State monitoring system and device of distribution transformer

Abstract

The invention provides a state monitoring system and device for a distribution transformer, and the system comprises an edge calculation device, a remote diagnosis service center, and a voiceprint monitoring device. The voiceprint monitoring device is used for monitoring a sound vibration signal on one side of the transformer in real time, and performing feature extraction and processing analysis on the obtained sound vibration signal to obtain a corresponding signal analysis result; the edge calculation device is used for storing and processing the received signal analysis result through a preset processing rule to obtain a corresponding transformer state monitoring result; the remote diagnosis service center is used for receiving a signal analysis result output by the voiceprint monitoring device and a transformer state monitoring result output by the edge computing device in real time; and the transformer is diagnosed and controlled according to the input control instruction, so that state monitoring and exception handling of the transformer are realized. According to the invention, the operation state of the transformer is monitored through voiceprint state perception, and whole-process and whole-stage monitoring and early warning are realized.

Description

Technical Field

[0001] The present invention relates to the technical field of transformer monitoring, and in particular to a state monitoring system and device for a distribution transformer. Background Art

[0002] Power transformers are key equipment in the power system. Noise and vibration are generated during the operation of transformers. The amplitude, time domain waveform, and spectrum characteristics of sound and vibration are closely related to their operating voltage, current, mechanical state, excitation state, insulation state, etc., and can timely reflect changes in the operating state of the equipment. The significance of voiceprint monitoring of distribution transformers: On the one hand, by continuously monitoring the voiceprint information of distribution transformers, abnormal sounds inside the transformer can be discovered in time, and intervention can be made before small problems expand into major faults, thereby avoiding more serious damage and expensive repair costs; on the other hand, voiceprint monitoring of distribution transformers does not require manual intervention, and automatically issues fault alarms, which improves the efficiency of operation and maintenance work and reduces operation and maintenance costs. At present, the main monitoring methods for distribution transformers include temperature monitoring, oil quality monitoring, gas monitoring, electrical quantity monitoring, etc., which have a lag in fault diagnosis and cannot provide early fault warnings. Summary of the invention

[0003] The purpose of the present invention is to provide a distribution transformer status monitoring system and device to solve the technical problem of how to monitor the transformer status through voiceprint.

[0004] In one aspect, a condition monitoring system for a distribution transformer is provided, comprising:

[0005] Edge computing devices, remote diagnostic service centers, and voiceprint monitoring devices;

[0006] The voiceprint monitoring device is used to monitor the sound vibration signal on one side of the transformer in real time, and perform feature extraction and processing analysis on the acquired sound vibration signal to obtain the corresponding signal analysis result;

[0007] The edge computing device is used to store and process the received signal analysis results according to preset processing rules to obtain corresponding transformer status monitoring results;

[0008] The remote diagnosis service center is used to receive in real time the signal analysis results output by the voiceprint monitoring device and the transformer status monitoring results output by the edge computing device; and diagnose and control the transformer according to the input control instructions to realize transformer status monitoring and abnormality processing.

[0009] Preferably, it also includes a fault detection module, which is used to identify the received sound vibration signal through a preset fault identification model, and match the identification result with a preset fault training database. If the fault training database contains a signal identical to the identification result, the corresponding fault signal is output.

[0010] Preferably, it also includes a fault alarm module and a fault recording module connected thereto; the fault recording module is used to record corresponding fault data after receiving a fault signal, and the fault recording module is used to back up and save the fault data.

[0011] Preferably, the voiceprint monitoring device includes at least a data sensing submodule, a data acquisition submodule and a data processing submodule; the data sensing submodule is used to receive the transformer sound vibration signal through a vibration acceleration sensor installed on one side of the transformer; the data acquisition submodule is used to collect the signal of the vibration acceleration sensor and send it to the data processing submodule; the data processing submodule is used to extract features and process and analyze the received data, and package the results and send them to the remote diagnosis service center.

[0012] Preferably, the remote diagnosis service center at least includes a voiceprint sample library management module, an algorithm model library management module, and a cloud server maintenance management module; the voiceprint sample library management module is used to pre-store voiceprint samples corresponding to the transformer according to historical records; the algorithm model library management module is used to pre-store corresponding algorithm programs; and the cloud server maintenance management module is used to control and manage the cloud server.

[0013] Preferably, the remote diagnosis service center also includes a cloud server; the cloud server is used to call the corresponding algorithm in the algorithm model library to process the received signal analysis results, and match the processing results with the voiceprints in the voiceprint sample library, and determine whether there is an abnormality based on the matching results.

[0014] Preferably, the remote diagnosis service center further includes a diagnosis analysis module; the diagnosis analysis module monitors the operation data of the cloud server and diagnoses the status of the cloud server based on the monitoring data.

[0015] On the other hand, a state monitoring device for a distribution transformer is also provided, and the transformer is monitored by the system.

[0016] In summary, the implementation of the embodiments of the present invention has the following beneficial effects:

[0017] The distribution transformer status monitoring system and device provided by the present invention can timely and reliably discover the inherent hidden dangers of power equipment, assist operation and maintenance personnel to quickly investigate and solve them, thereby ensuring the stable operation of power equipment and the continuous and reliable power supply. The importance of monitoring the operating status of the transformer and realizing early warning of faults through "voice print" and "vibration" status sensing elements is becoming increasingly prominent. It is of great significance to realize full-process and full-stage monitoring and early warning, effectively improve the safe and stable operation level of power transformers, realize comprehensive evaluation of the operating status of the transformer, and form effective evaluation and diagnosis results, provide scientific and effective reference basis for operation and maintenance personnel, and can effectively extend the equipment operation time, formulate reasonable maintenance plans, and prevent the occurrence of sudden failures. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, without paying creative labor, other drawings obtained based on these drawings still belong to the scope of the present invention.

[0019] Figure 1 Schematic diagram of a state monitoring system for a distribution transformer in an embodiment of the present invention. DETAILED DESCRIPTION

[0020] In order to make the objectives, technical solutions and advantages of the present invention more clear, the present invention will be further described in detail below with reference to the accompanying drawings.

[0021] like Figure 1 FIG. 1 is a schematic diagram of an embodiment of a state monitoring system for a distribution transformer provided by the present invention. In this embodiment, the system includes:

[0022] An edge computing device, a remote diagnosis service center and a voiceprint monitoring device; the voiceprint monitoring device is used to monitor the sound and vibration signals on one side of the transformer in real time, and perform feature extraction and processing analysis on the acquired sound and vibration signals to obtain corresponding signal analysis results; the edge computing device is used to store and process the received signal analysis results according to preset processing rules to obtain corresponding transformer status monitoring results; the remote diagnosis service center is used to receive the signal analysis results output by the voiceprint monitoring device and the transformer status monitoring results output by the edge computing device in real time; and diagnose and control the transformer according to the input control instructions to realize transformer status monitoring and abnormality processing. It can be understood that the voiceprint monitoring device includes a data sensing submodule, a data acquisition submodule and a data processing submodule; the distribution transformer voiceprint monitoring system monitors the operating status of the distribution transformer in real time; the data sensing submodule uses a vibration acceleration sensor, which is installed on one side of the transformer to receive the transformer sound vibration signal; the data acquisition submodule collects the signal of the vibration acceleration sensor and sends it to the data processing submodule, the data processing submodule extracts features and processes and analyzes the transmitted data, and packages it and sends it to the remote diagnosis service center; the voiceprint monitoring device is connected to the edge computing device, the edge computing device includes an Internet of Things module and an edge server, and the edge server is used to store and process the data of the voiceprint monitoring device. The remote diagnosis service center is connected to the personal terminal through a wireless communication module. The voiceprint monitoring device is connected to the remote diagnosis service center through a data transmission module.

[0023] One embodiment further includes a fault detection module, which is used to identify the received sound vibration signal through a preset fault recognition model, and match the recognition result with a preset fault training database. If the fault training database has a signal identical to the recognition result, the corresponding fault signal is output. The fault detection module uses the YOLOv11 model to construct a fault map, and the fault detection module includes a fault training database, which compares the acquired data with the fault training database and outputs a fault signal; the YOLOv11 model is a simple and easy-to-apply algorithm. The main architectural innovations of YOLOv11 revolve around the C3K2 block, the SPFF module, and the C2PSA block, all of which enhance its ability to process spatial information while maintaining high-speed reasoning. YOLO is an advanced target detection algorithm. It treats target detection tasks as regression problems. Its core is to divide the input image into multiple grids, each of which is responsible for predicting the bounding box and category probability of the object. This enables end-to-end fast detection, with obvious advantages in speed. It treats target detection tasks as regression problems. Its core is to divide the input image into multiple grids, each of which is responsible for predicting the bounding box and category probability of the object. This enables end-to-end rapid detection, with obvious speed advantages.

