CN118536409B - Method and system for predicting inter-gate short circuit of generator rotor winding based on correction prediction - Google Patents
Method and system for predicting inter-gate short circuit of generator rotor winding based on correction prediction Download PDFInfo
- Publication number
- CN118536409B CN118536409B CN202411010843.XA CN202411010843A CN118536409B CN 118536409 B CN118536409 B CN 118536409B CN 202411010843 A CN202411010843 A CN 202411010843A CN 118536409 B CN118536409 B CN 118536409B
- Authority
- CN
- China
- Prior art keywords
- prediction
- exciting current
- fault
- characteristic
- value
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 70
- 238000004804 winding Methods 0.000 title claims abstract description 25
- 238000012937 correction Methods 0.000 title claims abstract description 23
- 230000005284 excitation Effects 0.000 claims abstract description 44
- 238000004458 analytical method Methods 0.000 claims abstract description 26
- 230000008569 process Effects 0.000 claims abstract description 9
- 230000006870 function Effects 0.000 claims description 45
- 238000012549 training Methods 0.000 claims description 24
- 230000014509 gene expression Effects 0.000 claims description 14
- 238000004422 calculation algorithm Methods 0.000 claims description 13
- 238000005259 measurement Methods 0.000 claims description 11
- 230000001360 synchronised effect Effects 0.000 claims description 11
- 238000004590 computer program Methods 0.000 claims description 8
- 239000013598 vector Substances 0.000 claims description 8
- 238000012795 verification Methods 0.000 claims description 8
- 230000009466 transformation Effects 0.000 claims description 6
- 238000011156 evaluation Methods 0.000 claims description 5
- 238000009827 uniform distribution Methods 0.000 claims description 3
- 230000004913 activation Effects 0.000 claims description 2
- 238000007796 conventional method Methods 0.000 description 4
- 230000008901 benefit Effects 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 3
- 238000009826 distribution Methods 0.000 description 3
- 238000005070 sampling Methods 0.000 description 3
- 238000012360 testing method Methods 0.000 description 3
- 238000003491 array Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 238000002790 cross-validation Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000012417 linear regression Methods 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000004088 simulation Methods 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000001186 cumulative effect Effects 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005315 distribution function Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000010200 validation analysis 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
-
- 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
- G01R31/50—Testing of electric apparatus, lines, cables or components for short-circuits, continuity, leakage current or incorrect line connections
- G01R31/52—Testing for short-circuits, leakage current or ground faults
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/048—Activation functions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/09—Supervised learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/08—Probabilistic or stochastic CAD
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/02—Reliability analysis or reliability optimisation; Failure analysis, e.g. worst case scenario performance, failure mode and effects analysis [FMEA]
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- Control Of Eletrric Generators (AREA)
- Tests Of Circuit Breakers, Generators, And Electric Motors (AREA)
Abstract
The invention relates to the technical field of fault prediction, and discloses a method and a system for predicting a short circuit between generator rotor windings based on correction prediction, wherein the method comprises the following steps: acquiring electrical parameter information of equipment and acquiring excitation current data; predicting excitation current according to the electrical parameter information; judging a short circuit fault according to the predicted exciting current and the actually measured exciting current; performing reliability analysis on the fault judgment result by using a reliability judgment function, and compensating the predicted exciting current and the actually measured exciting current according to the analysis result; and repeating the reliability analysis and compensation processes until the fault judgment result is reliable, and outputting the judgment result of short circuit prediction. The model prediction error is effectively reduced, the inaccuracy of the data at a single moment is compensated, and the accuracy of the data is enhanced. The method has stronger robustness under different environments and fault types, and can effectively resist noise and interference.
Description
Technical Field
The invention relates to the technical field of fault prediction, in particular to a method and a system for predicting a short circuit between generator rotor windings based on correction prediction.
Background
With the continuous development of power systems, the operational reliability and stability of the transmission line become particularly important. Three-phase currents of the transmission line may be affected by various faults, such as a short-circuit fault, a ground fault, etc., during operation. These faults can cause abnormal changes in the current traveling wave, which in turn affect the safe operation of the power system.
Currently, a number of methods and techniques have been proposed for detecting and predicting a traveling fault wave in a transmission line. Conventional detection methods are mainly based on signal processing and pattern recognition techniques, such as wavelet transformation, fourier transformation, etc., which can detect the change of the fault traveling wave to some extent. However, due to the complexity of the power system and the variety of fault types, existing methods still have shortcomings in terms of accuracy and real-time.
The traditional method inputs the machine end voltage, active power and reactive power which are actually measured on line into an established prediction model, the output of the prediction model is an excitation current prediction value, and the output is compared with the excitation current of the generator which is actually measured on line, and if the error exceeds a set threshold value, the fault is diagnosed as the occurrence of turn-to-turn short circuit fault of the rotor winding of the synchronous generator.
However, the prediction model itself has prediction errors, and the generator excitation current itself actually measured on line is not necessarily acquired accurately enough, and errors exist, which may lead to inaccurate prediction results.
Disclosure of Invention
The present invention has been made in view of the above-described problems.
Therefore, the technical problems solved by the invention are as follows: the existing prediction judging method has the problems of low accuracy, large error and the like.
In order to solve the technical problems, the invention provides the following technical scheme: the method for predicting the inter-gate short circuit of the generator rotor winding based on correction prediction comprises the following steps:
acquiring electrical parameter information of equipment and acquiring excitation current data;
predicting excitation current according to the electrical parameter information;
Judging a short circuit fault according to the predicted exciting current and the actually measured exciting current;
Performing reliability analysis on the fault judgment result by using a reliability judgment function, and compensating the predicted exciting current and the actually measured exciting current according to the analysis result;
And repeating the reliability analysis and compensation processes until the fault judgment result is reliable, and outputting the judgment result of short circuit prediction.
