CN115374957B - Radial flow type small hydropower station multiscale missing measurement data reconstruction method - Google Patents
Radial flow type small hydropower station multiscale missing measurement data reconstruction method Download PDFInfo
- Publication number
- CN115374957B CN115374957B CN202210714048.3A CN202210714048A CN115374957B CN 115374957 B CN115374957 B CN 115374957B CN 202210714048 A CN202210714048 A CN 202210714048A CN 115374957 B CN115374957 B CN 115374957B
- Authority
- CN
- China
- Prior art keywords
- data
- measurement data
- missing
- neural network
- hydropower station
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000005259 measurement Methods 0.000 title claims abstract description 121
- 238000000034 method Methods 0.000 title claims abstract description 32
- 238000013528 artificial neural network Methods 0.000 claims abstract description 47
- 238000012549 training Methods 0.000 claims abstract description 29
- 239000013598 vector Substances 0.000 claims abstract description 9
- 239000011159 matrix material Substances 0.000 claims description 16
- 230000005856 abnormality Effects 0.000 claims description 10
- 238000000342 Monte Carlo simulation Methods 0.000 claims description 8
- 238000011478 gradient descent method Methods 0.000 claims description 7
- 238000002347 injection Methods 0.000 claims description 6
- 239000007924 injection Substances 0.000 claims description 6
- 238000012544 monitoring process Methods 0.000 claims description 5
- 238000007781 pre-processing Methods 0.000 claims description 5
- 238000010276 construction Methods 0.000 claims description 4
- 238000013480 data collection Methods 0.000 claims description 2
- 238000013507 mapping Methods 0.000 abstract description 2
- 230000006870 function Effects 0.000 description 5
- 230000002159 abnormal effect Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000005611 electricity Effects 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 230000007257 malfunction Effects 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 238000010248 power generation Methods 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000010998 test method Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
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
- G06Q10/00—Administration; Management
- G06Q10/20—Administration of product repair or maintenance
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/21—Design, administration or maintenance of databases
- G06F16/215—Improving data quality; Data cleansing, e.g. de-duplication, removing invalid entries or correcting typographical errors
-
- 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—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Economics (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- Health & Medical Sciences (AREA)
- Marketing (AREA)
- Databases & Information Systems (AREA)
- Quality & Reliability (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Operations Research (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Supply And Distribution Of Alternating Current (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a method for reconstructing radial flow type small hydropower station multiscale missing measurement data, which comprises the steps of constructing radial flow type small hydropower station operation data with missing at the current moment of an electric power system into vectors and inputting the vectors into a trained deep neural network to obtain complete operation data after reconstruction; the run-off data of the radial small hydropower station missing in the vector is set to be a fixed value far away from the rated value; the deep neural network is obtained by training historical operation data of the power system based on the distributed small hydropower stations. The invention uses the deep neural network to approximate the complex nonlinear mapping relation between the missing small hydropower station measurement data and other measurement data of the power system, thereby accurately and efficiently reconstructing the missing power system data. The invention considers the radial flow type small hydropower station multi-scale modeling and has important significance for guaranteeing the safe and stable operation of the power system.
Description
Technical Field
The invention relates to the technical field of power systems, in particular to a reconstruction method of missing measurement data of a power system containing radial-flow small hydropower stations.
Background
The rapid development of water and electricity resources, especially small water and electricity in China makes up the defect of the conventional thermal power resources, and plays an important role in promoting the rapid development of economy. However, the current built radial-flow medium-and-small hydropower stations generally have the idea of rebuilding light pipes, and cannot be effectively monitored and scheduled. The cascade power stations do not share information such as power generation, water quantity and the like, and the cooperative degree is low. This causes a series of unusual problems such as radial-flow type small hydropower station data loss, data delay and the like to often appear in the power system, and further influences the normal operation of the intelligent dispatching center. In order to ensure safe and stable operation of the power system, the method is important to multi-scale modeling of the radial-flow small hydropower station of the power system.
The intelligent scheduling established on the power data is one of important technical means for power guarantee. The intelligent power dispatching center is used for judging the safe operation state of the power grid according to the data fed back by the information acquisition equipment and the actual operation parameters of the power grid, adjusting the output and load distribution and enabling the power system to continuously and stably operate.
