information of the image.
Further, the fault diagnosis model is trained based on the normal operation data of the generator normal state database and the fault condition data of the generator fault mode database, and is used for judging the fault type of the generator based on the feature vector corresponding to the generator after the training is completed, wherein the fault diagnosis model is a machine learning model and comprises an SVM, a neural network model and a decision tree.
Further, the step of comparing the feature vectors based on the fault diagnosis model, the normal operation data of the generator normal state database and the fault condition data of the generator fault mode database to determine whether the generator has a fault and the fault type and the fault position of the generator includes:
inputting the feature vector corresponding to the generator into a fault diagnosis model to perform fault classification, and determining whether the generator has a rotor short-circuit fault or not;
when the short-circuit fault of the rotor exists in the generator, matching degree calculation is carried out on the feature vector corresponding to the generator and fault condition data of a generator fault mode database, short-circuit fault data with the highest matching degree are determined, and the fault type of the generator is determined;
and the position estimation sub-module is used for determining the fault position of the preset possibility corresponding to the fault type of the generator based on the short-circuit fault data with the highest matching degree and expert experience data.
Further, the method further comprises:
acquiring the collected feedback data in a preset period, wherein the feedback data comprises various characteristic parameter data collected in the process of monitoring the running state of the generator, real fault data confirmed and recorded by technicians in the process of overhauling the short-circuit fault of the rotor of the generator and data for recovering the normal running of the generator after overhauling;
supplementing and correcting normal operation data of a normal state database of the generator based on various characteristic parameter data collected in the monitoring process of the running state of the generator and the data of the normal operation recovery of the generator after maintenance to obtain an optimized and updated normal state database of the generator;
verifying and correcting fault condition data of a generator fault mode database based on real fault data confirmed and recorded by technicians in the process of short-circuit fault maintenance of the generator rotor, and obtaining an optimized and updated generator fault mode database;
and adjusting parameters and structures corresponding to the fault diagnosis model based on the feedback data to obtain the fault diagnosis model after optimization and updating.
In a third aspect, an embodiment of the present application further provides a computer readable storage medium, where a generator rotor short-circuit fault monitoring program is stored on the computer readable storage medium, where the generator rotor short-circuit fault monitoring program, when executed by a processor, implements the steps of the generator rotor short-circuit fault monitoring method as described above.
In summary, compared with the prior art, the technical scheme provided by the embodiment of the application has the following beneficial effects:
the system, the method and the readable storage medium for monitoring the short-circuit fault of the generator rotor provided by the embodiment of the application collect multi-source data in the operation of the generator by using a multi-element complementary detection method, have stronger system integrity, do not depend on a certain detection technology, and have higher application range and adaptability. In addition, by establishing a sound high-quality generator running state database, a generator fault mode database and an intelligent fault diagnosis model for fault diagnosis, the accurate positioning and classified diagnosis of the short-circuit fault of the generator rotor can be realized intelligently and accurately.
Drawings
FIG. 1 is a schematic diagram of a functional module of a generator rotor short-circuit fault monitoring system according to an embodiment of the present application;
fig. 2 is a flow chart of a method for monitoring a short-circuit fault of a generator rotor according to an embodiment of the application.
Detailed Description
The following description of the embodiments of the present application 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 application, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
Referring to fig. 1, an embodiment of the present application provides a system for monitoring a short-circuit fault of a generator rotor, the system specifically includes:
the data acquisition module 10 is used for acquiring target data when the generator is operated, wherein the target data comprises electric parameters, vibration signals, noise signals, temperature parameters, image information and electromagnetic parameters of the generator;
the feature extraction module 20 is configured to perform feature extraction on the target data to obtain a corresponding feature vector;
the fault diagnosis module 30 is configured to compare the feature vectors based on the fault diagnosis model, the normal operation data of the generator normal state database, and the fault condition data of the generator fault mode database, and determine whether the generator has a fault and the fault type and the fault location of the generator.
