CN117909323A - Power transmission network false data detector and detection method thereof - Google Patents
Power transmission network false data detector and detection method thereof Download PDFInfo
- Publication number
- CN117909323A CN117909323A CN202311679964.9A CN202311679964A CN117909323A CN 117909323 A CN117909323 A CN 117909323A CN 202311679964 A CN202311679964 A CN 202311679964A CN 117909323 A CN117909323 A CN 117909323A
- Authority
- CN
- China
- Prior art keywords
- data
- encoder
- state
- sparse self
- state estimation
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 57
- 230000005540 biological transmission Effects 0.000 title claims abstract description 25
- 238000005259 measurement Methods 0.000 claims abstract description 58
- 238000000354 decomposition reaction Methods 0.000 claims description 36
- 239000011159 matrix material Substances 0.000 claims description 36
- 239000013598 vector Substances 0.000 claims description 30
- 238000000034 method Methods 0.000 claims description 28
- 238000012549 training Methods 0.000 claims description 24
- 238000009826 distribution Methods 0.000 claims description 11
- 238000004458 analytical method Methods 0.000 claims description 7
- 238000010586 diagram Methods 0.000 claims description 7
- 230000008569 process Effects 0.000 claims description 7
- 238000005520 cutting process Methods 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 abstract description 8
- 238000004422 calculation algorithm Methods 0.000 description 3
- 238000004891 communication Methods 0.000 description 3
- 230000006872 improvement Effects 0.000 description 3
- 230000009466 transformation Effects 0.000 description 3
- 230000006978 adaptation Effects 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 206010033799 Paralysis Diseases 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 125000004122 cyclic group Chemical group 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000004880 explosion Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 230000003211 malignant effect Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
- 238000011426 transformation method Methods 0.000 description 1
- 230000001052 transient effect Effects 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Abstract
The invention discloses a transmission network false data detector, which comprises: the system comprises a state estimator module, a bad data identification module and a false data detection module which are sequentially connected in series, wherein the state estimator module is used for continuously acquiring measurement data and estimating the internal state of the power system based on the measurement data to generate a state estimator; the bad data identification module is used for carrying out iterative calculation based on the measured data and the state estimation quantity until the state estimation quantity converges, carrying out bad data detection based on residual errors, and correcting the suspicious bad data based on the residual errors if the suspicious bad data exists; the false data detection module is used for judging the measured data output by the bad data identification module based on the pre-trained sparse self-encoder to generate a false data detection result of the power transmission network, so that the limitation of the topological structure of the power system is effectively eliminated, the detection accuracy is improved, and the problem that the traditional false data detection method of the power transmission network cannot balance the calculation load and the detection accuracy is solved.
Description
Technical Field
The invention relates to the technical field of data cleaning of power systems, in particular to a false data detector of a power transmission network and a detection method thereof.
Background
With the development of information communication technology, more and more communication devices and computer technologies are integrated in a power system, but due to the dependence of a power information physical system (CPS) on data communication, the CPS is easily interfered by network attack, so that the problem of power system information security is also a hot spot of current research. Network security consists of confidentiality, integrity and availability, with different attack methods for three different features. False data bypasses the identification of bad data modules, and measurement information containing attack data is transmitted to a control center, so that decision of the control center can be influenced, and serious malignant results such as large-area paralysis of a power system are easily caused.
Current methods of detecting spurious data can be categorized into three types, state estimation based, trajectory prediction based, and artificial intelligence based. Most state estimation methods are an improvement on the original weighted least squares regression method, inevitably require collection of a large amount of measurement data in the power system network, and are limited by the topology of the system itself. The method based on track prediction uses a Kalman filter and the like to predict the state value, the data at the previous sampling time is required to be stable and reliable, and the defect of escaping state estimation is basically avoided. The artificial intelligence method is widely applied to practical scenes such as image recognition, sound processing and the like, has high data processing accuracy, can get rid of the limitation of an original structure when applied to a power system, starts from the characteristics of data, analyzes whether false data exist, but the semi-supervised learning method needs to mark the data, and has the problems of complex pre-training process and large workload; some intelligent algorithms (such as convolutional neural networks added with gating cyclic unit structures) can have gradient explosion conditions along with the increase of data scale; in the actual operation process of the machine learning method applied to false data detection, the method for detecting the false data at a single moment is not necessarily effective because of complex power grid topological structure, large data acquisition quantity and obvious nonlinear structural characteristics, and less false data can be collected.