In order to solve the problem of measuring missing values of the power system, such as the Chinese patent document CN109144987A, a deep learning-based power system missing value reconstruction method and application thereof are disclosed, wherein the reconstruction data is used as pseudo measurement by training through improving the WGAN after GAN. However, because the scheme does not consider the factor of poor data cooperativity among the cascade power stations, the power data is likely to be interfered or failed in the links of acquisition, measurement, transmission, conversion and the like, and the power system has serial abnormal problems of data loss, data delay and the like, thereby affecting the safe and stable operation of intelligent power dispatching and being not suitable for the application scene of the flow-through small hydropower station.
Disclosure of Invention
In view of the problem that the fault tolerance performance is poor due to the fact that factors of poor data cooperativity among the cascade power stations are not considered in the prior art, the invention aims to provide the radial flow type small hydropower station multiscale missing measurement data reconstruction method, and the complex time-space correlation among measurement data is learned from historical data by adopting a deep neural network, so that missing measurement data of a power system is accurately and effectively reconstructed, the problem of data abnormality in power scheduling is solved, and safe and stable operation of the power system is ensured.
The technical scheme adopted by the invention is as follows: a radial flow type small hydropower station multiscale missing measurement data reconstruction method comprises the following steps:
s1: performing characteristic engineering on the historical measurement data based on a historical measurement database of the radial-flow small hydropower system acquired by the power data acquisition monitoring system and the phasor measurement device;
s2: in an on-line training link, adopting a Monte Carlo method to randomly discard partial runoff type small hydropower station measurement data under a certain time section, and simulating measurement data loss caused by data loss, data delay and data abnormality in a power system to form measurement data containing the runoff type small hydropower station loss;
s3: constructing a cascade deep neural network, taking a system data vector containing the radial flow type small hydropower station missing measurement data as an input, and outputting the reconstructed measurement data;
s4: training the deep neural network reconstruction data on the simulated missing measurement data by using a gradient descent method;
s5: and (3) applying the deep neural network after training in the step (S4) to an online reconstruction link, reconstructing missing measurement data of the radial flow type small hydropower system in real time, and reconstructing the power measurement system.
The electric power measurement system comprises an electric power data acquisition monitoring system, a phasor measurement device, a measurement database and a deep neural network, and solves the problem of missing measurement data caused by data loss, data delay and data abnormality in the electric power measurement system through the training of the deep neural network, solves the problem of data abnormality in electric power scheduling, and ensures the safe and stable operation of the electric power system.
Further, the step S1 of performing feature engineering on the historical measurement data includes:
selecting m groups of complete historical measurement data in a historical measurement database to construct a data setData d of the m th group m The method comprises the following steps:
wherein V is i m ,P i m ,/>Respectively representing the voltage amplitude, the voltage phase angle, the node injection active power and the node injection reactive power at the node i in the m-th set of historical measurement data; n is the number of observable nodes of the system.
Further, the off-line training step of step S2 includes the specific steps of:
the mth set of historical measurement data d at the time of the time section t m 1-k metrology feature data are randomly discarded using the Monte Carlo method, and all discarded metrology data is marked as a particular value, e t Simulating measurement data missing phenomena caused by data loss, data delay and data abnormality of the power system, marking the processed measurement data set as D ', and marking the D'As an input to the subsequent deep neural network, the historical measurement dataset D is taken as a target value for the output of the subsequent deep neural network.
Further, the construction of the cascaded deep neural network in step S3 includes:
the input data of the deep neural network is the measurement data set D' processed in the step S2, the output data is the history measurement data set D, and the input-output relationship of the first hidden layer is:
wherein,and Y (l) FC Input and output of the first FC layer respectively; />And b (l) The weight matrix and the bias matrix of the first FC layer are trainable parameters respectively.
Further, training the deep neural network in step S4 includes:
and (3) training the trainable parameters of the deep neural network by using the input data and the output data obtained in the step S2 and using a gradient descent method by taking the mean square error as a loss function so as to minimize the mean square error.