As a preferable scheme of the method for predicting the inter-gate short circuit of the generator rotor winding based on correction prediction, the method comprises the following steps: the electrical parameter information comprises machine terminal voltage, active power and reactive power; and a time stamp is placed on each electrical parameter information.
As a preferable scheme of the method for predicting the inter-gate short circuit of the generator rotor winding based on correction prediction, the method comprises the following steps: the excitation current prediction comprises the steps of dividing a prediction model into a basic prediction part and a characteristic disturbance part, and training the basic prediction part through a training set to obtain a basic prediction model; applying the basic prediction model to a feature set, and training a feature disturbance model under the current feature through the difference between a prediction result and the real exciting current; verifying the prediction model by using a verification set, and adjusting a basic prediction part and a characteristic disturbance part according to a verification result;
The characteristic set comprises a plurality of sub-sets selected through the characteristic expression of the electrical parameter information, and the summarized sets of all the sub-sets; capturing the characteristic expression and the real exciting current of each time stamp in a history record, randomly selecting R exciting currents with the same characteristic expression, and taking the selected exciting currents with the same characteristic expression as a subset;
the characteristic representation comprises two dimensions which respectively take the characteristic of the exciting current value at the current moment and the waveform characteristic of the exciting current at the previous period as characteristic representations; the characteristic representation at time t is expressed as ；
The current exciting current numerical value characteristic is as follows:
，
Wherein, The exciting current value at the current moment is represented;
waveform characteristics of exciting current in the former period:
，
Wherein, The length of the time window is indicated,Time of presentationThe exciting current value at the position is equal to the exciting current value,Representing the frequency content of the signal,Representing the integral variable.
As a preferable scheme of the method for predicting the inter-gate short circuit of the generator rotor winding based on correction prediction, the method comprises the following steps: taking the machine end voltage, active power and reactive power of the synchronous generator under various working conditions as training input samples of an SVR algorithm when the synchronous generator normally operates, taking exciting current as training target output samples of the SVR algorithm, and establishing a basic prediction model of the exciting current of the synchronous generator through the SVR;
The SVR algorithm of the basic prediction model is expressed as:
，
Wherein, Representing the base predicted value, which is the input feature vectorOutputting a basic prediction model; n represents the nth sample index in the training data, for a total of N samples; Representing a feature transformation function; And Is a Lagrangian multiplier; For the kernel function part, RBF kernel function is used to represent the input features Support vectorSimilarity in feature space; parameters representing kernel functions, controlling the width of the RBF kernel functions; b represents a bias term;
，
Wherein, Representing a gaussian convolution integral for extracting a smoothed portion of the feature; Representing polynomial features for extracting nonlinear relationships;
the characteristic disturbance model is expressed as:
，
Wherein, The characteristic disturbance function is represented and is random disturbance adjustment under different characteristic states; And The method is characterized by controlling the contribution degree of each feature to the disturbance quantity; And Representing a disturbance proportion coefficient and controlling the amplitude of disturbance; And Is random disturbance variable and obeys uniform distributionIntroducing a random disturbance component;
，
Wherein, Representing the final predicted result.
As a preferable scheme of the method for predicting the inter-gate short circuit of the generator rotor winding based on correction prediction, the method comprises the following steps: the fault judgment comprises the steps of judging by using an excitation current predicted value and an excitation current online actual measurement value;
，
Wherein P represents the result of the fault evaluation, Indicating the on-line measured value of the exciting current,Indicating a predicted value of exciting current;
if P > 0.0025, then the fault is determined.
As a preferable scheme of the method for predicting the inter-gate short circuit of the generator rotor winding based on correction prediction, the method comprises the following steps: the reliability analysis comprises setting a predictive model for exciting current in the previous 100 timesPredicting an existing error W 1,W2,...,W100;
setting exciting current in the previous 100 times of online actual measurement The acquisition error is S 1,S2,...,S100;
establishing a reliability judging function:
，
Wherein rang (0.3, 0.5) represents a random number arbitrarily selected from 0.3 to 0.5; a minimum value of the error W 1,W2,...,W100 indicating the prediction exists; A maximum value representing the error W 1,W2,...,W100 that is predicted to exist; Indicating that the acquisition error is the maximum value of S 1,S2,...,S100; Indicating that the acquisition error is the minimum value of S 1,S2,...,S100;
when G is more than 0.7, judging that the fault judgment is reliable, and directly outputting a judgment result; otherwise, compensation correction is carried out.
As a preferable scheme of the method for predicting the inter-gate short circuit of the generator rotor winding based on correction prediction, the method comprises the following steps: the compensation includes, for the first 100 excitation currentsPredicting the existing errors W 1,W2,...,W100, finding out J errors greater than 0, and then 100-J errors smaller than 0;
when the error/reality is more than 17%, judging that the error is far away from the real value, wherein Q data are arranged in the data which are far away from the real value, and 100-Q data are arranged in the data which are close to the real value;
Let the compensation coefficient be B:
，
wherein m represents the number of cycles, [.] represents the rounding, Q represents the number of data with errors far from the true value, J represents the number of errors greater than 0, An i-th error is represented, i representing an index of the error;
The compensated excitation current predicted value is expressed as:
，
excitation current measured on line in the previous 100 times The acquisition error is S 1,S2,...,S100;
Let the compensation coefficient be B1:
，
the on-line actual measurement value of the compensated exciting current is expressed as:
，
Wherein, Rang (-m, m) represents random numbers arbitrarily chosen from-m to m;
Returning to the step of fault judgment after finishing correction, and setting the steps of initial cycle times m=1 cycle fault judgment and reliability judgment, wherein m is increased by 1 every cycle; the cycle is repeated until the result G of the reliability judging function is more than 0.7, the fault judgment is reliable, the cycle is stopped, and the judgment is output.
