CN110633874B - Transformer substation secondary equipment state monitoring method and readable storage medium - Google Patents

Transformer substation secondary equipment state monitoring method and readable storage medium Download PDF

Info

Publication number
CN110633874B
CN110633874B CN201910933444.3A CN201910933444A CN110633874B CN 110633874 B CN110633874 B CN 110633874B CN 201910933444 A CN201910933444 A CN 201910933444A CN 110633874 B CN110633874 B CN 110633874B
Authority
CN
China
Prior art keywords
model
feature
monitoring
dbn
transformer substation
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201910933444.3A
Other languages
Chinese (zh)
Other versions
CN110633874A (en
Inventor
王洪彬
张友强
李�杰
龚秋憬
余红欣
何燕
何荷
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Electric Power Research Institute of State Grid Chongqing Electric Power Co Ltd
State Grid Corp of China SGCC
Original Assignee
Electric Power Research Institute of State Grid Chongqing Electric Power Co Ltd
State Grid Corp of China SGCC
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Electric Power Research Institute of State Grid Chongqing Electric Power Co Ltd, State Grid Corp of China SGCC filed Critical Electric Power Research Institute of State Grid Chongqing Electric Power Co Ltd
Priority to CN201910933444.3A priority Critical patent/CN110633874B/en
Publication of CN110633874A publication Critical patent/CN110633874A/en
Application granted granted Critical
Publication of CN110633874B publication Critical patent/CN110633874B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

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
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"

Abstract

The invention discloses a method for monitoring the state of secondary equipment of a transformer substation and a readable storage medium, wherein the method comprises the steps of collecting monitoring signals of the secondary equipment of the transformer substation, and extracting the characteristics of the detection signals; and performing model recognition on the extracted features, and completing the trend prediction of the state information of the secondary equipment of the transformer substation according to the model recognition result. The method of the invention can realize more advanced evaluation by using the implicit information of the data point change trend and through trend prediction, and effectively reduce the difference between the evaluation result and the actual state of the equipment.

