CN115982616A - Relay fault detection and analysis system and method based on edge calculation - Google Patents
Relay fault detection and analysis system and method based on edge calculation Download PDFInfo
- Publication number
- CN115982616A CN115982616A CN202211501769.2A CN202211501769A CN115982616A CN 115982616 A CN115982616 A CN 115982616A CN 202211501769 A CN202211501769 A CN 202211501769A CN 115982616 A CN115982616 A CN 115982616A
- Authority
- CN
- China
- Prior art keywords
- fault
- edge
- data
- module
- relay
- 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
Landscapes
- Testing And Monitoring For Control Systems (AREA)
Abstract
The invention relates to a relay fault detection and analysis system and method based on edge calculation. Compared with the traditional relay fault detection and analysis system, the fault diagnosis and processing process is closer to a data source due to the introduction of edge calculation, the end-to-end processing time delay is effectively reduced, meanwhile, the pressure of mass data on network transmission and centralized data storage is avoided, and the system performance and the fault response speed are obviously improved; the edge computing equipment is used for desensitizing relay fault data of a plurality of working areas, so that the safety of local data can be ensured, the function of the data in optimizing a global comprehensive model can be fully exerted, and a more complete relay fault analysis and prevention guidance scheme is provided.
Description
Technical Field
The invention belongs to the technical field of component commonality, and particularly relates to a relay fault detection and analysis system and method based on edge calculation.
Background
The relay is an indispensable component in an electrical system, is mainly responsible for circuit conversion, safety protection and the like, and is a key for ensuring the reliable operation of the whole electrical system. In the fields of communication, navigation, aviation and the like, the relay has very high reliability requirements, and once the relay breaks down, serious consequences and great economic loss are often caused. At present, fault detection and analysis of relays mostly occur after faults occur, and the problems of certain time delay in fault alarming, diagnosis and control are solved.
Disclosure of Invention
The invention aims to overcome the defects of the prior art, and provides a relay fault detection and analysis system and method based on edge calculation, which can acquire fault data by utilizing various sensors installed at a fault detection end according to the types and reasons of relay faults, and meanwhile, introduce edge calculation equipment to deploy a deep learning neural network to edge nodes to realize rapid diagnosis and processing of the faults.
The technical problem to be solved by the invention is realized by adopting the following technical scheme:
a relay fault detection and analysis system based on edge calculation comprises a fault detection end, edge equipment, a data transmission module and a server end, wherein the fault detection end is connected with the edge equipment, the edge equipment is bidirectionally connected with the data transmission module, the data transmission module is bidirectionally connected with the server end,
the fault detection end is used for collecting and processing relay fault data of each working area;
the edge device is used for receiving relay fault data, training a fault model of a corresponding working area at the edge side, giving an alarm and responding to a fault condition in real time, and displaying the fault condition;
the data transmission module is used for transmitting the equipment between the edge equipment and the server side;
the server side is used for receiving the fault models, desensitizes each fault model by adopting a federal learning method, evaluates and updates the models, and finally integrates the models into a combined model and sends the combined model to the computing modules of each edge device.
And the fault detection end comprises a signal acquisition device and a data processing module, wherein the output end of the signal acquisition device is connected with the input end of the data processing module.
And, signal pickup assembly includes coil current sensor, coil voltage sensor, contact current sensor, contact voltage sensor, temperature sensor, humidity transducer and baroceptor for gather relay main performance parameter and environmental status signal.
And the data processing module comprises a data acquisition card and a mainboard, the data acquisition card and the mainboard are connected in a fastening and stacking mode, the data acquisition card is used for acquiring data sent by the signal acquisition device, and the mainboard is used for carrying out A/D conversion processing on the acquired data.
And the edge device comprises a calculation module and a front-end module, the calculation module is connected with the front-end module, the calculation module is used for receiving the relay fault data set of the working area where the calculation module is located, training the fault model of the corresponding working area at the edge side, giving an alarm and responding to the fault condition in real time, and the front-end module is used for displaying a visual interface.
