CN113449914A - Power system monitoring method and system - Google Patents
Power system monitoring method and system Download PDFInfo
- Publication number
- CN113449914A CN113449914A CN202110711578.8A CN202110711578A CN113449914A CN 113449914 A CN113449914 A CN 113449914A CN 202110711578 A CN202110711578 A CN 202110711578A CN 113449914 A CN113449914 A CN 113449914A
- Authority
- CN
- China
- Prior art keywords
- power system
- operating state
- probability
- classification model
- output
- 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
- 238000000034 method Methods 0.000 title claims abstract description 41
- 238000012544 monitoring process Methods 0.000 title claims abstract description 23
- 239000013598 vector Substances 0.000 claims abstract description 53
- 238000013145 classification model Methods 0.000 claims abstract description 38
- 238000012549 training Methods 0.000 claims abstract description 33
- 230000006870 function Effects 0.000 claims abstract description 22
- 230000004913 activation Effects 0.000 claims abstract description 16
- 230000008569 process Effects 0.000 claims abstract description 14
- 238000007781 pre-processing Methods 0.000 claims abstract description 11
- 238000002759 z-score normalization Methods 0.000 claims 1
- 230000008034 disappearance Effects 0.000 abstract description 6
- 238000004880 explosion Methods 0.000 abstract description 5
- 238000013528 artificial neural network Methods 0.000 description 6
- 230000010365 information processing Effects 0.000 description 6
- 238000007689 inspection Methods 0.000 description 4
- 238000005070 sampling Methods 0.000 description 4
- 230000002159 abnormal effect Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 230000010354 integration Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 238000011425 standardization method Methods 0.000 description 2
- 238000012546 transfer Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 239000002360 explosive Substances 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 230000000644 propagated effect Effects 0.000 description 1
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
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- 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
-
- 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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Marketing (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- Entrepreneurship & Innovation (AREA)
- Game Theory and Decision Science (AREA)
- General Health & Medical Sciences (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Development Economics (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Educational Administration (AREA)
- Life Sciences & Earth Sciences (AREA)
- Primary Health Care (AREA)
- Water Supply & Treatment (AREA)
- Public Health (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
本发明公开了一种电力系统监测方法及系统，通过采集电力系统各节点一段观测时间段内的观测数据；对观测数据进行预处理，得到特征向量；将特征向量输入训练好的运行状态分类模型，得到电力系统各节点当前时刻属于不同运行状态的概率，选取最高概率所对应的运行状态作为各节点当前时刻的运行状态；其中，运行状态分类模型在训练过程中，基于总误差更新学习率，在加快收敛速度的同时，提高了模型的预测精度；而且模型采用双曲正弦激活函数，在解决了传统激活函数导致的梯度爆炸和梯度消失的问题的同时，保证了模型收敛速度。
The invention discloses a power system monitoring method and system. By collecting the observation data of each node of the power system within a period of observation time; preprocessing the observation data to obtain a feature vector; and inputting the feature vector into a trained operating state classification model , obtain the probability that each node of the power system belongs to different operating states at the current time, and select the operating state corresponding to the highest probability as the operating state of each node at the current time; among them, the operating state classification model updates the learning rate based on the total error during the training process, While accelerating the convergence speed, the prediction accuracy of the model is improved; and the model adopts a hyperbolic sine activation function, which not only solves the problems of gradient explosion and gradient disappearance caused by the traditional activation function, but also ensures the model convergence speed.
Description
Technical Field
The invention belongs to the field of power system monitoring, and particularly relates to a power system monitoring method and system.
Background
The statements in this section merely provide background information related to the present disclosure and may not necessarily constitute prior art.
At present, the inspection work of an electric power system mainly adopts manpower, the manual inspection is a safety risk in a high altitude area, the manual inspection and maintenance are subject to a manual management strategy and personnel quality, the inspection quality cannot be guaranteed, and the cost is very high.
Therefore, it is necessary to provide a power system monitoring method and system to solve the above technical drawbacks.