In this embodiment, the generator rotor short-circuit fault monitoring system includes a data acquisition module, and by arranging a vibration sensor, an electromagnetic sensor, a temperature sensor and the like in the running process of the generator, the real-time monitoring and acquisition of data such as mechanical, electromagnetic and thermal parameters can be realized. Specifically, the multisource data monitored in the running process of the generator in the scheme mainly comprises: electrical parameters, vibration signals, noise signals, temperature parameters, image information, electromagnetic parameters, and the like. The electric parameters such as the generating capacity of the unit, the rotating speed of the generator, exciting current, current waveform and the like can be used for detecting the electric operation state of the generator; the vibration signal is analyzed, so that whether the mechanical condition of the generator is abnormal or not can be monitored, and whether the rotor short circuit condition exists or not is judged; noise signals, by analyzing the operation noise of the generator, the mechanical operation condition of the generator can be monitored; the temperature parameter can be used for detecting whether the generator has overheat phenomenon or not by monitoring the temperature of the key part of the generator; the image information can judge the mechanical damage degree of the generator or the installation condition of elements through the image, and the possible influence of the rotor short circuit is detected; electromagnetic parameters, through monitoring voltage and current changes, insulation faults of the generator and the like can be judged. The multisource data is acquired by using a multielement complementary detection method, so that the system is stronger in integrity, does not depend on a certain detection technology, and has a higher application range and adaptability. Meanwhile, the running state of the generator can be accurately and comprehensively reflected, whether the generator has an abnormal condition or not is detected, and particularly, the judgment of the short-circuit fault of the rotor can be more accurate and reliable.
After multi-source target data are acquired, a feature extraction module of the generator rotor short-circuit fault monitoring system can perform feature extraction on the target data to obtain feature vectors corresponding to the monitored target data, and the running state and fault information of the generator are better represented through the feature vectors. The extracted feature vector is input to a fault diagnosis module of the generator rotor short-circuit fault monitoring system, and the intelligent fault diagnosis model, the normal operation data of the generator normal state database and the fault condition data of the generator fault mode database are called through the fault diagnosis module and are compared in a classified mode, so that whether the generator generates a short-circuit fault or not and the type and the position of the short-circuit fault generated by the generator can be rapidly and highly accurately determined. In order to establish a sound high-quality generator running state database, a generator fault mode database and an intelligent fault diagnosis model, the system is used for intelligently and accurately positioning and classifying short-circuit faults of a generator rotor, and an electric power enterprise is required to provide detailed information of a generator structure and a working principle to design a reasonable monitoring scheme and establish an accurate generator normal state database and an accurate generator fault mode database. Meanwhile, a large amount of actual operation data of the generator and historical fault case information are required to be accessed to train a fault diagnosis model, and close matching and data support of an electric power enterprise are required.
After the short-circuit fault monitoring system of the generator rotor in the scheme is established, the accurate judgment and positioning of the short-circuit fault of the generator rotor can be realized, a more reliable and intelligent equipment state monitoring means is provided for a power plant, the safety and economic benefits of a unit are improved, the system has important significance for safe and stable operation of an electric power system, and a large amount of valuable data can be generated during the operation of the system.
Further, in an embodiment, the generator normal state database is constructed based on generator target data collected in a normal running state of the generator; the generator fault mode database is constructed based on the generator state data monitored under different short circuit types in the generator rotor short circuit simulation test.
In this embodiment, during normal operation of the generator, all the set sensors (such as a vibration sensor, an electromagnetic sensor, a temperature sensor, etc.) are recorded, a multidimensional database of a normal operation state of a rotor of the generator is constructed, and meanwhile, simulation tests are performed on the shunt-to-lease short-circuit faults of the generator under various short-circuit types to obtain monitoring data corresponding to the short-circuit parameters, the positions and the fault types, so as to construct a multidimensional database of a short-circuit fault mode of the generator. The multidimensional database of the normal running state of the generator rotor and the multidimensional database of the short-circuit fault mode of the generator are used for storing normal running data and fault condition data of equipment and serve as a reference for fault comparison analysis.