Further, before step S4, the method further includes a missing measurement data preprocessing step:
using the particular value e determined in step S2 t As a substitute for missing measurement data, preprocessing the real-time measurement data matrix d of the power system t 。
Further, the online reconstruction step in step S5 includes the specific steps of:
the preprocessed measurement data matrix d t And (4) inputting the data into the deep neural network trained in the step (S4) to reconstruct and output the complete measurement data of the power system.
Compared with the prior art, the method for reconstructing the multi-scale missing measurement data of the radial-flow small hydropower station solves the training difficulty of electric power data through deep neural network training, and can effectively improve the fault tolerance of the measurement data in an electric power operation system, so that the problem of missing measurement data caused by data loss, data delay and data abnormality in the electric power system is solved, the problem of safe and stable operation of a power grid in intelligent scheduling is effectively solved, and the guarantee capability of the power grid is improved.
Drawings
FIG. 1 is a flow chart of a reconstruction method according to the present invention.
Fig. 2 is a diagram of a deep neural network according to the present invention.
Fig. 3 is a diagram illustrating an IEEE30 node system architecture according to an exemplary embodiment of the present invention.
Fig. 4 is a graph showing the change of the mean square error index in the training process of the present invention.
Fig. 5 is measurement data to be reconstructed of the power system at a certain moment in time according to the present invention.
Fig. 6 shows measurement data after the reconstruction of the power system at a certain moment.
Detailed Description
The electric power system is related to national life, and the reconstruction of missing measurement data of the electric power system is important to guaranteeing safe and stable operation of the electric power system. The accuracy and real-time of the data collected by the power data collection and monitoring system (supervisory control and data acquisition, SCADA) and the phasor measurement device (Phasor Measurement Unit, PMU) has important effects on power system state estimation, stability analysis and operation optimization. The power data may be interfered or malfunction in the links of collection, measurement, transmission, conversion, etc., which may cause a series of abnormal problems of data loss, data delay, etc. in the power system.
Example 1:
the embodiment 1 provides a method for reconstructing multi-scale missing measurement data of a radial-flow small hydropower station, as shown in fig. 1, comprising the following steps:
step S1: and carrying out characteristic engineering on the historical measurement data based on a historical measurement database of the radial-flow small hydropower system acquired by the electric power data acquisition monitoring system and the phasor measurement device.
The feature engineering includes a feature acquisition step of starting from initial measurement data of a power system history measurement database, and establishing a model for providing power characterization digitization, resulting in better interpretability, thereby facilitating subsequent machine learning steps.
Comprising the following steps:
the feature engineering steps include: selecting m groups of complete historical measurement data in a historical measurement database to construct a data setData d of the m th group m The method comprises the following steps:
wherein V is i m ,P i m ,/>Respectively representing the voltage amplitude, the voltage phase angle, the node injection active power and the node injection reactive power at the node i in the m-th set of historical measurement data; n is the number of observable nodes of the system.
Step S2: in the on-line training link, part of measurement data under a certain time section is randomly discarded by adopting a Monte Carlo method, missing measurement data caused by data loss, data delay and data abnormality in a power system is simulated, and simulated missing measurement data is formed.
The Monte Carlo method is to calculate the integral by using a random test method, and obtain the arithmetic average value of the random variable by the test as the estimated value of the integral. When the test sample of the monte carlo method is large enough, the mean value approaches the expectations.
In the off-line training link, the specific steps include:
the mth set of historical measurement data d at the time of the time section t m 1-k metrology feature data are randomly discarded using the Monte Carlo method, and all discarded metrology data is marked as a particular value, e t Simulating measurement data missing phenomena caused by data loss, data delay and data abnormality of the power system, marking the processed measurement data set as D ', taking the D' as the input of the subsequent deep neural network, and taking the historical measurement data set D as the target value of the output of the subsequent deep neural network.
Step S3: and constructing a cascade deep neural network, taking the simulated system data vector of the missing measurement data as input, and outputting the reconstructed measurement data.
The cascade deep neural network structure is shown in fig. 2. The neural network structure is provided with a hidden layer with neuron depth, and the input layer and the output layer are both at one level. The network structure has high learning speed, and the original structure can be maintained after the training set is changed.
As shown in fig. 3, the present embodiment 1 adopts the IEEE30 node system to perform simulation verification of validity. The total active power output of the system generator is 191.6MW under a certain section in FIG. 3, and the total load is 189.2MW. A phase measuring device PMU is additionally arranged on each node in the 30-node system, and the electrical data of the node is measured at the frequency of 50 Hz. To simulate abnormal problems such as data loss and data delay of the power system, 1-20 pieces of sampling data are randomly set to 0 to represent abnormal values when all the time is processed.
The construction of the cascade deep neural network comprises the following steps: the input data of the deep neural network is the measurement data set D' processed in the step S2, the output data is the history measurement data set D, and the input-output relationship of the first hidden layer is:
wherein,and Y (l) FC Input and output of the first FC layer respectively; />And b (l) The weight matrix and the bias matrix of the first FC layer are trainable parameters respectively.
Step S4: the deep neural network reconstruction data is trained using gradient descent on the simulated missing measurement data.
The method comprises the following steps of: using the particular value e determined in step S2 t As a substitute for missing measurement data, preprocessing the real-time measurement data matrix d of the power system t 。
The method comprises the steps of training trainable parameters of the deep neural network by using the input data and the output data obtained in the step S2 and using a gradient descent method to minimize the mean square error as a loss function.
The gradient descent method is a first order optimization algorithm that finds the objective function minimization, i.e., the loss function minimization.
Step S5: and (3) applying the deep neural network after training in the step (S4) to an online reconstruction link, reconstructing missing measurement data of the power system in real time, and reconstructing the power measurement system.
The online reconstruction link comprises: the preprocessed measurement data matrix d t And (4) inputting the data into the deep neural network trained in the step (S4) to reconstruct and output the complete measurement data of the power system.
Example 2:
on the basis of embodiment 1, a method for reconstructing radial-flow small hydropower multi-scale missing measurement data is provided, in step S4, the depth neural network reconstruction data is trained on the simulated missing measurement data by using a gradient descent method, and the processed 10 6 And taking the system measurement data as training data, training the deep neural network by using the method, and obtaining a curve of the loss function changing along with the iteration number in the training process, such as a mean square error index change diagram in the training process shown in fig. 4. In FIG. 4The trend with the number of iterations can be seen.
In step S5: and (3) applying the deep neural network after training in the step (S4) to an online reconstruction link, reconstructing missing measurement data of the power system in real time, and reconstructing the power measurement system. At a certain time t, the system loses data due to communication failure, the lost data is represented by symbols, and at the moment, the characteristic matrix of the power system is shown in fig. 5 and is measurement data to be rebuilt of the power system.
The feature matrix is input into a trained deep neural network to obtain a reconstructed power system feature parameter matrix, and the reconstructed power system feature parameter matrix is measurement data of the power system, as shown in fig. 6.
The invention constructs the radial flow type small hydropower station operation data with the loss at the current moment of the power system into vectors and inputs the vectors into a trained deep neural network to obtain complete operation data after reconstruction; the run-off data of the radial small hydropower station missing in the vector is set to be a fixed value far away from the rated value; the deep neural network is obtained by training historical operation data of the power system based on the distributed small hydropower stations. The deep neural network is utilized to learn the complex space-time correlation between the measured data from the historical data, and the deep neural network is utilized to approximate the complex nonlinear mapping relation between the missing measured data and the complete measured data, so that the method has important significance for guaranteeing the safe and stable operation of the power system.
Claims (6)
1. The method for reconstructing the radial-flow type small hydropower multi-scale missing measurement data is characterized by comprising the following steps of:
s1: based on a runoff type small hydropower station electric power system history measurement database collected by an electric power data collection monitoring system and a phasor measurement device, carrying out characteristic engineering on the history measurement data, wherein: selecting a plurality of groups of complete history measurement data in a history measurement database to construct a data set;
s2: in an on-line training link, adopting a Monte Carlo method to randomly discard partial runoff type small hydropower station measurement data under a certain time section, and simulating measurement data loss caused by data loss, data delay and data abnormality in a power system to form measurement data containing the runoff type small hydropower station loss;
s3: a cascade deep neural network is constructed, a system data vector containing radial flow type small hydropower station missing measurement data is taken as input, the reconstructed measurement data is output, and the construction of the cascade deep neural network comprises the following steps: the input data of the deep neural network is the measurement data set D' processed in the step S2, the output data is the history measurement data set D, and the input-output relationship of the first hidden layer is:wherein (1)>And->Input and output of the first FC layer respectively; />And b (l) Respectively a weight matrix and a bias matrix of the first FC layer, which are trainable parameters;
s4: training the deep neural network reconstruction data on the simulated missing measurement data by using a gradient descent method;
s5: applying the deep neural network after training in the step S4 to an online reconstruction link, reconstructing missing measurement data of a radial flow type small hydropower station electric power system in real time, and reconstructing the electric power measurement system, wherein: and the online reconstruction link is used for reconstructing complete measurement data of the output power system.
2. The method for reconstructing multiscale missing measurement data of a radial flow type small hydropower station according to claim 1, wherein the method comprises the following steps: the step S1 of performing feature engineering on the historical measurement data comprises the following steps:
selecting m groups of complete historical measurement data construction in a historical measurement databaseData setData d of the m th group m The method comprises the following steps:
wherein,respectively representing the voltage amplitude, the voltage phase angle, the node injection active power and the node injection reactive power at the node i in the m-th set of historical measurement data; n is the number of observable nodes of the system.
3. The method for reconstructing multiscale missing measurement data of a radial flow type small hydropower station according to claim 1, wherein the method comprises the following steps: the off-line training link in step S2 specifically includes:
the mth set of historical measurement data d at the time of the time section t m 1-k metrology feature data are randomly discarded using the Monte Carlo method, and all discarded metrology data is marked as a particular value, e t Simulating measurement data missing phenomena caused by data loss, data delay and data abnormality of the power system, marking the processed measurement data set as D ', taking the D' as the input of the subsequent deep neural network, and taking the historical measurement data set D as the target value of the output of the subsequent deep neural network.
4. A method for reconstructing multi-scale missing measurement data of a small radial flow hydropower station according to claim 3, wherein training the deep neural network in step S4 comprises:
and (3) training the trainable parameters of the deep neural network by using the input data and the output data obtained in the step S2 and using a gradient descent method by taking the mean square error as a loss function so as to minimize the mean square error.
5. A method for reconstructing multiscale missing measurement data of a small radial flow hydropower station according to claim 3, further comprising the step of preprocessing missing measurement data before step S4:
using the particular value e determined in step S2 t As a substitute for missing measurement data, preprocessing the real-time measurement data matrix d of the power system t 。
6. The method for reconstructing multiscale missing measurement data of a radial flow type small hydropower station according to claim 5, wherein the method comprises the following steps: the online reconstruction step of step S5 includes the specific steps of:
the preprocessed measurement data matrix d t And (4) inputting the data into the deep neural network trained in the step (S4) to reconstruct and output the complete measurement data of the power system.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210714048.3A CN115374957B (en) | 2022-06-22 | 2022-06-22 | Radial flow type small hydropower station multiscale missing measurement data reconstruction method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210714048.3A CN115374957B (en) | 2022-06-22 | 2022-06-22 | Radial flow type small hydropower station multiscale missing measurement data reconstruction method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115374957A CN115374957A (en) | 2022-11-22 |
| CN115374957B true CN115374957B (en) | 2024-02-27 |
Family
ID=84061701
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210714048.3A Active CN115374957B (en) | 2022-06-22 | 2022-06-22 | Radial flow type small hydropower station multiscale missing measurement data reconstruction method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115374957B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109102110A (en) * | 2018-07-23 | 2018-12-28 | 云南电网有限责任公司临沧供电局 | A kind of radial-flow type small power station goes out force prediction method and device in short term |
| CN109144987A (en) * | 2018-08-03 | 2019-01-04 | 天津相和电气科技有限公司 | Electric system based on deep learning measures missing values method for reconstructing and its application |
| CN111401553A (en) * | 2020-03-12 | 2020-07-10 | 南京航空航天大学 | Missing data filling method and system based on neural network |
| CN113052469A (en) * | 2021-03-30 | 2021-06-29 | 贵州电网有限责任公司 | Method for calculating wind-solar-water-load complementary characteristic of small hydropower area lacking measurement runoff |
| CN114611590A (en) * | 2022-03-01 | 2022-06-10 | 浙江大学 | Graph neural network-based power system missing data reconstruction method and system |
-
2022
- 2022-06-22 CN CN202210714048.3A patent/CN115374957B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109102110A (en) * | 2018-07-23 | 2018-12-28 | 云南电网有限责任公司临沧供电局 | A kind of radial-flow type small power station goes out force prediction method and device in short term |
| CN109144987A (en) * | 2018-08-03 | 2019-01-04 | 天津相和电气科技有限公司 | Electric system based on deep learning measures missing values method for reconstructing and its application |
| CN111401553A (en) * | 2020-03-12 | 2020-07-10 | 南京航空航天大学 | Missing data filling method and system based on neural network |
| CN113052469A (en) * | 2021-03-30 | 2021-06-29 | 贵州电网有限责任公司 | Method for calculating wind-solar-water-load complementary characteristic of small hydropower area lacking measurement runoff |
| CN114611590A (en) * | 2022-03-01 | 2022-06-10 | 浙江大学 | Graph neural network-based power system missing data reconstruction method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115374957A (en) | 2022-11-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112348292B (en) | Short-term wind power prediction method and system based on deep learning network | |
| Kamruzzaman et al. | A convolutional neural network-based approach to composite power system reliability evaluation | |
| CN114429153B (en) | Gear box increment fault diagnosis method and system based on life learning | |
| CN110570122A (en) | Offshore wind power plant reliability assessment method considering wind speed seasonal characteristics and current collection system element faults | |
| Tao et al. | Abnormal detection of wind turbine based on SCADA data mining | |
| CN115693661A (en) | Cascading failure risk key line identification method based on graph neural network | |
| CN113937764A (en) | Low-voltage distribution network high-frequency measurement data processing and topology identification method | |
| CN111654392A (en) | Low-voltage distribution network topology identification method and system based on mutual information | |
| Kaplan et al. | Fault diagnosis of smart grids based on deep learning approach | |
| CN110389268B (en) | Online monitoring system of electric power system | |
| CN115640748A (en) | Dynamic frequency response prediction method for generators after disturbance of power system | |
| Gong et al. | Transient stability assessment of electric power system based on voltage phasor and cnn-lstm | |
| CN115374957B (en) | Radial flow type small hydropower station multiscale missing measurement data reconstruction method | |
| Sun et al. | Reliability prediction of distribution network based on PCA-GA-BP neural network | |
| CN114611590A (en) | Graph neural network-based power system missing data reconstruction method and system | |
| CN115510961A (en) | Community comprehensive energy system operation safety assessment method based on active learning | |
| CN114710116A (en) | Actual measurement modeling method and system of photovoltaic cell assembly based on fuzzy model | |
| CN114707789A (en) | Low-voltage distribution network high-frequency measurement data processing and topology identification method and system | |
| Wang et al. | Ensemble Learning Model of Power System Transient Stability Assessment Based on Bayesian Model Averaging Method | |
| CN112651628A (en) | Power system transient stability evaluation method based on capsule neural network | |
| Li et al. | Extraction of Abnormal Points from On-line Operation Data of Intelligent Meter Based on LSTM | |
| Qin et al. | Generative Adversarial Networks-based Distribution Network Operation Scenario Generation | |
| Diao et al. | Deep Integration and Learning of Measurement Data in Active Distribution Power Networks for Load Flow Estimation Considering Incomplete Information | |
| Cai et al. | Study on Wind Power Prediction Based on EEMD-LSTM | |
| Dwivedi et al. | Temporal Patterns and Seasonal Effects in Wind Power Forecasting: A Time Series Analysis |
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 |