Description

Transformer substation secondary equipment state monitoring method and readable storage medium
Technical Field
The invention relates to the technical field of transformer substation fault monitoring, in particular to a transformer substation secondary equipment state monitoring method and a readable storage medium.
Background
In recent years, with the continuous expansion of the scale of a power grid, the level of power transmission and transformation voltage is continuously improved, and the construction of a new generation of intelligent transformer substation is promoted. This causes the number and complexity of power equipment to increase continuously, the maintenance cost of the equipment to increase continuously, and the maintenance workload of the relay protection equipment to increase sharply. In order to ensure reliable operation of a power grid under the conditions of a large number of secondary devices and high device complexity, a maintenance strategy of power equipment is developing towards a state maintenance direction. The diagnosis of the running state of the secondary equipment of the intelligent substation is an important basis for risk assessment and state maintenance. Therefore, the accuracy of condition monitoring is critical to the safety and reliability of the power grid.
Disclosure of Invention
In view of the above-mentioned defects in the prior art, an object of the present invention is to provide a method for monitoring the state of a secondary device in a substation and a readable storage medium, which can realize more advanced evaluation by using implicit information of a data point variation trend and by means of trend prediction, and effectively reduce the difference between an evaluation result and the actual state of the device.
One of the objects of the present invention is achieved by a method for monitoring the state of a secondary device of a substation, the method comprising,
collecting monitoring signals of secondary equipment of a transformer substation, and extracting characteristics of the monitoring signals;
and carrying out model identification on the extracted features, and completing the trend prediction of the state information of the secondary equipment of the transformer substation according to the model identification result.
Optionally, the performing feature extraction on the monitoring signal includes:
dividing the acquired signal waveform into a training set and a test set;
preprocessing the signal waveforms of the training set and the test set;
the pretreatment comprises the following steps: filtering, calculating a waveform root mean square, extracting a sag data segment and resampling.
Optionally, the performing feature extraction on the monitoring signal further includes:
Establishing a Deep Belief Network (DBN) model according to the data dimensions of the acquired training set, and determining the node number of each visible layer and each hidden layer according to the DBN model;
inputting the time domain data of the acquired signal waveforms of the training set into a DBN model;
and extracting the signal characteristics of the DBN model by adopting an unsupervised layer-by-layer training method according to the determined node number.
Optionally, performing model identification on the extracted features, including:
inputting the extracted signal characteristics into a DBN model for model training;
and verifying the trained DBN model through a test set.
Optionally, the performing feature extraction on the monitoring signal further includes:
evaluating the number of hidden layer units and the influence of a multi-layer limited Boltzmann machine RBM on a DBN model through characteristic dispersion to highlight data characteristics;
the characteristic dispersion satisfies:
Figure GDA0003634089480000021
wherein D represents the feature dispersion, v represents the feature distance of the current class, m represents the feature number of the same class, and pv，mRepresenting the feature vector, p, of the current class of waveformsv，cThe feature center vector of the waveform is represented, n represents the number of features of the same category, and u represents the total number of data categories.
Optionally, the selection principle of the feature center vector satisfies:
Figure GDA0003634089480000022
Wherein l, i, j all represent waveform type, pv，i，pv，jA feature vector representing the i, j waveform of the current class.
Optionally, the trend prediction of the state information of the secondary device of the substation is completed according to the model identification result, and the trend prediction includes:
predicting the state information trend by adopting a quadratic exponential smoothing method according to the model identification result;
the prediction model of the quadratic exponential smoothing method satisfies the following conditions:
Figure GDA0003634089480000023
wherein x ist+TDenotes the prediction result,. epsilon.denotes the smoothing coefficient, at，btAre all intermediate quantities, T represents the number of iterations,
Figure GDA0003634089480000024
representing the first and second smoothed values, respectively.
Optionally, the trend prediction of the state information of the secondary device of the substation is completed according to the model identification result, further comprising:
selecting the range of the smoothing coefficient, determining the minimum variance and the error square sum between the monitoring value and the predicted value, and satisfying the following conditions:
Figure GDA0003634089480000031
wherein N is the number of training samples, k represents the number of groups,
Figure GDA0003634089480000032
is a predicted value, x, of the kth set of monitored valueskIs the actual value of the kth set of monitored values.
The second object of the present invention is achieved by the technical solution, which is a computer-readable storage medium, wherein an implementation program for information transfer is stored on the computer-readable storage medium, and the implementation program implements the steps of the foregoing method when executed by a processor.
Due to the adoption of the technical scheme, the invention has the following advantages: the traditional monitoring method only utilizes time period information and has certain hysteresis. The method of the invention utilizes the implicit information of the data point variation trend and can realize more advanced evaluation through trend prediction, thereby effectively reducing the difference between the evaluation result and the actual state of the equipment.
Additional advantages, objects, and features of the invention will be set forth in part in the description which follows and in part will become apparent to those having ordinary skill in the art upon examination of the following or may be learned from practice of the invention.
Drawings
The drawings of the invention are illustrated as follows:
FIG. 1 is a flow chart of an embodiment of the present invention;
fig. 2 is a schematic diagram of real-time status information of a secondary device according to an embodiment of the present invention;
FIG. 3 is a schematic diagram illustrating comparison between a predicted voltage value and a monitored voltage value according to an embodiment of the present invention.
Detailed Description
The invention is further illustrated by the following figures and examples.
A first embodiment of the present invention provides a method for monitoring the state of a secondary device of a substation, as shown in fig. 1, the method includes,