In summary, the conventional method for detecting the false data of the power transmission network has the problem that the calculation load and the detection accuracy cannot be balanced.
Disclosure of Invention
In view of the above, the invention provides a transmission network false data detector and a method thereof, which solve the problems that the traditional transmission network false data detection method cannot balance the calculation load and the detection accuracy by configuring a false data detection method based on the combination of wavelet transformation and a sparse self-encoder in a false data detection module.
In order to solve the above problems, the technical scheme of the invention is to adopt a transmission network false data detector, comprising: the system comprises a state estimator module, a bad data identification module and a false data detection module which are sequentially connected in series, wherein the state estimator module is used for continuously acquiring measurement data and estimating the internal state of the power system based on the measurement data to generate a state estimator; the bad data identification module is used for carrying out iterative computation based on the measured data and the state estimation quantity until the state estimation quantity is converged, carrying out bad data detection based on residual errors, and correcting the suspicious bad data based on the residual errors if the suspicious bad data exists; and the false data detection module is used for judging the measured data output by the bad data identification module based on the pre-trained sparse self-encoder and generating a transmission network false data detection result.
Optionally, the spurious data detection module is configured to: generating state estimation data of each node of the power system based on the measurement data output by the bad data identification module, cutting the continuous state estimation data by using a sliding time module, dividing the continuous state estimation data into continuous state estimation data under a plurality of sections, decomposing the divided continuous state estimation data by using a wavelet analysis method in time domain and frequency domain, and extracting wavelet decomposition energy values to form feature vectors; and constructing a decomposition matrix based on a plurality of feature vectors to draw an energy distribution diagram under a multi-scale, taking a vector formed by decomposition coefficients of the decomposition matrix as input of each row of the sparse self-encoder, training the sparse self-encoder, and dynamically adjusting a weight matrix and a bias vector of an encoder and a decoder of the sparse self-encoder to obtain the trained sparse self-encoder.
Optionally, a vector formed by the decomposition coefficients of the decomposition matrix is used as an input of each row of the sparse self-encoder, and in the process of training the sparse self-encoder, a reconstruction error of input data and output data of the sparse self-encoder is used as a constraint condition, and the reconstruction error is minimized as a training target.
Optionally, the state estimator module is configured to: after measurement data of the power system are obtained, calculating a state variable under an alternating current power flow model by using a formula z=h (x) +e, wherein h is a Jacobian matrix considering the topological structure of the alternating current system, x is the state variable, z is the measurement data obtained through a data acquisition center, e is a measurement error, and standard normal distribution is obeyed; based on the state variables, the formula is utilizedAnd calculating the state estimation quantity, wherein W is a weight positive definite matrix.
Optionally, the state estimator module is further configured to: after generating the state estimator, the residual of the metrology data is generated based on the state estimator and the metrology data.
Correspondingly, the invention provides a transmission network false data detection method, which comprises the following steps: acquiring measurement data and estimating the internal state of the power system based on the measurement data to generate a state estimator; performing iterative computation based on the measurement data and the state estimation quantity until the state estimation quantity converges, performing bad data detection based on residual errors, and if suspicious bad data exists, correcting the suspicious bad data based on the residual errors; and judging the measurement data output by the bad data identification module based on the pre-trained sparse self-encoder, and generating a transmission network false data detection result.
Optionally, the method of pre-training the sparse self-encoder comprises: generating state estimation data of each node of the power system based on the measurement data output by the bad data identification module, cutting the continuous state estimation data by using a sliding time module, dividing the continuous state estimation data into continuous state estimation data under a plurality of sections, decomposing the divided continuous state estimation data by using a wavelet analysis method in time domain and frequency domain, and extracting wavelet decomposition energy values to form feature vectors; and constructing a decomposition matrix based on a plurality of feature vectors to draw an energy distribution diagram under a multi-scale, taking a vector formed by decomposition coefficients of the decomposition matrix as input of each row of the sparse self-encoder, training the sparse self-encoder, and dynamically adjusting a weight matrix and a bias vector of an encoder and a decoder of the sparse self-encoder to obtain the trained sparse self-encoder.
