CN110632545A - Regression neural network-based error risk assessment method and device for electronic transformer - Google Patents
Regression neural network-based error risk assessment method and device for electronic transformer Download PDFInfo
- Publication number
- CN110632545A CN110632545A CN201910909225.1A CN201910909225A CN110632545A CN 110632545 A CN110632545 A CN 110632545A CN 201910909225 A CN201910909225 A CN 201910909225A CN 110632545 A CN110632545 A CN 110632545A
- Authority
- CN
- China
- Prior art keywords
- electronic transformer
- neural network
- difference
- model
- predicted
- 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
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 43
- 238000012502 risk assessment Methods 0.000 title claims abstract description 34
- 238000000034 method Methods 0.000 title claims abstract description 25
- 230000007613 environmental effect Effects 0.000 claims abstract description 46
- 230000000306 recurrent effect Effects 0.000 claims abstract description 30
- 238000012549 training Methods 0.000 claims description 14
- 238000012216 screening Methods 0.000 claims description 7
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000013210 evaluation model Methods 0.000 claims description 3
- 238000011156 evaluation Methods 0.000 abstract description 6
- 238000005070 sampling Methods 0.000 abstract description 2
- 210000002569 neuron Anatomy 0.000 description 8
- 238000005259 measurement Methods 0.000 description 6
- 230000000694 effects Effects 0.000 description 5
- 238000012360 testing method Methods 0.000 description 5
- 230000004913 activation Effects 0.000 description 2
- 230000006978 adaptation Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R35/00—Testing or calibrating of apparatus covered by the other groups of this subclass
- G01R35/02—Testing or calibrating of apparatus covered by the other groups of this subclass of auxiliary devices, e.g. of instrument transformers according to prescribed transformation ratio, phase angle, or wattage rating
-
- 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
- G06Q10/063—Operations research, analysis or management
- G06Q10/0635—Risk analysis of enterprise or organisation activities
-
- 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)
- Human Resources & Organizations (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Strategic Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- General Health & Medical Sciences (AREA)
- Public Health (AREA)
- Biomedical Technology (AREA)
- Operations Research (AREA)
- Game Theory and Decision Science (AREA)
- Educational Administration (AREA)
- Water Supply & Treatment (AREA)
- Development Economics (AREA)
- Primary Health Care (AREA)
- Power Engineering (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Quality & Reliability (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)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses an error risk assessment method and device for an electronic transformer based on a recurrent neural network, which are characterized in that a risk assessment model of the electronic transformer based on the recurrent neural network is trained according to the current and the voltage of the electronic transformer and screened environmental characteristic data; and inputting the characteristic data of the electronic transformer to be evaluated into the model to obtain the predicted specific difference and angular difference of the electronic transformer, and calculating the operation risk index of the electronic transformer according to the predicted specific difference and angular difference. The electronic transformer is modeled by using the sampling current and voltage values of the electronic transformer and combining the environmental characteristic data, the specific difference and the angular difference are predicted, risk evaluation is carried out, and the accuracy of transformer evaluation is verified.
Description
Technical Field
The invention belongs to the technical field of measurement, and particularly relates to an electronic transformer error risk assessment method and device based on a recurrent neural network.
Background
An electronic transformer is a power distribution device consisting of one or more voltage or current sensors connected to a transmission system and to a secondary converter, for transmitting quantities proportional to the quantities being measured, to measurement instruments, meters and relay protection or control devices. With the development of the intelligent power grid technology in China, the electronic transformer becomes a key device for realizing the digitization of the electrical parameter information in the intelligent power grid.
However, in order to evaluate the error performance most concerned by the electronic instrument transformer, a high-precision standard is required to be provided to calibrate the electronic instrument transformer to be tested, and the high-precision standard is difficult to be put into a high-voltage environment due to the restriction of the operating environment, and can be calibrated only in a short time even if put into an online environment.
Due to the lack of an evaluation means of the operation state of the electronic transformer, namely the measurement error, the potential hidden danger is difficult to find in time, the requirement on the safe and stable operation of the power system is high, if the hidden danger of the electronic transformer is not found in time, the power failure maintenance brings great economic loss, and the satisfaction degree of users is reduced.