collecting monitoring signals of secondary equipment of a transformer substation, and extracting characteristics of the monitoring signals;
And performing model recognition on the extracted features, and completing the trend prediction of the state information of the secondary equipment of the transformer substation according to the model recognition result.
The method of the invention utilizes the implicit information of the data point variation trend and can realize more advanced evaluation through trend prediction, thereby effectively reducing the difference between the evaluation result and the actual state of the equipment.
Optionally, in an optional embodiment of the present invention, the performing feature extraction on the monitoring signal includes:
dividing the acquired signal waveform into a training set and a test set;
preprocessing the signal waveforms of the training set and the test set;
the pretreatment comprises the following steps: filtering, calculating a waveform root mean square, extracting a sag data segment and resampling.
Specifically, the secondary equipment state feature extraction and Deep Belief Network (DBN) is composed of a plurality of layers of Restricted Boltzmann Machines (RBMs) and a layer of BP neural network, and has good classification effect and high-efficiency feature extraction capability. The core of the algorithm is to optimize interlayer connection weight by using a greedy learning algorithm layer by layer.
For a given set of states (v, h), the joint assigned energy function of the constrained boltzmann machine (RBM) is:
Figure GDA0003634089480000041
In the formula, viAnd aiThe state and offset of the i display layer unit are obtained; n and m are respectively rbm apparent layer unit number and hidden layer unit number; h is a total ofjAnd bjState and offset of hidden layer unit j; and theta is a parameter of the rbm model.
S＝{v1，v2，L，vsAnd the hidden layer is used as the characteristic of input data in the display layer by obtaining model parameters through a logarithm likelihood function of the RBM on the maximized sample and fitting the training sample.
Figure GDA0003634089480000042
Figure GDA0003634089480000043
The log-likelihood function uses a contrast divergence algorithm to obtain theta, and further obtains:
Figure GDA0003634089480000044
therefore, ωijCan be updated as:
Figure GDA0003634089480000045
where θ is the momentum and η is the learning speed.
On the basis of the above description, the secondary device state feature identification based on the DBN includes:
1) the monitoring system acquires time domain monitoring signals of various secondary devices;
2) dividing all data into a training set and a test set, and simultaneously preprocessing waveform data: filtering, calculating the root mean square of the waveform, extracting a sag data segment and resampling.
Optionally, the performing feature extraction on the monitoring signal further includes:
establishing a Deep Belief Network (DBN) model according to the data dimension of the collected training set, and determining the node number of each display layer and each hidden layer according to the DBN model;
Inputting the acquired time domain data of the signal waveforms of the training set into a DBN model;
and extracting the signal characteristics of the DBN model by adopting an unsupervised layer-by-layer training method according to the determined node number.
Specifically, the above scheme includes:
3) establishing a multi-hidden-layer DBN model according to data dimensions so as to accurately calculate the number of nodes of each display layer and each hidden layer;
4) and inputting the time domain data into the DBN model, and extracting the characteristics of the DBN model by adopting an unsupervised layer-by-layer training method.
Optionally, performing model identification on the extracted features, including:
inputting the extracted signal characteristics into a DBN model for model training;
and verifying the trained DBN model through a test set.
Specifically, 5) waveform data to be identified is input into the DBN model, and the state data of the secondary equipment is identified through the trained model, so that the identification accuracy of the method is verified.
Optionally, the performing feature extraction on the monitoring signal further includes:
evaluating the number of hidden layer units and the influence of a multi-layer limited Boltzmann machine (RBM) on a DBN model through feature dispersion to highlight data features;
the characteristic dispersion satisfies:
Figure GDA0003634089480000051
wherein D represents the feature dispersion, v represents the feature distance of the current class, m represents the feature number of the same class, and p v，mRepresenting the feature vector, p, of the current class of waveformsv，cThe feature center vector of the waveform is represented, n represents the number of features of the same category, and u represents the total number of data categories.
Optionally, the selection principle of the feature center vector satisfies:
Figure GDA0003634089480000052
wherein l, i, j all represent waveform type, pv，i，pv，jA feature vector representing the i, j waveform of the current class.
Specifically, in the embodiment, in order to evaluate the clustering degree of feature extraction under different parameters, feature dispersion is defined to evaluate the number of hidden layer units and the influence of the RBM on the DBN model. In general, to effectively highlight and make difficult to fit data features, it is desirable to extract features from the same class of data as closely as possible. Therefore, the smaller the feature dispersion D, the better the extracted features. The feature dispersion D is defined as the average value of the maximum distance between each type of feature vector and the feature center vector thereof, and the feature dispersion D satisfies the following conditions:
Figure GDA0003634089480000061
wherein D represents the characteristic dispersion, v representsDistance of features of the preceding class, m representing the number of features of the same class, pv，mRepresenting the feature vector, p, of the current class of waveformsv，cThe feature center vector of the waveform is represented, n represents the number of features of the same category, and u represents the total number of data categories.
The selection principle of the characteristic center vector meets the following requirements:
Figure GDA0003634089480000062
in the formula, m is the characteristic number of the same category; v is the characteristic distance of the current v class; p is a radical ofv，l，pv，i，pv，jRespectively, the feature vectors, p, of the v-type l, i, j waveformsv，cRepresenting a feature center vector of the waveform.
Optionally, the trend prediction of the state information of the secondary device of the substation according to the model identification result includes:
predicting the state information trend by adopting a quadratic exponential smoothing method according to the model identification result;
the prediction model of the quadratic exponential smoothing method satisfies the following conditions:
Figure GDA0003634089480000063
wherein x ist+TDenotes the prediction result,. epsilon.denotes the smoothing coefficient, at，btAre all intermediate quantities, T represents the number of iterations,
Figure GDA0003634089480000064
representing the first and second smoothed values, respectively.
Specifically, as shown in fig. 2, the function of the secondary equipment is greatly affected by the reliability of components, accidents occur randomly, and the traditional state evaluation method depends on the historical operating state of the equipment, and the accuracy and timeliness of the traditional state evaluation method are not guaranteed. In the embodiment, a state prediction information model is established based on the self-checking information and the alarm information of the secondary equipment of the intelligent substation and in combination with the key historical operation information,
the state evaluation results in the existing literature mainly depend on historical information and recent data, and the method has the following problems: some information points are good, but have obvious deterioration trend and should be paid attention; some of the spots of information are poor, but tend to improve, and do not need to be processed immediately. Therefore, the embodiment proposes a strategy of predicting the information trend by using real-time data and then performing fuzzy evaluation based on the prediction result, so that the evaluation error is improved to a certain extent. The information trend prediction enlarges the change of the equipment state, and hidden dangers can be found more timely by enlarging the state deterioration trend; by expanding the state optimization trend, inefficient processes can be avoided and costs can be saved.
Further, in the embodiment, a quadratic exponential smoothing method is adopted to predict the state information trend;
if the time series of the monitored parameters is x1，x2，L，xtTaking the arithmetic mean of 3-5 data as the initial value
Figure GDA0003634089480000071
And
Figure GDA0003634089480000072
the prediction model of the quadratic exponential smoothing method is as follows:
Figure GDA0003634089480000073
the prediction result is recorded as xt+1，xt+2，L，xt+T。
Optionally, the trend prediction of the state information of the secondary device of the substation is completed according to the model identification result, further comprising:
selecting the range of the smoothing coefficient, determining the minimum variance and the error square sum between the monitoring value and the predicted value, and satisfying the following conditions:
Figure GDA0003634089480000074
wherein N is the number of training samples, k represents the number of groups,
Figure GDA0003634089480000075
is a predicted value, x, of the kth set of monitored valueskIs the actual value of the kth set of monitored values.
Specifically, in the present embodiment, the smoothing coefficient ∈ determines the response sensitivity and the degree of smoothing between the predicted value and the actual value. The larger the epsilon value is, the smaller the contribution of the long-term monitoring parameters to the predicted value is; the smaller the value of epsilon, the greater the contribution of the long-term monitored parameter to the predicted value. The magnitude of epsilon depends on the variation of the device test parameters. Thus, in this embodiment, first, an approximate range of ε is selected; and then, calculating and selecting to obtain the minimum variance and the sum of squared errors between the monitored value and the predicted value.
Figure GDA0003634089480000076
Wherein N is the number of training samples, k is the number of groups,
Figure GDA0003634089480000077
is a predicted value, x, of the kth set of monitored valueskIs the actual value of the kth set of monitored values.
By the method, the method has the following advantages
1) The feature extraction capability of the deep information network is fully utilized, and manual feature extraction is converted into automatic feature extraction. The method integrates the extraction and classification of the features to solve the problems that the extraction of the artificial features depends on expert experience too much and the universality of unknown features is not enough.
2) The traditional monitoring method only utilizes time period information and has certain hysteresis. And by using the implicit information of the data point variation trend and through trend prediction, more advanced evaluation can be realized, and the difference between the evaluation result and the actual state of the equipment is effectively reduced.
3) The superiority and effectiveness of the method are verified by comparing with other methods.
The second embodiment of the invention provides an implementation case of a transformer substation secondary equipment state monitoring method
The state of the secondary equipment of a 220kV intelligent substation in a national grid is analyzed through an example and compared with a typical secondary equipment state monitoring method.
A. Secondary equipment state recognition result analysis based on DBN
In this embodiment, the input layer of the DBN model includes five layers, the number of units in each layer is 35, and the number of units in the last layer is the number of extracted features. The maximum number of iterations of the RBM is 12 and the momentum parameter is 4. The accuracy of the DBN-based secondary device identification method is shown in table 1. Experimental results show that the accuracy of the identification method for extracting and classifying the feature vectors by using the DBN model is 98.03%, and the identification accuracy of the network switch and the measurement and control equipment can reach 99.97%.
TABLE 1 accuracy of secondary equipment state monitoring method based on DBN
Class of secondary device Number of test samples Correct identification of quantity Percent accuracy%
Relay protection device 20 19 95
Measurement and control equipment 20 20 100
Intelligent terminal 20 19 95
Communication equipment 20 20 100
Total up to 80 78 98.03
B-state information trend prediction result analysis
And a quadratic exponential smoothing prediction method is adopted, so that one-step short-term time sequence prediction of the equipment information points is realized.
The accuracy and timeliness of evaluation are improved by using the change trend of the data, and the prediction result of the subsequent state is updated by using the real-time monitoring value. The device voltage prediction results are shown in fig. 3, and since only one-step prediction is performed, the prediction accuracy is high, and the value of ∈ is 0.4.
When the equipment voltage is larger than a normal value, the predicted value highlights the trend of increasing the voltage; the predicted value will highlight a trend of smaller voltage when the voltage decreases. The prediction results are penalized locally, i.e. the predicted values highlight the deterioration trend of the actual values, leaving a safety buffer margin, which is consistent with the meaning of using the state assessment to implement the state maintenance.
The second object of the present invention is achieved by the technical solution, which is a computer-readable storage medium, wherein an implementation program for information transfer is stored on the computer-readable storage medium, and the implementation program implements the steps of the foregoing method when executed by a processor.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solutions of the present invention and not for limiting the same, and although the present invention is described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: modifications and equivalents may be made to the embodiments of the invention without departing from the spirit and scope of the invention, which is intended to be covered thereby.

Claims (5)

1. A method for monitoring the state of secondary equipment of a transformer substation is characterized by comprising the following steps,
collecting monitoring signals of secondary equipment of a transformer substation, and extracting characteristics of the monitoring signals;
performing model identification on the extracted features, and completing trend prediction of the state information of the secondary equipment of the transformer substation according to the model identification result;
wherein the performing feature extraction on the monitoring signal comprises:
dividing the acquired signal waveform into a training set and a test set;
preprocessing the signal waveforms of the training set and the test set;
the pretreatment comprises the following steps: filtering, calculating a waveform root mean square, extracting a sag data segment and resampling;
the performing feature extraction on the monitoring signal further includes:
establishing a Deep Belief Network (DBN) model according to the data dimension of the collected training set, and determining the node number of each display layer and each hidden layer according to the DBN model;
inputting the acquired time domain data of the signal waveforms of the training set into a DBN model;
extracting the signal characteristics of the DBN model by adopting an unsupervised layer-by-layer training method according to the determined node number;
the model identification of the extracted features comprises the following steps:
Inputting the extracted signal characteristics into a DBN model for model training;
verifying the trained DBN model through a test set;
the performing feature extraction on the monitoring signal further includes:
evaluating the number of hidden layer units and the influence of a multi-layer limited Boltzmann machine RBM on a DBN model through characteristic dispersion to highlight data characteristics;
the characteristic dispersion satisfies the following conditions:
Figure FDA0003634089470000011
wherein D represents the feature dispersion, v represents the feature distance of the current class, m represents the feature number of the same class, and pv,mFeature vector, p, representing the current class of waveformsv,cThe feature center vector of the waveform is represented, n represents the number of features of the same category, and u represents the total number of data categories.
2. The method of claim 1, wherein the feature center vector is selected based on the following criteria:
Figure FDA0003634089470000012
wherein l, i, j all represent waveform type, pv,i，pv,jA feature vector representing the i, j waveform of the current class.
3. The method of claim 1, wherein performing trend prediction of substation secondary device status information based on model identification results comprises:
predicting the state information trend by adopting a quadratic exponential smoothing method according to the model identification result;
the prediction model of the quadratic exponential smoothing method satisfies the following conditions:
Figure FDA0003634089470000021
Wherein x ist+TDenotes the prediction result,. epsilon.denotes the smoothing coefficient, at，btAre all intermediate quantities, T represents the number of iterations,
Figure FDA0003634089470000022
representing the first and second smoothed values, respectively.
4. The method of claim 3, wherein the trend prediction of the substation secondary device status information is performed based on the model identification result, further comprising:
selecting the range of the smoothing coefficient, determining the minimum variance and the error square sum between the monitoring value and the predicted value, and satisfying the following conditions:
Figure FDA0003634089470000023
wherein N is the number of training samples, k represents the number of groups,
Figure FDA0003634089470000024
is a predicted value, x, of the kth set of monitored valueskIs the actual value of the kth set of monitored values.
5. A computer-readable storage medium, characterized in that it has stored thereon a program for implementing the transfer of information, which program, when being executed by a processor, implements the steps of the method according to any one of claims 1 to 4.
CN201910933444.3A 2019-09-29 2019-09-29 Transformer substation secondary equipment state monitoring method and readable storage medium Active CN110633874B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910933444.3A CN110633874B (en) 2019-09-29 2019-09-29 Transformer substation secondary equipment state monitoring method and readable storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910933444.3A CN110633874B (en) 2019-09-29 2019-09-29 Transformer substation secondary equipment state monitoring method and readable storage medium

Publications (2)

Publication Number Publication Date
CN110633874A CN110633874A (en) 2019-12-31
CN110633874B true CN110633874B (en) 2022-06-28

Family

ID=68973756

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910933444.3A Active CN110633874B (en) 2019-09-29 2019-09-29 Transformer substation secondary equipment state monitoring method and readable storage medium

Country Status (1)

Country Link
CN (1) CN110633874B (en)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5210704A (en) * 1990-10-02 1993-05-11 Technology International Incorporated System for prognosis and diagnostics of failure and wearout monitoring and for prediction of life expectancy of helicopter gearboxes and other rotating equipment
CN106908674A (en) * 2017-02-17 2017-06-30 国网上海市电力公司 A kind of Transformer condition evaluation based on the prediction of multimode amount
CN106980922A (en) * 2017-03-03 2017-07-25 国网天津市电力公司 A kind of power transmission and transformation equipment state evaluation method based on big data
CN109242205A (en) * 2018-10-09 2019-01-18 国网上海市电力公司 A kind of transformer state prediction technique and system

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5210704A (en) * 1990-10-02 1993-05-11 Technology International Incorporated System for prognosis and diagnostics of failure and wearout monitoring and for prediction of life expectancy of helicopter gearboxes and other rotating equipment
CN106908674A (en) * 2017-02-17 2017-06-30 国网上海市电力公司 A kind of Transformer condition evaluation based on the prediction of multimode amount
CN106980922A (en) * 2017-03-03 2017-07-25 国网天津市电力公司 A kind of power transmission and transformation equipment state evaluation method based on big data
CN109242205A (en) * 2018-10-09 2019-01-18 国网上海市电力公司 A kind of transformer state prediction technique and system

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
A Novel Deep Recurrent Belief Network Model for Trend Prediction of Transformer DGA Data;Bo Qi等;《IEEE Access》;20190614;第7卷;第80069-80078页 *
基于深度信念网络和多信息融合的复杂机电装备故障诊断方法;刘秀丽等;《机床与液压》;20190115;第47卷(第1期);第151-156页 *

Also Published As

Publication number Publication date
CN110633874A (en) 2019-12-31

Similar Documents

Publication Publication Date Title
CN110647539B (en) Prediction method and system for vehicle faults
US20190285517A1 (en) Method for evaluating health status of mechanical equipment
CN104392390B (en) A kind of secondary equipment of intelligent converting station appraisal procedure based on TOPSIS models
CN109583520B (en) State evaluation method of cloud model and genetic algorithm optimization support vector machine
CN104966161B (en) A kind of power quality recorder data calculation and analysis methods based on gauss hybrid models
CN105279365A (en) Method for learning exemplars for anomaly detection
CN106934237A (en) Radar cross-section redaction measures of effectiveness creditability measurement implementation method
CN108304567B (en) Method and system for identifying working condition mode and classifying data of high-voltage transformer
Shi et al. Improving power grid monitoring data quality: An efficient machine learning framework for missing data prediction
CN104052612A (en) Method and system for fault identification and positioning of telecommunication service
CN113901977A (en) Deep learning-based power consumer electricity stealing identification method and system
CN110794254B (en) Power distribution network fault prediction method and system based on reinforcement learning
CN112150304A (en) Power grid running state track stability prejudging method and system and storage medium
CN114429152A (en) Rolling bearing fault diagnosis method based on dynamic index antagonism self-adaption
CN115600136A (en) High-voltage bushing fault diagnosis method, system and medium based on multiple sensors
CN115793590A (en) Data processing method and platform suitable for system safety operation and maintenance
CN105488598A (en) Medium-and-long time electric power load prediction method based on fuzzy clustering
CN113742993B (en) Method, device, equipment and storage medium for predicting life loss of dry-type transformer
CN105335763A (en) Fabric defect classification method based on improved extreme learning machine
CN110633874B (en) Transformer substation secondary equipment state monitoring method and readable storage medium
CN105404973A (en) Power transmission and transformation equipment state prediction method and system
CN114692729A (en) New energy station bad data identification and correction method based on deep learning
CN113139332A (en) Automatic model construction method, device and equipment
CN111126694A (en) Time series data prediction method, system, medium and device
CN116227172A (en) Rolling bearing performance degradation evaluation method based on convolutional neural network

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