Artificial intelligence distribution network line fault positioning system based on dual-mode fusion adaptive wireless communication

Abstract

The invention provides an artificial intelligence distribution network line fault positioning system based on dual-mode fusion adaptive wireless communication. The artificial intelligence distribution network line fault positioning system comprises a power grid cloud platform and a plurality of terminal devices deployed in a distribution network line. The terminal equipment comprises a power management module, a signal acquisition module, an artificial intelligence fault algorithm module, a signal processing module and a dual-mode fusion adaptive wireless communication module; the signal acquisition module acquires a power frequency signal, a traveling wave signal, a temperature and humidity signal and a GPS signal; the artificial intelligence fault algorithm module performs fault feature analysis on the power frequency signal and the traveling wave signal data, and determines a fault type and a fault position in combination with a GPS and a double-end traveling wave monitoring principle; and the dual-mode fusion adaptive wireless communication module is combined with LoRa communication and LTE communication to upload data information of the terminal equipment to a power grid cloud platform. According to the invention, the fault occurrence point and the fault type of the distribution network line can be accurately judged, and the fault information is timely, stably and safely uploaded to the cloud platform through dual-mode fusion adaptive wireless communication.

Description

Technical Field

[0001] The present invention relates to the field of power fault detection and the field of the Internet of Things, and in particular to an artificial intelligence distribution network line fault locating system with dual-mode fusion adaptive wireless communication. Background Art

[0002] In the power system, devices with wired connections need to be connected to each other through data cables. When a device in the middle fails, all device information after the device will be lost, resulting in the power grid platform being unable to make timely judgments, thus affecting emergency rescue and maintenance work. Devices that use a single mobile communication technology such as GSM and LTE cannot guarantee stable transmission of fault information when the natural environment is harsh, the infrastructure is weak, and full signal coverage cannot be achieved. Devices that use a single wireless communication technology such as LoRa and ZigBee face low transmission data rates, small transmission data capacity, and the risk of data privacy leakage and data tampering.

[0003] In addition, traditional equipment terminals do not have the ability to judge fault types and locate faults. They often only use sensors to capture physical information of distribution network lines and then send the data to the cloud platform for fault processing. This system has problems such as huge data volume, high power consumption, large network traffic, and high data costs.

[0004] In view of this, the present invention provides an artificial intelligence distribution network line fault location system for dual-mode fusion adaptive wireless communication. Summary of the invention

[0005] The purpose of the present invention is to propose an artificial intelligence distribution network line fault location system with dual-mode fusion adaptive wireless communication, which adopts artificial intelligence algorithm to accurately determine the fault occurrence point and fault type of the distribution network line, and upload the fault information such as fault occurrence time, fault type, longitude and latitude coordinates of the fault occurrence point in a timely, stable and secure manner to the local power grid management cloud platform through dual-mode fusion adaptive wireless communication, so as to improve the efficiency and accuracy of distribution network line fault monitoring.

[0006] To achieve the above-mentioned purpose, the technical solution of the present invention is: an artificial intelligence distribution network line fault location system with dual-mode fusion adaptive wireless communication, comprising a power grid cloud platform and a plurality of terminal devices deployed in the distribution network line; the terminal device comprises a power management module, a signal acquisition module, an artificial intelligence fault algorithm module, a signal processing module and a dual-mode fusion adaptive wireless communication module;

[0007] The power management module is used to supply power to the terminal device;

[0008] The signal acquisition module collects power frequency signals, traveling wave signals, temperature and humidity signals and GPS signals;

[0009] The signal processing module preprocesses and stores the collected original signals, and transmits the processed data to the artificial intelligence fault algorithm module;

[0010] The artificial intelligence fault algorithm module uses artificial intelligence algorithms to analyze and judge fault characteristics of power frequency signals and traveling wave signal data, combines GPS and double-end traveling wave monitoring principles, determines the fault type and fault location, and generates early warning information;

[0011] The dual-mode fusion adaptive wireless communication module combines LoRa communication and LTE communication to build a terminal network, connect to the power grid cloud platform and upload data information of all terminal devices in the terminal network.

[0012] Preferably, the preprocessing uses digital filtering technology to remove noise interference in the original signal and retain useful signal components.

[0013] Preferably, the power management module includes a battery charging and discharging unit, a CT power supply unit and a solar power supply unit; the CT power supply unit directly obtains electrical energy from the power grid using a current transformer; the solar power supply unit converts solar energy into electrical energy through an integrated solar panel; the CT power supply unit and the solar power supply unit both power the terminal device through the battery charging and discharging unit.

[0014] Preferably, the power frequency signal and the traveling wave signal are collected through a full-bandwidth current transformer CT.

[0015] Preferably, the artificial intelligence algorithm is implemented through a deep learning model, specifically:

[0016] Collect the power frequency waveform and traveling wave waveform when the historical distribution network fault occurs, and mark the characteristic information of each fault category to form model data of different fault categories; input the model data of different fault categories into the deep learning model for deep learning; the fault categories include short circuit fault, open circuit fault, tree obstacle, wind deviation and wildfire;

[0017] The deep learning model is used to analyze the waveform feature points of the real-time collected power frequency signal and traveling wave signal data, including the peak pattern, peak duration, and peak interval, to determine the fault type. The fault occurrence point is then calculated by combining the GPS location coordinate information of the terminal device and the captured waveform time. The specific calculation is as follows:

[0018] Lm=(L+V*(Tm-Tn)) / 2;

[0019] Ln=(LV*(Tm-Tn)) / 2;

[0020] Wherein, Lm is the distance between terminal device M and fault point C, Ln is the distance between terminal device N and fault point C, terminal device M and terminal device N are configured at both ends of the distribution network line where fault point C occurs, L is the length between terminal device M and terminal device N, Tm and Tn are the times when terminal device M and terminal device N detect the arrival of the fault, respectively, and V is the propagation speed of the traveling wave on the wire.

[0021] Preferably, the dual-mode fusion adaptive wireless communication module combines LoRa communication and LTE communication to build a terminal network, connect to the power grid cloud platform and upload data information of all terminal devices in the terminal network, specifically:

[0022] When a terminal device cannot search for the LoRa signal of other terminal devices, it determines itself as a single device, and establishes a connection with the power grid cloud platform through the LTE network and uploads its own device data to the power grid cloud platform;

[0023] When there are multiple terminal devices, multiple terminal devices automatically build a terminal network based on the hybrid network topology of LoRa, and the operating condition information is shared between the terminal devices in the terminal network. The central device of the terminal network establishes a connection with the power grid cloud platform through LTE mobile communication and uploads the data of the entire terminal network.

[0024] Preferably, the central device is determined specifically as follows:

[0025] If there is already a central device in the terminal network, the central device sends out working condition information containing the central device identifier. After other terminal devices detect this identifier, the promotion process will not be started;

[0026] If there is no broadcast message from the central device in the terminal network, the terminal device will start the promotion process and send its own working condition information and promotion request to the terminal network; after receiving the working condition information and promotion request of the requesting device, other terminal devices in the terminal network will compare the status of the requesting device with their own status. When their own status is lower than that of the requesting device, they will send an approval message to the terminal network, otherwise they will send a rejection message and start the promotion process;

[0027] When the requesting device receives a rejection message from other terminal devices, it stops the request promotion process. When the requesting device receives an approval message from all other terminal devices in the terminal network, it becomes the central device of the terminal network.

[0028] Preferably, the operating condition information includes device ID, remaining power, operating status, and location coordinates, and the operating status includes LTE signal quality and GPS search satellite signal quality.

[0029] Preferably, comparing the state of the requesting device with the state of the device itself is specifically comparing the remaining power and LTE signal quality of the requesting device with the remaining power and LTE signal quality of the device itself:

[0030] First, compare the remaining power. If the remaining power is less than or equal to 5%, it will not participate in the promotion process;

[0031] When the power levels are both greater than 5%, the LTE signal quality is compared, and the terminal device with the stronger LTE signal quality is preferred; when the LTE signal quality is equal, the terminal device that starts the promotion process first is preferred. The signal quality range is 0 to 31, with 31 being the strongest.

[0032] Preferably, the dual-mode fusion adaptive wireless communication module allows a remote control center to configure and adjust the system.

[0033] Compared with the prior art, the present invention has the following beneficial effects:

[0034] The present invention breaks away from the limitations of traditional limited and single wireless communication through dual-mode fusion adaptive wireless communication, gives full play to the advantages of dual modes, forms a complementary effect, and is suitable for a wider range of application scenarios.

[0035] The present invention uses artificial intelligence fault location of terminal equipment to perform artificial intelligence algorithm judgment on the collected data on the device side, without the need to transmit a large amount of data to the cloud platform and then perform fault identification data processing through the cloud platform. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 This is a diagram of the fault location system architecture of the present invention. DETAILED DESCRIPTION

[0037] The following is combined with Figure 1 , the technical solution of the present invention is specifically described.

[0038] The present invention proposes an artificial intelligence distribution network line fault location system with dual-mode fusion adaptive wireless communication, including a power grid cloud platform and multiple terminal devices deployed in the distribution network line; the terminal devices include a power management module, a signal acquisition module, an artificial intelligence fault algorithm module, a signal processing module and a dual-mode fusion adaptive wireless communication module; these modules jointly realize accurate monitoring, accurate fault location and efficient data transmission in complex environments;

[0039] The power management module is used to supply power to the terminal device to ensure continuous operation of the system;

[0040] The signal acquisition module collects power frequency signals, traveling wave signals, temperature and humidity signals and GPS signals;

[0041] Industrial frequency signal: monitors the basic frequency and waveform of the power system to reflect the operating status of the power grid; traveling wave signal: captures transient voltage or current fluctuations in the power line for fault location and type identification; temperature and humidity signal: monitors changes in the temperature and humidity of the equipment environment to assess whether the equipment operating environment is suitable; GPS signal: receives global positioning system signals, provides accurate time synchronization and geographic location information, and supports remote positioning and tracking.

[0042] The signal processing module preprocesses and stores the collected original signals, and transmits the processed data to the artificial intelligence fault algorithm module; efficient storage of processed data can provide data support for fault backtracking and system optimization;

[0043] The artificial intelligence fault algorithm module uses artificial intelligence algorithms to analyze and judge fault characteristics of power frequency signals and traveling wave signal data, combines GPS and double-end traveling wave monitoring principles, determines the fault type and fault location, and generates early warning information;

[0044] The dual-mode fusion adaptive wireless communication module combines LoRa communication and LTE communication to build a terminal network, connect to the power grid cloud platform and upload data information of all terminal devices in the terminal network.

[0045] The device information uploaded to the cloud platform includes the following:

[0046] Equipment working conditions: battery power status, GPS signal latitude and longitude, 4G signal quality, temperature and humidity;

[0047] Equipment parameters: traveling wave threshold, power frequency threshold, heartbeat cycle, sleep cycle;

[0048] Calibration information;

[0049] Intranet card information: server port, APN, CCID

[0050] Fault information: fault type, fault waveform.

[0051] Compared with the prior art terminal equipment that performs fault identification by uploading a large amount of waveform data in a short time, the present invention performs waveform fault identification on the local device side and determines the fault type, so the amount of transmitted data is small.

[0052] In this embodiment, the preprocessing adopts digital filtering technology to remove noise interference in the original signal and retain useful signal components, thereby improving data quality and availability.

[0053] In this embodiment, the power management module includes a battery charging and discharging unit, a CT power taking unit and a solar power taking unit;

[0054] The CT power supply unit uses a current transformer to directly obtain electric energy from the power grid, which is suitable for environments with stable current sources and achieves self-sufficiency;

[0055] The solar power unit converts solar energy into electrical energy through integrated solar panels, providing green energy solutions for remote areas or areas that are difficult to access the power grid;

[0056] The CT power supply unit and the solar power supply unit both supply power to the terminal device through the battery charging and discharging unit.

[0057] In this embodiment, the power frequency signal and the traveling wave signal are collected through a full-bandwidth current transformer CT.

[0058] In this embodiment, the artificial intelligence algorithm is implemented through a deep learning model, specifically:

[0059] Collect the power frequency waveform and traveling wave waveform when the historical distribution network fault occurs, and mark the characteristic information of each fault category to form model data of different fault categories; input the model data of different fault categories into the deep learning model for deep learning; the fault categories include short circuit fault, open circuit fault, tree obstacle, wind deviation and wildfire;

[0060] The deep learning model is used to analyze the waveform feature points of the real-time collected power frequency signal and traveling wave signal data, including the peak pattern, peak duration, and peak interval, to determine the fault type. The fault occurrence point is then calculated by combining the GPS location coordinate information of the terminal device and the captured waveform time. The specific calculation is as follows:

[0061] Lm=(L+V*(Tm-Tn)) / 2;

[0062] Ln=(LV*(Tm-Tn)) / 2;

[0063] Wherein, Lm is the distance between terminal device M and fault point C, Ln is the distance between terminal device N and fault point C, terminal device M and terminal device N are configured at both ends of the distribution network line where fault point C occurs, L is the length between terminal device M and terminal device N, Tm and Tn are the times when terminal device M and terminal device N detect the arrival of the fault, respectively, and V is the propagation speed of the traveling wave on the wire.

[0064] In this embodiment, the dual-mode fusion adaptive wireless communication module combines LoRa communication and LTE communication to build a terminal network, connect to the power grid cloud platform and upload data information of all terminal devices in the terminal network, specifically:

[0065] When a terminal device cannot search for the LoRa signal of other terminal devices, it determines itself as a single device, and establishes a connection with the power grid cloud platform through the LTE network and uploads its own device data to the power grid cloud platform;

[0066] When there are multiple terminal devices, multiple terminal devices automatically build a terminal network based on the hybrid network topology of LoRa, and the operating condition information is shared between the terminal devices in the terminal network. The central device of the terminal network establishes a connection with the power grid cloud platform through LTE mobile communication and uploads the data of the entire terminal network.

[0067] Dual-mode fusion means that hardware devices are compatible with both LTE communication and LORA communication. In the software network architecture, a hybrid network architecture is built to combine LTE and LORA, and multiple communication protocols and data formats are integrated in the software driver layer.

[0068] In this embodiment, the central device is determined specifically as follows:

[0069] If there is already a central device in the terminal network, the central device sends out working condition information containing the central device identifier. After other terminal devices detect this identifier, the promotion process will not be started;

[0070] If there is no broadcast message from the central device in the terminal network, the terminal device will start the promotion process and send its own working condition information and promotion request to the terminal network; after receiving the working condition information and promotion request of the requesting device, other terminal devices in the terminal network will compare the status of the requesting device with their own status. When their own status is lower than that of the requesting device, they will send an approval message to the terminal network, otherwise they will send a rejection message and start the promotion process;

[0071] When the requesting device receives a rejection message from other terminal devices, it stops the request promotion process. When the requesting device receives an approval message from all other terminal devices in the terminal network, it becomes the central device of the terminal network.

[0072] Each terminal device has the ability to become a central node, reducing dependence on a single node and providing network robustness and fault tolerance. When the network topology changes, such as when a central device is damaged by external forces, the terminal network can adapt, start the arbitration process, and automatically adjust the network to ensure data connection with the power grid cloud platform and ensure data stability.

[0073] In this embodiment, the operating condition information includes device ID, remaining power, operating status, and location coordinates, and the operating status includes LTE signal quality and GPS search satellite signal quality.

[0074] In this embodiment, comparing the state of the requesting device with its own state specifically includes comparing the remaining power and LTE signal quality of the requesting device with its own remaining power and LTE signal quality:

[0075] First, compare the remaining power. If the remaining power is less than or equal to 5%, it will not participate in the promotion process;

[0076] When the power levels are both greater than 5%, the LTE signal quality is compared, and the terminal device with the stronger LTE signal quality is preferred; when the LTE signal quality is equal, the terminal device that starts the promotion process first is preferred. The signal quality range is 0 to 31, with 31 being the strongest.

[0077] In this embodiment, the dual-mode fusion adaptive wireless communication module allows the remote control center to configure and adjust the system.

[0078] In summary, the present invention supplies power to the entire system through the power management module to ensure the normal operation of all modules; continuously collects environmental and equipment signals through the signal acquisition module, and sends the data to the signal processing module; filters and stores the received signals through the signal processing module, and passes the processed data to the artificial intelligence fault algorithm module for analysis; the artificial intelligence fault algorithm module processes the collected data by algorithm, and determines whether there is a fault based on the data analysis results, generates early warning information, and sends it to the remote monitoring center through the dual-mode fusion wireless communication module; the dual-mode fusion adaptive wireless communication module is not only responsible for data transmission, but also allows the remote control center to configure and adjust the system, forming a closed-loop monitoring and management process. At the same time, its flexible communication protocol and network topology design ensure the adaptability and reliability of the system in different environments.

[0079] The above are preferred embodiments of the present invention. Any changes made according to the technical solution of the present invention, as long as the resulting functions do not exceed the scope of the technical solution of the present invention, belong to the protection scope of the present invention.

Claims

1. An artificial intelligence distribution network line fault location system with dual-mode fusion adaptive wireless communication, characterized in that: It includes a power grid cloud platform and multiple terminal devices deployed in the distribution network lines; the terminal devices include a power management module, a signal acquisition module, an artificial intelligence fault algorithm module, a signal processing module and a dual-mode fusion adaptive wireless communication module; The power management module is used to supply power to the terminal device; The signal acquisition module collects power frequency signals, traveling wave signals, temperature and humidity signals and GPS signals; The signal processing module preprocesses and stores the collected raw signals, and transmits the processed data to the artificial intelligence fault algorithm module; The artificial intelligence fault algorithm module uses artificial intelligence algorithms to analyze and judge fault characteristics of power frequency signals and traveling wave signal data, combines GPS and double-end traveling wave monitoring principles, determines the fault type and fault location, and generates early warning information; The dual-mode fusion adaptive wireless communication module combines LoRa communication and LTE communication to build a terminal network, connect to the power grid cloud platform and upload data information of all terminal devices in the terminal network.

2. According to claim 1, an artificial intelligence distribution network line fault location system for dual-mode fusion adaptive wireless communication is characterized in that: The preprocessing adopts digital filtering technology to remove noise interference in the original signal and retain useful signal components.

3. According to claim 1, the artificial intelligence distribution network line fault location system of dual-mode fusion adaptive wireless communication is characterized in that: The power management module includes a battery charging and discharging unit, a CT power supply unit and a solar power supply unit; the CT power supply unit uses a current transformer to directly obtain electrical energy from the power grid; the solar power supply unit converts solar energy into electrical energy through an integrated solar panel; the CT power supply unit and the solar power supply unit both power the terminal device through the battery charging and discharging unit.

4. According to claim 1, the artificial intelligence distribution network line fault location system of dual-mode fusion adaptive wireless communication is characterized in that: The power frequency signal and the traveling wave signal are collected through a full-bandwidth current transformer CT.

5. According to claim 1, the artificial intelligence distribution network line fault location system of dual-mode fusion adaptive wireless communication is characterized in that: The artificial intelligence algorithm is implemented through a deep learning model, specifically: Collect the power frequency waveform and traveling wave waveform of the historical distribution network when faults occur, and mark the characteristic information of each fault category to form model data of different fault categories; input the model data of different fault categories into the deep learning model for deep learning; The fault categories include short circuit fault, open circuit fault, tree obstacle, wind deviation and wildfire; The deep learning model is used to analyze the waveform feature points of the real-time collected power frequency signal and traveling wave signal data, including the peak pattern, peak duration, and peak interval, to determine the fault type. The fault occurrence point is then calculated by combining the GPS location coordinate information of the terminal device and the captured waveform time. The specific calculation is as follows: Lm=(L+V*(Tm-Tn)) / 2; Ln=(LV*(Tm-Tn)) / 2; Wherein, Lm is the distance between terminal device M and fault point C, Ln is the distance between terminal device N and fault point C, terminal device M and terminal device N are configured at both ends of the distribution network line where fault point C occurs, L is the length between terminal device M and terminal device N, Tm and Tn are the times when terminal device M and terminal device N detect the arrival of the fault, respectively, and V is the propagation speed of the traveling wave on the wire.

6. According to claim 1, the artificial intelligence distribution network line fault location system of dual-mode fusion adaptive wireless communication is characterized in that: The dual-mode fusion adaptive wireless communication module combines LoRa communication and LTE communication to build a terminal network, connect to the power grid cloud platform and upload data information of all terminal devices in the terminal network, specifically: When a terminal device cannot search for the LoRa signal of other terminal devices, it determines itself as a single device, and establishes a connection with the power grid cloud platform through the LTE network and uploads its own device data to the power grid cloud platform; When there are multiple terminal devices, multiple terminal devices automatically build a terminal network based on the hybrid network topology of LoRa, and the operating condition information is shared between the terminal devices in the terminal network. The central device of the terminal network establishes a connection with the power grid cloud platform through LTE mobile communication and uploads the data of the entire terminal network.

7. The artificial intelligence distribution network line fault location system of dual-mode fusion adaptive wireless communication according to claim 6 is characterized in that: The determination of the central device is specifically as follows: If there is already a central device in the terminal network, the central device sends out working condition information containing the central device identifier. After other terminal devices detect this identifier, the promotion process will not be started; If there is no broadcast message from the central device in the terminal network, the terminal device will start the promotion process and send its own working condition information and promotion request to the terminal network; after receiving the working condition information and promotion request of the requesting device, other terminal devices in the terminal network will compare the status of the requesting device with their own status. When their own status is lower than that of the requesting device, they will send an approval message to the terminal network, otherwise they will send a rejection message and start the promotion process; When the requesting device receives a rejection message from other terminal devices, it stops the request promotion process. When the requesting device receives an approval message from all other terminal devices in the terminal network, it becomes the central device of the terminal network.

8. The artificial intelligence distribution network line fault location system of dual-mode fusion adaptive wireless communication according to claim 7 is characterized in that: The operating condition information includes device ID, remaining power, operating status, and location coordinates. The operating status includes LTE signal quality and GPS search satellite signal quality.

9. The artificial intelligence distribution network line fault location system of dual-mode fusion adaptive wireless communication according to claim 8 is characterized in that: The comparing the state of the requesting device with the state of the device itself specifically includes comparing the remaining power and LTE signal quality of the requesting device with the remaining power and LTE signal quality of the device itself: First, compare the remaining power. If the remaining power is less than or equal to 5%, it will not participate in the promotion process; When the power levels are both greater than 5%, the LTE signal quality is compared, and the terminal device with the stronger LTE signal quality is preferred; when the LTE signal quality is equal, the terminal device that starts the promotion process first is preferred. The signal quality range is 0 to 31, with 31 being the strongest.

10. The artificial intelligence distribution network line fault location system of dual-mode fusion adaptive wireless communication according to claim 1 is characterized in that: The dual-mode fusion adaptive wireless communication module allows a remote control center to configure and adjust the system.

Images