CN-121980187-A - Transformer cooling system fault detection method and related device

CN121980187ACN 121980187 ACN121980187 ACN 121980187ACN-121980187-A

Abstract

The application discloses a fault detection method of a transformer cooling system and a related device, wherein the method comprises the steps of collecting multi-source heterogeneous data in the running process of the transformer, and carrying out normalization processing to obtain multi-mode data, wherein the multi-mode data comprises temperature, oil flow rate, pressure and load current; the method comprises the steps of carrying out real-time reconstruction on temperature distribution in a transformer according to temperature and load current in multi-mode data to obtain a dynamic temperature gradient field, carrying out vortex dynamics analysis according to oil flow rate, pressure and the dynamic temperature gradient field to determine vortex characteristics including pressure-flow phase difference, turbulence intensity and oil flow vorticity, constructing a space-time diagram structural model based on the multi-mode data, the dynamic temperature gradient field and the vortex characteristics, and solving a cooling system fault detection result. The application can solve the technical problems that the prior art has poor sensing capability, can not capture early vortex abnormality, and needs shutdown detection, so that complex working conditions are difficult to process and real-time working condition change can not be dealt with.

Inventors

TANG QI
FAN XINMING
ZENG QINGHUI
GAO XUE
WANG ZHIJIAO

Assignees

广东电网有限责任公司佛山供电局

Dates

Publication Date: 20260505
Application Date: 20260127

Claims (10)

1. A method for detecting a fault in a cooling system of a transformer, comprising: The method comprises the steps of collecting multi-source heterogeneous data in the running process of a transformer, and carrying out normalization processing to obtain multi-mode data, wherein the multi-mode data comprise temperature, oil flow rate, pressure and load current; Reconstructing the internal temperature distribution of the transformer in real time according to the temperature and the load current in the multi-mode data to obtain a dynamic temperature gradient field; performing vortex dynamics analysis according to the oil flow rate, the pressure and the dynamic temperature gradient field, and determining vortex characteristics, wherein the vortex characteristics comprise pressure-flow phase difference, turbulence intensity and oil flow vorticity; and constructing a space-time diagram structural model based on the multi-mode data, the dynamic temperature gradient field and the vortex characteristics, and solving a cooling system fault detection result.
2. The method for detecting a fault in a cooling system of a transformer according to claim 1, wherein the reconstructing the internal temperature distribution of the transformer in real time according to the temperature and the load current in the multi-mode data to obtain a dynamic temperature gradient field comprises: Performing heat transfer analysis according to the temperature in the multi-mode data to construct a thermal resistance item; carrying out loss calculation according to the load current to obtain a loss heat source vector; constructing a temperature control equation according to the thermal resistance term, the loss heat source vector and the heat capacity term; and reconstructing the temperature distribution in the transformer in real time based on the temperature control equation to obtain a dynamic temperature gradient field.
3. The transformer cooling system fault detection method of claim 1, wherein the performing a vortex dynamics analysis based on the oil flow rate, the pressure, and the dynamic temperature gradient field to determine a vortex signature comprises: performing Euclidean norm-based time correlation calculation according to the pressure and the flow to obtain a pressure-flow phase difference; Constructing a vortex equation according to the oil flow velocity and the dynamic temperature gradient field, performing vortex dynamics analysis, and calculating the oil flow vortex; and quantitatively analyzing the turbulence state of the oil flow based on the pressure to obtain the turbulence intensity.
4. The method of claim 1, wherein constructing a space-time diagram structural model based on the multi-modal data, the dynamic temperature gradient field, and the vortex characteristics and solving for cooling system fault detection results comprises: respectively constructing a graph convolution layer and a time sequence convolution layer based on the multi-mode data, the dynamic temperature gradient field and the vortex characteristics; Based on a cooperative attention mechanism and a preset loss function, constructing a space-time diagram structural model by combining the diagram convolution layer and the time sequence convolution layer; and carrying out iterative optimization solving on the space-time diagram structural model to obtain a cooling system fault detection result, wherein the cooling system fault detection result comprises a fault probability vector.
5. The method for detecting the fault of the cooling system of the transformer according to claim 1, wherein the constructing a space-time diagram structural model based on the multi-mode data, the dynamic temperature gradient field and the vortex characteristics and solving the cooling system fault detection result further comprises: and triggering an adaptive fault control strategy according to different fault types in the fault detection result of the cooling system.
6. A transformer cooling system fault detection device, comprising: The data processing unit is used for collecting multi-source heterogeneous data in the running process of the transformer and carrying out normalization processing to obtain multi-mode data, wherein the multi-mode data comprises temperature, oil flow rate, pressure and load current; The temperature reconstruction unit is used for reconstructing the internal temperature distribution of the transformer in real time according to the temperature and the load current in the multi-mode data to obtain a dynamic temperature gradient field; The vortex analysis unit is used for carrying out vortex dynamics analysis according to the oil flow rate, the pressure and the dynamic temperature gradient field to determine vortex characteristics, wherein the vortex characteristics comprise pressure-flow phase difference, turbulence intensity and oil flow vorticity; and the fault detection unit is used for constructing a space-time diagram structural model based on the multi-mode data, the dynamic temperature gradient field and the vortex characteristics and solving a cooling system fault detection result.
7. The transformer cooling system fault detection device according to claim 6, wherein the temperature reconstruction unit is specifically configured to: Performing heat transfer analysis according to the temperature in the multi-mode data to construct a thermal resistance item; carrying out loss calculation according to the load current to obtain a loss heat source vector; constructing a temperature control equation according to the thermal resistance term, the loss heat source vector and the heat capacity term; and reconstructing the temperature distribution in the transformer in real time based on the temperature control equation to obtain a dynamic temperature gradient field.
8. The transformer cooling system fault detection device according to claim 6, wherein the fault detection unit is specifically configured to: respectively constructing a graph convolution layer and a time sequence convolution layer based on the multi-mode data, the dynamic temperature gradient field and the vortex characteristics; Based on a cooperative attention mechanism and a preset loss function, constructing a space-time diagram structural model by combining the diagram convolution layer and the time sequence convolution layer; and carrying out iterative optimization solving on the space-time diagram structural model to obtain a cooling system fault detection result, wherein the cooling system fault detection result comprises a fault probability vector.
9. A transformer cooling system fault detection device, the device comprising a processor and a memory; the memory is used for storing program codes and transmitting the program codes to the processor; The processor is configured to perform the transformer cooling system fault detection method of any one of claims 1-5 according to instructions in the program code.
10. A computer readable storage medium for storing program code for performing the transformer cooling system fault detection method of any one of claims 1-5.

Description

Transformer cooling system fault detection method and related device Technical Field The application relates to the field of transformer fault detection, in particular to a transformer cooling system fault detection method and a related device. Background The transformer is used as a core hub device of the power system and bears key responsibilities of power transmission and voltage conversion, and the running stability of the transformer directly determines the safety and reliability of power supply of a power grid. The cooling system is used as a heat dissipation center of the transformer, takes away heat generated by the winding and the iron core in the modes of oil circulation, forced air cooling and the like, avoids the problems of insulation aging, insulation oil cracking and the like caused by overheating of equipment, and is a core support for guaranteeing long-term safe operation of the transformer. Along with the development of the power system to high voltage, large capacity and intelligent directions, the load fluctuation of the transformer is more frequent, the working condition complexity faced by the cooling system is obviously improved, various fault risks such as blockage, leakage and oil pump faults are greatly increased, and higher requirements are provided for the accuracy, instantaneity and comprehensiveness of the fault detection technology. The prior transformer cooling system fault detection technology has a plurality of bottlenecks which are difficult to break through, and severely restricts the operation and maintenance efficiency and the power grid safety. For example, the sensing layer depends on more than 15 sensors to be densely distributed, so that the hardware cost is high, the whole-field temperature gradient is difficult to cover in whole, the oil flow analysis only monitors a single pressure parameter and cannot capture early vortex anomalies, the traditional methods such as SVM algorithm and in-oil gas analysis need to be stopped for detection, concurrent faults are difficult to distinguish, the physical model is complex in calculation and cannot respond to working condition changes in real time, the fault processing delay is caused by manual decision in control response, and serious consequences such as insulation accelerated aging, winding overheating and burning are easy to be caused after a cooling system is stopped. These problems restrict the accuracy, real-time performance and economy of the fault detection of the transformer cooling system, and cannot meet the actual requirements of the intelligent operation and maintenance of the power system, so that a high-efficiency, accurate and real-time fault detection technology is needed to break through the limitation of the prior art and ensure the safe operation of the power grid. Disclosure of Invention The application provides a fault detection method and a related device for a transformer cooling system, which are used for solving the technical problems that the prior art is poor in sensing capability, early vortex abnormality cannot be captured, and complex working conditions are difficult to process due to the fact that shutdown detection is needed, and real-time working condition change cannot be dealt with. In view of this, a first aspect of the present application provides a method for detecting a fault in a cooling system of a transformer, including: The method comprises the steps of collecting multi-source heterogeneous data in the running process of a transformer, and carrying out normalization processing to obtain multi-mode data, wherein the multi-mode data comprise temperature, oil flow rate, pressure and load current; Reconstructing the internal temperature distribution of the transformer in real time according to the temperature and the load current in the multi-mode data to obtain a dynamic temperature gradient field; performing vortex dynamics analysis according to the oil flow rate, the pressure and the dynamic temperature gradient field, and determining vortex characteristics, wherein the vortex characteristics comprise pressure-flow phase difference, turbulence intensity and oil flow vorticity; and constructing a space-time diagram structural model based on the multi-mode data, the dynamic temperature gradient field and the vortex characteristics, and solving a cooling system fault detection result. Preferably, the reconstructing the internal temperature distribution of the transformer in real time according to the temperature and the load current in the multi-mode data to obtain a dynamic temperature gradient field includes: Performing heat transfer analysis according to the temperature in the multi-mode data to construct a thermal resistance item; carrying out loss calculation according to the load current to obtain a loss heat source vector; constructing a temperature control equation according to the thermal resistance term, the loss heat source vector and the heat capacity term; and reconstruc