Disclosure of Invention
The invention provides a method and a system for monitoring a power system to solve the problems, wherein observation data of each node of the power system in a certain observation time period are collected; preprocessing observation data to obtain a characteristic vector; inputting the feature vectors into a trained operation state classification model to obtain the probability that each node of the power system belongs to different operation states at the current moment, and selecting the operation state corresponding to the highest probability as the operation state of each node at the current moment; the learning rate of the operation state classification model is updated based on the total error in the training process, so that the convergence speed is accelerated, and the prediction precision of the model is improved; and the model adopts a hyperbolic sine activation function, so that the problems of gradient explosion and gradient disappearance caused by the traditional activation function are solved, and the convergence speed of the model is ensured.
According to some embodiments, the invention adopts the following technical scheme:
a power system monitoring method, comprising:
acquiring observation data of a certain node of the power system in a certain observation time period;
preprocessing observation data to obtain a characteristic vector;
inputting the characteristic vector into a trained operation state classification model to obtain the probability that the current moment of the node of the power system belongs to different operation states, and selecting the operation state corresponding to the highest probability as the current moment of the node; and in the training process of the operation state classification model, if the total error of the output probability of the output layer and the expected probability does not meet the convergence condition, updating the learning rate based on the total error, and transferring to back propagation until the total error meets the convergence condition.
Further, the training process of the operation state classification model specifically includes:
constructing a training set;
initializing a network;
an input feature vector is taken from the training set and added to the network, a target probability vector corresponding to the input feature vector is obtained, and hidden layer output is calculated;
calculating a probability vector output by an output layer according to the hidden layer output;
calculating a network prediction error and an error item of a hidden layer;
updating the adjustment quantity of each weight and threshold according to the network prediction error and the error item of the hidden layer;
calculating a total error according to the probability vector output by the output layer and the target probability vector, and updating the learning rate;
judging whether a convergence condition is met, if not, returning to calculate hidden layer output; if so, saving the operation state classification model.
Further, the pretreatment comprises the following specific steps:
interpolating missing values by mean interpolation;
carrying out standardization treatment by using a z-score standardization method;
and converting the standardized observation data into a feature vector.
Further, the operation state classification model adopts a hyperbolic sine activation function.
Furthermore, when the training is finished, the weight and the threshold are stored in the file by the operation state classification model, and when the training is performed next time, the weight and the threshold are directly derived from the file for training without initialization.
A power system monitoring system comprising:
a data acquisition module configured to: acquiring observation data of a certain node of the power system in a certain observation time period;
a data pre-processing module configured to: preprocessing observation data to obtain a characteristic vector;
an operating state prediction module configured to: inputting the characteristic vector into a trained operation state classification model to obtain the probability that the current moment of the node of the power system belongs to different operation states, and selecting the operation state corresponding to the highest probability as the current moment of the node; and in the training process of the operation state classification model, if the total error of the output probability of the output layer and the expected probability does not meet the convergence condition, updating the learning rate based on the total error, and transferring to back propagation until the total error meets the convergence condition.
An electronic device comprising a memory and a processor, and computer instructions stored on the memory and executed on the processor, the computer instructions, when executed by the processor, performing the steps of the method of the first aspect.
A computer readable storage medium storing computer instructions which, when executed by a processor, perform the steps of the method of the first aspect.
The invention has the beneficial effects that:
1. according to the monitoring method of the power system, the operation state of each node of the power system at the current moment is obtained by adopting the operation state classification model based on the observation data in a period of observation time, so that the organic integration of historical operation data and the current moment data is realized, the operation state of the system is comprehensively described, and the accuracy of the operation state prediction of the power system is improved.
2. The embodiment of the invention provides an operation state classification model for predicting the operation state of a power system, which updates the learning rate in each iteration process based on the total error, and when the total error is larger, the learning rate is larger, and the convergence speed is accelerated; when the total error is small, the learning rate is small, and the local optimization capability is enhanced; the convergence speed is accelerated, and meanwhile, the prediction precision of the model is improved.