A modified prediction-based generator rotor winding inter-gate short circuit prediction system employing the method of the present invention is characterized by:
The acquisition unit acquires electrical parameter information of equipment and acquires excitation current data;
a prediction unit for predicting excitation current according to the electrical parameter information;
The compensation unit is used for judging a short circuit fault according to the predicted exciting current and the actually measured exciting current; performing reliability analysis on the fault judgment result by using a reliability judgment function, and compensating the predicted exciting current and the actually measured exciting current according to the analysis result; and repeating the reliability analysis and compensation processes until the fault judgment result is reliable, and outputting the judgment result of short circuit prediction.
A computer device, comprising: a memory and a processor; the memory stores a computer program, wherein: the processor, when executing the computer program, implements the steps of the method of any of the present invention.
A computer-readable storage medium having stored thereon a computer program, wherein: the computer program, when executed by a processor, implements the steps of the method of any of the present invention.
The invention has the beneficial effects that: the method for predicting the inter-gate short circuit of the generator rotor winding based on correction prediction effectively reduces model prediction errors, extracts key features from exciting current waveforms in the previous period, compensates inaccuracy of single-moment data, and enhances accuracy of the data. By designing a complex characteristic disturbance function, random disturbance is proportionally increased and decreased according to the difference of characteristic expression, so that the model has stronger robustness under different environments and fault types, and noise and interference can be effectively resisted.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a general flow chart of a method for predicting an inter-gate short circuit of a generator rotor winding based on a modified prediction according to a first embodiment of the present invention;
fig. 2 is a logic block diagram of a method for predicting an inter-gate short circuit of a rotor winding of a generator based on a modified prediction according to a first embodiment of the present invention.
Detailed Description
So that the manner in which the above recited objects, features and advantages of the present invention can be understood in detail, a more particular description of the invention, briefly summarized above, may be had by reference to the embodiments, some of which are illustrated in the appended drawings. All other embodiments, which can be made by one of ordinary skill in the art based on the embodiments of the present invention without making any inventive effort, shall fall within the scope of the present invention.
Example 1
Referring to fig. 1, for one embodiment of the present invention, a method for predicting an inter-gate short circuit of a generator rotor winding based on a modified prediction is provided, comprising:
S1: and acquiring electrical parameter information of the equipment and collecting exciting current data.
Further, the electrical parameter information comprises machine terminal voltage, active power and reactive power; and a time stamp is placed on each electrical parameter information.
The sampling frequency of the data acquisition system is set to ensure that real-time electrical parameter information can be acquired at an appropriate frequency (e.g., once per second or higher). And adding a time stamp for the electrical parameter information collected each time, and ensuring the time sequence and the accuracy of data recording. The timestamp may be obtained by synchronizing to a standard time server, such as an NTP server. And storing the acquired electrical parameter information (machine terminal voltage, active power and reactive power) and the corresponding time stamp thereof in a database or a data record file, so as to ensure the integrity and traceability of the data.
An excitation current sensor is arranged in an excitation system of the generator and is used for measuring excitation current in real time. Setting the sampling frequency of the exciting current data acquisition system, and ensuring the consistency with the sampling frequency of the electrical parameter information so as to facilitate the synchronous analysis of the follow-up data. The excitation current here includes not only the true value (which can be obtained by theoretical calculation) but also the acquired measured value, because the sensor may generate an error.
S2: and predicting the exciting current according to the electrical parameter information.
The excitation current prediction comprises the steps of dividing a prediction model into a basic prediction part and a characteristic disturbance part, and training the basic prediction part through a training set to obtain a basic prediction model; applying the basic prediction model to a feature set, and training a feature disturbance model under the current feature through the difference between a prediction result and the real exciting current; and verifying the prediction model by using a verification set, and adjusting a basic prediction part and a characteristic disturbance part according to a verification result.
Further, the characteristic set comprises a plurality of subsets selected through the characteristic expression of the electrical parameter information, and summarizing all the subsets; the characteristic representation and the real exciting current of each time stamp are captured in the history, R exciting currents with the same characteristic representation are randomly selected (if the exciting currents have no same characteristic representation, 1 exciting current is directly selected), and the selected exciting currents with the same characteristic representation are used as a subset.
The characteristic representation comprises two dimensions which respectively take the characteristic of the exciting current value at the current moment and the waveform characteristic of the exciting current at the previous period as characteristic representations; the characteristic representation at time t is expressed as。
The current exciting current numerical value characteristic is as follows:
，
Wherein, The excitation current value at the present time is indicated.
Waveform characteristics of exciting current in the former period:
，
Wherein, The length of the time window is indicated,Time of presentationThe exciting current value at the position is equal to the exciting current value,Representing the frequency content of the signal,Representing the integral variable.
It is known that the basic prediction part is trained by a training set, and a basic prediction model is constructed by using a Support Vector Regression (SVR) algorithm, so that the model prediction error is reduced. The characteristic disturbance model under the current characteristic is trained by dividing the characteristic expression of the electrical parameter information into a plurality of subsets and utilizing the difference of the characteristic expression and the real exciting current in the historical data, so that the characteristic changes under different fault modes and environments are effectively captured. And designing a characteristic disturbance model, and carrying out random disturbance proportional increase and decrease through the difference of characteristic expression to enhance the robustness of the model under different environments and fault types. Through division of feature sets and utilization of historical data, effective information in the historical data is fully mined, data utilization efficiency is improved, and performance of a prediction model is optimized.