Further, in an embodiment, the feature extraction module further includes:
the time domain feature extraction sub-module is used for extracting the features of the time domain of the target data to obtain the time domain features corresponding to the target data, wherein the time domain features comprise the maximum value, the minimum value, the average value, the variance, the median value, the slope, the peak value and the valley value of the data corresponding to the electrical parameters, the vibration signals, the noise signals and the temperature parameters;
the frequency domain feature extraction submodule is used for calculating frequency spectrums of vibration signals, noise signals and current waveforms based on the fast Fourier transform, and carrying out frequency domain feature extraction on data corresponding to the vibration signals, the noise signals and the current waveforms based on the frequency spectrums to obtain frequency domain features corresponding to target data, wherein the frequency domain features comprise main frequencies, frequency zero crossings, cut-off frequencies, ratios of low-frequency signal power and high-frequency signal power of the frequency spectrums and power in frequency sub-bands;
the spatial domain feature extraction sub-module is used for extracting the spatial domain features of the data corresponding to the image information to obtain the spatial domain features corresponding to the target data, wherein the spatial domain features comprise gray distribution of the image, geometric features in the image and edge and contour information of the image.
In this embodiment, the feature extraction module further includes a time domain feature extraction sub-module, a frequency domain feature extraction sub-module, and a spatial domain feature extraction sub-module. On the basis of comprehensively collecting mechanical, electrical and thermal signals, image information and other multi-source target data of the generator, when the target data is subjected to feature extraction, time domain, frequency domain and space domain features are extracted from the target data, so that more comprehensive and detailed feature information is obtained.
Specifically, when the time domain feature extraction submodule extracts the time domain feature, the maximum value, the minimum value, the average value, the variance, the median, the slope, the peak value (the maximum value in the signal) and the valley value (the minimum value in the signal) of the electrical parameter, the vibration signal, the noise signal and the temperature parameter are calculated and obtained from the data corresponding to the electrical parameter, the vibration signal, the noise signal and the temperature parameter, wherein the peak value and the valley value serve as fault features. When the frequency domain feature extraction submodule extracts the frequency domain features, the frequency spectrums of vibration signals, noise signals and current waveforms are calculated through Fast Fourier Transform (FFT), main frequencies, frequency zero crossings, cut-off frequencies and the like of the frequency spectrums are selected as features, the ratio of low-frequency signal power to high-frequency signal power is calculated as the features, and the power in a frequency sub-band is selected as the features. When the spatial domain feature extraction submodule extracts the spatial domain features, gray distribution of images, such as variance, entropy, contrast and the like, can be counted to judge the mechanical damage degree of the generator; detecting geometrical characteristics in the image, such as the number, the area and the like of surface cracks, so as to judge the mechanical damage degree of the generator; and selecting the edge and contour information of the image as characteristics to judge the installation condition of the element.
The extracted features are combined to form a multidimensional feature vector. The multidimensional feature vector contains information of a corresponding time domain, a frequency domain and a space domain, has strong representation capability on the current state of the generator, can comprehensively and completely describe the dynamic change process and mechanical structure characteristics of the generator, and provides effective feature variables for diagnosis of short-circuit faults of the generator rotor, so that the method can further realize rapid identification and accurate positioning of the short-circuit faults of the generator rotor based on the multidimensional feature vector, and improves the accuracy and efficiency of diagnosis of the generator faults.
Further, in an embodiment, the fault diagnosis model is trained based on the normal operation data of the generator normal state database and the fault condition data of the generator fault mode database, and is used for judging the fault type of the generator based on the feature vector corresponding to the generator after the training is completed, wherein the fault diagnosis model is a machine learning model, and comprises an SVM, a neural network model and a decision tree.
In this embodiment, the fault diagnosis model used is a machine learning model, including an SVM (support vector machine), a neural network model, a decision tree, and the like, and is obtained by training normal operation data of a generator normal state database and fault condition data of a generator fault mode database, and is used for judging a generator fault type based on a feature vector corresponding to the generator after training is completed. The mode utilizes richer data, and can construct a more comprehensive and accurate intelligent fault diagnosis model.
Further, in an embodiment, the fault diagnosis module further includes:
the classifying sub-module is used for inputting the feature vector corresponding to the generator into the fault diagnosis model to classify faults and determining whether the generator has a rotor short circuit fault or not;
the matching sub-module is used for calculating the matching degree of the feature vector corresponding to the generator and fault condition data of the generator fault mode database when the rotor short-circuit fault of the generator is determined, determining the short-circuit fault data with the highest matching degree and determining the fault type of the generator;
and the position estimation sub-module is used for determining the fault position of the preset possibility corresponding to the fault type of the generator based on the short-circuit fault data with the highest matching degree and expert experience data.