An analysis method of a relay fault detection analysis system based on edge calculation comprises the following steps:
step 1: at the fault detection end, the signal acquisition device acquires main performance parameters and environmental state signals of the relay and transmits the main performance parameters and the environmental state signals to the data processing module;
and 2, step: the data processing module performs A/D conversion on the acquired data and sends the data to a computing module in the edge device;
and step 3: the calculation module of each edge device receives the relay fault data set of the working area, trains the fault model of the corresponding working area at the edge side, gives an alarm and responds to the fault condition in real time, and displays the fault condition in a visual interface through the front-end module; meanwhile, the calculation module uploads the trained fault model to a server side through a data transmission module;
and 4, step 4: the server side receives the relay fault models sent by the edge devices, desensitizes each fault model by adopting a federal learning method, evaluates and updates the models, and finally integrates the models into a combined model and sends the combined model to the computing module of each edge device;
and 5: and after the calculation modules of the edge devices receive the combined model, performing a new iteration on the models to form a more optimized fault diagnosis model.
The invention has the advantages and positive effects that:
compared with the traditional relay fault detection and analysis system, the fault diagnosis and processing process is closer to a data source due to the introduction of edge calculation, the end-to-end processing time delay is effectively reduced, meanwhile, the pressure of mass data on network transmission and centralized data storage is avoided, and the system performance and the fault response speed are obviously improved; the edge computing equipment is used for desensitizing relay fault data of a plurality of working areas, so that the safety of local data can be ensured, the function of the data in optimizing a global comprehensive model can be fully exerted, and a more complete relay fault analysis and prevention guidance scheme is provided.
Drawings
FIG. 1 is a system connection diagram of the present invention.
Detailed Description
The present invention is further described in detail below with reference to the accompanying drawings.
A relay fault detection and analysis system based on edge calculation is shown in figure 1 and comprises a fault detection end, edge equipment, a data transmission module and a server end, wherein the fault detection end is connected with the edge equipment, the edge equipment is connected with the data transmission module in a bidirectional mode, and the data transmission module is connected with the server end in a bidirectional mode.
The fault detection terminal is used for collecting and processing relay fault data of each working area; the fault detection end comprises a signal acquisition device and a data processing module, wherein the output end of the signal acquisition device is connected with the input end of the data processing module.
The signal acquisition device comprises a coil current sensor, a coil voltage sensor, a contact current sensor, a contact voltage sensor, a temperature sensor, a humidity sensor and an air pressure sensor and is used for acquiring main performance parameters of the relay and environmental state signals.
The data processing module comprises a PC104 data acquisition card and a PC104 mainboard which are connected in a fastening and stacking mode, wherein the PC104 data acquisition card is used for acquiring data sent by various sensors, and the PC104 mainboard runs a VxWork operating system and is used for carrying out A/D conversion processing on the acquired data.
The edge device is used for receiving relay fault data, training a fault model of a corresponding working area on the edge side, giving an alarm and responding to the fault condition in real time, and displaying the fault condition at the same time. In the system, a plurality of Jetson Nano devices are selected as edge devices and are widely deployed in each working area. The edge device comprises a computing module and a front-end module, the computing module provides basic computing power by utilizing the higher performance and energy efficiency of the Jetson Nano device, receives real-time data transmitted by a fault detection end, trains a neural network of a relay fault model in a working area with a certain scale, the front-end module is provided with a visual interface, the port is 5001, the current state of the edge device can be displayed, the current state includes whether the edge device is started or not, whether the edge device is offline, whether the edge device is calculating or not and the like, and meanwhile, the performance index of the neural network currently operated by the edge device is dynamically displayed in real time.
The data transmission module is used for transmitting equipment between the edge equipment and the server side; the system uses zeroTie to build a Virtual Local Area Network (VLAN), the IP ratio of the local area network is 10.147.20.X, and the equipment connection uses a TCP protocol.
The server side is used for receiving the fault models, desensitizes each fault model by adopting a federal learning method, evaluates and updates the models, and finally integrates the models into a combined model and sends the combined model to the computing modules of each edge device. In the system, jetson XavierNX is used as a server end, the strong operation performance of the system can provide accelerated AI calculation of up to 21TOPS for the edge, a plurality of neural networks are operated in parallel, and a plurality of data from high-resolution sensors are processed.
An analysis method of a relay fault detection and analysis system based on edge calculation comprises the following steps:
step 1: at the fault detection end, the signal acquisition device acquires main performance parameters and environmental state signals of the relay and transmits the main performance parameters and the environmental state signals to the data processing module;
step 2: the data processing module performs A/D conversion on the acquired data and sends the data to a computing module in the edge device;
and 3, step 3: the calculation module of each edge device receives the relay fault data set of the working area, trains the fault model of the corresponding working area at the edge side, gives an alarm and responds to the fault condition in real time, and displays the fault condition in a visual interface through the front-end module; meanwhile, the calculation module uploads the trained fault model to a server side through a data transmission module;
and 4, step 4: the server side receives the relay fault models sent by the edge devices, desensitizes each fault model by adopting a federal learning method, evaluates and updates the models, and finally integrates the models into a combined model and sends the combined model to the computing module of each edge device;
and 5: and after the calculation modules of the edge devices receive the combined model, carrying out a new iteration on the models of the edge devices to form a more optimized fault diagnosis model.
It should be emphasized that the embodiments described herein are illustrative rather than restrictive, and thus the present invention is not limited to the embodiments described in the detailed description, but also includes other embodiments that can be derived from the technical solutions of the present invention by those skilled in the art.
Claims (6)
1. The utility model provides a relay fault detection analytic system based on edge calculation which characterized in that: comprises a fault detection end, edge equipment, a data transmission module and a server end, wherein the fault detection end is connected with the edge equipment, the edge equipment is bidirectionally connected with the data transmission module, the data transmission module is bidirectionally connected with the server end,
the fault detection end is used for collecting and processing relay fault data of each working area;
the edge device is used for receiving relay fault data, training a fault model of a corresponding working area at the edge side, giving an alarm and responding to a fault condition in real time, and displaying the fault condition;
the data transmission module is used for transmitting the equipment between the edge equipment and the server side;
and the server side is used for receiving the fault models, desensitizing each fault model by adopting a federal learning method, evaluating and updating the models, and finally synthesizing the models into a combined model and transmitting the combined model to the computing modules of each edge device.
2. The relay fault detection and analysis system based on edge calculation according to claim 1, wherein: the fault detection end comprises a signal acquisition device and a data processing module, wherein the output end of the signal acquisition device is connected with the input end of the data processing module.
3. The relay fault detection and analysis system based on edge calculation according to claim 2, characterized in that: the signal acquisition device comprises a coil current sensor, a coil voltage sensor, a contact current sensor, a contact voltage sensor, a temperature sensor, a humidity sensor and an air pressure sensor and is used for acquiring main performance parameters of the relay and environmental state signals.
4. The relay fault detection and analysis system based on edge calculation according to claim 2, characterized in that: the data processing module comprises a data acquisition card and a mainboard, the data acquisition card and the mainboard are connected in a fastening and stacking mode, the data acquisition card is used for acquiring data sent by the signal acquisition device, and the mainboard is used for carrying out A/D conversion processing on the acquired data.
5. The relay fault detection and analysis system based on edge calculation of claim 1, wherein: the edge device comprises a calculation module and a front-end module, the calculation module is connected with the front-end module, the calculation module is used for receiving a relay fault data set of a working area where the calculation module is located, training a fault model of the corresponding working area on the edge side, giving an alarm and responding to a fault condition in real time, and the front-end module is used for displaying a visual interface.
6. An analysis method of the relay fault detection and analysis system based on the edge calculation according to any one of claims 1to 5, characterized in that: the method comprises the following steps:
step 1: at the fault detection end, the signal acquisition device acquires main performance parameters and environmental state signals of the relay and transmits the main performance parameters and the environmental state signals to the data processing module;
step 2: the data processing module performs A/D conversion on the acquired data and sends the data to a computing module in the edge device;
and step 3: the calculation module of each edge device receives the relay fault data set of the working area, trains the fault model of the corresponding working area at the edge side, gives an alarm and responds to the fault condition in real time, and displays the fault condition in a visual interface through the front-end module; meanwhile, the calculation module uploads the trained fault model to a server side through a data transmission module;
and 4, step 4: the server side receives the relay fault models sent by the edge devices, desensitizes each fault model by adopting a federal learning method, evaluates and updates the models, and finally synthesizes the models into a combined model and sends the combined model to the computing module of each edge device;
and 5: and after the calculation modules of the edge devices receive the combined model, performing a new iteration on the models to form a more optimized fault diagnosis model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211501769.2A CN115982616A (en) | 2022-11-28 | 2022-11-28 | Relay fault detection and analysis system and method based on edge calculation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211501769.2A CN115982616A (en) | 2022-11-28 | 2022-11-28 | Relay fault detection and analysis system and method based on edge calculation |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115982616A true CN115982616A (en) | 2023-04-18 |
Family
ID=85972836
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211501769.2A Pending CN115982616A (en) | 2022-11-28 | 2022-11-28 | Relay fault detection and analysis system and method based on edge calculation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115982616A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118152970A (en) * | 2024-05-11 | 2024-06-07 | 国网山东省电力公司烟台供电公司 | Equipment state trend sensing method based on edge calculation algorithm |
-
2022
- 2022-11-28 CN CN202211501769.2A patent/CN115982616A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118152970A (en) * | 2024-05-11 | 2024-06-07 | 国网山东省电力公司烟台供电公司 | Equipment state trend sensing method based on edge calculation algorithm |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103823433B (en) | Method for realizing relay protection equipment on-line monitoring by use of communication process analysis | |
| CN108072806A (en) | Track circuit fault diagnosis system and method | |
| CN102841296A (en) | Online monitoring system and method for partial discharge of intelligent switch cabinet based on ultra-high frequency detection | |
| CN202326011U (en) | State-monitoring and fault-diagnosis system of wind-power set | |
| CN111884347B (en) | Power data centralized control system for multi-source power information fusion | |
| CN115982616A (en) | Relay fault detection and analysis system and method based on edge calculation | |
| CN101840629B (en) | Monitoring system and method of aerobridge | |
| CN111327468A (en) | Operation method and system for edge computing platform of power system | |
| CN112697267A (en) | Abnormal vibration detection device for industrial equipment | |
| CN114167315A (en) | Intelligent online monitoring system and method for transformer | |
| CN112614130A (en) | Unmanned aerial vehicle power transmission line insulator fault detection method based on 5G transmission and YOLOv3 | |
| CN114204677B (en) | Remote monitoring method and system for power generation equipment | |
| CN100524367C (en) | Subsway rail car pulling circuit fault diagnosis system based on wave form identification | |
| CN113460884A (en) | Remote fault diagnosis system for crane | |
| CN100585366C (en) | A kind of on-line monitoring of synchronous gear box initial failure and diagnostic device | |
| CN111722612A (en) | Electric power oil-shortage monitoring and predicting system | |
| CN116512264A (en) | Remote robot inspection system | |
| CN114025249B (en) | Substation mass data monitoring device and method based on Internet of things technology | |
| CN105424387A (en) | Steering gear real-time remote intelligent monitoring system | |
| CN216434701U (en) | Integrated 5G's online fortune dimension monitoring system | |
| CN212518914U (en) | Fault detection system of photovoltaic module | |
| CN104392102A (en) | Automatic extraction and calculation method for online monitoring indexes of relaying protection equipment | |
| CN104215321A (en) | Circuit breaker vibration monitor system and method based on optical fiber sensors | |
| CN111208803A (en) | Tank gun control system fault online detection device and method | |
| CN115291059B (en) | Voiceprint monitoring system of power 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 |