At present, electronic transformers of equipment manufacturers have uneven production level and different design methods.
Disclosure of Invention
In order to solve the problems, the invention discloses an electronic transformer error risk assessment method and device based on a recurrent neural network, and solves the problem that the measurement error of an electronic transformer is difficult to assess.
The technical scheme of the invention is as follows: an error risk assessment method of an electronic transformer based on a recurrent neural network,
training a risk assessment model of the electronic transformer based on a recurrent neural network according to the current and the voltage of the electronic transformer and the screened environmental characteristic data;
and inputting the characteristic data of the electronic transformer to be evaluated into the model to obtain the predicted specific difference and angular difference of the electronic transformer, and calculating the operation risk index of the electronic transformer according to the predicted specific difference and angular difference.
Further, the environment feature data includes: ambient temperature, ambient humidity, intelligent substation space magnetic field, electronic transformer operational environment's vibration.
Further, the screening process of the environmental characteristic data of the electronic transformer comprises the following steps:
analyzing the maximum value and the minimum value of each environmental characteristic data, and determining a threshold Th of the difference between the maximum value and the minimum value;
for a certain environmental characteristic, the following conditions are satisfied:
Mmax-Mmin<Th
Mmax,Mminrespectively, the maximum value and the minimum value of a certain environmental characteristic M.
Further, the regression neural network-based electronic transformer risk assessment model comprises four full-connection layers, wherein the input of the four full-connection layers is the environmental characteristics, current and voltage of the electronic transformer, the real specific difference and angular difference of the electronic transformer are used as labels, and the predicted specific difference and angular difference of the electronic transformer are output.
Further, the electronic transformer operation risk index R is calculated by the following method:
R=α(|p′|-0.2)/0.2+β(|q′|-10)/10
wherein p 'is a specific difference predicted by the electronic transformer to be evaluated through the model, and q' is an angular difference predicted by the electronic transformer to be evaluated through the model; alpha and beta are weight coefficients.
The utility model provides an electronic transformer error risk assessment device based on recurrent neural network which characterized in that includes:
the electronic transformer risk assessment model training module based on the recurrent neural network is used for training an electronic transformer risk assessment model based on the recurrent neural network according to the current and the voltage of the electronic transformer and the screened environmental characteristic data;
and the electronic transformer operation risk index calculation module is used for inputting the characteristic data of the electronic transformer to be evaluated into the model to obtain the predicted specific difference and angular difference of the electronic transformer and calculating the operation risk index of the electronic transformer according to the predicted specific difference and angular difference.
Further, the environment feature data includes: ambient temperature, ambient humidity, intelligent substation space magnetic field, electronic transformer operational environment's vibration.
Further, the screening process of the environmental characteristic data of the electronic transformer comprises the following steps:
analyzing the maximum value and the minimum value of each environmental characteristic data, and determining a threshold Th of the difference between the maximum value and the minimum value;
for a certain environmental characteristic, the following conditions are satisfied:
Mmax-Mmin<Th
Mmax,Mminrespectively, the maximum value and the minimum value of a certain environmental characteristic M.
Further, the regression neural network-based electronic transformer risk assessment model comprises four full-connection layers, wherein the input of the four full-connection layers is the environmental characteristics, current and voltage of the electronic transformer, the real specific difference and angular difference of the electronic transformer are used as labels, and the predicted specific difference and angular difference of the electronic transformer are output.
Further, the electronic transformer operation risk index R is calculated by the following method:
R=α(|p′|-0.2)/0.2+β(|q′|-10)/10
wherein p 'is a specific difference predicted by the electronic transformer to be evaluated through the model, and q' is an angular difference predicted by the electronic transformer to be evaluated through the model; alpha and beta are weight coefficients.
The invention has the beneficial effects that:
1) the electronic transformer is modeled by using the sampling current and voltage values of the electronic transformer and combining environmental characteristic data, the specific difference and the angular difference are predicted, risk evaluation is carried out, and the accuracy of transformer evaluation is verified;
2) according to the invention, by removing invalid data in the environmental characteristic data of the electronic transformer, the effects of improving the value density of the data and optimizing the data access speed are achieved.
Drawings
Fig. 1 is a flowchart of an error risk assessment method for an electronic transformer according to an embodiment of the present invention;
FIG. 2 is a predicted scatter plot of the ratio differences according to an embodiment of the present invention;
FIG. 3 is a predicted scatter plot of angular differences in accordance with an embodiment of the present invention.
Detailed Description
The technical solutions of the present invention are further described in detail below with reference to specific examples so that those skilled in the art can better understand the present invention and can implement the present invention, but the examples are not intended to limit the present invention.
Example 1:
as shown in fig. 1, an error risk assessment method for an electronic transformer based on a recurrent neural network includes the steps of:
the method comprises the following steps of firstly, screening the effectiveness of environmental characteristic data where an electronic transformer is located to obtain screened effective environmental characteristic data, and specifically comprises the following steps:
1) the environmental characteristic data includes: analyzing the maximum value and the minimum value of each environmental characteristic data to determine a threshold Th of the difference between the maximum value and the minimum value; the value of Th is required to accurately reflect the critical state of unstable environment for subsequent data screening.
2) For a certain environmental characteristic M (temperature, humidity, magnetic field or vibration), it needs to satisfy:
Mmax-Mmin<Th (1)
Mmax,Mminrespectively taking the maximum value and the minimum value of a certain environmental characteristic M, removing all data groups with difference values larger than a threshold value, and eliminating the influence of environmental instability on the operation of the electronic transformer so as to better establish and train a model;
calculating the real specific difference and the angular difference of the electronic transformer according to the current and the voltage of the electronic transformer,
JC=(J-J′)*60 (2)
wherein BC is a specific difference, B is a voltage (current) amplitude measured by the electronic transformer, and B' is a voltage (current) amplitude of the traditional electromagnetic standard transformer, and is dimensionless; JC is the angular difference, J is the voltage (current) phase measured by the electronic transformer, and J' is the voltage (current) phase of the conventional electromagnetic standard transformer, and the unit is minutes.
Secondly, training a risk assessment model of the electronic transformer based on a recurrent neural network according to current and voltage of the electronic transformer and environmental characteristic data;
the electronic transformer risk assessment model based on the recurrent neural network comprises four full-connection layers, four environmental characteristics (temperature, humidity, vibration and magnetic field) and current and voltage of the electronic transformer are input, the real specific difference and the angular difference of the electronic transformer are used as labels, the predicted specific difference and the predicted angular difference of the electronic transformer are output, and the neural network learns all the characteristics of the electronic transformer by adjusting the weight among neurons. Training an electronic transformer risk assessment model based on a recurrent neural network, and specifically comprises the following steps:
1) 70% of the screened valid data are classified as a training set, and the remaining 30% are classified as a testing set. The training set is used for learning the model, the temperature (DEG C), the humidity (RH), the vibration (g) and the magnetic field (Gs) of each group of electronic transformers when working, and current and voltage data are used as input samples of the model, and the labels of the neural network are real specific difference and angular difference.
For neural networks, it is assumed that the l-1 st layer has m neurons, the l-1 st layer has n neurons, k is 1 to m, j is 1 to n,for the output before the activation of the jth neuron of the ith layer, the output after the activation of the jth neuron of the ith layer isComprises the following steps:
wherein the content of the first and second substances,is the weight between the j and k neurons of the l layer,bias for the jth neuron at level i;
the weighting coefficients of layer I form an n x m matrix W, and the bias of layer I forms an n x 1 vector blThe output of layer l-1 constitutes onem x 1 vector al-1The linear output z before the first layer is deactivated forms an n x 1 vector zlThe output a of the l-th layerlThe first step is:
al=σ(zl)=σ(Wlal-1+bl) (4)
it can be seen that the output of the l-th layer constitutes an n x 1 vector al。
2) By forward and backward propagation, i.e.
Continuously adjusting the weight of the connection between the neurons by using a gradient descent algorithm until the loss function is reduced to the minimum, and converging the model:
wherein, p and q are the real ratio difference and the angular difference calculated in the step (2) of the step one, namely the labels used in the model training,andas predicted ratio and angle differences; training the model to enable the predicted ratio difference and angle difference to be close to the real ratio difference and angle difference in the label;
3) the test set is used to verify the effect of the model. And inputting each group of data (temperature, humidity, vibration, magnetic field, current and voltage of the electronic transformer) of the test set into the model to obtain a predicted specific difference and angular difference, and comparing the predicted specific difference and angular difference with a real angular difference and specific difference calculated by the current and voltage in each group of data of the test set to obtain errors of the specific difference and the angular difference. Here we use the root mean square error in statistics to measure the effectiveness of the model, and if the model is more effective, the error is smaller, and the prediction accuracy of the specific difference and the angular difference is higher. If the model effect is not good, the network depth is deepened, and after the two to three times of operations, one model structure with the minimum error is selected as a well-trained model.
And thirdly, inputting the characteristic data of the electronic transformer to be evaluated into the trained electronic transformer risk evaluation model based on the recurrent neural network to obtain the predicted specific difference and angular difference of the electronic transformer, and calculating the operation risk index of the electronic transformer according to the specific difference and the angular difference.
And inputting the temperature, humidity, vibration, magnetic field, current and voltage data of the electronic transformer to be evaluated into the trained risk evaluation model of the electronic transformer based on the recurrent neural network, so that the ratio difference and the angular difference of the electronic transformer during operation can be predicted, and risk evaluation is carried out through the predicted ratio difference and angular difference. According to the voltage error (specific difference) and phase error (angular difference) limits of the electronic voltage transformer for measurement specified in GB/T2.8407.7-2007, since the data source of the invention requires an accuracy level of 0.2, i.e. the data is not reliable when the specific difference is greater than 0.2% or the angular difference is greater than 10 degrees, the running risk index of the transformer is calculated by multiplying different weight coefficients alpha and beta according to the respective exceeding percentages by the following formula:
R=α(|p′|-0.2)/0.2+β(|q′|-10)/10 (6)
and p 'is the specific difference predicted by the electronic transformer to be evaluated through the model, and q' is the angular difference predicted by the electronic transformer to be evaluated through the model. Similarly, if the ratio difference and the angle difference are smaller, the risk index of the electronic transformer is lower, and the operation effect is more reliable. On the contrary, if the ratio difference and the angle difference are larger, the risk index of the electronic transformer is higher.
The electronic transformer error risk assessment method solves the problem that the error state of the electronic transformer cannot be assessed without depending on a standard device, and meanwhile, the risk that the electronic transformer measurement data are used for engineering needs to be further assessed. The working environment data (current, voltage, temperature, humidity, vibration and magnetic field) of the electronic transformer is input, so that the ratio difference and the angle difference of the transformer during working can be obtained through prediction, the working reliability of the electronic transformer is indirectly obtained, and the smaller the ratio difference and the angle difference is, the higher the working reliability of the transformer is, and the lower the working reliability of the transformer is otherwise.
As shown in fig. 2 and 3, a scatter diagram is drawn by comparing the actual ratio difference and angular difference in the test set with the ratio difference and angular difference predicted by the neural network, and ideally, the scatter point should approach a straight line passing through the origin, so that the model can achieve a better prediction effect. Then, the risk index of the mutual inductor can be further estimated by using the ratio difference and the angle difference obtained by prediction, if 300 groups of data are sampled, the ratio of the number of each risk level under the real and predicted conditions is calculated, and if the R value is below 5%, no risk exists; if between 5% and 10%, it is a secondary risk; more than 10% is the first order risk. The results can be seen in the following table.
TABLE 1 true and predicted occupation of risk classes
The invention provides an error risk assessment method of an electronic transformer based on a recurrent neural network, so that the screened effective data is learned, and the operation ratio difference and the angle difference of the electronic transformer can be predicted under the condition of no standard device. On the basis, the operation reliability and the risk of the electronic transformer are further evaluated by analyzing the predicted ratio difference and the angle difference.
Example 2:
an electronic transformer error risk assessment device based on a recurrent neural network comprises:
the electronic transformer risk assessment model training module based on the recurrent neural network is used for training an electronic transformer risk assessment model based on the recurrent neural network according to the current and the voltage of the electronic transformer and the screened environmental characteristic data;
and the electronic transformer operation risk index calculation module is used for inputting the characteristic data of the electronic transformer to be evaluated into the model to obtain the predicted specific difference and angular difference of the electronic transformer and calculating the operation risk index of the electronic transformer according to the predicted specific difference and angular difference.
The environmental characteristic data includes: ambient temperature, ambient humidity, intelligent substation space magnetic field, electronic transformer operational environment's vibration.
The screening process of the environmental characteristic data of the electronic transformer comprises the following steps:
analyzing the maximum value and the minimum value of each environmental characteristic data, and determining a threshold Th of the difference between the maximum value and the minimum value;
for a certain environmental characteristic, the following conditions are satisfied:
Mmax-Mmin<Th
Mmax,Mminrespectively, the maximum value and the minimum value of a certain environmental characteristic M.
The electronic transformer risk assessment model based on the recurrent neural network comprises four full-connection layers, wherein the input of the four full-connection layers is the environmental characteristics, current and voltage of the electronic transformer, the real specific difference and angular difference of the electronic transformer are used as labels, and the predicted specific difference and angular difference of the electronic transformer are output.
The electronic transformer operation risk index R is calculated by the following steps:
R=α(|p′|-0.2)/0.2+β(|q′|-10)/10
wherein p 'is a specific difference predicted by the electronic transformer to be evaluated through the model, and q' is an angular difference predicted by the electronic transformer to be evaluated through the model; alpha and beta are weight coefficients.
The above is only a preferred embodiment of the present invention, and it should be noted that: it will be apparent to those skilled in the art that various modifications and adaptations can be made without departing from the principles of the invention and these are intended to be within the scope of the invention.
Claims (10)
1. An electronic transformer error risk assessment method based on a recurrent neural network is characterized in that,
training a risk assessment model of the electronic transformer based on a recurrent neural network according to the current and the voltage of the electronic transformer and the screened environmental characteristic data;
and inputting the characteristic data of the electronic transformer to be evaluated into the model to obtain the predicted specific difference and angular difference of the electronic transformer, and calculating the operation risk index of the electronic transformer according to the predicted specific difference and angular difference.
2. The method for evaluating the error risk of the electronic transformer based on the recurrent neural network as claimed in claim 1, wherein the environmental characteristic data comprises: ambient temperature, ambient humidity, intelligent substation space magnetic field, electronic transformer operational environment's vibration.
3. The method for evaluating the error risk of the electronic transformer based on the recurrent neural network as claimed in claim 1, wherein the screening process of the environmental characteristic data of the electronic transformer comprises:
analyzing the maximum value and the minimum value of each environmental characteristic data, and determining a threshold Th of the difference between the maximum value and the minimum value;
for a certain environmental characteristic, the following conditions are satisfied:
Mmax-Mmin<Th
Mmax,Mminrespectively, the maximum value and the minimum value of a certain environmental characteristic M.
4. The method for evaluating the error risk of the electronic transformer based on the recurrent neural network as claimed in claim 1, wherein the electronic transformer risk evaluation model based on the recurrent neural network comprises four fully connected layers, the input is the environmental characteristics and the current and the voltage of the electronic transformer, the real specific difference and the angular difference of the electronic transformer are used as labels, and the predicted specific difference and the angular difference of the electronic transformer are output.
5. The method for evaluating the error risk of the electronic transformer based on the recurrent neural network as claimed in claim 1, wherein the operation risk index R of the electronic transformer is calculated by:
R=α(|p′|-0.2)/0.2+β(|q′|-10)/10
wherein p 'is a specific difference predicted by the electronic transformer to be evaluated through the model, and q' is an angular difference predicted by the electronic transformer to be evaluated through the model; alpha and beta are weight coefficients.
6. The utility model provides an electronic transformer error risk assessment device based on recurrent neural network which characterized in that includes:
the electronic transformer risk assessment model training module based on the recurrent neural network is used for training an electronic transformer risk assessment model based on the recurrent neural network according to the current and the voltage of the electronic transformer and the screened environmental characteristic data;
and the electronic transformer operation risk index calculation module is used for inputting the characteristic data of the electronic transformer to be evaluated into the model to obtain the predicted specific difference and angular difference of the electronic transformer and calculating the operation risk index of the electronic transformer according to the predicted specific difference and angular difference.
7. The regression neural network-based electronic transformer error risk assessment device according to claim 6, wherein the environmental characteristic data comprises: ambient temperature, ambient humidity, intelligent substation space magnetic field, electronic transformer operational environment's vibration.
8. The regression neural network-based electronic transformer error risk assessment device according to claim 6, wherein the screening process of the environmental characteristic data of the electronic transformer is as follows:
analyzing the maximum value and the minimum value of each environmental characteristic data, and determining a threshold Th of the difference between the maximum value and the minimum value;
for a certain environmental characteristic, the following conditions are satisfied:
Mmax-Mmin<Th
Mmax,Mminrespectively, the maximum value and the minimum value of a certain environmental characteristic M.
9. The regression neural network-based electronic transformer error risk assessment device according to claim 6, wherein the regression neural network-based electronic transformer risk assessment model comprises four fully-connected layers, the input of the four fully-connected layers is the environmental characteristics, the current and the voltage of the electronic transformer, the real specific difference and the angular difference of the electronic transformer are used as labels, and the predicted specific difference and the angular difference of the electronic transformer are output.
10. The regression neural network-based error risk assessment device for the electronic transformer according to claim 6, wherein the operation risk index R of the electronic transformer is calculated by the following method:
R=α(|p′|-0.2)/0.2+β(|q′|-10)/10
wherein p 'is a specific difference predicted by the electronic transformer to be evaluated through the model, and q' is an angular difference predicted by the electronic transformer to be evaluated through the model; alpha and beta are weight coefficients.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910909225.1A CN110632545B (en) | 2019-09-25 | 2019-09-25 | Regression neural network-based error risk assessment method and device for electronic transformer |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910909225.1A CN110632545B (en) | 2019-09-25 | 2019-09-25 | Regression neural network-based error risk assessment method and device for electronic transformer |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110632545A true CN110632545A (en) | 2019-12-31 |
CN110632545B CN110632545B (en) | 2021-09-21 |
Family
ID=68974210
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910909225.1A Active CN110632545B (en) | 2019-09-25 | 2019-09-25 | Regression neural network-based error risk assessment method and device for electronic transformer |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110632545B (en) |
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111537938A (en) * | 2020-03-31 | 2020-08-14 | 国网江西省电力有限公司电力科学研究院 | Error short-time prediction method for electronic transformer based on intelligent algorithm |
CN111814390A (en) * | 2020-06-18 | 2020-10-23 | 三峡大学 | Voltage transformer error prediction method based on transfer entropy and wavelet neural network |
CN112014785A (en) * | 2020-08-06 | 2020-12-01 | 三峡大学 | Error compensation method for air-core coil current transformer based on elastic network |
CN113297797A (en) * | 2021-06-02 | 2021-08-24 | 东南大学 | XGboost-based electronic transformer measurement error state evaluation method and device |
CN113702895A (en) * | 2021-10-28 | 2021-11-26 | 华中科技大学 | Online quantitative evaluation method for error state of voltage transformer |
CN113887846A (en) * | 2021-12-07 | 2022-01-04 | 武汉格蓝若智能技术有限公司 | Out-of-tolerance risk early warning method for capacitive voltage transformer |
CN114114126A (en) * | 2021-11-22 | 2022-03-01 | 浙江华采科技有限公司 | Digital calibration method and system for low-permeability magnetic powder core current transformer |
CN115542064A (en) * | 2022-12-01 | 2022-12-30 | 威胜集团有限公司 | Real-time monitoring method and device for error state of mutual inductor |
CN115542230A (en) * | 2022-11-24 | 2022-12-30 | 武汉格蓝若智能技术股份有限公司 | Current transformer error estimation method and device based on diffusion model |
CN117131458A (en) * | 2023-10-26 | 2023-11-28 | 国网江西省电力有限公司电力科学研究院 | Data validity analysis method and system for optical fiber current transformer |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107255792A (en) * | 2017-06-20 | 2017-10-17 | 华中科技大学 | A kind of electronic type voltage transformer error on-line monitoring method and system |
CN108414848A (en) * | 2018-01-08 | 2018-08-17 | 浙江工业大学 | Electric energy quality multi-period comprehensive early warning method for power distribution network with distributed power supply |
CN109031130A (en) * | 2018-05-31 | 2018-12-18 | 轻客智能科技(江苏)有限公司 | A kind of electricity calibration method and device |
CN109581269A (en) * | 2018-12-25 | 2019-04-05 | 中国电力科学研究院有限公司 | A kind of electronic mutual inductor error characteristics fast appraisement method and system |
CN109583044A (en) * | 2018-11-09 | 2019-04-05 | 中国直升机设计研究所 | Lifting airscrew flight load prediction technique based on RBF neural |
CN110095744A (en) * | 2019-04-04 | 2019-08-06 | 国网江苏省电力有限公司电力科学研究院 | A kind of electronic mutual inductor error prediction method |
-
2019
- 2019-09-25 CN CN201910909225.1A patent/CN110632545B/en active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107255792A (en) * | 2017-06-20 | 2017-10-17 | 华中科技大学 | A kind of electronic type voltage transformer error on-line monitoring method and system |
CN108414848A (en) * | 2018-01-08 | 2018-08-17 | 浙江工业大学 | Electric energy quality multi-period comprehensive early warning method for power distribution network with distributed power supply |
CN109031130A (en) * | 2018-05-31 | 2018-12-18 | 轻客智能科技(江苏)有限公司 | A kind of electricity calibration method and device |
CN109583044A (en) * | 2018-11-09 | 2019-04-05 | 中国直升机设计研究所 | Lifting airscrew flight load prediction technique based on RBF neural |
CN109581269A (en) * | 2018-12-25 | 2019-04-05 | 中国电力科学研究院有限公司 | A kind of electronic mutual inductor error characteristics fast appraisement method and system |
CN110095744A (en) * | 2019-04-04 | 2019-08-06 | 国网江苏省电力有限公司电力科学研究院 | A kind of electronic mutual inductor error prediction method |
Non-Patent Citations (3)
Title |
---|
WU QIAO,ET AL.: "Error prediction method of electronic transformers based on RBF neural network", 《 2017 13TH IEEE INTERNATIONAL CONFERENCE ON ELECTRONIC MEASUREMENT & INSTRUMENTS (ICEMI)》 * |
卢树峰 等: "直流偏磁对电流互感器的作用机理研究", 《电测与仪表》 * |
黄玲芳: "基于回归神经网络误差配准算法的仿真实现", 《计算机仿真》 * |
Cited By (16)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111537938B (en) * | 2020-03-31 | 2022-12-09 | 国网江西省电力有限公司电力科学研究院 | Error short-time prediction method for electronic transformer based on intelligent algorithm |
CN111537938A (en) * | 2020-03-31 | 2020-08-14 | 国网江西省电力有限公司电力科学研究院 | Error short-time prediction method for electronic transformer based on intelligent algorithm |
CN111814390A (en) * | 2020-06-18 | 2020-10-23 | 三峡大学 | Voltage transformer error prediction method based on transfer entropy and wavelet neural network |
CN111814390B (en) * | 2020-06-18 | 2023-07-28 | 三峡大学 | Voltage transformer error prediction method based on transfer entropy and wavelet neural network |
CN112014785A (en) * | 2020-08-06 | 2020-12-01 | 三峡大学 | Error compensation method for air-core coil current transformer based on elastic network |
CN113297797A (en) * | 2021-06-02 | 2021-08-24 | 东南大学 | XGboost-based electronic transformer measurement error state evaluation method and device |
CN113297797B (en) * | 2021-06-02 | 2023-12-15 | 东南大学 | XGBoost-based electronic transformer measurement error state evaluation method and device |
CN113702895A (en) * | 2021-10-28 | 2021-11-26 | 华中科技大学 | Online quantitative evaluation method for error state of voltage transformer |
CN113702895B (en) * | 2021-10-28 | 2022-02-08 | 华中科技大学 | Online quantitative evaluation method for error state of voltage transformer |
CN114114126A (en) * | 2021-11-22 | 2022-03-01 | 浙江华采科技有限公司 | Digital calibration method and system for low-permeability magnetic powder core current transformer |
CN113887846B (en) * | 2021-12-07 | 2022-02-22 | 武汉格蓝若智能技术有限公司 | Out-of-tolerance risk early warning method for capacitive voltage transformer |
CN113887846A (en) * | 2021-12-07 | 2022-01-04 | 武汉格蓝若智能技术有限公司 | Out-of-tolerance risk early warning method for capacitive voltage transformer |
CN115542230A (en) * | 2022-11-24 | 2022-12-30 | 武汉格蓝若智能技术股份有限公司 | Current transformer error estimation method and device based on diffusion model |
CN115542230B (en) * | 2022-11-24 | 2023-03-24 | 武汉格蓝若智能技术股份有限公司 | Current transformer error estimation method and device based on diffusion model |
CN115542064A (en) * | 2022-12-01 | 2022-12-30 | 威胜集团有限公司 | Real-time monitoring method and device for error state of mutual inductor |
CN117131458A (en) * | 2023-10-26 | 2023-11-28 | 国网江西省电力有限公司电力科学研究院 | Data validity analysis method and system for optical fiber current transformer |
Also Published As
Publication number | Publication date |
---|---|
CN110632545B (en) | 2021-09-21 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110632545B (en) | Regression neural network-based error risk assessment method and device for electronic transformer | |
CN111814390B (en) | Voltage transformer error prediction method based on transfer entropy and wavelet neural network | |
CN112051481B (en) | Alternating current-direct current hybrid power grid fault area diagnosis method and system based on LSTM | |
CN106443285B (en) | Multiple-harmonic-source harmonic responsibility quantitative analysis method based on total least square method | |
CN113297797B (en) | XGBoost-based electronic transformer measurement error state evaluation method and device | |
CN113702895B (en) | Online quantitative evaluation method for error state of voltage transformer | |
CN113792495B (en) | Lightning arrester operation state identification method and device based on online monitoring data | |
CN112731260B (en) | Online evaluation method for error state of voltage transformer based on concept drift recognition | |
Liu et al. | Combined forecasting method of dissolved gases concentration and its application in condition-based maintenance | |
CN117748507B (en) | Distribution network harmonic access uncertainty assessment method based on Gaussian regression model | |
CN110889207A (en) | System combination model credibility intelligent evaluation method based on deep learning | |
CN113536662B (en) | Electronic transformer error state prediction method based on firefly optimized LightGBM algorithm | |
Wang et al. | Multi‐harmonic sources location based on sparse component analysis and complex independent component analysis | |
CN109193639B (en) | Robust estimation method for power system | |
CN115951292A (en) | Electronic voltage transformer error state online evaluation method and device | |
CN115878963A (en) | Capacitance voltage transformer metering error prediction method, system, terminal and medium | |
CN115408944A (en) | Hot spot temperature prediction method of dry type smoothing reactor | |
CN112464848B (en) | Information flow abnormal data monitoring method and device based on density space clustering | |
CN115062533A (en) | Construction and prediction method and medium of multiple regression prediction model | |
Yu et al. | Knowledge reasoning of transmission line component detection using CRITIC and TOPSIS approaches | |
Xia et al. | Measurement error estimation for distributed smart meters through a modified BP neural network | |
CN110321650B (en) | Structural reliability analysis method based on novel test design and weight response surface | |
CN113537338A (en) | Robust line parameter identification method based on LSTM neural network and improved SCADA data | |
Mazon et al. | Fault location system on double circuit two-terminal transmission lines based on ANNs | |
CN112464436A (en) | Step length adjusting method for parameter identification of power simulation model |
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 |