In this embodiment, the fault diagnosis module further includes a classification sub-module, a matching sub-module, and a position estimation sub-module. After extracting the feature vector corresponding to the generator in real time based on the target data, the classifying sub-module inputs the feature vector into the fault diagnosis model to perform fault classification, whether the category of the current generator state is a short-circuit fault type or not is obtained, if so, the existence of the short circuit of the generator rotor is confirmed, and if not, the current state of the generator is normal. When the rotor short-circuit fault of the generator is determined based on the classification submodule, in order to determine the fault type of the generator more accurately, the matching submodule calculates the matching degree of the feature vector and fault condition data of the generator fault mode database, and finds out the short-circuit fault data with the highest matching degree, namely one or more groups of most similar short-circuit fault data, so that the fault type of the generator is determined. And the position estimation sub-module is used for synthesizing short-circuit fault data with highest matching degree and expert experience data to reasoning and determining a preset possible fault position corresponding to the fault type of the generator so as to be referred by maintenance personnel. By classifying, matching and estimating the position of the faults, the intelligent and accurate positioning and classified diagnosis of the short-circuit faults of the generator rotor can be further realized.
Further, in an embodiment, the system further comprises an optimization module for:
acquiring the collected feedback data in a preset period, wherein the feedback data comprises various characteristic parameter data collected in the process of monitoring the running state of the generator, real fault data confirmed and recorded by technicians in the process of overhauling the short-circuit fault of the rotor of the generator and data for recovering the normal running of the generator after overhauling;
supplementing and correcting normal operation data of a normal state database of the generator based on various characteristic parameter data collected in the monitoring process of the running state of the generator and the data of the normal operation recovery of the generator after maintenance to obtain an optimized and updated normal state database of the generator;
verifying and correcting fault condition data of a generator fault mode database based on real fault data confirmed and recorded by technicians in the process of short-circuit fault maintenance of the generator rotor, and obtaining an optimized and updated generator fault mode database;
and adjusting parameters and structures corresponding to the fault diagnosis model based on the feedback data to obtain the fault diagnosis model after optimization and updating.
In this embodiment, after the normal state database, the fault mode database and the fault diagnosis model of the generator are built, real-time data, fault data and normal working data of the generator are collected as feedback data through the equipment state monitoring and maintenance process at regular intervals (for example, at regular intervals of 3 months), so as to optimize and update the normal state database, the fault mode database and the fault diagnosis model of the generator, realize online self-learning and optimization upgrading, and continuously improve the self-adaptive capacity and fault cognition level of the short-circuit fault monitoring system of the generator rotor, thereby achieving higher fault diagnosis performance (higher judgment efficiency and better judgment accuracy). The feedback data reflects the real-time state, fault condition and normal working condition of the equipment, and specifically comprises the following steps: 1) Various characteristic parameter data collected in the equipment running state monitoring process can reflect the real-time performance of equipment and possible abnormal conditions; 2) The technician confirms and records the real fault data in the overhaul process, including the type, the occurrence position, the possible reasons and the like of the fault, and the data can be used as important feedback to verify and optimize the fault mode database and the fault diagnosis model of the generator; 3) After maintenance, the equipment recovers the data of normal operation, which can be used as the data feedback of the normal state and is used for optimizing the database of the normal state of the generator and improving the accuracy of the identification of the normal operation state of the generator.
These feedback data can be used to: supplementing and optimizing a normal state database of the generator to improve the accuracy of identifying the normal state of the generator; verifying and correcting a generator fault mode database to optimize fault classification judgment; the intelligent fault diagnosis model is retrained, and model parameters (such as SVM) and model structures (such as neural network model) are adjusted to improve the fault diagnosis accuracy. And updating the optimized and updated generator normal state database, the generator fault mode database and the fault diagnosis model to the generator rotor short-circuit fault monitoring system, so that continuous intelligent online learning optimization and upgrading of the model and the database can be realized, and the fault diagnosis performance of the fault diagnosis module is improved.
The embodiment of the application also provides a method for monitoring the short-circuit fault of the generator rotor.
Referring to fig. 2, a flow chart of a first embodiment of a method for monitoring a short circuit fault of a generator rotor is shown.
In this embodiment, the method for monitoring a short-circuit fault of a generator rotor includes:
step S10, collecting target data when the generator operates, wherein the target data comprises electric parameters, vibration signals, noise signals, temperature parameters, image information and electromagnetic parameters of the generator;
step S20, extracting features of the target data to obtain corresponding feature vectors;
and step S30, classifying the feature vectors based on the fault diagnosis model, the normal operation data of the generator normal state database and the fault condition data of the generator fault mode database, and determining whether the generator has a fault, and the fault type and the fault position of the generator.
Further, in an embodiment, the generator normal state database is constructed based on generator target data collected in a normal running state of the generator; the generator fault mode database is constructed based on the generator state data monitored under different short circuit types in the generator rotor short circuit simulation test.
Further, in an embodiment, the step S20 includes:
performing time domain feature extraction on data corresponding to the electrical parameters, the vibration signals, the noise signals and the temperature parameters to obtain time domain features corresponding to the target data, wherein the time domain features comprise maximum values, minimum values, average values, variances, medians, slopes, peaks and valleys of the corresponding data;
calculating frequency spectrums of data corresponding to vibration signals, noise signals and current waveforms based on fast Fourier transformation, and carrying out frequency domain feature extraction on the data corresponding to the vibration signals, the noise signals and the current waveforms based on the frequency spectrums to obtain frequency domain features corresponding to target data, wherein the frequency domain features comprise main frequencies, frequency zero crossings, cut-off frequencies, ratios of low-frequency signal power and high-frequency signal power of the frequency spectrums and power in frequency sub-bands;
and extracting the spatial domain features of the data corresponding to the image information to obtain the spatial domain features corresponding to the target data, wherein the spatial domain features comprise gray distribution of the image, geometric features in the image and edge and contour information of the image.
Further, in an embodiment, the fault diagnosis model is trained based on the normal operation data of the generator normal state database and the fault condition data of the generator fault mode database, and is used for judging the fault type of the generator based on the feature vector corresponding to the generator after the training is completed, wherein the fault diagnosis model is a machine learning model, and comprises an SVM, a neural network model and a decision tree.
Further, in an embodiment, the step S30 includes:
inputting the feature vector corresponding to the generator into a fault diagnosis model to perform fault classification, and determining whether the generator has a rotor short-circuit fault or not;
when the short-circuit fault of the rotor exists in the generator, matching degree calculation is carried out on the feature vector corresponding to the generator and fault condition data of a generator fault mode database, short-circuit fault data with the highest matching degree are determined, and the fault type of the generator is determined;
and the position estimation sub-module is used for determining the fault position of the preset possibility corresponding to the fault type of the generator based on the short-circuit fault data with the highest matching degree and expert experience data.
Further, in an embodiment, the method further comprises:
acquiring the collected feedback data in a preset period, wherein the feedback data comprises various characteristic parameter data collected in the process of monitoring the running state of the generator, real fault data confirmed and recorded by technicians in the process of overhauling the short-circuit fault of the rotor of the generator and data for recovering the normal running of the generator after overhauling;
supplementing and correcting normal operation data of a normal state database of the generator based on various characteristic parameter data collected in the monitoring process of the running state of the generator and the data of the normal operation recovery of the generator after maintenance to obtain an optimized and updated normal state database of the generator;
verifying and correcting fault condition data of a generator fault mode database based on real fault data confirmed and recorded by technicians in the process of short-circuit fault maintenance of the generator rotor, and obtaining an optimized and updated generator fault mode database;
and adjusting parameters and structures corresponding to the fault diagnosis model based on the feedback data to obtain the fault diagnosis model after optimization and updating.
In the embodiment of the method for monitoring the short-circuit fault of the generator rotor, each step corresponds to the functional implementation of each module of the system for monitoring the short-circuit fault of the generator rotor, and the functions and the implementation process of the system are not repeated here.
Embodiments of the present application provide a computer readable storage medium, in which an operating system, a network communication module, a user interface module, and a generator rotor short-circuit fault monitoring program may be included in a memory. The processor may call a generator rotor short-circuit fault monitoring program stored in the memory, and execute the steps of the generator rotor short-circuit fault monitoring method provided by the embodiment of the application.
The method implemented when the generator rotor short-circuit fault monitoring program is executed may refer to various embodiments of the generator rotor short-circuit fault monitoring method of the present application, and will not be described herein.
The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.
The above examples illustrate only a few embodiments of the application, which are described in detail and are not to be construed as limiting the scope of the application. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the application, which are all within the scope of the application. Accordingly, the scope of protection of the present application is to be determined by the appended claims.
Claims (13)
1. A generator rotor short circuit fault monitoring system, the system comprising:
the data acquisition module is used for acquiring target data when the generator operates, wherein the target data comprises electric parameters, vibration signals, noise signals, temperature parameters, image information and electromagnetic parameters of the generator;
the feature extraction module is used for extracting features of the target data to obtain corresponding feature vectors;
the fault diagnosis module is used for comparing the feature vectors based on the fault diagnosis model, the normal operation data of the generator normal state database and the fault condition data of the generator fault mode database, and determining whether the generator is faulty or not and the fault type and the fault position of the generator.
2. The system according to claim 1, wherein: the generator normal state database is constructed based on generator target data collected under the normal running state of the generator; the generator fault mode database is constructed based on the generator state data monitored under different short circuit types in the generator rotor short circuit simulation test.
3. The system of claim 1, wherein the feature extraction module further comprises:
the time domain feature extraction sub-module is used for extracting the features of the time domain of the target data to obtain the time domain features corresponding to the target data, wherein the time domain features comprise the maximum value, the minimum value, the average value, the variance, the median value, the slope, the peak value and the valley value of the data corresponding to the electrical parameters, the vibration signals, the noise signals and the temperature parameters;
the frequency domain feature extraction submodule is used for calculating frequency spectrums of vibration signals, noise signals and current waveforms based on the fast Fourier transform, and carrying out frequency domain feature extraction on data corresponding to the vibration signals, the noise signals and the current waveforms based on the frequency spectrums to obtain frequency domain features corresponding to target data, wherein the frequency domain features comprise main frequencies, frequency zero crossings, cut-off frequencies, ratios of low-frequency signal power and high-frequency signal power of the frequency spectrums and power in frequency sub-bands;
the spatial domain feature extraction sub-module is used for extracting the spatial domain features of the data corresponding to the image information to obtain the spatial domain features corresponding to the target data, wherein the spatial domain features comprise gray distribution of the image, geometric features in the image and edge and contour information of the image.
4. The system according to claim 1, wherein: the fault diagnosis model is trained based on the normal operation data of the generator normal state database and the fault condition data of the generator fault mode database, and is used for judging the fault type of the generator based on the feature vector corresponding to the generator after the training is completed, wherein the fault diagnosis model is a machine learning model and comprises an SVM, a neural network model and a decision tree.
5. The system of claim 1, wherein the fault diagnosis module further comprises:
the classifying sub-module is used for inputting the feature vector corresponding to the generator into the fault diagnosis model to classify faults and determining whether the generator has a rotor short circuit fault or not;
the matching sub-module is used for calculating the matching degree of the feature vector corresponding to the generator and fault condition data of the generator fault mode database when the rotor short-circuit fault of the generator is determined, determining the short-circuit fault data with the highest matching degree and determining the fault type of the generator;
and the position estimation sub-module is used for determining the fault position of the preset possibility corresponding to the fault type of the generator based on the short-circuit fault data with the highest matching degree and expert experience data.