The method comprises the steps of taking machine end voltage, active power and reactive power under various working conditions when a synchronous generator normally operates as training input samples of an SVR algorithm, taking exciting current as training target output samples of the SVR algorithm, and establishing a basic prediction model of the exciting current of the synchronous generator through the SVR.
The SVR algorithm of the basic prediction model is expressed as:
，
Wherein, Representing the base predicted value, which is the input feature vectorOutputting a basic prediction model; n represents the nth sample index in the training data, for a total of N samples; Representing a feature transformation function; And Is a Lagrangian multiplier; For the kernel function part, RBF kernel function is used to represent the input features Support vectorSimilarity in feature space; parameters representing kernel functions, controlling the width of the RBF kernel functions; b represents the bias term.
，
Wherein, Representing a gaussian convolution integral for extracting a smoothed portion of the feature; And expressing polynomial characteristics for extracting nonlinear relation.
The characteristic disturbance model is expressed as:
，
Wherein, The characteristic disturbance function is represented and is random disturbance adjustment under different characteristic states; And The method is characterized by controlling the contribution degree of each feature to the disturbance quantity; And Representing a disturbance proportion coefficient and controlling the amplitude of disturbance; And Is random disturbance variable and obeys uniform distributionA random disturbance component is introduced.
，
Wherein, Representing the final predicted result.
It is noted that VR algorithms can handle complex nonlinear relationships by introducing kernel functions to map input features into high-dimensional space. Through the characteristics (machine side voltage, active power and reactive power) and the target (exciting current) in the training data, the SVR can effectively capture the mode in the data, and the prediction accuracy is improved. Through SVR optimization of the training data set, the basic prediction model can provide excitation current prediction as accurate as possible under given input characteristics, and prediction deviation caused by errors of the model is reduced. The disturbance factors of the two dimensions are used for generating disturbance items, so that the disturbance items can be adapted to the fault environment, and accurate prediction can be made (the strength of the two dimensions can influence the fluctuation condition to a certain extent, and the accuracy of prediction can be improved by respectively considering the characteristics of the two dimensions).
S3: and judging the short circuit fault according to the predicted exciting current and the actually measured exciting current.
The fault judgment comprises the step of judging by using an excitation current predicted value and an excitation current online actual measurement value.
，
Wherein P represents the result of the fault evaluation,Indicating the on-line measured value of the exciting current,Indicating the excitation current predicted value. If P > 0.0025, then the fault is determined.
The fault judgment is realized by the difference between the predicted value and the online measured value of the exciting current. Specifically, a fault evaluation result P is set to quantify the error between the predicted value and the measured value. If P exceeds a predetermined threshold value of 0.0025, a fault is determined. Determining this 0.0025 threshold needs to be accomplished through extensive experimentation and data analysis. The experimental procedure was as follows:
a large amount of normal operation data and fault data is collected. These data include measured values of machine side voltage, active power, reactive power and exciting current, and corresponding predicted values.
The prediction error for each data point is calculated, namely:
，
Analyzing error distribution: a histogram of errors and a Cumulative Distribution Function (CDF) are plotted against normal operation data and fault data, respectively. By comparing the error distributions of the two types of data, their differences are observed.
A threshold value is selected from the error distribution map that is effective to distinguish between normal and fault conditions.
The error of the normal operation data is found to be mostly below 0.0025 through statistical analysis, and the error of the fault data is mostly above 0.0025. Thus 0.0025 is chosen as the threshold to distinguish between normal and fault conditions.
The effect of the selected threshold in the actual application is verified using cross-validation or a separate set of validations. And calculating confusion matrix, wherein the confusion matrix comprises indexes such as True Positive Rate (TPR), false Positive Rate (FPR), true Negative Rate (TNR), false Negative Rate (FNR) and the like. The rationality and the actual effect of the threshold are evaluated by these indices.
S4: performing reliability analysis on the fault judgment result by using a reliability judgment function, and compensating the predicted exciting current and the actually measured exciting current according to the analysis result; and repeating the reliability analysis and compensation processes until the fault judgment result is reliable, and outputting the judgment result of short circuit prediction.
The reliability analysis comprises setting a predictive model for exciting current in the previous 100 timesThe error W 1,W2,...,W100 that exists is predicted. Setting exciting current in the previous 100 times of online actual measurementThe acquisition error that exists is S 1,S2,...,S100.
Establishing a reliability judging function:
，
Wherein rang (0.3, 0.5) represents a random number arbitrarily selected from 0.3 to 0.5; a minimum value of the error W 1,W2,...,W100 indicating the prediction exists; A maximum value representing the error W 1,W2,...,W100 that is predicted to exist; Indicating that the acquisition error is the maximum value of S 1,S2,...,S100; indicating that the acquisition error is the minimum of S 1,S2,...,S100.
When G is more than 0.7, judging that the fault judgment is reliable, and directly outputting a judgment result; otherwise, compensation correction is carried out.
It should be noted that, when the reliability judging function G is calculated, a random number between 0.3 and 0.5 is introduced, so that the function has a certain randomness, and the excessively stiff judging standard is avoided.
By evaluating the reliability of the prediction error and the acquisition error, the prediction and actual measurement data are ensured to be reliable enough when fault judgment is carried out, so that the possibility of misjudgment is reduced. When a large error exists in the prediction model or data acquisition, the judgment result is improved through a compensation and correction mechanism, and the robustness and stability of fault judgment are enhanced.
