CN108152676B - Power distribution network fault positioning method and system based on intelligent agent - Google Patents
Power distribution network fault positioning method and system based on intelligent agent Download PDFInfo
- Publication number
- CN108152676B CN108152676B CN201711403320.1A CN201711403320A CN108152676B CN 108152676 B CN108152676 B CN 108152676B CN 201711403320 A CN201711403320 A CN 201711403320A CN 108152676 B CN108152676 B CN 108152676B
- Authority
- CN
- China
- Prior art keywords
- agent
- feature vector
- power distribution
- distribution network
- location
- 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
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/08—Locating faults in cables, transmission lines, or networks
- G01R31/088—Aspects of digital computing
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Locating Faults (AREA)
- Remote Monitoring And Control Of Power-Distribution Networks (AREA)
Abstract
The invention discloses a power distribution network fault positioning method based on an intelligent agent, which comprises the following steps: when a power distribution network fault occurs, firstly, initializing the initial position of an intelligent agent in the power distribution network; the intelligent agent is enabled to move in the power distribution network, and when the intelligent agent moves to a new position, the characteristic vector of the position is obtained; determining how the agent processes the feature vector of the current position and the feature vector of the internal storage part of the agent by utilizing a decision table in the agent according to the feature vector of the current position and the feature vector stored in the agent; and traversing each position in the power distribution network by the intelligent agent, wherein after the intelligent agent traverses all the positions, the position corresponding to the characteristic vector stored in the intelligent agent is the position of the fault point.
Description
Technical Field
The invention relates to the technical field of electric power, in particular to a power distribution network fault positioning method and system based on an intelligent agent.
Background
The power distribution network is an important component in a power system, and with the rapid development of the smart power grid, a large number of distributed power supplies are not determined to be connected, so that the fault information of the power distribution network is more complex, and the accurate and rapid analysis of the fault becomes more difficult. In order to ensure highly intelligent operation of the power distribution network, real-time monitoring, timely early warning of abnormal conditions and rapid positioning and processing of faults need to be carried out on feeder line operation data. Therefore, a power distribution network is usually provided with devices such as a line fault indicator and a feeder terminal, and the devices are used for recording the operating condition of the power distribution network. In the prior art, a method for positioning a power distribution fault generally needs to extract wave recording characteristics by means of artificial setting, and then identify and position the fault by using machine learning models such as a neural network. For example, CN104101812 discloses a method and a system for detecting and locating faults of a low-current grounded power distribution network, in which a system master station extracts transient signals of zero-sequence voltage and zero-sequence current from wave records of multiple points of a power distribution network, calculates similarity between a characteristic value and waveforms of various positions, and judges suspected fault areas by using a machine learning model according to a single-point characteristic value and a correlation characteristic value. In CN103454559, a small current ground fault on-line positioning method and positioning apparatus based on a BP neural network multi-criterion fusion algorithm accurately capture zero sequence current transient signals of zero sequence current exceeding 1 cycle before a starting value and 4 cycles after the starting value through terminals installed at different positions on a line, extract signal characteristics of fault transient signals by using a Prony algorithm, a wavelet packet algorithm, a HHT algorithm and a fractal algorithm, analyze the characteristics extracted by multiple algorithms by using a BP neural network, and determine a section where a fault point is located.
When various machine learning models are used for fault location judgment, the used machine learning models must be trained firstly, a large amount of data is needed for training the machine training models, if the data generated by the monitoring terminals only suitable for one power distribution network can cause insufficient training data amount, the data output by the monitoring terminals of different power distribution networks have different characteristics, and if the data belonging to the monitoring terminals of different power distribution networks are mixed and used for training the same model, the training result deviation can be caused very easily. On the other hand, in the prior art, the fault location point corresponds to the position of the monitoring terminal, and when the machine learning model generates a plurality of fault point results by mistake, a new machine learning model is required to be further screened.