[0024] One embodiment further includes a fault alarm module and a fault recording module connected thereto; the fault recording module is used to record the corresponding fault data after receiving the fault signal, and the fault recording module is used to back up and save the fault data. The fault recording module is used to record the fault data, and the fault recording module is connected to a fault data backup module. The fault detection module and the fault alarm module can realize automatic evaluation and monitoring: no manual intervention is required, and the normal operation of the equipment is not affected. It has a clear and unambiguous alarm signal: once an insulation problem is detected, an alarm signal is output in time; and the operation mode is optional: it can meet both grounded conditions and ungrounded conditions.

[0025] In one embodiment, the voiceprint monitoring device at least includes a data sensing submodule, a data acquisition submodule and a data processing submodule; the data sensing submodule is used to receive the transformer sound vibration signal through a vibration acceleration sensor installed on one side of the transformer; the data acquisition submodule is used to collect the signal of the vibration acceleration sensor and send it to the data processing submodule; the data processing submodule is used to extract features and process and analyze the received data, and package the obtained results and send them to the remote diagnosis service center. The method for preprocessing the data by the data processing submodule is: preprocessing the voiceprint data, then performing SVM test, LFA test, I-VECTOR test, and then inputting it into the preprocessing library for training, and the training method includes GMM training and UBM training, and performing channel compensation, channel training and SVM training on the data after GMM training and UBM training, and then transmitting the data to the edge computing device and the remote diagnosis server center. By using SVM test, LFA test, I-VECTOR test, GMM training, and UBM training, the data can be fully tested and trained through GMM training and UBM training to improve the monitoring accuracy.

[0026] In one embodiment, the remote diagnosis service center at least includes a voiceprint sample library management module, an algorithm model library management module, and a cloud server maintenance management module; the voiceprint sample library management module is used to pre-store voiceprint samples corresponding to the transformer according to historical records; the algorithm model library management module is used to pre-store corresponding algorithm programs; and the cloud server maintenance management module is used to control and manage the cloud server. It also includes a cloud server; the cloud server is used to call the corresponding algorithm in the algorithm model library to process the received signal analysis results, and match the processing results with the voiceprints in the voiceprint sample library, and determine whether it is abnormal according to the matching results. It also includes a diagnosis analysis module; the diagnosis analysis module monitors the operating data of the cloud server and diagnoses the status of the cloud server according to the monitoring data. It should be noted that the algorithm model library management module includes a deep learning model library and an image processing model library; the deep learning model library includes a convolutional neural network and a recurrent neural network; the image processing model library includes an image classification image processing algorithm, a target detection image processing algorithm, and a semantic segmentation image processing algorithm. The neural network and the recurrent neural network are composed of a large number of neurons connected to each other. After each neuron receives the input of the linear combination, it is initially simply linearly weighted. Later, a nonlinear activation function is added to each neuron, so that the output is nonlinearly transformed. The connection between each two neurons represents a weighted value, which is called a weight. Different weights and activation functions will result in different outputs of the neural network. The algorithm model library management module includes a natural language processing module and a reinforcement learning module; the natural language processing module includes text classification, text generation, and named entity recognition algorithms; the reinforcement learning module includes a model library based on a reinforcement learning algorithm for solving decision-making problems.

[0027] Specific embodiments, through the deep integration of sensing technology, artificial intelligence, edge computing, and cloud computing, timely and reliably discover the inherent hidden dangers of power equipment, assist operation and maintenance personnel to quickly investigate and solve them, thereby ensuring the stable operation of power equipment and the continuous and reliable power supply. Through the "voiceprint" and "vibration" state sensing elements, the operating state of the transformer is monitored. The voiceprint state monitoring system of the distribution transformer based on voiceprint recognition realizes early warning of faults, and its importance is increasingly prominent. The present invention is used to monitor the operating state of the distribution transformer in real time and extract the characteristic information of the operating sound of the distribution transformer; the second is to provide alarm information of the distribution transformer fault in a timely manner and give the fault type diagnosis result. The present invention does not need to contact the live equipment, does not change the operating state of the equipment, and is easy to implement. The present invention can track the changes in the operating state of the equipment in real time, and has high detection sensitivity; the present invention also supplements the lack of existing monitoring state quantities and adds a fault diagnosis scheme within the low frequency range of 20Hz to 20kHz of sound and vibration. The present invention can realize a full-process and full-stage monitoring and early warning system, effectively improve the safe and stable operation level of power transformers, and is of great significance. The system monitors a large amount of real-time data such as sound pattern vibration online, combined with offline input parameters such as transformer structure and factory test, to complete further analysis and diagnosis of transformer internal faults. It can achieve a comprehensive evaluation of the transformer operation status and form effective evaluation and diagnosis results, providing a scientific and effective reference basis for operation and maintenance personnel, and can effectively extend the equipment operation time, formulate reasonable maintenance plans, and prevent sudden failures.

[0028] An embodiment of the present invention further provides a state monitoring device for a distribution transformer, and the transformer is monitored by the system.

[0029] It should be noted that the device described in the above embodiment corresponds to the system described in the above embodiment. Therefore, the part of the device described in the above embodiment that is not described in detail can be obtained by referring to the content of the system described in the above embodiment, and will not be repeated here.

[0030] In summary, the implementation of the embodiments of the present invention has the following beneficial effects:

[0031] The distribution transformer status monitoring system and device provided by the present invention can timely and reliably discover the inherent hidden dangers of power equipment, assist operation and maintenance personnel to quickly investigate and solve them, thereby ensuring the stable operation of power equipment and the continuous and reliable power supply. The importance of monitoring the operating status of the transformer and realizing early warning of faults through "voice print" and "vibration" status sensing elements is becoming increasingly prominent. It is of great significance to realize full-process and full-stage monitoring and early warning, effectively improve the safe and stable operation level of power transformers, realize comprehensive evaluation of the operating status of the transformer, and form effective evaluation and diagnosis results, provide scientific and effective reference basis for operation and maintenance personnel, and can effectively extend the equipment operation time, formulate reasonable maintenance plans, and prevent the occurrence of sudden failures.

[0032] The above disclosure is only the preferred embodiment of the present invention, which certainly cannot be used to limit the scope of the present invention. Therefore, equivalent changes made according to the claims of the present invention are still within the scope of the present invention.

Claims

1. A distribution transformer condition monitoring system, characterized in that: include: Edge computing devices, remote diagnostic service centers, and voiceprint monitoring devices; The voiceprint monitoring device is used to monitor the sound vibration signal on one side of the transformer in real time, and perform feature extraction and processing analysis on the acquired sound vibration signal to obtain the corresponding signal analysis result; The edge computing device is used to store and process the received signal analysis results according to preset processing rules to obtain corresponding transformer status monitoring results; The remote diagnosis service center is used to receive in real time the signal analysis results output by the voiceprint monitoring device and the transformer status monitoring results output by the edge computing device; and diagnose and control the transformer according to the input control instructions to realize transformer status monitoring and abnormality processing.

2. The system according to claim 1, characterized in that It also includes a fault detection module, which is used to identify the received sound vibration signal through a preset fault recognition model, and match the recognition result with a preset fault training database. If the fault training database contains a signal that is the same as the recognition result, the corresponding fault signal is output.

3. The system according to claim 2, characterized in that It also includes a fault alarm module and a fault recording module connected thereto; the fault recording module is used to record corresponding fault data after receiving a fault signal, and the fault recording module is used to back up and save the fault data.

4. The system according to claim 3, characterized in that The voiceprint monitoring device at least includes a data sensing submodule, a data acquisition submodule and a data processing submodule; the data sensing submodule is used to receive the transformer sound vibration signal through a vibration acceleration sensor installed on one side of the transformer; the data acquisition submodule is used to collect the signal of the vibration acceleration sensor and send it to the data processing submodule; the data processing submodule is used to extract features and process and analyze the received data, and package the results and send them to the remote diagnosis service center.

5. The system according to claim 4, characterized in that The remote diagnosis service center at least includes a voiceprint sample library management module, an algorithm model library management module, and a cloud server maintenance management module; The voiceprint sample library management module is used to pre-store voiceprint samples corresponding to the transformer according to historical records; The algorithm model library management module is used to pre-store the corresponding algorithm program; the cloud server maintenance management module is used to control and manage the cloud server.

6. The system according to claim 5, characterized in that The remote diagnosis service center also includes a cloud server; the cloud server is used to call the corresponding algorithm in the algorithm model library to process the received signal analysis results, and match the processed results with the voiceprints in the voiceprint sample library, and determine whether there is an abnormality based on the matching results.

7. The system according to claim 6, characterized in that The remote diagnosis service center also includes a diagnosis analysis module; the diagnosis analysis module monitors the operation data of the cloud server and diagnoses the status of the cloud server based on the monitoring data.

8. A state monitoring device for a distribution transformer, characterized in that: The transformer is monitored by the system according to any one of claims 1 to 7.

Images