And automatically evaluating the data quality by calculating the error range and fluctuation condition, and determining whether further processing is required according to the evaluation result, thereby improving the automation degree and the intelligent level of the system. The relative volatility of the data is assessed by comparing the range of prediction errors with the range of acquisition errors. If the fluctuation range of the prediction error is too large relative to the acquisition error, it is explained that the prediction model may not be reliable enough in the present case. And the prediction result is corrected by adjusting the output of the prediction model or collecting data again, so that the influence caused by errors is reduced.
Further, for the first 100 exciting currentsPredicting the existing errors W 1,W2,...,W100, finding out J errors greater than 0, and then 100-J errors smaller than 0. When the error/reality is more than 17%, the error is far from the real value, Q data are arranged in the data far from the real value, and 100-Q data are arranged in the data near to the real value.
Let the compensation coefficient be B:
，
wherein m represents the number of cycles, [.] represents the rounding, Q represents the number of data with errors far from the true value, J represents the number of errors greater than 0, The i-th error is represented, i representing the index of the error.
The compensated excitation current predicted value is expressed as:
，
excitation current measured on line in the previous 100 times The acquisition error that exists is S 1,S2,...,S100.
Let the compensation coefficient be B1:
，
the on-line actual measurement value of the compensated exciting current is expressed as:
，
Wherein, Representing the activation function, rang (-m, m) represents a random number chosen arbitrarily from-m to m. Returning to the step of fault judgment after finishing correction, and setting the steps of initial cycle times m=1 cycle fault judgment and reliability judgment, wherein m is increased by 1 every cycle; the cycle is repeated until the result G of the reliability judging function is more than 0.7, the fault judgment is reliable, the cycle is stopped, and the judgment is output.
In another aspect, the present embodiment also provides a system for predicting an inter-gate short circuit of a generator rotor winding based on a modified prediction, comprising:
And the acquisition unit acquires the electrical parameter information of the equipment and acquires excitation current data.
And the prediction unit predicts the exciting current according to the electrical parameter information.
The compensation unit is used for judging a short circuit fault according to the predicted exciting current and the actually measured exciting current; performing reliability analysis on the fault judgment result by using a reliability judgment function, and compensating the predicted exciting current and the actually measured exciting current according to the analysis result; and repeating the reliability analysis and compensation processes until the fault judgment result is reliable, and outputting the judgment result of short circuit prediction.
The above functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a usb disk, a removable hard disk, a Read-only memory (ROM), a random access memory (RAM, randomAccessMemory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
Logic and/or steps represented in the flowcharts or otherwise described herein, e.g., a ordered listing of executable instructions for implementing logical functions, can be embodied in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, processor-containing system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions. For the purposes of this description, a "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device.
More specific examples (a non-exhaustive list) of the computer-readable medium would include the following: an electrical connection (electronic device) having one or more wires, a portable computer diskette (magnetic device), a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). In addition, the computer readable medium may even be paper or other suitable medium on which the program is printed, as the program may be electronically captured, via, for instance, optical scanning of the paper or other medium, then compiled, interpreted or otherwise processed in a suitable manner, if necessary, and then stored in a computer memory.
It is to be understood that portions of the present invention may be implemented in hardware, software, firmware, or a combination thereof. In the above-described embodiments, the various steps or methods may be implemented in software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, may be implemented using any one or combination of the following techniques, as is well known in the art: discrete logic circuits having logic gates for implementing logic functions on data signals, application specific integrated circuits having suitable combinational logic gates, programmable Gate Arrays (PGAs), field Programmable Gate Arrays (FPGAs), and the like.
Example 2
In the following, for one embodiment of the invention, a method for predicting the short circuit between generator rotor windings based on correction prediction is provided, and in order to verify the beneficial effects of the invention, scientific demonstration is performed through economic benefit calculation and simulation experiments.
In order to verify the effectiveness of the fault judging and predicting method of the present invention, experimental tests were conducted.
First, the test is performed on a synchronous generator of a certain power plant. The machine end voltage, active power, reactive power and exciting current data of the generator in normal operation and fault state are collected. The data acquisition frequency is set to once per second and the duration is one hour, ensuring a sufficient amount of data for model training and verification.
Secondly, the acquired data is divided into a training set and a verification set. In the training set, a Support Vector Regression (SVR) algorithm is used, the machine side voltage, the active power and the reactive power are used as input features, the exciting current is used as target output, and a basic prediction model is trained. The feature transformation function comprises a Gaussian function convolution integral and a smoothing part and polynomial features, and is used for enhancing the capturing capability of the model on the nonlinear relation. The best super parameters (including kernel function parameters and regularization parameters) are selected by cross-validation using Radial Basis Function (RBF) kernel functions.
The invention and the traditional method are utilized to respectively carry out error test on the same data sample, and the traditional method generally uses a linear regression model or a simple neural network model to predict exciting current. The data statistics recorded for the two methods are shown in tables 1 and 2, respectively.
Table 1 error data table of the present method
| Experimental objects | Terminal voltage (V) | Active power (kW) | Reactive power (kVar) | Excitation current actual measurement value (A) | Exciting current predictive value (A) | Prediction error (%) | Acquisition error (%) |
| Data point 1 | 235 | 110 | 55 | 6.5 | 6.4 | 1.54 | 1.00 |
| Data point 2 | 230 | 108 | 53 | 6.3 | 6.2 | 1.59 | 0.95 |
| Data point 3 | 240 | 115 | 60 | 6.8 | 6.6 | 2.94 | 1.23 |
| Data point 4 | 238 | 112 | 58 | 6.7 | 6.5 | 2.99 | 1.10 |
| Data point 5 | 232 | 109 | 54 | 6.4 | 6.3 | 1.56 | 1.02 |
| Data point 6 | 236 | 111 | 56 | 6.6 | 6.4 | 3.03 | 1.20 |
| Data point 7 | 234 | 113 | 57 | 6.5 | 6.3 | 3.08 | 1.15 |
Table 2 error data table of conventional method
| Data points | Terminal voltage (V) | Active power (kW) | Reactive power (kVar) | Excitation current actual measurement value (A) | Exciting current predictive value (A) | Prediction error (%) |
| Data point 1 | 235 | 110 | 55 | 6.5 | 6.1 | 6.15 |
| Data point 2 | 230 | 108 | 53 | 6.3 | 5.9 | 6.35 |
| Data point 3 | 240 | 115 | 60 | 6.8 | 6.2 | 8.82 |
| Data point 4 | 238 | 112 | 58 | 6.7 | 6.3 | 5.97 |
| Data point 5 | 232 | 109 | 54 | 6.4 | 6.0 | 6.25 |
| Data point 6 | 236 | 111 | 56 | 6.6 | 6.1 | 7.58 |
| Data point 7 | 234 | 113 | 57 | 6.5 | 6.0 | 7.69 |
By comparing the experimental data of the invention with that of the traditional method, the obvious improvement of the invention in the aspects of prediction precision and robustness can be clearly seen. The average prediction error of the method of the invention is about 2.12%, which is significantly lower than 6.83% of the conventional method. The prediction error of the method of the present invention ranges from 1.54% to 3.08%, whereas the prediction error of the conventional method ranges from 5.97% to 8.82%.
Through the analysis and comparison, the innovation and the advantages of the invention in the aspects of exciting current prediction and fault judgment can be clearly seen. The invention utilizes the advanced SVR algorithm and the characteristic disturbance model, and remarkably improves the prediction precision and the robustness of the model. Meanwhile, the reliability of the fault judgment result is ensured through the reliability judgment function, and a powerful guarantee is provided for the safe operation of the power system. When facing a complex nonlinear relation and a noise environment, the traditional method has low prediction precision and unstable result, and is difficult to meet the requirements of practical application. The invention solves the defects in the traditional method by improving and innovating the prior art, provides more accurate and stable fault prediction and judgment results, and reflects the remarkable innovation and practical value.
Through a large number of simulation experiments, the fault is judged by adopting the method and the general prediction method respectively. The accuracy of the recorded fault determination is shown in table 3.
Table 3 accuracy vs. table
| Rotor state | Traditional judgment accuracy | The judging accuracy of the invention |
| Normal state | 90% | 100% |
| Short circuit | 85% | 95% |
It can be seen that the conventional method causes low accuracy in judging the rotor state under normal conditions and short circuit conditions by a simple judgment strategy.
The invention can not generate misjudgment under the condition of normal rotor state, and has higher stability. Under the short circuit fault, the accuracy of the judging result is up to 95%, which is obviously superior to the traditional method.
The traditional method is difficult to process complex nonlinear characteristics and noise environments due to the dependence on a simple linear regression or a basic machine learning model, so that the judgment accuracy is low, and misjudgment and missed judgment are easy to generate. The invention provides more accurate and stable fault prediction and judgment results, and represents remarkable practical value.
It should be noted that the above embodiments are only for illustrating the technical solution of the present invention and not for limiting the same, and although the present invention has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical solution of the present invention may be modified or substituted without departing from the spirit and scope of the technical solution of the present invention, which is intended to be covered in the scope of the claims of the present invention.
Claims (8)
1. The method for predicting the inter-gate short circuit of the generator rotor winding based on correction prediction is characterized by comprising the following steps:
acquiring electrical parameter information of equipment and acquiring excitation current data;
predicting excitation current according to the electrical parameter information;
Judging a short circuit fault according to the predicted exciting current and the actually measured exciting current;
Performing reliability analysis on the fault judgment result by using a reliability judgment function, and compensating the predicted exciting current and the actually measured exciting current according to the analysis result;
repeating the reliability analysis and compensation processes until the fault judgment result is reliable, and outputting a judgment result of short circuit prediction;
the electrical parameter information comprises machine terminal voltage, active power and reactive power; and put a time stamp on each electrical parameter information;
The excitation current prediction comprises the steps of dividing a prediction model into a basic prediction part and a characteristic disturbance part, and training the basic prediction part through a training set to obtain a basic prediction model; applying the basic prediction model to a feature set, and training a feature disturbance model under the current feature through the difference between a prediction result and the real exciting current; verifying the prediction model by using a verification set, and adjusting a basic prediction part and a characteristic disturbance part according to a verification result;
The characteristic set comprises a plurality of sub-sets selected through the characteristic expression of the electrical parameter information, and the summarized sets of all the sub-sets; capturing the characteristic expression and the real exciting current of each time stamp in a history record, randomly selecting R exciting currents with the same characteristic expression, and taking the selected exciting currents with the same characteristic expression as a subset;
The characteristic representation comprises two dimensions which respectively take the characteristic of the exciting current value at the current moment and the waveform characteristic of the exciting current at the previous period as characteristic representations; the characteristic representation at time t is denoted (L t,Ft);
The current exciting current numerical value characteristic is as follows:
Lt＝I(t)
Wherein, I (t) represents the exciting current value at the current moment;
waveform characteristics of exciting current in the former period:
Where T represents the time window length, I (τ) represents the excitation current value at time τ, ω represents the frequency component of the signal, and τ represents the integral variable.
2. The correction prediction-based method for predicting inter-gate short circuit of generator rotor winding of claim 1, wherein: taking the machine end voltage, active power and reactive power of the synchronous generator under various working conditions as training input samples of an SVR algorithm when the synchronous generator normally operates, taking exciting current as training target output samples of the SVR algorithm, and establishing a basic prediction model of the exciting current of the synchronous generator through the SVR;
The SVR algorithm of the basic prediction model is expressed as:
Wherein f base (x) represents a basic prediction value, which is the output of a basic prediction model under the input characteristic vector x; n represents the nth sample index in the training data, for a total of N samples; phi (x) represents a feature transformation function; alpha n Is a Lagrangian multiplier; exp (-gamma||phi (x n)-φ(x)||2) is a kernel function part, an RBF kernel function is used to represent similarity of input features x and support vectors x n in a feature space, gamma represents parameters of the kernel function, controls the width of the RBF kernel function, and b represents a bias term;
Wherein, Representing a gaussian convolution integral for extracting a smoothed portion of the feature; Representing polynomial features for extracting nonlinear relationships;
the characteristic disturbance model is expressed as:
D(x)＝β1F1(1+δ1sin(∈1π))+β2F2(1+δ2sin(∈2π))
Wherein, (x) represents a characteristic disturbance function, which is random disturbance adjustment under different characteristic expressions; beta 1 and beta 2 are the weights of the features, and the contribution degree of each feature to the disturbance quantity is controlled; δ 1 and δ 2 represent disturbance scaling factors, controlling the amplitude of the disturbance; e 1 and E 2 are random disturbance variables and obey uniform distribution Introducing a random disturbance component;
Wherein, Representing the final predicted result.
3. The correction prediction-based method for predicting inter-gate short circuit of generator rotor winding as claimed in claim 2, wherein: the fault judgment comprises the steps of judging by using an excitation current predicted value and an excitation current online actual measurement value;
Wherein P represents a fault evaluation result, i f2 represents an excitation current online measured value, and i f1 represents an excitation current predicted value;
if P > 0.0025, then the fault is determined.
4. A method for predicting an inter-gate short circuit of a generator rotor winding based on modified predictions as set forth in claim 3, wherein: the reliability analysis comprises the steps of setting an error W 1,W2,...,W100 existing in the prediction of the excitation current i f1 by a prediction model in the previous 100 processes;
Setting the acquisition error of the excitation current i f2 which is actually measured on line in the previous 100 times to be S 1,S2,...,S100;
establishing a reliability judging function:
Wherein rang (0.3, 0.5) represents a random number arbitrarily selected from 0.3 to 0.5; min (W i) represents the minimum value of the error W 1,W2,...,W100 that is predicted to exist; max (W i) represents the maximum value of the error W 1,W2,...,W100 that is predicted to exist; max (S i) represents the maximum value of S 1,S2,...,S100 of the acquisition error; min (S i) represents that the acquisition error is the minimum value of S 1,S2,...,S100;
when G is more than 0.7, judging that the fault judgment is reliable, and directly outputting a judgment result; otherwise, compensation correction is carried out.
5. The method for predicting an inter-gate short circuit of a generator rotor winding based on modified prediction of claim 4, wherein: the compensation comprises the steps of predicting the existing errors W 1,W2,...,W100 of the exciting current i f1 for the first 100 times, finding out J errors greater than 0, and finding out 100-J errors smaller than 0;
when the error/reality is more than 17%, judging that the error is far away from the real value, wherein Q data are arranged in the data which are far away from the real value, and 100-Q data are arranged in the data which are close to the real value;
Let the compensation coefficient be B:
Wherein m represents the number of cycles, [.] represents the rounding, Q represents the number of data with errors far from the true value, J represents the number of errors greater than 0, W i represents the ith error, i represents the index of the error;
The compensated excitation current predicted value is expressed as:
if1＝B*if1
The acquisition error of the excitation current i f2 which is actually measured on line in the previous 100 times is S 1,S2,...,S100;
Let the compensation coefficient be B1:
B1＝(max(s1,s2,...,s100)-min((s1+s2),(s2+s3),...,(s99+s100)))rand(-m,m)*sigmoid(t)
the on-line actual measurement value of the compensated exciting current is expressed as:
if2＝B1*if2
wherein sigmoid (t) represents an activation function, rang (-m, m) represents a random number arbitrarily selected from-m to m;
Returning to the step of fault judgment after finishing correction, and setting the steps of initial cycle times m=1 cycle fault judgment and reliability judgment, wherein m is increased by 1 every cycle; the cycle is repeated until the result G of the reliability judging function is more than 0.7, the fault judgment is reliable, the cycle is stopped, and the judgment is output.
6. A corrective-prediction-based motor rotor winding inter-gate short prediction system employing the method of any one of claims 1-5, characterized by:
The acquisition unit acquires electrical parameter information of equipment and acquires excitation current data;
a prediction unit for predicting excitation current according to the electrical parameter information;
The compensation unit is used for judging a short circuit fault according to the predicted exciting current and the actually measured exciting current; performing reliability analysis on the fault judgment result by using a reliability judgment function, and compensating the predicted exciting current and the actually measured exciting current according to the analysis result; and repeating the reliability analysis and compensation processes until the fault judgment result is reliable, and outputting the judgment result of short circuit prediction.
7. A computer device, comprising: a memory and a processor; the memory stores a computer program characterized in that: the processor, when executing the computer program, implements the steps of the method according to any one of claims 1-5.
8. A computer-readable storage medium having stored thereon a computer program, characterized by: the computer program implementing the steps of the method according to any of claims 1-5 when executed by a processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202411010843.XA CN118536409B (en) | 2024-07-26 | 2024-07-26 | Method and system for predicting inter-gate short circuit of generator rotor winding based on correction prediction |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202411010843.XA CN118536409B (en) | 2024-07-26 | 2024-07-26 | Method and system for predicting inter-gate short circuit of generator rotor winding based on correction prediction |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118536409A CN118536409A (en) | 2024-08-23 |
| CN118536409B true CN118536409B (en) | 2024-10-18 |
Family
ID=92383042
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202411010843.XA Active CN118536409B (en) | 2024-07-26 | 2024-07-26 | Method and system for predicting inter-gate short circuit of generator rotor winding based on correction prediction |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118536409B (en) |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107247230A (en) * | 2017-07-03 | 2017-10-13 | 浙江大学 | A kind of electric rotating machine state monitoring method based on SVMs and data-driven |
Family Cites Families (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105977948B (en) * | 2016-05-16 | 2018-07-10 | 山东大学 | Multivariable effect dive towards single-pole reclosing supplies arc characteristic Forecasting Methodology |
| CN106777606A (en) * | 2016-12-02 | 2017-05-31 | 上海电机学院 | A kind of gearbox of wind turbine failure predication diagnosis algorithm |
| CN108681633B (en) * | 2018-05-11 | 2022-03-29 | 上海电力学院 | Condensate pump fault early warning method based on state parameters |
| CN109061462A (en) * | 2018-09-14 | 2018-12-21 | 广西电网有限责任公司电力科学研究院 | A kind of High Voltage Circuit Breaker Contacts ablation assessment of failure method |
| CN110531266B (en) * | 2019-10-14 | 2021-08-24 | 国网山东省电力公司电力科学研究院 | Synchronous motor excitation winding turn-to-turn short circuit fault early warning method |
| US11971790B2 (en) * | 2020-08-25 | 2024-04-30 | Nvidia Corporation | Online fault detection in ReRAM-based AI/ML |
| CN113850320A (en) * | 2021-09-24 | 2021-12-28 | 沈阳理工大学 | Transformer fault detection method based on improved support vector machine regression algorithm |
| US20230108916A1 (en) * | 2021-10-05 | 2023-04-06 | Indian Institute Of Technology Bombay | Method and system for forecasting non-stationary time-series |
| CN114636900B (en) * | 2022-05-11 | 2022-09-30 | 广东电网有限责任公司东莞供电局 | Power distribution network multiple fault diagnosis method |
| CN116307014A (en) * | 2022-09-07 | 2023-06-23 | 南方电网数字电网研究院有限公司 | Fault prediction model training method and device and fault prediction method |
| CN115951218A (en) * | 2023-01-13 | 2023-04-11 | 山西京玉发电有限责任公司 | Double-check generator rotor turn-to-turn short circuit diagnosis method and diagnosis device |
| CN118194699A (en) * | 2024-02-29 | 2024-06-14 | 天津国能盘山发电有限责任公司 | On-line evaluation method and system for excitation system of thermal power generating unit |
-
2024
- 2024-07-26 CN CN202411010843.XA patent/CN118536409B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107247230A (en) * | 2017-07-03 | 2017-10-13 | 浙江大学 | A kind of electric rotating machine state monitoring method based on SVMs and data-driven |
Non-Patent Citations (1)
| Title |
|---|
| 基于PSO-SVR的同步发电机励磁电流预测;李红连;唐炬;;西北农林科技大学学报(自然科学版);20130531(06);第188-193页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118536409A (en) | 2024-08-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Zio | A study of the bootstrap method for estimating the accuracy of artificial neural networks in predicting nuclear transient processes | |
| CN111459700A (en) | Method and apparatus for diagnosing device failure, diagnostic device, and storage medium | |
| CN112327191B (en) | Battery diving probability evaluation method based on geometric feature fusion decision | |
| CN112101554A (en) | Anomaly detection method and device, equipment and computer-readable storage medium | |
| CN112926636A (en) | Method and device for detecting abnormal temperature of traction converter cabinet body | |
| CN114912640A (en) | Method and system for detecting abnormal mode of generator set based on deep learning | |
| CN115408864B (en) | Electronic transformer error state self-adaptive prediction method, system and equipment | |
| US7949497B2 (en) | Machine condition monitoring using discontinuity detection | |
| CN117828413A (en) | Transformer oil temperature prediction method and system based on LSTM neural network | |
| CN116736214A (en) | CVT stability evaluation method and system | |
| CN116306806A (en) | Fault diagnosis model determining method and device and nonvolatile storage medium | |
| CN115793590A (en) | Data processing method and platform suitable for system safety operation and maintenance | |
| CN118536409B (en) | Method and system for predicting inter-gate short circuit of generator rotor winding based on correction prediction | |
| CN115166625A (en) | Intelligent ammeter error estimation method and device | |
| CN114720901A (en) | Method for estimating health state of lithium ion battery | |
| US20210080924A1 (en) | Diagnosis Method and Diagnosis System for a Processing Engineering Plant and Training Method | |
| JP5949032B2 (en) | Pre-processing method and abnormality diagnosis device | |
| CN115423370A (en) | Relay protection equipment health state assessment method and device | |
| CN116777109A (en) | Wind generating set equipment state evaluation method and system | |
| CN115659271A (en) | Sensor abnormality detection method, model training method, system, device, and medium | |
| CN115982591A (en) | Method and device for detecting health state of transformer, equipment and storage medium | |
| CN116627759B (en) | Financial payment equipment circuit safety detection device | |
| CN117851953B (en) | Water use abnormality detection method, device, electronic apparatus, and storage medium | |
| CN118330470B (en) | Super-capacitor coupling lithium battery energy storage capacity distribution method considering life cycle | |
| CN118757347B (en) | Intelligent identification method and system for wind generating set |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |