CN112215722A - Dominant instability mode discrimination model construction method and dominant instability mode discrimination method - Google Patents
Dominant instability mode discrimination model construction method and dominant instability mode discrimination method Download PDFInfo
- Publication number
- CN112215722A CN112215722A CN202011067784.1A CN202011067784A CN112215722A CN 112215722 A CN112215722 A CN 112215722A CN 202011067784 A CN202011067784 A CN 202011067784A CN 112215722 A CN112215722 A CN 112215722A
- Authority
- CN
- China
- Prior art keywords
- instability
- power angle
- voltage
- dominant
- instability mode
- 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.)
- Granted
Links
Images
Classifications
-
- 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
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/06—Electricity, gas or water supply
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/16—Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization
-
- 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
- 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
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/067—Enterprise or organisation modelling
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J2203/00—Indexing scheme relating to details of circuit arrangements for AC mains or AC distribution networks
- H02J2203/20—Simulating, e g planning, reliability check, modelling or computer assisted design [CAD]
Abstract
The invention discloses a method for constructing a dominant instability mode discrimination model and a method for discriminating a dominant instability mode, and belongs to the field of stability analysis of power systems. Aiming at the coupling interweaving characteristic of power angle instability and voltage instability, the invention applies deep learning to the judgment of a dominant instability mode, constructs a double-input convolution neural network model containing an SENet module, introduces the information between neural network channels which can be modeled by the SENet module, adaptively learns and adjusts the weight when each channel is added, amplifies useful characteristics and inhibits the characteristics with little use for the current task, and can quickly and accurately judge whether the system is unstable or not and whether the system belongs to the power angle instability or the voltage instability according to bus voltage after large disturbance, power angle measurement data or simulation data after the model training is finished, thereby providing a basis for the selection and coordination of control measures, greatly improving the safety of a power system and having stronger practicability.
Description
Technical Field
The invention belongs to the field of stability analysis of power systems, and particularly relates to a method for constructing a dominant instability mode discrimination model and a method for discriminating a dominant instability mode.
Background
The safe and stable operation of the power system is important for the national energy safety and the development of the economic society, and the safety and stability problems are the important concerns of system planning, operation and worker protection. The instability of the power system seriously jeopardizes the safe operation of the system, and causes great economic loss and even personal injury. In early small grids, transient power angle instability was the most dominant form of instability. However, as the interconnection scale of the power grid becomes larger and the load becomes heavier, the operating point of the power grid is closer to the transmission limit of the power grid, and a relatively large threat is caused to the voltage stability. In urban and industrial areas, the voltage stability problem is further aggravated by the heavy use of dynamic loads such as air conditioners and motors. Therefore, in modern power systems, the forms of power system instability have become more diverse and complex. Different forms of instability require different emergency control measures to restore the system to stability: the power angle instability needs to adopt a generator switching measure, and the voltage instability needs to switch the load. Therefore, the stability and the instability of the power system after the large disturbance and the leading instability mode are quickly and accurately judged, time is won for emergency control measures, and meanwhile, a basis is provided for which measures are taken, so that the safety and the stability of the system are effectively guaranteed.
In practice, however, two destabilizing modes often appear simultaneously interleaved coupled together: the power angle instability can cause low voltage of an oscillation center, and the power angle instability can be caused by the voltage instability, so that great difficulty is brought to distinguishing and distinguishing the dominance of the power angle instability and the voltage instability. The existing method has the following problems in aspects of rapidity, accuracy and adaptability:
the time domain simulation method has the advantages of early development, maturity, strong adaptability and rich provided information, but has heavy calculation burden and long time consumption, and is difficult to meet the requirements during online application; and the analysis of the simulation result is difficult and depends heavily on the expert experience. Methods such as energy functions, bifurcation analysis, etc. are difficult to apply to actual large power grids and power grids containing complex controllers.
With the development of WAMS, data in a power system are accumulated continuously, and meanwhile, the deep learning method is applied to the dominant instability mode judgment to be possible due to the progress of the deep learning method. Deep learning can get rid of dependence on expert experience, automatically extract required features from original measurement data or simulation data and complete the tasks of instability mode discrimination and classification. At present, some researches apply deep learning to single power angle stability judgment or single voltage stability judgment to obtain good effects, but mutual influence and coupling of power angle stability and voltage stability are not considered, and how to distinguish dominance of power angle stability and voltage stability is not considered, so that selection of control measures is not targeted, and safe and stable operation of a power system is influenced.
Therefore, how to quickly and accurately judge the dominant instability mode is a technical problem to be solved urgently at present.
Disclosure of Invention
Aiming at the defects or improvement requirements of the prior art, the invention provides a method for constructing a dominant instability mode discrimination model and a method for discriminating the dominant instability mode, and aims to solve the technical problem that the safe operation of a power system cannot be reliably ensured because the dominant discrimination of the instability mode cannot be realized by the conventional discrimination method.
To achieve the above object, according to an aspect of the present invention, there is provided a method for constructing a dominant instability mode discrimination model, including:
s1, respectively setting different tide operation working conditions, load motor component proportions, fault lines, fault positions and fault duration time to perform multiple groups of transient stability simulation;
s2, discretely sampling bus voltage and a power angle of the generator from the measurement data output after the fault occurs, and arranging the bus voltage and the power angle of the generator into two matrix forms after unifying the dimensions of the bus voltage and the power angle obtained by sampling to form a voltage matrix and a power angle matrix corresponding to each group of measurement data;
s3, determining the label information of the dominant instability mode of each group of measurement data by combining with expert experience;
s4, inputting a voltage matrix and a power angle matrix corresponding to each group of measurement data and corresponding labels into a dominant instability mode discrimination network, and training to obtain a dominant instability mode discrimination model through supervised learning;
the dominant instability mode discrimination network comprises a double-input convolutional neural network, a SENet network and a full-connection neural network;
the double-input convolutional neural network is used for respectively and automatically extracting required characteristics from the input voltage matrix and the input power angle matrix to obtain a characteristic vector corresponding to each sample;
the SENet network is used for explicitly modeling the correlation relation between the channels of the double-input convolutional neural network, adaptively learning and adjusting the weight value when each channel is added to form an output characteristic diagram;
and the fully-connected neural network is used for receiving the feature vector corresponding to each sample and classifying the sample through nonlinear transformation and a softmax function.
Further, unifying the dimensions of the bus voltage and the generator power angle in step S2, specifically, standardizing the bus voltage and the generator power angle by using Z-score standardization.
Further, the dominant instability mode label information includes a stable mode, power angle instability, and voltage instability.
Further, the expert experience specifically includes that the power angle is firstly turned on to be power angle instability, and the voltage is firstly broken down to be voltage instability; the generator end fault is power angle instability; judging from the effectiveness of the control measures, the generator is effectively switched to be power angle instability, and the load is effectively switched to be voltage instability; the instability of the local isolated part of the system does not affect the stability of the rest main systems, and the system is in a stable mode.
Further, the SENET network specifically executes processes including squeezing and excitation; the method comprises the following steps of extruding, gathering information in each input channel through global average pooling to form a multi-dimensional statistic for representing the correlation among the channels; and (3) excitation, namely firstly compressing the input multidimensional statistic through a fully-connected bottleneck link, then expanding the compressed multidimensional statistic into the original size, and limiting the original size through a nonlinear sigmoid function to generate weights when different channels are added.
Further, in step S4, the supervised learning specifically includes:
and adopting an Adam optimization algorithm, taking a cross entropy loss function as an objective function to perform supervised learning, and dynamically adjusting the learning rate in the process of the supervised learning.
According to another aspect of the present invention, a method for determining a dominant instability mode is provided, including:
s1, after a power system fails, collecting bus voltage and power angle data after the failure;
s2, after the dimensions of the bus voltage and the power angle of the generator obtained by sampling are unified, arranging the dimensions into a form of two matrixes to form a group of voltage matrixes and power angle matrixes;
and S3, inputting the voltage matrix and the power angle matrix into the trained dominant instability mode discrimination model to obtain the failed dominant instability mode.
In general, the above technical solutions contemplated by the present invention can achieve the following advantageous effects compared to the prior art.
Because measures such as generator cutting and the like are needed for power angle instability, measures such as load cutting and the like are needed for voltage instability, and the two measures are mutually interwoven, the control measures can have due effects only by rapidly and accurately judging the dominance of the power angle instability and the voltage instability. Therefore, the invention applies deep learning to the dominant instability mode judgment for the first time, constructs a dual-input convolution neural network model containing a SENet module, introduces the SENet module to model information among the neural network channels, adaptively learns and adjusts the weight when each channel is added, amplifies useful characteristics and inhibits the characteristics with little use for the current task, and can quickly and accurately judge whether the system is unstable and belongs to power angle instability or voltage instability according to the voltage after large disturbance, power angle measurement data or simulation data after the model training is finished, thereby providing basis for the selection and coordination of control measures, greatly improving the safety of the power system and having strong practicability.
Drawings
Fig. 1 is a flowchart of a method for constructing a dominant instability mode discrimination model according to an embodiment of the present invention;
fig. 2 is a schematic diagram of an architecture of a dominant instability mode determination network according to an embodiment of the present invention;
FIG. 3 is a diagram of a SENET structure provided by an embodiment of the present invention;
FIG. 4 is a diagram of a neural network architecture for a specific implementation of SENEt according to an embodiment of the present invention;
fig. 5 is a single line diagram of a system of 36 nodes of a chinese electric academy of sciences 8 machine according to an embodiment of the present invention;
fig. 6 is a graph illustrating a loss function variation of a training process according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention. In addition, the technical features involved in the embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.
Referring to fig. 1, an embodiment of the present invention provides a method for constructing a dominant instability mode discrimination model, where the method includes:
s1, respectively setting different tide operation working conditions, load motor component proportions, fault lines, fault positions and fault duration time to perform multiple groups of transient stability simulation;
for example, the load level may be set to fluctuate from 90% to 110% of the standard operating conditions; the motor components in the load account for 50%, 60%, 70%, 80% and 90%; traversing all non-transformer lines in the system by the fault line; the fault positions are 2%, 20%, 50%, 80% and 98% of the total length of each line; the failure duration was set to 0.05s/0.25s, 0.15s/0.3 s.
S2, discretely sampling bus voltage and a power angle of the generator from the measurement data output after the fault occurs, and arranging the bus voltage and the power angle of the generator into two matrix forms after unifying the dimensions of the bus voltage and the power angle obtained by sampling to form a voltage matrix and a power angle matrix corresponding to each group of measurement data;
note that, since the power angle instability mode is strongly correlated with the power angle variable and the voltage instability is strongly correlated with the voltage variable, both of them need to be input. Specifically, a group of bus voltage and generator power angle discrete point sets in the corresponding observation window are extracted from each group of original measurement data or simulation data, a point is sampled every 0.01 second, and the two discrete point sets are arranged into a form of two matrixes. And standardizing all voltage matrixes and all power angle matrixes by adopting a Z-score standardization method to form a standardized voltage matrix and a standardized power angle matrix.
It should be noted that the voltage matrix and the power angle matrix are in the following forms:
in the formula Vi，jAnd deltai，k(i 1, 2, …, T, j 1, 2, …, B, k 1, 2, … M) are the voltage of the jth bus and the power angle value of the kth generator at the ith sampling time, T is the number of sampling points corresponding to the length of the observation window, B is the number of buses in the power system, and M is the number of generators in the power system.
S3, determining the label information of the dominant instability mode of each group of measurement data by combining with expert experience;
the expert experience includes: the power angle is opened firstly, usually the power angle is unstable, and the voltage is broken firstly, usually the voltage is unstable; the generator end fault is often power angle instability; judging from the effectiveness of the control measures, the generator is effectively switched to be power angle instability, and the load is effectively switched to be voltage instability; the instability of the locally isolated part of the system does not affect the stability of the rest main system and the like.
S4, inputting a voltage matrix and a power angle matrix corresponding to each group of measurement data and corresponding labels into a dominant instability mode discrimination network, and training to obtain a dominant instability mode discrimination model through supervised learning;
the dominant instability mode discrimination network proposed by the present invention is shown in fig. 2, and comprises three parts: a two-input convolutional neural network, SENET, and a fully-connected neural network. The squashed and excitation (SEnet) part is shown in FIG. 3, and comprises four steps, which are:
transformation step (transformation): convolution transformation to form unweighted feature maps;
extrusion step (squeeze): gathering information in the same channel to form a multi-dimensional statistic for representing correlation between channels, which is generally implemented by global average pooling, as shown in fig. 4;
excitation step (excitation): firstly, compressing input multidimensional statistics through a fully-connected bottleneck link, then expanding the compressed multidimensional statistics into an original size, and limiting the original size through a nonlinear sigmoid function to generate weights when different channels are added, wherein the weights are shown in fig. 4;
scale step (scale): and multiplying the generated weight by the unweighted feature map, and adding to form a final output.
SENet carries out explicit modeling on the correlation among the channels of the convolutional neural network, self-adaptive learning and adjustment are carried out on the weight when each channel is added to form an output characteristic diagram, useful characteristics are amplified, characteristics with little use for the current task are restrained, and the characteristic extraction effect can be improved.
In the training process, an Adam optimization algorithm is adopted, a cross entropy loss function is used as a target function for supervised learning, and in the supervised learning process, the learning rate is dynamically adjusted.
The embodiment of the invention also provides a method for judging the dominant instability mode, which comprises the following steps:
s1, after a power system fails, collecting bus voltage and power angle data after the failure;
s2, after the dimensions of the bus voltage and the power angle of the generator obtained by sampling are unified, arranging the dimensions into a form of two matrixes to form a group of voltage matrixes and power angle matrixes;
and S3, inputting the voltage matrix and the power angle matrix into the trained dominant instability mode discrimination model to obtain the failed dominant instability mode.
The method is specifically described below by taking a 36-machine node test system of a Chinese academy of electrical sciences as an example, and a single line diagram of the system is shown in fig. 5. In the stage of sample set generation, transient stability simulation is carried out on PSASP software, and voltage data collected by a PMU and a WAMS system are simulated. The operation condition before the fault comprises three load levels of 90%, 100% and 110%, the output level of the generator is correspondingly adjusted according to the load levels, and the voltage of each bus is ensured to be within an allowable range. Three-phase metallic short-circuit faults are set on all 26 non-transformer lines, the short-circuit positions are respectively 2%, 20%, 50%, 80% and 98% of the line positions, and the faults respectively last for 0.05s, 0.15s, 0.25s or 0.3 s. The simulation time duration is set to 20s, and finally 7800 samples are obtained.
And determining the label information of the dominant instability mode of each group of original measurement data according to the power grid topology, the fault information, the curve form and the like corresponding to the transient stability sample corresponding to each group of original measurement data and by combining with expert experience.
The obtained marked samples are randomly divided into 70%, 15% and 15% which are respectively a training set, a verification set and a test set.
In the off-line training stage, Z-score standardization is carried out on training set data according to step S2, then a dominant instability mode discrimination network is input for training, and a verification set and a test set are standardized according to the same parameters. During training, an Adam optimization algorithm is adopted, a cross entropy loss function is used as a target function to perform supervised learning, and if the loss of a verification set is not reduced within 50 epochs in the training process, the learning rate is reduced (divided by the loss of the verification set
) The training process loss function changes as shown in fig. 6.
The trained model is applied to an online or offline scene, and the model can give a judgment result of a stable or dominant instability mode according to actually measured or simulated voltage and power angle data. And dispatching personnel or calculating personnel can take corresponding control measures according to the judgment result to ensure the safety and stability of the system.
In order to verify the effect of the model, the accuracy of the dominant instability mode discrimination model based on the SENet-containing convolutional neural network disclosed by the invention is up to 97.4% through testing by using a test sample simulation model application scene, and the testing accuracy of the corresponding dominant instability mode discrimination model (keeping other conditions unchanged) of the convolutional neural network without the SENet structure is only 96.9%. Therefore, the SEnet is used, the capability of convolutional network feature learning is improved, and the effect of distinguishing the dominant instability mode is improved.
Other machine learning models, including decision trees, discriminant analysis, support vector machines, nearest neighbor classifiers, etc., were trained using the same normalized input data, and the test results are listed in table 1. The test result shows that the method has more obvious advantages in evaluation accuracy compared with other machine learning models based on the dominant instability mode discrimination model containing the SENET convolutional neural network.
TABLE 1
It will be understood by those skilled in the art that the foregoing is only a preferred embodiment of the present invention, and is not intended to limit the invention, and that any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the scope of the present invention.
Claims (7)
1. A method for constructing a dominant instability mode discrimination model is characterized by comprising the following steps:
s1, respectively setting different tide operation working conditions, load motor component proportions, fault lines, fault positions and fault duration time to perform multiple groups of transient stability simulation;
s2, discretely sampling bus voltage and a power angle of the generator from the measurement data output after the fault occurs, and arranging the bus voltage and the power angle of the generator into two matrix forms after unifying the dimensions of the bus voltage and the power angle obtained by sampling to form a voltage matrix and a power angle matrix corresponding to each group of measurement data;
s3, determining the label information of the dominant instability mode of each group of measurement data by combining with expert experience;
s4, inputting a voltage matrix and a power angle matrix corresponding to each group of measurement data and corresponding labels into a dominant instability mode discrimination network, and training to obtain a dominant instability mode discrimination model through supervised learning;
the dominant instability mode discrimination network comprises a double-input convolutional neural network, a SENet network and a full-connection neural network;
the double-input convolutional neural network is used for respectively and automatically extracting required characteristics from the input voltage matrix and the input power angle matrix to obtain a characteristic vector corresponding to each sample;
the SENet network is used for explicitly modeling the correlation relation between the channels of the double-input convolutional neural network, adaptively learning and adjusting the weight value when each channel is added to form an output characteristic diagram;
and the fully-connected neural network is used for receiving the feature vector corresponding to each sample and classifying the sample through nonlinear transformation and a softmax function.
2. The method as claimed in claim 1, wherein in step S2, the dimensions of the bus voltage and the generator power angle are unified, specifically, Z-score normalization is adopted to normalize the bus voltage and the generator power angle.
3. The method according to claim 1, wherein the dominant instability mode signature information includes a stable mode, a power angle instability, and a voltage instability.
4. The method for constructing the dominant instability mode discriminant model according to claim 3, wherein the expert experience specifically includes that a power angle is firstly turned on to be power angle instability, and a voltage is firstly broken down to be voltage instability; the generator end fault is power angle instability; judging from the effectiveness of the control measures, the generator is effectively switched to be power angle instability, and the load is effectively switched to be voltage instability; the instability of the local isolated part of the system does not affect the stability of the rest main systems, and the system is in a stable mode.
5. The method for constructing the discriminant model of dominant instability mode according to claim 1, wherein the specific implementation process of the SENET network includes extrusion and excitation; the method comprises the following steps of extruding, gathering information in each input channel through global average pooling to form a multi-dimensional statistic for representing the correlation among the channels; and (3) excitation, namely firstly compressing the input multidimensional statistic through a fully-connected bottleneck link, then expanding the compressed multidimensional statistic into the original size, and limiting the original size through a nonlinear sigmoid function to generate weights when different channels are added.
6. The method for constructing a dominant instability mode discriminant model according to claim 1, wherein the supervised learning in step S4 specifically includes:
and adopting an Adam optimization algorithm, taking a cross entropy loss function as an objective function to perform supervised learning, and dynamically adjusting the learning rate in the process of the supervised learning.
7. A method for discriminating a dominant instability mode is characterized by comprising the following steps:
s1, after a power system fails, collecting bus voltage and power angle data after the failure;
s2, after the dimensions of the bus voltage and the power angle of the generator obtained by sampling are unified, arranging the dimensions into a form of two matrixes to form a group of voltage matrixes and power angle matrixes;
s3, inputting the voltage matrix and the power angle matrix into the dominant instability mode discrimination model trained according to any one of claims 1-6 to obtain the dominant instability mode after the fault.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011067784.1A CN112215722B (en) | 2020-10-07 | 2020-10-07 | Dominant instability mode discrimination model construction method and dominant instability mode discrimination method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011067784.1A CN112215722B (en) | 2020-10-07 | 2020-10-07 | Dominant instability mode discrimination model construction method and dominant instability mode discrimination method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112215722A true CN112215722A (en) | 2021-01-12 |
| CN112215722B CN112215722B (en) | 2022-06-14 |
Family
ID=74053472
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011067784.1A Active CN112215722B (en) | 2020-10-07 | 2020-10-07 | Dominant instability mode discrimination model construction method and dominant instability mode discrimination method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112215722B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113285452A (en) * | 2021-05-31 | 2021-08-20 | 四川大学 | Method for prejudging transient instability of power system and generating generator tripping control strategy |
| CN113435492A (en) * | 2021-06-21 | 2021-09-24 | 华中科技大学 | Power system dominant instability mode discrimination method based on active learning |
| CN114034957A (en) * | 2021-11-12 | 2022-02-11 | 广东电网有限责任公司江门供电局 | Transformer vibration abnormity detection method based on working condition division |
| CN114091705A (en) * | 2021-11-26 | 2022-02-25 | 国网四川省电力公司电力科学研究院 | Power system instability analysis method and device, electronic equipment and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103632043A (en) * | 2013-11-19 | 2014-03-12 | 国家电网公司 | Dominant power system instability mode recognition method based on real-time measurement response information |
| CN105224721A (en) * | 2015-09-08 | 2016-01-06 | 中国电力科学研究院 | A kind of automatic identifying method of electromechanical transient Failure Model |
| CN106651096A (en) * | 2016-09-30 | 2017-05-10 | 华中电网有限公司 | Fuzzy comprehensive evaluation method for transient power angle instability and voltage instability |
| CN109428327A (en) * | 2017-09-01 | 2019-03-05 | 中国电力科学研究院 | Power grid key branch and leading stable mode recognition methods and system based on response |
| CN111539459A (en) * | 2020-04-03 | 2020-08-14 | 中国电力科学研究院有限公司 | Method and system for identifying instability type of power system |
-
2020
- 2020-10-07 CN CN202011067784.1A patent/CN112215722B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103632043A (en) * | 2013-11-19 | 2014-03-12 | 国家电网公司 | Dominant power system instability mode recognition method based on real-time measurement response information |
| CN105224721A (en) * | 2015-09-08 | 2016-01-06 | 中国电力科学研究院 | A kind of automatic identifying method of electromechanical transient Failure Model |
| CN106651096A (en) * | 2016-09-30 | 2017-05-10 | 华中电网有限公司 | Fuzzy comprehensive evaluation method for transient power angle instability and voltage instability |
| CN109428327A (en) * | 2017-09-01 | 2019-03-05 | 中国电力科学研究院 | Power grid key branch and leading stable mode recognition methods and system based on response |
| CN111539459A (en) * | 2020-04-03 | 2020-08-14 | 中国电力科学研究院有限公司 | Method and system for identifying instability type of power system |
Non-Patent Citations (2)
| Title |
|---|
| F. SENESOULIN: ""Estimation of Dominant Power Oscillation Mode using LSTM-RNN based on Synchrophasor Data"", 《19TH IEEE INTERNATIONAL CONFERENCE ON ENVIRONMENT AND ELECTRICAL ENGINEERING (EEEIC)》, 31 December 2019 (2019-12-31), pages 1 - 6 * |
| ZHONGTUO SHI: ""Convolutional neural network-based power system transient stability assessment and instability mode prediction"", 《APPLIED ENERGY》, 14 February 2020 (2020-02-14), pages 1 - 13 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113285452A (en) * | 2021-05-31 | 2021-08-20 | 四川大学 | Method for prejudging transient instability of power system and generating generator tripping control strategy |
| CN113285452B (en) * | 2021-05-31 | 2023-03-10 | 四川大学 | Method for prejudging transient instability of power system and generating generator tripping control strategy |
| CN113435492A (en) * | 2021-06-21 | 2021-09-24 | 华中科技大学 | Power system dominant instability mode discrimination method based on active learning |
| CN113435492B (en) * | 2021-06-21 | 2022-04-22 | 华中科技大学 | Power system dominant instability mode discrimination method based on active learning |
| CN114034957A (en) * | 2021-11-12 | 2022-02-11 | 广东电网有限责任公司江门供电局 | Transformer vibration abnormity detection method based on working condition division |
| CN114034957B (en) * | 2021-11-12 | 2023-10-03 | 广东电网有限责任公司江门供电局 | Transformer vibration anomaly detection method based on working condition division |
| CN114091705A (en) * | 2021-11-26 | 2022-02-25 | 国网四川省电力公司电力科学研究院 | Power system instability analysis method and device, electronic equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112215722B (en) | 2022-06-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112215722B (en) | Dominant instability mode discrimination model construction method and dominant instability mode discrimination method | |
| CN107340456B (en) | Power distribution network operating condition intelligent identification Method based on multiple features analysis | |
| CN106777984B (en) | A method of photovoltaic array Working state analysis and fault diagnosis are realized based on density clustering algorithm | |
| CN102074955B (en) | Method based on knowledge discovery technology for stability assessment and control of electric system | |
| CN100536411C (en) | An improved adaptive boosting algorithm based Internet intrusion detection method | |
| CN110009208B (en) | Method and device for evaluating health state of pole-mounted switch complete equipment based on hybrid intelligent algorithm | |
| CN111401749A (en) | Dynamic safety assessment method based on random forest and extreme learning regression | |
| CN112200694B (en) | Dominant instability mode identification model construction and application method based on graph neural network | |
| CN107633349A (en) | Fault impact factor quantitative analysis method based on high-voltage switch gear | |
| CN111652479A (en) | Data driving method for dynamic security assessment of power system | |
| CN110866366A (en) | XGboost algorithm-based island detection method for photovoltaic microgrid containing PHEV | |
| CN111273125A (en) | RST-CNN-based power cable channel fault diagnosis method | |
| CN112821424A (en) | Power system frequency response analysis method based on data-model fusion drive | |
| CN111062569A (en) | Low-current fault discrimination method based on BP neural network | |
| CN105954616B (en) | Photovoltaic module method for diagnosing faults based on external characteristics electric parameter | |
| CN113177357B (en) | Transient stability assessment method for power system | |
| CN111585277B (en) | Power system dynamic security assessment method based on hybrid integration model | |
| CN111244937B (en) | Method for screening serious faults of transient voltage stability of power system | |
| CN109684749B (en) | Photovoltaic power station equivalent modeling method considering operating characteristics | |
| CN114169395A (en) | Construction method and application of dominant instability mode recognition model of power system | |
| CN108988319B (en) | Rapid emergency control method based on deep feedforward neural network and numerical integration sensitivity | |
| CN111814394A (en) | Power system safety assessment method based on correlation and redundancy detection | |
| CN111401481A (en) | Power grid operation mode automatic identification method based on multi-source information fusion technology | |
| Ekka et al. | Fault identification using fuzzy in renewable energy interfaced IEEE 13 bus system | |
| Thapa et al. | A machine learning-based approach to detection of fault locations in transmission networks |
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 |