CN111190072A - Centralized meter reading system diagnosis model establishing method, fault diagnosis method and fault diagnosis device - Google Patents
Centralized meter reading system diagnosis model establishing method, fault diagnosis method and fault diagnosis device Download PDFInfo
- Publication number
- CN111190072A CN111190072A CN201911267803.2A CN201911267803A CN111190072A CN 111190072 A CN111190072 A CN 111190072A CN 201911267803 A CN201911267803 A CN 201911267803A CN 111190072 A CN111190072 A CN 111190072A
- Authority
- CN
- China
- Prior art keywords
- fault
- meter reading
- reading system
- diagnosis model
- centralized
- 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.)
- Pending
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
- G01R31/081—Locating faults in cables, transmission lines, or networks according to type of conductors
- G01R31/086—Locating faults in cables, transmission lines, or networks according to type of conductors in power transmission or distribution networks, i.e. with interconnected conductors
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Remote Monitoring And Control Of Power-Distribution Networks (AREA)
Abstract
The invention provides a centralized meter reading system diagnosis model establishing method, a fault diagnosis method and a fault diagnosis device. Because the number of power users is large and the system topology is complex, the centralized meter reading system diagnosis model can obtain the position of fault equipment according to fault information, and the workload of manual operation and maintenance for removing faults is reduced.
Description
Technical Field
The invention relates to the field of power distribution network equipment fault diagnosis, and provides a method for establishing a diagnosis model of a fault position of a low-voltage centralized meter reading system.
Background
The electric power user electricity meter reading is to read, record and account the electric energy meter of the user in the power distribution network and form an account type charging flow, and is a main mode for the user to master the electricity utilization condition of each user in the power distribution network in the power system. With the increasing attention paid to the automation transformation of the power distribution network in the construction of the power distribution network, the collection of various data in the power distribution network is more and more intelligent, and the meter reading of a power user is changed from a manual mode to a computer remote automatic meter reading mode by virtue of the development of a computer technology and a communication technology. The central metering automation system of the power master station communicates with meter reading concentrators in different areas through wireless communication modes such as 4G and 3G, electricity utilization data of electric meters of all power users in a distribution area where the concentrators are responsible are read from the concentrators, and the data are arranged and then sent to a power data cloud platform to wait for further processing or use.
The low-voltage centralized meter reading system is composed of a master station, a communication channel and field devices, wherein the field devices mainly comprise concentrators, collectors and electric meters, and communication modes among the devices comprise RS485 wired communication, low-voltage carrier communication, 4G wireless communication and the like. The electric meters of all users send the electricity utilization data of the users to the collectors which are responsible for data acquisition of the plurality of electric meters, the collectors send the data to the concentrators which are responsible for one distribution area, and finally the concentrators send the data to the master station which is responsible for one distribution network.
Because the number of power consumers is huge, each power consumer is provided with a separate electric meter, and a collector, a concentrator and other equipment are considered. In a commercial power distribution area, the equipment of the low-voltage meter reading system can reach the level of millions, and the equipment is often widely distributed and is remote in installation position, so that the workload of the traditional operation and maintenance inspection mode is huge.
Disclosure of Invention
The invention aims to provide a centralized meter reading system diagnosis model establishing method, a fault diagnosis method and a fault diagnosis device, which aim to solve the technical problems.
The embodiment of the invention is realized by the following steps: a centralized meter reading system diagnosis model building method comprises the following steps:
acquiring system topology information which comprises equipment information and equipment link information;
generating a fault event according to the system topology information, and generating a fault section feature vector of the fault event;
generating a training sample according to the system topology information and the fault section characteristic vector;
and inputting the training sample into a defined neural network to obtain the diagnosis model of the meter reading system.
A fault diagnosis method comprises the steps of diagnosing a centralized meter reading system by using a centralized meter reading system diagnosis model obtained by a centralized meter reading system diagnosis model establishing method, inputting a test sample into the centralized meter reading system diagnosis model, and obtaining fault position information.
A fault diagnosis apparatus comprising:
the data acquisition module is used for acquiring system topology information which comprises equipment information and equipment link information, generating a fault event according to the system topology information, generating a fault section characteristic vector of the fault event, and generating a training sample according to the system topology information and the fault section characteristic vector;
the model establishing module is used for inputting the training sample into a defined neural network to obtain the diagnosis model of the meter reading system;
and the fault information output module is used for generating a test sample according to the system topology information and the fault section characteristic vector, inputting the test sample into a centralized reading system diagnosis model to obtain fault position information and outputting the fault position information.
The embodiment of the invention has the beneficial effects that: the method comprises the steps of obtaining system topology information of a low-voltage meter reading system, generating a fault event according to the system topology information, further generating a fault section characteristic vector, generating a training sample according to the fault section characteristic vector and the system topology information, and inputting the training sample into a defined neural network to obtain a meter reading system diagnosis model. Because the number of power users is large, the system topology is complex, and the position of the fault equipment can be obtained by the centralized meter reading system diagnosis model according to the fault information, the workload of manual operation and maintenance for removing faults is reduced.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present invention and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings can be obtained according to the drawings without inventive efforts.
Fig. 1 is a flowchart of a method for establishing a centralized meter reading system diagnostic model according to an embodiment of the present invention;
fig. 2 is a topological diagram of a centralized meter reading system according to an embodiment of the present invention.
Detailed Description
Referring to fig. 1, an embodiment of the present invention provides a method for establishing a centralized meter reading system diagnostic model, including the following steps:
acquiring system topology information which comprises equipment information and equipment link information;
generating a fault event according to the system topology information, and generating a fault section feature vector of the fault event;
generating a training sample according to the system topology information and the fault section characteristic vector;
and inputting the training sample into a defined neural network to obtain the diagnosis model of the meter reading system.
The fault section feature vector calculation method comprises the following steps:
fault section characteristic vector X ═ H, V, phi, psi ]
In the formula: h represents an ammeter state vector, common hardware fault types of the ammeter comprise flying, falling and forward and reverse creeping, the state of the ammeter is judged according to real-time monitoring of the ammeter and a fault identification model, and any element H in the state vector HiRepresenting the state of the electric meter numbered i in the system, and if the electric meter is in failure, Hi1, otherwise Hi＝0；
V generationState vector of electricity information of meter master station user, wherein any element ViThe state that the master station acquires the data of the number i electric meter is shown, and if the master station cannot acquire the data of the number i electric meter, Vi1, otherwise Vi＝0；
Phi represents field equipment communication vector, wherein any element phijThe state that the field equipment receives data from the lower layer equipment except the ammeter is shown, when the lower layer of the equipment number j has a fault, the quantity of the data received by the equipment is reduced, and the phi at the momentj1, elsej＝0；
Ψ represents a link listening state vector, and by monitoring the data traffic of a communication line, when the communication line between the device i and the device j fails or the device itself fails, the data traffic through the line is greatly reduced, and Ψij1, otherwise Ψij＝0。
In a specific implementation, the topology of a low-voltage meter reading system of a distribution network in a certain area is shown in fig. 2, and the innermost layer is a master station layer. The outer layer is a concentrator layer, the power distribution network in the whole area is divided into 2 distribution areas, and each distribution area corresponds to one concentrator. The secondary outer layer and the outermost layer are an acquisition layer and an electric energy surface layer, 5 electric energy meters are arranged in the platform area 1, the electric energy meters M1 and M2 are connected with a collector C1, the electric energy meters M3 and M4 are connected with a collector C2, the electric energy meters M5 are directly connected with a concentrator J1, and the collectors C1 and C2 are connected with the concentrator J1. The electric energy meters M6 and M7 in the station area 2 are directly connected with the concentrator J2.
According to the topology of fig. 2, the surge generates events where each fault occurs. Taking the example of the failure of the line L8, the direct effect is that data passing through the link is reduced or zeroed, the indirect effect is that the master station cannot acquire user data of M1 and M2, and the acquisition of information by the J1 is reduced.
And acquiring a fault section characteristic vector and a fault position characteristic vector corresponding to the fault event by combining the user electric energy data flow path. The fault section characteristic quantities comprise 7 electric energy meter state quantities, 7 main station user electricity utilization information state quantities, 4 field communication state quantities, 11 link monitoring state quantities and 29 total fault section characteristic quantities. The number of output positions is 22, which is consistent with the sum of the numbers of field devices and communication channels, and a feature vector is extracted from the fault of the line L8, so that:
X＝[0,0,0,0,0,0,0,1,1,0,0,0,0,0,0,0,1,0,0,0,0,0,0,0,0,1,0,0,0]
the neural network is a Deep Belief Network (DBN), the number of the neural network layers is set to be 3, the number of hidden neurons is 50, and the rest parameters are shown in a table 1 for training.
TABLE 1 DBN parameter settings
The bottom layer of the deep confidence network (DBN) is formed by stacking a plurality of Boltzmann machines (RBMs), and the top layer is a BP neural network.
Compared with the traditional neural network, the unsupervised pre-training realizes the cleaning and classification of the fault section feature vectors in advance, and avoids the situation that the subsequent supervised learning falls into the local optimal solution.
The method for establishing the centralized reading system diagnosis model further comprises a neural network testing method, after the centralized reading system diagnosis model is obtained, a testing sample is generated according to the system topology information and the fault section characteristic vector, the testing sample is input into the centralized reading system diagnosis model, a fault position characteristic vector Y is output, each element Yi in the fault position characteristic vector Y corresponds to the position of each device in the system, if the position has a fault, Yi is 1, otherwise Yi is 0.
In a specific implementation:
Y＝[0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,0,0,0]
after the deep belief network training is completed, the model is tested by using the test sample, and the test result is shown in table 2.
Table 2 model test results
The state value of each neuron of the output layer is the probability of the occurrence of the corresponding fault, the output maximum value is used as a diagnosis result, and the result shows that the model can accurately diagnose the fault position.
In the embodiment, the system topology information of the low-voltage meter reading system is obtained, the fault event is generated according to the system topology information, the fault section characteristic vector is further generated, meanwhile, the training sample is generated according to the fault section characteristic vector and the system topology information, the training sample is input into the defined neural network to obtain the meter reading system diagnosis model, the meter reading system diagnosis model can obtain the position of the fault equipment according to the fault information, and the workload of manual operation, maintenance and fault removal is reduced.
The embodiment of the invention also provides a fault diagnosis method, which diagnoses the centralized meter reading system by using the centralized meter reading system diagnosis model obtained by the centralized meter reading system diagnosis model establishing method, and inputs the test sample into the centralized meter reading system diagnosis model to obtain fault position information.
In this embodiment, fault events of different devices or communication channels are obtained according to automatic emergence of system topology, test samples are extracted from the fault events, a centralized copy system diagnosis model is used, a low-voltage centralized copy system fault section feature vector X is input, a fault position feature vector Y is output, state values of neurons in an output layer are analyzed to be probabilities of corresponding faults, and the maximum output value is used as a fault diagnosis result.
An embodiment of the present invention also provides a fault diagnosis apparatus, including:
the data acquisition module is used for acquiring system topology information which comprises equipment information and equipment link information, generating a fault event according to the system topology information, generating a fault section characteristic vector of the fault event, and generating a training sample according to the system topology information and the fault section characteristic vector;
the model establishing module is used for inputting the training sample into a defined neural network to obtain the diagnosis model of the meter reading system;
and the fault information output module is used for generating a test sample according to the system topology information and the fault section characteristic vector, inputting the test sample into a centralized reading system diagnosis model to obtain fault position information and outputting the fault position information.
In the embodiment, a fault diagnosis device is connected to a low-voltage centralized meter reading system, a data acquisition module acquires topological information of the low-voltage centralized meter reading system, a fault event and a fault section characteristic vector are generated according to the topological information of the system, and a model building module obtains a diagnosis model of the centralized meter reading system according to a training sample; the fault information output module can input the test sample into the centralized reading system diagnosis model and finally output fault position information, and the centralized reading system diagnosis model can obtain the position of fault equipment according to the fault information due to the large number of power users and the complex system topology, so that the workload of manual operation and maintenance for fault removal is reduced.
In the description of the embodiments of the present invention, the term "and/or" herein is only one kind of association relationship describing an associated object, and means that there may be three kinds of relationships, for example, a and/or B, and may mean: a exists alone, A and B exist simultaneously, and B exists alone. In addition, the character "/" herein generally indicates that the former and latter related objects are in an "or" relationship.
Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that changes, modifications, substitutions and alterations can be made in these embodiments without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
Claims (8)
1. The method for establishing the centralized meter reading system diagnosis model is characterized by comprising the following steps of:
acquiring system topology information which comprises equipment information and equipment link information;
generating a fault event according to the system topology information, and generating a fault section feature vector of the fault event;
generating a training sample according to the system topology information and the fault section characteristic vector;
and inputting the training sample into a defined neural network to obtain the diagnosis model of the meter reading system.
2. The method for establishing a centralized meter reading system diagnostic model according to claim 1, wherein the method for calculating the fault section feature vector comprises the following steps:
fault section characteristic vector X ═ H, V, phi, psi ]
In the formula: h represents the state vector of the electric meter, and any element H in the state vector HiRepresenting the state of the electric meter numbered i in the system, and if the electric meter is in failure, Hi1, otherwise Hi＝0；
V represents the state vector of the power consumption information of the master station user, wherein any element ViThe state that the master station acquires the data of the number i electric meter is shown, and if the master station cannot acquire the data of the number i electric meter, Vi1, otherwise Vi＝0；
Phi represents field equipment communication vector, wherein any element phijThe state that the field equipment receives data from the lower layer equipment except the ammeter is shown, when the lower layer of the equipment number j has a fault, the quantity of the data received by the equipment is reduced, and the phi at the momentj1, elsej＝0；
Ψ represents a link listening state vector, and by monitoring the data traffic of a communication line, when the communication line between the device i and the device j fails or the device itself fails, the data traffic through the line is greatly reduced, and Ψij1, otherwise Ψij＝0。
3. The centralized meter reading system diagnosis model establishment method according to claim 1, characterized in that: the neural network is a Deep Belief Network (DBN).
4. The centralized meter reading system diagnosis model establishment method according to claim 3, characterized in that: the bottom layer of the deep confidence network (DBN) is formed by stacking a plurality of Boltzmann machines (RBMs), and the top layer is a BP neural network.
5. The centralized meter reading system diagnosis model establishment method according to claim 1, characterized in that: the method for establishing the centralized meter reading system diagnosis model further comprises a neural network testing method, after the centralized meter reading system diagnosis model is obtained, a test sample is generated according to the system topology information and the fault section characteristic vector, the test sample is input into the centralized meter reading system diagnosis model, and a fault position characteristic vector Y is output.
6. The centralized meter reading system diagnosis model establishment method according to claim 5, characterized in that: each element Y in the fault position feature vector YiCorresponding to the position of each device in the system, if the position fails, Yi1, otherwise Yi＝0。
7. A fault diagnosis method, characterized in that a centralized copy system is diagnosed by using the centralized copy system diagnosis model obtained by the centralized copy system diagnosis model establishment method of any one of claims 1 to 6, and a test sample is input into the centralized copy system diagnosis model to obtain fault location information.
8. A failure diagnosis device characterized by comprising:
the data acquisition module is used for acquiring system topology information which comprises equipment information and equipment link information, generating a fault event according to the system topology information, generating a fault section characteristic vector of the fault event, and generating a training sample according to the system topology information and the fault section characteristic vector;
the model establishing module is used for inputting the training sample into a defined neural network to obtain the diagnosis model of the meter reading system;
and the fault information output module is used for generating a test sample according to the system topology information and the fault section characteristic vector, inputting the test sample into a centralized reading system diagnosis model to obtain fault position information and outputting the fault position information.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911267803.2A CN111190072A (en) | 2019-12-11 | 2019-12-11 | Centralized meter reading system diagnosis model establishing method, fault diagnosis method and fault diagnosis device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911267803.2A CN111190072A (en) | 2019-12-11 | 2019-12-11 | Centralized meter reading system diagnosis model establishing method, fault diagnosis method and fault diagnosis device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111190072A true CN111190072A (en) | 2020-05-22 |
Family
ID=70705769
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911267803.2A Pending CN111190072A (en) | 2019-12-11 | 2019-12-11 | Centralized meter reading system diagnosis model establishing method, fault diagnosis method and fault diagnosis device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111190072A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112382076A (en) * | 2020-11-12 | 2021-02-19 | 贵州电网有限责任公司 | Equipment and operation method for acquiring topological information of district electric energy meter |
| CN112595918A (en) * | 2020-12-26 | 2021-04-02 | 广东电网有限责任公司广州供电局 | Low-voltage meter reading fault detection method and device |
| CN113128117A (en) * | 2021-04-20 | 2021-07-16 | 河南能创电子科技有限公司 | Low-voltage centralized reading, operation and maintenance simulation device based on AI artificial neural network research |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103023719A (en) * | 2012-12-04 | 2013-04-03 | 宁夏隆基宁光仪表有限公司 | Specified transformer acquisition terminal failure diagnosis method based on bayesian network |
| CN110429713A (en) * | 2019-08-27 | 2019-11-08 | 青岛东软载波科技股份有限公司 | A kind of collecting and copying device fault diagnosis system based on terminal |
-
2019
- 2019-12-11 CN CN201911267803.2A patent/CN111190072A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103023719A (en) * | 2012-12-04 | 2013-04-03 | 宁夏隆基宁光仪表有限公司 | Specified transformer acquisition terminal failure diagnosis method based on bayesian network |
| CN110429713A (en) * | 2019-08-27 | 2019-11-08 | 青岛东软载波科技股份有限公司 | A kind of collecting and copying device fault diagnosis system based on terminal |
Non-Patent Citations (1)
| Title |
|---|
| 罗步升等: "基于拓扑解析与深度学习融合的低压集抄***故障诊断方法", 《电测与仪表》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112382076A (en) * | 2020-11-12 | 2021-02-19 | 贵州电网有限责任公司 | Equipment and operation method for acquiring topological information of district electric energy meter |
| CN112382076B (en) * | 2020-11-12 | 2021-08-27 | 贵州电网有限责任公司 | Equipment and operation method for acquiring topological information of district electric energy meter |
| CN112595918A (en) * | 2020-12-26 | 2021-04-02 | 广东电网有限责任公司广州供电局 | Low-voltage meter reading fault detection method and device |
| CN113128117A (en) * | 2021-04-20 | 2021-07-16 | 河南能创电子科技有限公司 | Low-voltage centralized reading, operation and maintenance simulation device based on AI artificial neural network research |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110929918B (en) | 10kV feeder fault prediction method based on CNN and LightGBM | |
| WO2021073462A1 (en) | 10 kv static load model parameter identification method based on similar daily load curves | |
| US8868985B2 (en) | Supervised fault learning using rule-generated samples for machine condition monitoring | |
| CN110829417B (en) | Electric power system transient stability prediction method based on LSTM double-structure model | |
| CN103901882B (en) | A kind of system and method for train dynamics system on-line monitoring fault diagnosis | |
| CN111190072A (en) | Centralized meter reading system diagnosis model establishing method, fault diagnosis method and fault diagnosis device | |
| CN109856299A (en) | A kind of transformer online monitoring differentiation threshold value dynamic setting method, system | |
| CN107340492A (en) | Electric power meter failure analysis methods with scene anticipation are excavated based on big data | |
| CN104052612B (en) | A kind of Fault Identification of telecommunication service and the method and system of positioning | |
| CN104966161B (en) | A kind of power quality recorder data calculation and analysis methods based on gauss hybrid models | |
| CN110703057A (en) | Power equipment partial discharge diagnosis method based on data enhancement and neural network | |
| CN111444169A (en) | Transformer substation electrical equipment state monitoring and diagnosis system and method | |
| CN109490814A (en) | Metering automation terminal fault diagnostic method based on deep learning and Support Vector data description | |
| CN108287327A (en) | Metering automation terminal fault diagnostic method based on Bayes's classification | |
| CN110503977A (en) | A kind of substation equipment audio signal sample analysis system | |
| CN110443481B (en) | Power distribution automation terminal state evaluation system and method based on hybrid K-nearest neighbor algorithm | |
| CN110837532A (en) | Method for detecting electricity stealing behavior of charging pile based on big data platform | |
| CN115453267A (en) | Fault diagnosis system for electric power information system | |
| CN115614292A (en) | Vibration monitoring device and method for vertical water pump unit | |
| CN116467648A (en) | Early monitoring method for nonlinear platform power failure based on Internet of things table | |
| CN116664095B (en) | Security inspection prediction method based on deep learning | |
| CN107016410B (en) | Power information acquisition system method for diagnosing faults and trouble-shooter | |
| CN115545240A (en) | Method, system, equipment and medium for diagnosing abnormal line loss of low-voltage distribution network transformer area | |
| CN117764547A (en) | Photovoltaic string fault diagnosis method and system | |
| CN113131487B (en) | Transformer area identification method and device based on voltage regulation pattern, storage medium and electronic equipment |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200522 |
|
| RJ01 | Rejection of invention patent application after publication |