3. The operation state classification model for predicting the operation state of the power system provided by the embodiment of the invention adopts a hyperbolic sine activation function, solves the problems of gradient explosion and gradient disappearance caused by gradient reverse transfer of a sigmod function or a logistic function in a deep neural network, and ensures the training convergence speed.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, are included to provide a further understanding of the invention, and are incorporated in and constitute a part of this specification, illustrate exemplary embodiments of the invention and together with the description serve to explain the invention and not to limit the invention.
FIG. 1 is a flow chart of a power system monitoring method of the present invention.
Detailed Description
The invention is further described with reference to the following figures and examples.
It is to be understood that the following detailed description is exemplary and is intended to provide further explanation of the invention as claimed. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
It is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of exemplary embodiments according to the invention. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, and it should be understood that when the terms "comprises" and/or "comprising" are used in this specification, they specify the presence of stated features, steps, operations, devices, components, and/or combinations thereof, unless the context clearly indicates otherwise.
Example 1
As shown in fig. 1, in the monitoring method for the power system provided by this embodiment, observation data of each node of the power system in an observation time period is collected; preprocessing observation data to obtain a characteristic vector; inputting the feature vectors into a trained operation state classification model to obtain the probability that each node of the power system belongs to different operation states at the current moment, and selecting the operation state corresponding to the highest probability as the operation state of the power system at the current moment; the learning rate of the operation state classification model is updated based on the total error in the training process, so that the convergence speed is accelerated, and the prediction precision of the model is improved; and the model adopts a hyperbolic sine activation function, so that the problems of gradient explosion and gradient disappearance caused by the traditional activation function are solved, and the convergence speed of the model is ensured.
A power system monitoring method, as shown in fig. 1, includes the following steps:
step 1: the method comprises the steps of obtaining observation data of any node of the power system at different observation moments in an observation time period before the current moment and the current moment, wherein the observation data comprise voltage, current, active power, reactive power, frequency and the like.
Specifically, the acquisition of the observation data at different observation times is performed by an acquisition device, and the acquisition device includes an analog quantity data acquisition unit, a temperature sensor, and the like.
The input end of the analog quantity data acquisition unit is connected with a PT/CT port of the power system for acquisition, and a voltage transformer inputs voltage and a current transformer inputs current; the ambient temperature of each node of the power system is measured by a temperature sensor.
The specific steps of acquiring observation data of a certain node of the power system at different observation times in an observation time period before the current time are as follows:
(1) the acquisition device acquires observation data of a certain node of the power system at the current moment, and the output end of the acquisition device is connected with the information processing device and outputs digital acquisition information at the current moment after AD conversion;
(2) the information processing device receives clock information from a system clock and digital acquisition information at the current moment, combines the clock information and the digital acquisition information at the current moment to generate a sampling information message at the current moment with a timestamp, and sends the sampling information message to the communication device;
(3) the system clock is connected with the communication device and receives and updates clock information, the storage device is connected with the information processing device and receives a current time sampling information message with a time stamp, meanwhile, the storage device outputs a historical information message to the information processing device, and at the moment, the information processing device obtains observation data of different observation times in the current time and an observation time period before the current time.
After receiving the sampling information message and the historical information message at the current moment, the information processing device firstly preprocesses data, and then predicts and obtains the current-moment operation state of the node of the power system based on a trained operation state classification model.
Step 2: and preprocessing the observation data to obtain a feature vector.
Firstly, interpolating a missing value by adopting mean interpolation;
secondly, standardizing the interpolated data by a z-score standardization method;
and finally, converting the standardized observation data into a characteristic vector X ═ X1,x2,...,xT) Wherein, the vector xtThe data acquired at time T after normalization, T1, …, T, is assumed to be a vector xtIs P, then the dimension of the feature vector X is TP.
And step 3: inputting the collected characteristic vectors into a trained operation state classification model to obtain the probability that the node belongs to different operation states at the current moment of the power system, and selecting the operation state corresponding to the highest probability as the operation state of the node at the current moment, wherein the operation state comprises a normal state, a general abnormal state and a serious abnormal state.
Wherein the operating state classification model utilizes a BP neural network structure.
The BP neural network is a multilayer feedforward neural network, mainly comprises an input layer, a hidden layer and an output layer, and is mainly characterized in that signals are transmitted in a forward direction, and errors are propagated in a reverse direction.
In the training process of the operation state classification model, if the total error of the output probability of the output layer and the expected probability does not meet the convergence condition, the learning rate is updated based on the total error, and the operation state classification model is transmitted to the back propagation until the total error of the output probability of the model and the expected probability meets the convergence condition. Specifically, the training process of the operation state classification model is as follows:
(1) and constructing a training set.
Assume that m training samples are constructed { (X (1), Y (1)), (X (2), Y (2)),1,y2,...,yK) The expected output of the feature vector x (i) corresponding to the ith sample at different observation times, i.e. the probability that the ith sample belongs to different operation states at time T, where K is 3 in this embodiment; the feature vector of the ith sample from 1 to T observation times is denoted as x (i) ═ x1(i),x2(i),...,xT(i) The characteristic index at the time t) can be a vector x with a dimension of PtWhen T is 1, …, and T, the dimension of the feature vector x (i) is TP, and TP is N, so the feature vector of the ith sample from 1 to T observation time can be represented as x (i) (x) and (i) ═ N1,x2,...,xN)。
(2) And (5) initializing the network.
Initializing connection weights V between input layer, hidden layer, and output layer neuronsnlAnd WlkHidden layer threshold alOutput layer threshold bkAnd initializes the learning rate β.
(3) An input feature vector X (i) is taken from the training set and added to the network, and a target output vector Y (i) corresponding to the input feature vector is obtained.
(4) The hidden layer output is computed.
According to the input characteristic vector X, the connection weight V between the input layer and the hidden layernlAnd an implicit layer threshold alAnd calculating hidden layer output:
wherein, N is the number of nodes of the network input layer, L is 1, …, L is the number of nodes of the hidden layer, f (x) is the activation function, the activation function of the invention adopts the hyperbolic sine activation function, specifically, it is the hyperbolic sine activation function
The activation function is replaced by a hyperbolic sine activation function from the original sigmod function or logetics function, so that the problems of gradient explosion and gradient disappearance caused by gradient reverse transfer of the sigmod function or the logetics function in the deep neural network are solved, and the training convergence speed is ensured.
(5) And calculating output layer output.
Outputting H and connecting weight W according to hidden layerlkAnd a threshold value bkAnd calculating the probability vector O ═ O (O) output by the output layer of the BP neural network1,o2,…,ok)
Wherein K is 1, …, and K is the number of nodes in the output layer.
(6) Computing network prediction error
Will output the element o in the vectorkWith elements in the target vectorElement ykAnd comparing, and calculating the network prediction error:
δk＝(yk-ok)ok(1-ok)
(7) calculating the error term of the hidden layer:
(8) updating the adjustment quantity of each weight and threshold according to the network prediction error and the error item of the hidden layer:
Wlk＝Wlk+βhlδk
bk＝bk+βδk
(9) calculating a total error according to the probability vector output by the output layer and the target probability vector;
after obtaining the probability vectors output by all output layers every time K goes through 1 to K, a total error is calculated according to the probability vectors output by the output layers and the target probability vector (i.e., the target output vector Y), the total error E can be expressed as,
(10) updating learning rate
β＝γE
Wherein gamma is an adjusting parameter. The traditional method adopting a fixed learning rate value has the following disadvantages: a small learning rate may take a significant amount of time to converge, or may fail to converge due to the disappearance of the gradient; a large learning rate puts the model at risk of exceeding a minimum value, so it will not converge, the so-called explosive gradient. Therefore, the learning rate is updated in each iteration process based on the total error, and when the total error is larger, the learning rate is larger, and the convergence speed is accelerated; when the total error is small, the learning rate is small, and the local optimization capability is enhanced, so that the convergence speed is accelerated and the prediction accuracy of the model is improved by the method for updating the learning rate based on the total error.
(11) Judging whether the total error meets a convergence condition: e is less than or equal to epsilon, epsilon is a convergence threshold value, if not, the step (3) is returned, and iteration is continued; if yes, go to the next step.
(12) And after training, storing the weight values and the threshold values in a file, confirming that each weight value and each threshold value have reached stability, forming an operation state classification model, and storing the operation state classification model.
And when training is performed next time, the weight and the threshold are directly exported from the file for training without initialization.
According to the monitoring method of the power system, the operation state of each node of the power system at the current moment is obtained by adopting the operation state classification model based on the observation data of each node in a period of observation time, so that the organic integration of historical operation data and current moment data is realized, the operation state of the system is comprehensively described, and the accuracy of the operation state prediction of the power system is improved.
Example 2
The present embodiment provides a power system monitoring system, including:
a data acquisition module configured to: acquiring observation data of a certain node of the power system in a certain observation time period;
a data pre-processing module configured to: preprocessing observation data to obtain a characteristic vector;
an operating state prediction module configured to: inputting the characteristic vector into a trained operation state classification model to obtain the probability that the current moment of the node of the power system belongs to different operation states, and selecting the operation state corresponding to the highest probability as the current moment of the node; and in the training process of the operation state classification model, if the total error of the output probability of the output layer and the expected probability does not meet the convergence condition, updating the learning rate based on the total error, and transferring to back propagation until the total error meets the convergence condition.
Example 3
The present embodiment also provides an electronic device, which includes a memory, a processor, and computer instructions stored in the memory and executed on the processor, wherein the computer instructions, when executed by the processor, perform the steps of the method of embodiment 1.
Example 4
The present embodiment also provides a computer-readable storage medium for storing computer instructions, which when executed by a processor, perform the steps of the method of embodiment 1.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (10)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110711578.8A CN113449914B (en) | 2021-06-25 | 2021-06-25 | Power system monitoring method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110711578.8A CN113449914B (en) | 2021-06-25 | 2021-06-25 | Power system monitoring method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113449914A true CN113449914A (en) | 2021-09-28 |
| CN113449914B CN113449914B (en) | 2023-05-26 |
Family
ID=77812801
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110711578.8A Active CN113449914B (en) | 2021-06-25 | 2021-06-25 | Power system monitoring method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113449914B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117216673A (en) * | 2023-11-08 | 2023-12-12 | 国网江西省电力有限公司电力科学研究院 | Current transformer monitoring evaluation overhauls platform |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102510059A (en) * | 2011-11-28 | 2012-06-20 | 山东电力集团公司潍坊供电公司 | Super short-term wind power forecasting method based on back propagation (BP) neural network |
| CN105095962A (en) * | 2015-07-27 | 2015-11-25 | 中国汽车工程研究院股份有限公司 | Method for predicting dynamic mechanical property of material based on BP artificial neural network |
| CN109272190A (en) * | 2018-08-03 | 2019-01-25 | 国网山东省电力公司烟台供电公司 | A kind of power cable state evaluation method based on momentum BP neural network |
| CN110571792A (en) * | 2019-07-29 | 2019-12-13 | 中国电力科学研究院有限公司 | A method and system for analyzing and evaluating the operating state of a power grid control system |
| CN111797974A (en) * | 2020-06-01 | 2020-10-20 | 武汉大学 | A Power System State Estimation Method Combined with Particle Filter and Convolutional Neural Network |
| CN112101530A (en) * | 2020-11-10 | 2020-12-18 | 南京集成电路产业服务中心有限公司 | Neural network training method, device, equipment and storage medium |
-
2021
- 2021-06-25 CN CN202110711578.8A patent/CN113449914B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102510059A (en) * | 2011-11-28 | 2012-06-20 | 山东电力集团公司潍坊供电公司 | Super short-term wind power forecasting method based on back propagation (BP) neural network |
| CN105095962A (en) * | 2015-07-27 | 2015-11-25 | 中国汽车工程研究院股份有限公司 | Method for predicting dynamic mechanical property of material based on BP artificial neural network |
| CN109272190A (en) * | 2018-08-03 | 2019-01-25 | 国网山东省电力公司烟台供电公司 | A kind of power cable state evaluation method based on momentum BP neural network |
| CN110571792A (en) * | 2019-07-29 | 2019-12-13 | 中国电力科学研究院有限公司 | A method and system for analyzing and evaluating the operating state of a power grid control system |
| CN111797974A (en) * | 2020-06-01 | 2020-10-20 | 武汉大学 | A Power System State Estimation Method Combined with Particle Filter and Convolutional Neural Network |
| CN112101530A (en) * | 2020-11-10 | 2020-12-18 | 南京集成电路产业服务中心有限公司 | Neural network training method, device, equipment and storage medium |
Non-Patent Citations (2)
| Title |
|---|
| 何林凯.: "缺陷漏磁信号的分类识别研究", 《中国优秀博硕士学位论文全文数据库（硕士）工程科技I辑》 * |
| 焦李成，等.: "《人工智能、类脑计算与图像解译前沿》", 30 November 2019 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117216673A (en) * | 2023-11-08 | 2023-12-12 | 国网江西省电力有限公司电力科学研究院 | Current transformer monitoring evaluation overhauls platform |
| CN117216673B (en) * | 2023-11-08 | 2024-03-12 | 国网江西省电力有限公司电力科学研究院 | A current transformer monitoring, evaluation and maintenance platform |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113449914B (en) | 2023-05-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113486578B (en) | Method for predicting residual life of equipment in industrial process | |
| CN112632972B (en) | Method for rapidly extracting fault information in power grid equipment fault report | |
| CN110288046B (en) | Fault prediction method based on wavelet neural network and hidden Markov model | |
| CN106022521B (en) | Short-term load prediction method of distributed BP neural network based on Hadoop architecture | |
| CN113554148A (en) | A BiLSTM Voltage Bias Prediction Method Based on Bayesian Optimization | |
| CN110222371A (en) | Engine residual life on-line prediction method based on Bayes and neural network | |
| CN113704075B (en) | Fault log-based high-performance computing system fault prediction method | |
| CN113722985A (en) | Method and system for evaluating health state and predicting residual life of aircraft engine | |
| Zhang et al. | Wind speed prediction research considering wind speed ramp and residual distribution | |
| CN114358192B (en) | Multi-source heterogeneous landslide data monitoring and fusing method | |
| CN114547974A (en) | Dynamic soft measurement modeling method based on input variable selection and LSTM neural network | |
| CN115730509B (en) | Spacecraft cabin temperature field reconstruction task research benchmark method based on machine learning | |
| CN109255726A (en) | A kind of ultra-short term wind power prediction method of Hybrid Intelligent Technology | |
| CN117407770A (en) | Classification and prediction method of high-voltage switchgear failure modes based on neural network | |
| CN118131044A (en) | A deep learning lithium battery SOH prediction method based on transformer architecture | |
| CN108879732A (en) | Transient stability evaluation in power system method and device | |
| CN115982534A (en) | Processing method of river hydrological monitoring data | |
| CN113449914A (en) | Power system monitoring method and system | |
| Yao et al. | Fault Diagnosis of Rolling Bearing based on VMD Feature Energy Reconstruction and ADE-ELM | |
| CN114492199A (en) | Method and system for analyzing performance of voltage transformer, electronic equipment and medium | |
| CN113887570A (en) | A Two-Classification Prediction Method of Solar Flare Based on Neural Network | |
| CN108446506A (en) | A kind of uncertain system modeling method based on section Feedback Neural Network | |
| Ge et al. | Research on temperature and humidity prediction model of granary based on rnn-lstm | |
| CN116957157A (en) | Production benefit prediction method and system based on multi-source data analysis | |
| CN115062762A (en) | Ocean current trajectory prediction method |
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 |