6. The system of claim 1, further comprising an optimization module for:
acquiring the collected feedback data in a preset period, wherein the feedback data comprises various characteristic parameter data collected in the process of monitoring the running state of the generator, real fault data confirmed and recorded by technicians in the process of overhauling the short-circuit fault of the rotor of the generator and data for recovering the normal running of the generator after overhauling;
supplementing and correcting normal operation data of a normal state database of the generator based on various characteristic parameter data collected in the monitoring process of the running state of the generator and the data of the normal operation recovery of the generator after maintenance to obtain an optimized and updated normal state database of the generator;
verifying and correcting fault condition data of a generator fault mode database based on real fault data confirmed and recorded by technicians in the process of short-circuit fault maintenance of the generator rotor, and obtaining an optimized and updated generator fault mode database;
and adjusting parameters and structures corresponding to the fault diagnosis model based on the feedback data to obtain the fault diagnosis model after optimization and updating.
7. A method for monitoring a short circuit fault of a generator rotor, the method comprising:
collecting target data when the generator operates, wherein the target data comprises electric parameters, vibration signals, noise signals, temperature parameters, image information and electromagnetic parameters of the generator;
extracting the characteristics of the target data to obtain corresponding characteristic vectors;
and comparing the feature vectors in a classified mode based on the fault diagnosis model, the normal operation data of the generator normal state database and the fault condition data of the generator fault mode database, and determining whether the generator is faulty or not and the fault type and the fault position of the generator.
8. The method according to claim 7, wherein: the generator normal state database is constructed based on generator target data collected under the normal running state of the generator; the generator fault mode database is constructed based on the generator state data monitored under different short circuit types in the generator rotor short circuit simulation test.
9. The method of claim 7, wherein the step of extracting features from the target data to obtain corresponding feature vectors comprises:
performing time domain feature extraction on data corresponding to the electrical parameters, the vibration signals, the noise signals and the temperature parameters to obtain time domain features corresponding to the target data, wherein the time domain features comprise maximum values, minimum values, average values, variances, medians, slopes, peaks and valleys of the corresponding data;
calculating frequency spectrums of data corresponding to vibration signals, noise signals and current waveforms based on fast Fourier transformation, and carrying out frequency domain feature extraction on the data corresponding to the vibration signals, the noise signals and the current waveforms based on the frequency spectrums to obtain frequency domain features corresponding to target data, wherein the frequency domain features comprise main frequencies, frequency zero crossings, cut-off frequencies, ratios of low-frequency signal power and high-frequency signal power of the frequency spectrums and power in frequency sub-bands;
and extracting the spatial domain features of the data corresponding to the image information to obtain the spatial domain features corresponding to the target data, wherein the spatial domain features comprise gray distribution of the image, geometric features in the image and edge and contour information of the image.
10. The method according to claim 7, wherein: the fault diagnosis model is trained based on the normal operation data of the generator normal state database and the fault condition data of the generator fault mode database, and is used for judging the fault type of the generator based on the feature vector corresponding to the generator after the training is completed, wherein the fault diagnosis model is a machine learning model and comprises an SVM, a neural network model and a decision tree.
11. The method of claim 7, wherein the step of comparing the feature vectors based on the fault diagnosis model, the normal operation data of the generator normal state database, and the fault condition data of the generator fault pattern database, and determining whether the generator has a fault and the type and location of the fault of the generator comprises:
inputting the feature vector corresponding to the generator into a fault diagnosis model to perform fault classification, and determining whether the generator has a rotor short-circuit fault or not;
when the short-circuit fault of the rotor exists in the generator, matching degree calculation is carried out on the feature vector corresponding to the generator and fault condition data of a generator fault mode database, short-circuit fault data with the highest matching degree are determined, and the fault type of the generator is determined;
and the position estimation sub-module is used for determining the fault position of the preset possibility corresponding to the fault type of the generator based on the short-circuit fault data with the highest matching degree and expert experience data.
12. The method of claim 7, wherein the method further comprises:
acquiring the collected feedback data in a preset period, wherein the feedback data comprises various characteristic parameter data collected in the process of monitoring the running state of the generator, real fault data confirmed and recorded by technicians in the process of overhauling the short-circuit fault of the rotor of the generator and data for recovering the normal running of the generator after overhauling;
supplementing and correcting normal operation data of a normal state database of the generator based on various characteristic parameter data collected in the monitoring process of the running state of the generator and the data of the normal operation recovery of the generator after maintenance to obtain an optimized and updated normal state database of the generator;
verifying and correcting fault condition data of a generator fault mode database based on real fault data confirmed and recorded by technicians in the process of short-circuit fault maintenance of the generator rotor, and obtaining an optimized and updated generator fault mode database;
and adjusting parameters and structures corresponding to the fault diagnosis model based on the feedback data to obtain the fault diagnosis model after optimization and updating.