Optionally, a vector formed by the decomposition coefficients of the decomposition matrix is used as an input of each row of the sparse self-encoder, and in the process of training the sparse self-encoder, a reconstruction error of input data and output data of the sparse self-encoder is used as a constraint condition, and the reconstruction error is minimized as a training target.
Optionally, obtaining the measurement data and estimating the internal state of the power system based on the measurement data, generating the state estimator includes: after measurement data of the power system are obtained, calculating a state variable under an alternating current power flow model by using a formula z=h (x) +e, wherein h is a Jacobian matrix considering the topological structure of the alternating current system, x is the state variable, z is the measurement data obtained through a data acquisition center, e is a measurement error, and standard normal distribution is obeyed; based on the state variables, the formula is utilizedAnd calculating the state estimation quantity, wherein W is a weight positive definite matrix.
Optionally, the transmission network false data detection method further includes: after generating the state estimator, the residual of the metrology data is generated based on the state estimator and the metrology data.
The primary improvement of the invention is that the state estimator module generates the state estimator of the power system and corrects the suspicious bad data through the bad data identification module, and then the false data detection result of the power transmission network is generated based on the configuration of the false data detection method based on the combination of wavelet transformation and sparse self-encoder in the false data detection module, thereby effectively getting rid of the limitation of the topological structure of the power system, improving the detection accuracy rate, effectively solving the actual problem of small false data quantity during pre-training, and solving the problems of unbalanced calculation force load and detection accuracy existing in the traditional false data detection method of the power transmission network.
Drawings
FIG. 1 is a simplified block diagram of a grid spurious data detector of the present invention;
fig. 2 is a simplified flow chart of a method of grid spurious data detection of the present invention.
Detailed Description
In order to make the technical solution of the present invention better understood by those skilled in the art, the present invention will be further described in detail with reference to the accompanying drawings and specific embodiments.
As shown in fig. 1, a grid dummy data detector, comprising: the system comprises a state estimator module, a bad data identification module and a false data detection module which are sequentially connected in series, wherein the state estimator module is used for continuously acquiring measurement data and estimating the internal state of the power system based on the measurement data to generate a state estimator; the bad data identification module is used for carrying out iterative computation based on the measured data and the state estimation quantity until the state estimation quantity is converged, carrying out bad data detection based on residual errors, and correcting the suspicious bad data based on the residual errors if the suspicious bad data exists; and the false data detection module is used for judging the measured data output by the bad data identification module based on the pre-trained sparse self-encoder and generating a transmission network false data detection result. The measurement data may include network connection parameters, branch parameters, and measurement system parameters.
Further, the present invention configures the dummy data detection module to: generating state estimation data of each node of the power system based on the measurement data output by the bad data identification module, cutting the continuous state estimation data by using a sliding time module, dividing the continuous state estimation data into continuous state estimation data under a plurality of sections, decomposing the divided continuous state estimation data by using a wavelet analysis method in time domain and frequency domain, and extracting wavelet decomposition energy values to form feature vectors; and constructing a decomposition matrix based on a plurality of feature vectors to draw an energy distribution diagram under a multi-scale, taking a vector formed by decomposition coefficients of the decomposition matrix as input of each row of the sparse self-encoder, training the sparse self-encoder, and dynamically adjusting a weight matrix and a bias vector of an encoder and a decoder of the sparse self-encoder to obtain the trained sparse self-encoder.
Furthermore, a vector formed by the decomposition coefficients of the decomposition matrix is used as the input of each row of the sparse self-encoder, and in the process of training the sparse self-encoder, the reconstruction error of the input data and the output data of the sparse self-encoder is used as a constraint condition, and the reconstruction error is minimized as a training target.
Still further, the spurious data detection module is further configured to: the wavelet transformation method is used for carrying out time-frequency domain analysis on state variables under the alternating current power flow model, a cluster of wavelet function systems is used for representing or approaching a certain function signal, sequence details can be displayed in a concentrated mode in the frequency domain, and the system has the capability of displaying local features in the time domain and the frequency domain so as to detect transient or singular points of the signal.