Disclosure of Invention
The technical purpose of the invention is to utilize an Agent to inspect the power distribution network, and the machine training data of the Agent is in units of each fault event in the power distribution network, so that the machine training deviation caused by the mixing of different power distribution network data is avoided.
In order to solve the technical problem, the invention provides an intelligent agent-based power distribution network fault positioning method, which comprises the following steps: when a power distribution network fault occurs, firstly, initializing the initial position of an intelligent agent in the power distribution network; the intelligent agent is enabled to move in the power distribution network, and when the intelligent agent moves to a new position, the characteristic vector of the position is obtained; determining how the agent processes the feature vector of the current position and the feature vector of the internal storage part of the agent by utilizing a decision table in the agent according to the feature vector of the current position and the feature vector stored in the agent; and traversing each position in the power distribution network by the intelligent agent, wherein after the intelligent agent traverses all the positions, the position corresponding to the characteristic vector stored in the intelligent agent is the position of the fault point.
In one embodiment, the position where the intelligent agent moves in the power distribution network refers to an interval position between two adjacent monitoring terminals in the power distribution network.
In one embodiment, the feature vector of the current position refers to a feature vector formed by feature values acquired by two monitoring terminals in front and at back of the current position in the power distribution direction.
In one embodiment, the characteristics collected by the monitoring terminal include, but are not limited to, zero sequence current maximum amplitude, power maximum amplitude, fault region average power, and fault region current wavelet energy entropy.
In one embodiment, the features collected by the monitoring terminal are determined by a deep neural network.
In one embodiment, the initializing agent location may set a start location of the agent to a power distribution line start point.
In one embodiment, the agent randomly jumps to other non-traversed locations if it moves to the end of the distribution line while traversing the distribution network location.
In one embodiment, the decision table of the agent decides to write the location feature vector of the current location into the storage unit to replace the original feature vector, or to retain the original data of the storage unit and discard the location feature vector of the current location, or to randomly select one of the two feature vectors to write into the storage unit without using either of the two feature vectors.
According to another aspect of the present invention, there is also provided a training method of an intelligent agent decision table for power distribution network fault location, the method including:
initializing all weight values Q (S, A) in a decision table;
the following process is repeated for each set of fault event data:
a. the location feature vector S of the agent is initialized.
b. Selecting an action A by adopting an epsilon-Greedy strategy according to the position characteristic vector S and the weight value Q (S,);
c. the data in the location feature vector memory is updated according to the selected action a and the agent is given a reward r and the agent is brought to the next location while the location state vector S' is obtained.
d. According to the position feature vector S ' and the weight value Q (S ',), selecting an action A ' by adopting an epsilon-Greedy strategy, updating Q (S, A) to be (1- α) Q (S, A) + α [ r + gamma Q (S ', A ') ], wherein α represents a learning rate, gamma represents an attenuation coefficient, 0 is more than or equal to α and is less than or equal to 1, and gamma is more than or equal to 0 and is less than or equal to 1, updating the position feature vector S to be the position feature vector S ', and updating the action A to be the action A '.
e. And repeating the steps b to d until all positions in the power distribution network are traversed.
The present invention will be further explained below.
As shown in fig. 1, a schematic diagram of a power distribution network fault location process according to the present invention is shown, when a fault trigger signal occurs in a power distribution network, the power distribution network fault location system according to the present invention is started, and first, a location initialization is performed on an intelligent agent, where the initialization location is selected as a power distribution starting location in a power distribution network topology structure, such as a substation power distribution starting location. The intelligent agent comprises a decision table and a position characteristic vector storage part, and when the initialization of the position of the intelligent agent is completed, the position characteristic vector of the initial position is stored in the position characteristic vector storage part of the intelligent agent.
The position characteristics refer to various characteristics of wave recording collected by a front monitoring terminal and a rear monitoring terminal of the position of the intelligent agent in the power distribution direction, such as the maximum amplitude of zero sequence current, the maximum amplitude of power, the mean power of a fault area, the wavelet energy entropy of the current of the fault area, and the like, for example, four characteristics of the maximum amplitude of zero sequence current, the maximum amplitude of power, the mean power of the fault area and the wavelet energy entropy of the current of the fault area are selected, and then a position characteristic vector can be expressed as (50, 50); (60, 70); (20, 30); (40, 50), wherein (50, 50) is a characteristic value representing the maximum amplitude of the zero sequence current collected by the monitoring terminal before and after the position of the intelligent agent, and (60, 70); (20, 30); and (40, 50) respectively representing the maximum power amplitude, the average power of the fault area and the characteristic values of the wavelet energy entropy of the current of the fault area of the monitoring terminal before and after the position of the intelligent agent. In the invention, data discretization processing is required on the original numerical values of a plurality of characteristics, the original numerical values are firstly scaled within the range of 0-100, and then discretization is carried out on the original numerical values in a unit of 10 bits.
After initialization is completed, the intelligent agent starts to patrol in the power distribution network topological structure according to the power distribution direction, the patrol needs to sequentially pass through the interval positions between each monitoring terminal along one power distribution line, and when the intelligent agent moves to the tail end of one power distribution line, the intelligent agent randomly selects the starting points of other lines which are not patrolled to continue patrolling.
Each time the agent moves to a new location, the agent first obtains the location feature vector at that location. Then, the agent performs table lookup operation in the decision table according to the position feature vector of the current position and the feature vector stored in the position feature vector storage unit. And selecting the action with the largest weight value from the action types corresponding to the two eigenvectors in the decision table for operation. The action types in the invention refer to: an action a1 of writing the position feature vector of the current position into the position feature vector storage unit to replace the original data; an action a2 of retaining the original data of the position feature vector storage unit and discarding the position feature vector of the current position; in act a3, one of two data types is randomly selected and written into the location feature vector storage. After the intelligent agent traverses the interval positions among all the monitoring terminals of the power distribution network, the position corresponding to the feature vector stored in the position feature vector storage part is the position of a fault point.
< machine training of agent >
It can be seen from the above-mentioned fault location process of the present invention that the decision table in the agent is the core part of the agent, so the purpose of machine learning of the agent is to train an effective decision table to control the agent to accurately find the fault location in the routing inspection process of the agent in the power distribution network.
In the present invention, a decision table of an agent is trained by machine Learning in a Q-Learning reinforcement Learning manner, as shown in a flow chart of fig. 2, the training process of the decision table is as follows:
first, all weight values Q (S, a) in the decision table are initialized, where S is a location feature vector of the location of the smart avatar, and action a is action a1, action a2, and action a 3. When the weight values Q (S, a) are initialized, all weight values are randomly generated.
The following process is repeated for each set of fault event data:
a. a location feature vector S of the agent is initialized, the location feature vector S including a location feature vector S1 at the agent and a feature vector S2 in a location feature vector store.
b. And selecting the action A by adopting an epsilon-Greedy strategy according to the position feature vector S and the weight value Q (S, the epsilon-Greedy strategy refers to that in the state S, the agent selects the action A corresponding to the maximum weight value Q with the epsilon probability to execute, and randomly selects one of the action a1, the action a2 and the action a3 with the probability of 1-epsilon to execute.
c. The data in the location feature vector memory is updated according to the selected action a and the agent is given a reward r and the agent is brought to the next location while the location state vector S' is obtained.
d. According to the position feature vector S ' and the weight value Q (S ',), selecting an action A ' by adopting an epsilon-Greedy strategy, updating Q (S, A) to be (1- α) Q (S, A) + α [ r + gamma Q (S ', A ') ], wherein α represents a learning rate, gamma represents an attenuation coefficient, 0 is more than or equal to α and is less than or equal to 1, and gamma is more than or equal to 0 and is less than or equal to 1, updating the position feature vector S to be the position feature vector S ', and updating the action A to be the action A '.
e. And repeating steps c to d until all positions in the power distribution network are traversed.
In the process, when the intelligent agent finds the fault point, r is 1, if the fault point is not found, r is-1, and in other states, r is 0.
One or more embodiments of the present invention may have the following advantages over the prior art: the intelligent Agent (Agent) is utilized to patrol the power distribution network, and machine training data of the intelligent Agent is in units of each fault event in the power distribution network, so that machine training deviation caused by mixing of different power distribution network data is avoided. .
Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the principles of the invention and not to limit the invention. In the drawings:
FIG. 1 is a schematic diagram of a power distribution network fault location process of the present invention;
FIG. 2 is a schematic flow chart of the present invention for reinforcement learning of an agent;
FIG. 3 is a schematic diagram of a power distribution network topology according to an embodiment of the present invention;
FIG. 4 is an agent decision representation of an embodiment of the present invention;
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail below with reference to the accompanying drawings.
Fig. 3 is a schematic diagram of a power distribution network topology according to an embodiment of the invention, and fig. 4 is an agent decision representation intent according to an embodiment of the invention; the method is described below with reference to fig. 3 and 4.
In the embodiment, only the maximum amplitude of the zero sequence current of the front detection point and the rear detection point is taken as a characteristic.
According to the distribution network topology structure shown in fig. 3, 5000 groups of fault event data are selected by using the intelligent machine training method in the invention to perform Q-Learning training on the decision table of the intelligent agent, so as to obtain the intelligent agent decision table for the distribution network topology structure shown in fig. 4.
When a fault occurs in the power distribution network, the power distribution network fault positioning system of the embodiment is triggered.
First, the agent is configured at the location S1, and the agent looks up the decision table according to the location feature vector S1 at the location S1 and the location feature vector S' in the memory to obtain the result to perform action a 1. The agent stores the location feature vector S1 in memory and proceeds to the next location S2.
The agent looks up the decision table according to the location feature vector S2 at the location of S2 and the location feature vector S' in memory to get the result to execute action a 1. The agent stores the location feature vector S2 in memory and proceeds to the next location S3.
The agent looks up the decision table according to the location feature vector S3 at the location of S3 and the location feature vector S' in memory to get the result to execute action a 1. The agent stores the location feature vector S3 in memory and proceeds to the next location S4.
The agent looks up the decision table according to the location feature vector S4 at the location of S4 and the location feature vector S' in memory to get the result to execute action a 3. The agent selects one of the location feature vector S3 and the location feature vector S4 for storage and proceeds to the next location S5.
The agent looks up the decision table according to the location feature vector S5 at the location of S5 and the location feature vector S' in memory to get the result to execute action a 1. The agent stores the location feature vector S5 in memory and proceeds to the next location S6.
The agent looks up the decision table according to the location feature vector S6 at the location of S6 and the location feature vector S' in memory to get the result to execute action a 1. The agent stores the location feature vector S6 in memory and proceeds to the next location S7.
The agent looks up the decision table according to the location feature vector S7 at the location of S7 and the location feature vector S' in memory to get the result to execute action a 2. The agent keeps the memory data from updating and proceeds to the next location s 8.
The agent looks up the decision table according to the location feature vector S8 at the location of S8 and the location feature vector S' in memory to get the result to execute action a 2. And the intelligent agent keeps the data of the memory from being updated, quits the positioning judgment and outputs the physical position corresponding to the position characteristic vector S' in the memory.
According to the judging process, the intelligent agent can accurately judge that the fault position is at the position s6 by traversing the whole power distribution network topology.
The above description is only an embodiment of the present invention, and the protection scope of the present invention is not limited thereto, and any person skilled in the art should modify or replace the present invention within the technical specification of the present invention.
Claims (7)
1. A power distribution network fault positioning method based on an intelligent agent is characterized by comprising the following steps:
when a power distribution network fault occurs, firstly, initializing the initial position of an intelligent agent in the power distribution network;
the intelligent agent performs inspection according to the power distribution direction in the topological structure of the power distribution network, the inspection needs to sequentially pass through the interval position between every two monitoring terminals along one power distribution line, and when the intelligent agent moves to the tail end of one power distribution line, the intelligent agent randomly selects the starting points of other lines which are not inspected to continue inspection; when the intelligent agent moves to a new position, acquiring a feature vector of the position; the characteristic vector is formed by a plurality of characteristic values of the recording waves collected by the front monitoring terminal and the rear monitoring terminal of the intelligent agent in the power distribution direction;
determining how the agent processes the feature vector of the current position and the feature vector of the internal storage part of the agent by utilizing a decision table in the agent according to the feature vector of the current position and the feature vector stored in the agent; selecting the action type with the largest weight value from the action types corresponding to the two feature vectors in the decision table for operation; the action type is as follows: an action a1 of writing the position feature vector of the current position into the position feature vector storage unit to replace the original data; an action a2 of retaining the original data of the position feature vector storage unit and discarding the position feature vector of the current position; act a3, randomly selecting one of two data to write into the location feature vector storage;
and traversing each position in the power distribution network by the intelligent agent, wherein after the intelligent agent traverses all the positions, the position corresponding to the characteristic vector stored in the intelligent agent is the position of the fault point.
2. The power distribution network fault location method according to claim 1, wherein the feature vector of the current position is a feature vector formed by feature values acquired by two monitoring terminals in front and at back of the current position in the power distribution direction.
3. The power distribution network fault location method of claim 2, wherein the characteristics collected by the monitoring terminal include, but are not limited to, zero sequence current maximum amplitude, power maximum amplitude, fault region average power, and fault region current wavelet energy entropy.
4. The method of claim 1, wherein initializing the agent location is setting a start location of the agent to a distribution line start point.
5. The distribution network fault location method of claim 1, wherein the agent randomly jumps to the starting location of other distribution lines that are not traversed if the agent moves to the end of the distribution line while traversing the distribution network location.
6. A training method of an intelligent agent decision table for power distribution network fault location is characterized by comprising the following steps:
initializing all weight values Q (S, A) in a decision table;
the following process is repeated for each set of fault event data:
a. initializing a position feature vector S of the agent;
b. selecting an action A by adopting an epsilon-Greedy strategy according to the position characteristic vector S and the weight value Q (S,);
c. updating data in a position feature vector memory according to the selected action A, giving a reward r to the agent, enabling the agent to enter the next position, and simultaneously obtaining a position state vector S';
d. according to the position feature vector S ' and the weight value Q (S ',), selecting an action A ' by adopting an epsilon-Greedy strategy, updating Q (S, A) into (1- α) Q (S, A) + α [ r + gamma Q (S ', A ') ], wherein α represents a learning rate, gamma represents an attenuation coefficient, the attenuation coefficient is more than or equal to 0 and less than or equal to α and less than or equal to 1, and the position feature vector S is updated into a position feature vector S ' and the action A is updated into the action A ';
e. and repeating steps c to d until all positions in the power distribution network are traversed.
7. An intelligent agent-based power distribution network fault location system, characterized in that the system uses the location method of any one of claims 1-5 to locate the fault occurrence point.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711403320.1A CN108152676B (en) | 2017-12-22 | 2017-12-22 | Power distribution network fault positioning method and system based on intelligent agent |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711403320.1A CN108152676B (en) | 2017-12-22 | 2017-12-22 | Power distribution network fault positioning method and system based on intelligent agent |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108152676A CN108152676A (en) | 2018-06-12 |
| CN108152676B true CN108152676B (en) | 2020-02-11 |
Family
ID=62465239
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711403320.1A Active CN108152676B (en) | 2017-12-22 | 2017-12-22 | Power distribution network fault positioning method and system based on intelligent agent |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108152676B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109245069A (en) * | 2018-10-18 | 2019-01-18 | 南京我的电气科技有限公司 | Road control device is drawn in a kind of power distribution network small current neutral grounding based on machine learning automatically |
| CN109507527B (en) * | 2018-11-06 | 2021-03-19 | 国网天津市电力公司电力科学研究院 | Fault positioning method for power distribution network containing distributed power supply of SNP system based on synaptic plasticity |
| CN110456231B (en) * | 2019-08-29 | 2020-07-28 | 南方电网科学研究院有限责任公司 | Power distribution network multiple fault diagnosis method under non-sound fault information |
| CN110988601A (en) * | 2019-12-24 | 2020-04-10 | 云南电力技术有限责任公司 | Distribution network branch line fault positioning method and device |
| CN113486873B (en) * | 2021-09-07 | 2021-12-07 | 南通高精数科机械有限公司 | Transformer substation equipment inspection method and system based on big data and artificial intelligence |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102411117B (en) * | 2011-08-09 | 2014-07-02 | 中国电力科学研究院 | Method for positioning short-circuit fault of distribution network based on distributed intelligent multi-calibration |
| CN102508103B (en) * | 2011-11-11 | 2014-07-23 | 中国电力科学研究院 | Distribution network hidden short-circuit fault risk early-warning and positioning method based on distributed intelligent agents |
| CN102707194B (en) * | 2012-05-17 | 2015-03-25 | 中国电力科学研究院 | Power distribution network broken line fault location method |
| CN102841582B (en) * | 2012-08-08 | 2015-11-18 | 中国电力科学研究院 | A kind of power distribution network self-healing control system and its implementation |
| CN105515045A (en) * | 2015-12-25 | 2016-04-20 | 国家电网公司 | Multi agent-based power transmission and distribution network and distributed type supply source coordinated control system and method |
-
2017
- 2017-12-22 CN CN201711403320.1A patent/CN108152676B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN108152676A (en) | 2018-06-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108152676B (en) | Power distribution network fault positioning method and system based on intelligent agent | |
| CN110082640B (en) | Distribution network single-phase earth fault identification method based on long-time memory network | |
| EP3422518B1 (en) | A method for recognizing contingencies in a power supply network | |
| CN108562821B (en) | Method and system for determining single-phase earth fault line selection of power distribution network based on Softmax | |
| CN109324241B (en) | Corrosion diagnosis early warning method and system for transformer substation grounding grid | |
| CN108414896B (en) | Power grid fault diagnosis method | |
| CN111064180B (en) | Medium-voltage distribution network topology detection and identification method based on AMI (advanced mechanical arm) power flow matching | |
| CN113447766A (en) | Method, device, equipment and storage medium for detecting high-resistance ground fault | |
| CN112465124B (en) | Twin depth space-time neural network model acquisition/fault diagnosis method and device | |
| CN110794254B (en) | Power distribution network fault prediction method and system based on reinforcement learning | |
| Sun et al. | A novel life prediction method for railway safety relays using degradation parameters | |
| CN216848010U (en) | Cable partial discharge online monitoring device for edge calculation | |
| CN105425105A (en) | Method and system for comprehensive positioning of fault line of power grid | |
| CN106443380A (en) | Distribution cable partial discharge signal identifying method and apparatus | |
| Zhu et al. | Faulty line identification method based on bayesian optimization for distribution network | |
| Moloi et al. | High impedance fault classification and localization method for power distribution network | |
| CN113657644B (en) | False data detection method, device, equipment and storage medium in smart power grid | |
| CN112710923B (en) | Data-driven single-phase earth fault line selection method based on post-fault steady-state information | |
| CN113945862A (en) | Method, device and equipment for identifying high-resistance grounding fault of power distribution network | |
| CN105656577A (en) | Clustering method and device for channel impulse response | |
| CN106646106B (en) | Electric network fault detection method based on outlier's detection technology | |
| CN113688857A (en) | Method for detecting foreign matters in power inspection image based on generation countermeasure network | |
| CN117590172A (en) | Partial discharge acoustic-electric combined positioning method, device and equipment applied to transformer | |
| CN111898446A (en) | Single-phase earth fault studying and judging method based on multi-algorithm normalization analysis | |
| Yan et al. | Fault location for 10kV distribution line based on traveling wave-ANN theory |
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 |