13. A computer readable storage medium, characterized in that the computer readable storage medium has stored thereon a generator rotor short circuit fault monitoring program, wherein the generator rotor short circuit fault monitoring program, when executed by a processor, implements the steps of the generator rotor short circuit fault monitoring method according to any of claims 7 to 12.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310708701.XA CN116699400A (en) | 2023-06-14 | 2023-06-14 | Generator rotor short-circuit fault monitoring system, method and readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310708701.XA CN116699400A (en) | 2023-06-14 | 2023-06-14 | Generator rotor short-circuit fault monitoring system, method and readable storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116699400A true CN116699400A (en) | 2023-09-05 |
Family
ID=87838862
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310708701.XA Pending CN116699400A (en) | 2023-06-14 | 2023-06-14 | Generator rotor short-circuit fault monitoring system, method and readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116699400A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116991145A (en) * | 2023-09-25 | 2023-11-03 | 上海纳信实业有限公司 | Main control unit preparation test method and device applied to exciting current |
-
2023
- 2023-06-14 CN CN202310708701.XA patent/CN116699400A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116991145A (en) * | 2023-09-25 | 2023-11-03 | 上海纳信实业有限公司 | Main control unit preparation test method and device applied to exciting current |
| CN116991145B (en) * | 2023-09-25 | 2023-12-15 | 上海纳信实业有限公司 | Main control unit preparation test method and device applied to exciting current |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2021137760A1 (en) | Method and apparatus for inspecting wind turbine blade, and device and storage medium thereof | |
| CN107992665B (en) | Online fault diagnosis and analysis method for alternating current filter of extra-high voltage converter station | |
| CN116699400A (en) | Generator rotor short-circuit fault monitoring system, method and readable storage medium | |
| CN117076955A (en) | Fault detection method and system for high-voltage frequency converter | |
| CN116625683A (en) | Wind turbine generator system bearing fault identification method, system and device and electronic equipment | |
| CN115374842A (en) | Mechanical bearing fault classification method and system based on generation countermeasure network | |
| CN115128345A (en) | Power grid safety early warning method and system based on harmonic monitoring | |
| CN115079052A (en) | Transformer fault diagnosis method and system | |
| CN117782198B (en) | Highway electromechanical equipment operation monitoring method and system based on cloud edge architecture | |
| CN117668751B (en) | High-low voltage power system fault diagnosis method and device | |
| CN117458955A (en) | Operation control method and system of motor | |
| CN116884432A (en) | VMD-JS divergence-based power transformer fault voiceprint diagnosis method | |
| CN117435908A (en) | Multi-fault feature extraction method for rotary machine | |
| Bouchareb et al. | Advanced diagnosis strategy for incipient stator faults in synchronous reluctance motor | |
| US20240052810A1 (en) | Blade fault diagnosis method, apparatus and system, and storage medium | |
| CN117949871B (en) | Sound collection and abnormal state identification system and method for transformer | |
| CN117854245B (en) | Abnormal equipment monitoring method and system based on equipment operation audio | |
| CN114637793B (en) | Equipment fault frequent region positioning method based on big data analysis | |
| CN117672255B (en) | Abnormal equipment identification method and system based on artificial intelligence and equipment operation sound | |
| CN117454315B (en) | Man-machine terminal picture data interaction method and system | |
| CN117977817B (en) | Control method, device, equipment and storage medium based on AI current fingerprint | |
| CN117952600B (en) | New energy automobile motor evaluation method and system based on acoustic data | |
| CN116484263B (en) | Intelligent self-service machine fault detection system and method | |
| CN112001326B (en) | Vibration fingerprint-based high-voltage switch equipment fault diagnosis and prediction method and system | |
| CN118260648A (en) | Pumped storage power station equipment supervision system for fault wave recording identification |
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 |