Further, since the working condition of the telemechanical device of the power system is constantly changed, the state estimation cannot work when the telemechanical information amount is seriously insufficient. Therefore, it is necessary to analyze the topology of the power system for observability verification prior to state estimation. If some part of the system is judged to be unobservable and a real-time database cannot be built up by state estimation, it should be backed out of the calculation of state estimation or made observable by adding a manually set virtual quantity measurement or called pseudo-metrology data. The present invention therefore configures the state estimator module to: after measurement data of the power system are obtained, calculating a state variable under an alternating current power flow model by using a formula z=h (x) +e, wherein h is a Jacobian matrix considering the topological structure of the alternating current system, x is a state variable, at least comprises voltage values and voltage phase angles of all nodes, z is measurement data obtained through a data acquisition center, and the measurement data are uploaded by measurement equipment such as a smart meter and a PMU; and e is measurement error and obeys standard normal distribution. Formula built by weighted least squares based on state variablesAnd calculating the state estimation quantity, wherein W is a weight positive definite matrix.
Further, the state estimator module is further configured to: after generating the state estimator, generating the residual error r of the measurement data based on the state estimator and the measurement data, wherein a calculation formula is as follows: r=z (t) -h [ x (t) ].
The primary improvement of the invention is that the state estimator module generates the state estimator of the power system and corrects the suspicious bad data through the bad data identification module, and then the false data detection result of the power transmission network is generated based on the configuration of the false data detection method based on the combination of wavelet transformation and sparse self-encoder in the false data detection module, thereby effectively getting rid of the limitation of the topological structure of the power system, improving the detection accuracy rate, effectively solving the actual problem of small false data quantity during pre-training, and solving the problems of unbalanced calculation force load and detection accuracy existing in the traditional false data detection method of the power transmission network.
Accordingly, as shown in fig. 2, the present invention provides a method for detecting false data of a power transmission network, including:
s1: measurement data is acquired and an internal state of the power system is estimated based on the measurement data, generating a state estimator.
Further, obtaining measurement data and estimating an internal state of the power system based on the measurement data, generating a state estimator, comprising: after measurement data of the power system are obtained, calculating a state variable under an alternating current power flow model by using a formula z=h (x) +e, wherein h is a Jacobian matrix considering the topological structure of the alternating current system, x is the state variable, and z is the measurement data obtained through a data acquisition center; based on the state variables, the formula is utilizedAnd calculating the state estimation quantity, wherein W is a weight positive definite matrix.
Further, the transmission network false data detection method further comprises the following steps: after generating the state estimator, the residual of the metrology data is generated based on the state estimator and the metrology data.
S2: performing iterative computation based on the measurement data and the state estimation quantity until the state estimation quantity converges, performing bad data detection based on residual errors, if the residual errors are higher than a threshold value, suspicious bad data exist, correcting the suspicious bad data based on the residual errors, and outputting corrected measurement data; if the residual error is lower than the threshold value, no suspicious bad data exists, and directly outputting measurement data;
S3: and judging the measurement data output by the bad data identification module based on the pre-trained sparse self-encoder, and generating a transmission network false data detection result.
Further, the method of pre-training the sparse self-encoder includes: generating state estimation data of each node of the power system based on the measurement data output by the bad data identification module, cutting the continuous state estimation data by using a sliding time module, dividing the continuous state estimation data into continuous state estimation data under a plurality of sections, decomposing the divided continuous state estimation data by using a wavelet analysis method in time domain and frequency domain, and extracting wavelet decomposition energy values to form feature vectors; and constructing a decomposition matrix based on a plurality of feature vectors to draw an energy distribution diagram under a multi-scale, taking a vector formed by decomposition coefficients of the decomposition matrix as input of each row of the sparse self-encoder, training the sparse self-encoder, and dynamically adjusting a weight matrix and a bias vector of an encoder and a decoder of the sparse self-encoder to obtain the trained sparse self-encoder.
Furthermore, a vector formed by the decomposition coefficients of the decomposition matrix is used as the input of each row of the sparse self-encoder, and in the process of training the sparse self-encoder, the reconstruction error of the input data and the output data of the sparse self-encoder is used as a constraint condition, and the reconstruction error is minimized as a training target.
The power transmission network false data detector and the detection method thereof provided by the embodiment of the invention. In the description, each embodiment is described in a progressive manner, and each embodiment is mainly described by the differences from other embodiments, so that the same similar parts among the embodiments are mutually referred. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant points refer to the description of the method section. It should be noted that it will be apparent to those skilled in the art that various modifications and adaptations of the invention can be made without departing from the principles of the invention and these modifications and adaptations are intended to be within the scope of the invention as defined in the following claims.
Those of skill would further appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both, and that the various illustrative elements and steps are described above generally in terms of functionality in order to clearly illustrate the interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention. The steps of a method or algorithm described in connection with the embodiments disclosed herein may be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. The software modules may be disposed in Random Access Memory (RAM), memory, read Only Memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.
Claims (10)
1. A grid spurious data detector, comprising: a state estimator module, a bad data identification module and a false data detection module which are connected in series, wherein,
The state estimator module is used for continuously acquiring measurement data and estimating the internal state of the power system based on the measurement data to generate a state estimator;
The bad data identification module is used for carrying out iterative computation based on the measured data and the state estimation quantity until the state estimation quantity is converged, carrying out bad data detection based on residual errors, and correcting the suspicious bad data based on the residual errors if the suspicious bad data exists;
And the false data detection module is used for judging the measured data output by the bad data identification module based on the pre-trained sparse self-encoder and generating a transmission network false data detection result.
2. The grid dummy data detector of claim 1, wherein the dummy data detection module is configured to: generating state estimation data of each node of the power system based on the measurement data output by the bad data identification module, cutting the continuous state estimation data by using a sliding time module, dividing the continuous state estimation data into continuous state estimation data under a plurality of sections, decomposing the divided continuous state estimation data by using a wavelet analysis method in time domain and frequency domain, and extracting wavelet decomposition energy values to form feature vectors;
And constructing a decomposition matrix based on a plurality of feature vectors to draw an energy distribution diagram under a multi-scale, taking a vector formed by decomposition coefficients of the decomposition matrix as input of each row of the sparse self-encoder, training the sparse self-encoder, and dynamically adjusting a weight matrix and a bias vector of an encoder and a decoder of the sparse self-encoder to obtain the trained sparse self-encoder.
3. The grid dummy data detector according to claim 2, wherein a vector of decomposition coefficients of a decomposition matrix is used as an input of each row of a sparse self-encoder, and a reconstruction error of input data and output data of the sparse self-encoder is used as a constraint condition in training the sparse self-encoder, and the reconstruction error is minimized as a training target.
4. The grid spurious data detector of claim 1, wherein the state estimator module is configured to: after measurement data of the power system are obtained, a state variable under an alternating current power flow model is calculated by using a formula z=h (x) +e, wherein h is a Jacobian matrix considering the topological structure of the alternating current system, x is the state variable, z is the measurement data obtained through a data acquisition center, and e is a measurement error and obeys standard normal distribution. Based on the state variables, the formula is utilizedAnd calculating the state estimation quantity, wherein W is a weight positive definite matrix.
5. The grid spurious data detector of claim 4, wherein the state estimator module is further configured to: after generating the state estimator, the residual of the metrology data is generated based on the state estimator and the metrology data.
6. A method for detecting grid spurious data, comprising:
acquiring measurement data and estimating the internal state of the power system based on the measurement data to generate a state estimator;
Performing iterative computation based on the measurement data and the state estimation quantity until the state estimation quantity converges, performing bad data detection based on residual errors, and if suspicious bad data exists, correcting the suspicious bad data based on the residual errors;
and judging the measurement data output by the bad data identification module based on the pre-trained sparse self-encoder, and generating a transmission network false data detection result.
7. The method of grid spurious data detection according to claim 6, wherein the method of pre-training the sparse self-encoder comprises:
Generating state estimation data of each node of the power system based on the measurement data output by the bad data identification module, cutting the continuous state estimation data by using a sliding time module, dividing the continuous state estimation data into continuous state estimation data under a plurality of sections, decomposing the divided continuous state estimation data by using a wavelet analysis method in time domain and frequency domain, and extracting wavelet decomposition energy values to form feature vectors;
And constructing a decomposition matrix based on a plurality of feature vectors to draw an energy distribution diagram under a multi-scale, taking a vector formed by decomposition coefficients of the decomposition matrix as input of each row of the sparse self-encoder, training the sparse self-encoder, and dynamically adjusting a weight matrix and a bias vector of an encoder and a decoder of the sparse self-encoder to obtain the trained sparse self-encoder.
8. The method according to claim 7, wherein a vector composed of decomposition coefficients of a decomposition matrix is used as an input of each row of a sparse self-encoder, and in the process of training the sparse self-encoder, a reconstruction error of input data and output data of the sparse self-encoder is used as a constraint condition, and the reconstruction error is minimized as a training target.
9. The method of claim 6, wherein obtaining the metrology data and estimating the internal state of the power system based on the metrology data, generating the state estimator, comprises:
After measurement data of the power system are obtained, a state variable under an alternating current power flow model is calculated by using a formula z=h (x) +e, wherein h is a Jacobian matrix considering the topological structure of the alternating current system, x is the state variable, z is the measurement data obtained through a data acquisition center, and e is a measurement error and obeys standard normal distribution.
Based on the state variables, the formula is utilizedAnd calculating the state estimation quantity, wherein W is a weight positive definite matrix.
10. The grid dummy data detection method according to claim 6, further comprising: after generating the state estimator, the residual of the metrology data is generated based on the state estimator and the metrology data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311679964.9A CN117909323A (en) | 2023-12-07 | 2023-12-07 | Power transmission network false data detector and detection method thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311679964.9A CN117909323A (en) | 2023-12-07 | 2023-12-07 | Power transmission network false data detector and detection method thereof |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117909323A true CN117909323A (en) | 2024-04-19 |
Family
ID=90680717
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311679964.9A Pending CN117909323A (en) | 2023-12-07 | 2023-12-07 | Power transmission network false data detector and detection method thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117909323A (en) |
-
2023
- 2023-12-07 CN CN202311679964.9A patent/CN117909323A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Mestav et al. | Bayesian state estimation for unobservable distribution systems via deep learning | |
| US11418029B2 (en) | Method for recognizing contingencies in a power supply network | |
| CN112149554B (en) | Training and fault detection methods of fault classification model and related devices | |
| CN110942109A (en) | PMU false data injection attack prevention method based on machine learning | |
| CN109145516B (en) | Analog circuit fault identification method based on improved extreme learning machine | |
| CN110866448A (en) | Flutter signal analysis method based on convolutional neural network and short-time Fourier transform | |
| EP3894872A1 (en) | Condition monitoring via energy consumption audit in electrical devices and electrical waveform audit in power networks | |
| CN112651290A (en) | Hydroelectric generating set vibration trend prediction method and system | |
| CN111046327A (en) | Prony analysis method suitable for low-frequency oscillation and subsynchronous oscillation identification | |
| CN113765880B (en) | Power system network attack detection method based on space-time correlation | |
| CN110991776A (en) | Method and system for realizing water level prediction based on GRU network | |
| CN115935139A (en) | Space field interpolation method for ocean observation data | |
| US11657148B2 (en) | Event analysis in an electric power system | |
| Wu et al. | Online identification of bad synchrophasor measurements via spatio-temporal correlations | |
| CN105228185A (en) | A kind of method for Fuzzy Redundancy node identities in identification communication network | |
| Huang et al. | False phasor data detection under time synchronization attacks: A neural network approach | |
| Horvath et al. | Sensor fault diagnosis of inland navigation system using physical model and pattern recognition approach | |
| CN111738348B (en) | Power data anomaly detection method and device | |
| CN117312769A (en) | BiLSTM-based method for detecting abnormality of time sequence data of Internet of things | |
| CN117909323A (en) | Power transmission network false data detector and detection method thereof | |
| CN116400168A (en) | Power grid fault diagnosis method and system based on depth feature clustering | |
| CN111695501B (en) | Equipment soft fault detection method based on operating system kernel calling data | |
| Qu et al. | Detection of False Data Injection Attack in Power System Based on Hellinger Distance | |
| Hu et al. | Detection of False Data Injection Attacks in Smart Grids Under Power Fluctuation Uncertainty Based on Deep Learning | |
| Feng et al. | Detection of False Data Injection Attacks in Cyber-Physical Power Systems: An Adaptive Adversarial Dual Autoencoder With Graph Representation Learning Approach |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination |