Intelligent fan for electric power gas insulation transformer

Abstract

The invention discloses an intelligent fan for an electric power gas insulation transformer. The intelligent fan comprises a circulating centrifugal fan body, a sensor system, a data collecting and processing module, an intelligent control module and a remote monitoring platform. The circulating centrifugal fan body is connected with a gas circuit pipeline of the gas insulation transformer through an inlet flange and an outlet flange and is directly driven by a motor; the sensor system is arranged on the circulating centrifugal fan body, and the arrangement follows the principle that the sensor system is close to a heat source, reflects truly and is convenient to maintain. According to the intelligent draught fan for the electric power gas insulation transformer, environment-friendly mixed gas is adopted as a circulating medium, and the carbon footprint is remarkably reduced while the insulation performance is guaranteed; dynamic matching of fan power and transformer thermal load is achieved, and the overall operation energy efficiency is improved; and multi-dimensional sensing information is fused, and an artificial intelligence technology is combined to construct a fault early warning model, so that the real-time sensing of the equipment state and the accurate identification of fault forebodes are realized.

Description

Technical Field

[0001] This invention relates to the field of power equipment cooling and intelligent operation and maintenance technology, and in particular to a smart fan for gas-insulated transformers. Background Technology

[0002] As a core piece of equipment in the power grid system, the safe, stable, and efficient operation of gas-insulated transformers directly affects the reliability and economy of the entire power system. Traditional power transformer cooling systems generally employ forced air cooling with fans to promptly dissipate heat generated during operation and prevent accelerated aging of insulation materials due to excessive temperature. However, existing fan systems are mostly passively operating, lacking the dynamic response capability to the actual load conditions and temperature rise changes of the transformer. This results in fans operating under suboptimal conditions for extended periods, leading not only to energy waste but also potential equipment fatigue due to excessive start-stop cycles or continuous high-load operation, thus affecting system lifespan.

[0003] Furthermore, sulfur hexafluoride (SF6), widely used in some high-end electrical equipment as an insulating and cooling medium, possesses excellent dielectric properties, but its global warming potential (GWP) is as high as 23,500 (with CO2 as 1), making it a potent greenhouse gas explicitly restricted under the Kyoto Protocol. Simultaneously, SF6 easily decomposes under electric arc conditions, producing toxic byproducts, posing safety hazards, and its recycling and disposal are complex and costly, facing increasingly stringent environmental regulations. With the deepening implementation of the "dual-carbon" strategy, promoting the transformation of power equipment towards green and low-carbon directions has become an inevitable trend in the industry's development.

[0004] Meanwhile, driven by the wave of intelligentization, traditional operation and maintenance models are evolving from "reactive maintenance" to "predictive maintenance." However, most existing wind turbine systems only have basic start-stop control functions and lack multi-source perception capabilities for their own and transformer's status, making it difficult to achieve early fault identification and health status assessment. This results in potential risks not being detected in a timely manner, increasing the probability of unplanned shutdowns and major accidents. Summary of the Invention

[0005] In view of the aforementioned deficiencies in the prior art, the technical problem to be solved by this invention is the poor environmental performance, low operating efficiency, lack of condition sensing capabilities, outdated operation and maintenance methods, noise and structural design limitations, and unreasonable sensor placement of existing circulating centrifugal fans widely used in gas-insulated power transformers. This invention provides a smart fan for cooling gas-insulated power transformers. It eliminates high GWP working fluids, using an environmentally friendly mixed gas as the circulating medium, significantly reducing the carbon footprint while ensuring insulation performance. By introducing intelligent control algorithms, it achieves dynamic matching between fan output and transformer heat load, improving overall operating efficiency. Furthermore, it integrates multi-dimensional sensor information such as temperature, pressure pulsation, and vibration, combined with artificial intelligence technology to construct a fault early warning model, enabling real-time perception of equipment status and accurate identification of fault precursors.

[0006] To achieve the above objectives, the present invention provides a smart fan for a gas-insulated power transformer, comprising a circulating centrifugal fan body, a sensor system, a data acquisition and processing module, an intelligent control module, and a remote monitoring platform; the circulating centrifugal fan body is connected to the gas pipeline of the gas-insulated transformer through an inlet flange and an outlet flange, and is directly driven by an electric motor connected to the circulating fan impeller through the motor shaft;

[0007] The sensor system is located on the body of the circulating centrifugal fan, and its arrangement follows the principles of "closeness to the heat source, accurate reflection, and ease of maintenance";

[0008] The data acquisition and processing module collects data through a sensor system;

[0009] The intelligent control module integrates the inverter control interface, digital output, and communication interface based on the data collected by the data acquisition and processing module.

[0010] The remote monitoring platform has a built-in 4G / 5G DTU or industrial Ethernet module, and all operating data is packaged and uploaded to the cloud server or remote monitoring center in 1-minute cycles.

[0011] Furthermore, the circulating centrifugal fan body is designed for environmentally friendly mixed gases of 8% C4F7N / 92% CO2, while also being compatible with SF6 gas. Because SF6 possesses inherent advantages such as high specific heat capacity, high density (i.e., high heat capacity), high heat transfer efficiency, and good insulation performance, it is suitable for applications where environmentally friendly gases can be used.

[0012] Furthermore, the circulating centrifugal fan body includes an impeller, which includes a front cover plate, a rear cover plate, blades, and a hub. The blades are disposed between the front cover plate and the rear cover plate, and the hub is disposed on the rear cover plate.

[0013] After the blades are riveted to the front and rear cover plates, argon arc welding is used to repair and overlay the joints, forming a smooth transition reinforced area.

[0014] Furthermore, the impeller flow channel adopts a high-intensity acceleration flow channel, which has the characteristics of high efficiency and low noise. Previous studies have also proven that using a high-intensity acceleration flow channel can suppress interference at the blade inlet.

[0015] Furthermore, the impeller's rear cover plate also includes balance holes.

[0016] Furthermore, the impeller's static pressure, volumetric flow rate, impeller diameter, and rotational speed in the air medium remain consistent with the original impeller.

[0017] Furthermore, the impeller is a forward-curved centrifugal impeller, comprising 52 blades with a blade outlet angle of 35° and an inlet angle of 17.27°.

[0018] Furthermore, the sensor system includes a temperature sensor inside the fan, a pressure sensor inside the fan, a temperature sensor inside the transformer, and an environmental sensor;

[0019] The internal temperature sensor of the fan is embedded in the volute, the internal pressure sensor of the fan is installed on the tongue of the volute, the internal temperature sensor of the transformer is arranged on the insulation support of the hottest section of the high voltage winding, and the environmental sensor is an integrated temperature and humidity transmitter, which is installed on the inner side of the collector pipe at the outermost and lowest point of the transformer radiator.

[0020] Furthermore, all sensor cables are led out through a dedicated conduit and converged into a waterproof and shockproof multi-core aviation connector, which is connected to the corresponding socket in the data acquisition box of the data acquisition and processing module.

[0021] Furthermore, the data acquisition and processing module continuously reads the transformer's internal temperature and load current signals. A built-in PID algorithm calculates the required cooling airflow and outputs a 4-20mA signal to the frequency converter, steplessly adjusting the fan speed to stabilize the transformer's hot spot temperature within the set range.

[0022] Read the ambient temperature and humidity, and calculate the current dew point temperature. T d Estimate the surface temperature of the radiator T s ;like If a condensation risk is detected, the intelligent control module will prioritize implementing anti-condensation strategies.

[0023] Furthermore, the intelligent control module includes a fault diagnosis module, which is used to determine the probability of the current state belonging to various types of faults. If the probability of any fault exceeds a set threshold, the impeller fault alarm module will be activated immediately, issuing an audible and visual alarm, and sending a warning message to the remote monitoring center through the data remote transmission module, including the fault type, probability, and timestamp.

[0024] The present invention has the following outstanding advantages:

[0025] (1) Green and environmentally friendly, emission reduction at the source

[0026] The fan design of this invention is adapted to a new type of environmentally friendly mixed gas (8%C4F7N / 92%CO2), which fundamentally avoids the high carbon emissions and toxicity risks brought by SF6, conforms to the national environmental protection policy, and significantly reduces the carbon footprint and disposal costs throughout the product's life cycle.

[0027] (2) Excellent interchangeability and compatibility

[0028] With only the impeller replaced and without altering any other transformer components, the external structure, interface dimensions, and mounting flanges of the new air purifier are completely identical to the original equipment, achieving a "plug-and-play" replacement that greatly simplifies the retrofit process and is suitable for simultaneous promotion of existing equipment upgrades and new projects.

[0029] (3) Strong adaptability of gas properties

[0030] The impeller has undergone specialized aerodynamic optimization, taking into full account the high-density, low-velocity diffusion characteristics of the C4F7N / CO2 mixed gas, ensuring that it can still provide sufficient static pressure and airflow under the same power input, effectively overcoming system flow resistance, and ensuring that the cooling effect is not affected by changes in the medium.

[0031] (4) High efficiency and energy saving, intelligent speed regulation

[0032] By introducing a variable frequency control mechanism based on load and temperature feedback, the fan speed can be continuously and smoothly adjusted, avoiding frequent start-stop and ensuring that the fan always operates in the high-efficiency range. Compared with traditional fixed-speed fans, energy savings can reach 20%-35%, significantly reducing the energy consumption of auxiliary systems.

[0033] (5) Multi-dimensional sensing for precise temperature control

[0034] Innovatively, the temperature sensor inside the transformer is placed in the hot spot area near the high-voltage winding, and an ambient temperature and humidity sensor is added at the lowest outer edge of the radiator to achieve accurate monitoring of the core heat point and high-risk condensation area, providing high-quality data support for intelligent control.

[0035] (6) Active anti-condensation control strategy

[0036] The system can intelligently start or speed up the fan based on the difference between the ambient dew point and the surface temperature of the radiator, preventing moisture from condensing and forming dirt or reducing insulation strength, thus improving the equipment's adaptability to humid, underground, or harsh outdoor environments.

[0037] (7) High-reliability impeller structure design

[0038] By performing welding and overlay welding on the bent root of traditional riveted blades, a smooth transition reinforcement zone is formed, which effectively eliminates stress concentration, significantly improves the impeller's fatigue resistance, and greatly extends its service life without significantly increasing manufacturing costs.

[0039] (8) Intelligent diagnosis and predictive maintenance

[0040] The first GADF-CNN deep learning diagnostic model based on pressure pulsation signals is created: by performing Gram Angular Field Transform (GADF) on high-frequency pressure pulsation signals to generate images, and inputting them into a convolutional neural network for training and recognition, it can achieve high-precision early warning of initial faults such as blade cracks, loosening, and imbalance, with high diagnostic accuracy and fast response speed.

[0041] (9) Integrated data acquisition and connection solution

[0042] The modular design and multi-core aviation plug integrate all sensor signal lines, enabling "one-click quick connection" and reducing on-site wiring time from several hours to tens of seconds. It also features splash-proof, shock-proof, and anti-misplugging functions, greatly improving engineering deployment efficiency and connection reliability.

[0043] (10) Remote monitoring and digital twin support

[0044] Equipped with a 4G / 5G or Ethernet communication module, it supports remote uploading and cloud storage of operating data, fault alarms, and historical curves, facilitating the construction of digital archives for transformer cooling systems and enabling unattended operation, remote diagnostics, and full lifecycle management.

[0045] In summary, this invention not only achieves a profound evolution from traditional functional components to a green, efficient, and intelligent core system, but also achieves a unified goal of environmental responsibility, energy efficiency improvement, and ultimate reliability through the deep integration of "hardware innovation + software empowerment." It represents the development direction of next-generation power gas-insulated transformer cooling technology and has significant technological advancements and broad market application prospects.

[0046] The following will further explain the concept, specific structure, and technical effects of the present invention in conjunction with the accompanying drawings, so as to fully understand the purpose, features, and effects of the present invention. Attached Figure Description

[0047] Figure 1 This is a schematic diagram of the structure of an embodiment of the present invention;

[0048] Figure 2 This is a schematic diagram of the structure of the circulating centrifugal fan impeller in this invention;

[0049] Figure 3 This is a schematic diagram of the intelligent integrated system for wind turbine operation and fault diagnosis of the present invention;

[0050] Figure 4 This is a schematic diagram showing the arrangement of the temperature sensor and ambient humidity sensor inside the transformer according to the present invention;

[0051] Figure 5 This is a schematic diagram of the multi-core aviation connector of the present invention. Detailed Implementation

[0052] To make the technical problems to be solved, the technical solutions, and the beneficial effects of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0053] In the following description, specific details, such as particular internal procedures and techniques, are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of the invention. However, those skilled in the art will appreciate that the invention may be practiced in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of the invention with unnecessary detail.

[0054] like Figure 1-4 As shown, the present invention provides a smart fan for a gas-insulated transformer, including a circulating centrifugal fan body, a sensor system, a data acquisition and processing module, an intelligent control module, and a remote monitoring platform; the circulating centrifugal fan body is connected to the gas pipeline of the gas-insulated transformer 18 through an inlet flange and an outlet flange, and is directly driven by a motor through a motor shaft connected to the circulating fan impeller via a flange.

[0055] The sensor system is located on the body of the circulating centrifugal fan, and its arrangement follows the principles of "closeness to the heat source, accurate reflection, and ease of maintenance";

[0056] The data acquisition and processing module collects data through a sensor system;

[0057] The intelligent control module integrates the inverter control interface, digital output, and communication interface based on the data collected by the data acquisition and processing module.

[0058] The remote monitoring platform has a built-in 4G / 5G DTU or industrial Ethernet module, and all operating data is packaged and uploaded to the cloud server or remote monitoring center in 1-minute cycles.

[0059] The circulating centrifugal fan body is designed for environmentally friendly mixed gases of 8% C4F7N / 92% CO2, and is also compatible with sulfur hexafluoride gas in emergency situations or special applications. Due to the inherent advantages of SF6, such as high specific heat capacity, high density (i.e., high heat capacity), high heat transfer efficiency, and good insulation properties, SF6 is suitable for applications where environmentally friendly gases can be used.

[0060] like Figure 1 As shown, the inlet flange and inlet cone section 1 of the circulating centrifugal fan are connected to the circulating gas pipeline via sealing rings and bolts. After the airflow is accelerated and depressurized by the inlet cone section, it enters the impeller area 2. The impeller performs work on the airflow, which is then pressurized and decelerated by the fan casing 3 before entering the airflow circulation pipeline. The outlet flange 4 is connected to the circulating gas pipeline via sealing rings and bolts. The fan is directly driven by the motor 11, which is connected to the fan via a connecting flange 8, with a sealing gasket 6 between the flanges. To facilitate the acquisition of temperature and pressure pulsations inside the fan, fine-threaded holes are drilled in the fan casing 3 for the installation of temperature sensor 5 and pressure pulsation sensor 7, respectively. To ensure a compact system structure, the intelligent integrated platform 10 for fan operation and fault diagnosis is installed on a platform on one side of the motor 11. The sensor wires of temperature sensor 5 and pressure pulsation sensor 7 are routed through conduit 9.

[0061] In this embodiment of the invention, the main body of the circulating centrifugal fan is largely the same as that of an existing circulating centrifugal fan, the difference being the impeller, such as... Figure 2 As shown, the impeller includes a front cover plate 13, a rear cover plate 14, blades 15, and a hub 16. The blades are positioned between the front cover plate 13 and the rear cover plate 14, and the hub 16 is positioned on the rear cover plate 14. After the blades 15 are riveted to the front cover plate 13 and the rear cover plate 14, argon arc welding is used at the riveting point 12 to perform repair welding and overlay welding, forming a smooth transition reinforced zone. This significantly reduces the stress concentration factor and improves the fatigue life of the impeller under high-speed rotation and gas pulsation loads. The overlay welding process requires control of heat input to avoid deformation of the cover plate.

[0062] To ensure the interchangeability of the newly designed impeller with the original impeller as much as possible, and to minimize the cost of equipment upgrades, the static pressure, volumetric flow rate, impeller diameter, and rotational speed of the newly designed impeller in the air medium are kept consistent with those of the original impeller. The fan is driven by a direct electric motor. When the working fluid is converted into a low-heat-capacity environmentally friendly gas (a gas with low density or low specific heat capacity), if the power of the centrifugal fan exceeds the rated power of the electric motor, the fan speed can be appropriately increased. In this embodiment, the impeller adopts a forward-curved multi-bladed centrifugal impeller, and the specific design parameters are as follows:

[0063] Table 1 Design parameters of the wind turbine impeller

[0064]

[0065] (1) Aerodynamic design and media adaptability optimization

[0066] This fan is specifically designed for environmentally friendly mixed gases of 8% C4F7N / 92% CO2. The density of this mixed gas at 20℃ and 0.14MPa is approximately 2.694 kg / m³. 3 It is 2.25 times that of air, but only about 1 / 3 that of SF6. Through CFD simulation and aerodynamic redesign, it is ensured that the impeller can still provide a static pressure of no less than 2450 Pa and 28 m³ in environmentally friendly gases while maintaining the original rotational speed (900 r / min) and installation dimensions. 3 The flow rate is [value] / min to meet the cooling requirements of the transformer. Finally, the design parameters of the fan impeller are summarized in Table 3.

[0067] Table 2 Summary of blade geometric parameters

[0068]

[0069] The impeller's rear cover plate also includes a balance hole 17.

[0070] The impeller's static pressure, volumetric flow rate, impeller diameter, and rotational speed in the air medium remain consistent with the original impeller. The impeller is a forward-curved centrifugal impeller with a high-acceleration flow channel. High-acceleration flow channels are characterized by high efficiency and low noise; previous studies have also demonstrated that using a high-acceleration flow channel suppresses interference at the blade inlet. The impeller comprises 52 blades with a 35° blade exit angle and a 17.27° inlet angle.

[0071] Secondly, the sensor system includes internal temperature sensors for the wind turbine, internal pressure sensors for the wind turbine, internal temperature sensors for the transformer, and environmental sensors; specifically,

[0072] Sensor systems are the foundation for intelligent sensing, and their deployment follows the principles of "closeness to heat sources, accurate reflection, and ease of maintenance," specifically as follows: Figure 1 and 3 As shown.

[0073] (1) Internal sensors of the fan

[0074] Temperature sensor 5 inside the fan: It adopts a PT100 platinum resistance thermometer and is embedded in the non-direct scouring position inside the volute (3) to monitor the temperature of the gas at the fan outlet.

[0075] Pressure pulsation sensor 7 inside the fan: It adopts a high-frequency piezoelectric sensor and is installed near the tongue of the volute. It is used to collect pressure pulsation signals with the same frequency and harmonic frequency as the impeller rotation. The sampling frequency is not less than 10 kHz.

[0076] (2) Sensors inside the transformer

[0077] Transformer internal temperature sensor 23: It adopts a fiber optic temperature sensor or an electromagnetic interference resistant platinum resistance thermometer, and is directly arranged on the insulation support near the hottest section of the high voltage winding 21 to monitor the temperature of the most sensitive part of the transformer in real time.

[0078] (3) Environmental sensors

[0079] Ambient temperature and humidity sensor 25: An integrated temperature and humidity transmitter installed on the inner side of the manifold at the outermost and lowest point of the transformer radiator 24, used to accurately monitor the microenvironment at the location most prone to condensation.

[0080] (4) Signal integration and transmission

[0081] All sensor cables are led out through a dedicated conduit 9 and converged into a waterproof and shockproof multi-core aviation connector (e.g., Figure 5 (As shown). This plug connects to the corresponding socket on the data acquisition box, enabling a quick and reliable "one-click" connection for all signals, greatly simplifying on-site installation and subsequent maintenance.

[0082] The intelligent control module includes a fault diagnosis module, which is used to determine the probability of the current state belonging to various types of faults. If the probability of any fault exceeds the set threshold, the impeller fault alarm module will be activated immediately, issuing an audible and visual alarm, and sending a warning message to the remote monitoring center through the data remote transmission module, including the fault type, probability and timestamp.

[0083] The implementation details of the data acquisition and processing module and the intelligent control system are as follows:

[0084] 1) Hardware Architecture

[0085] The data acquisition and processing module is built on an industrial-grade embedded system and has multi-channel synchronous acquisition capabilities. The control module uses a high-performance PLC or industrial microcontroller and integrates a frequency converter control interface, digital output (alarm), and communication interface.

[0086] (2) Intelligent operation control process

[0087] like Figure 4 As shown, after the system is powered on, the control module executes the following logic in parallel:

[0088] a. Heat load tracking mode

[0089] The system continuously reads the transformer's internal temperature (23°C) and load current signal. A built-in PID algorithm calculates the required cooling airflow and outputs a 4-20mA signal to the frequency converter, continuously adjusting the fan speed to stabilize the transformer's hot spot temperature within the set range.

[0090] b. Anti-condensation control mode

[0091] Read the value from the ambient temperature and humidity sensor 25 and calculate the current dew point temperature. T d Estimate the surface temperature of the heat sink. T s (This can be achieved by using a model or by adding a surface temperature sensor). If If a condensation risk is detected, the control module will prioritize implementing anti-condensation strategies, such as appropriately increasing the minimum fan speed or starting the fan earlier, to ensure... .

[0092] c. Fault Diagnosis Mode

[0093] This mode operates independently and continuously.

[0094] (3) Specific implementation of the fault diagnosis system

[0095] Fault diagnosis is the core innovation of this invention, and its process is as follows: Figure 5 As shown:

[0096] a. Signal preprocessing

[0097] The raw signal from the pressure pulsation sensor 7 inside the fan is collected and filtered out by a bandpass filter (usually 0.5 times to 10 times the blade frequency) to remove irrelevant noise.

[0098] b. Feature Image Generation

[0099] A 1-second signal segment is subjected to Gram Angular Field Transform (GADF) to convert it into a 128×128 pixel grayscale image. This image encodes the temporal correlation of the signal into the geometric structure of the image.

[0100] c. Intelligent recognition

[0101] The generated GADF image is input into a pre-trained convolutional neural network (CNN) model. This CNN model structure, which includes two convolutional pooling layers and one fully connected classification layer, has been trained offline using a large amount of historical data (normal, leaf crack, loose, dust accumulation, etc.).

[0102] d. Decision-making and alarm: The CNN model outputs the probability of the current state belonging to various types of faults. If the probability of any fault exceeds a set threshold (e.g., 85%), the impeller fault alarm module will be activated immediately, issuing an audible and visual alarm, and sending a warning message, including the fault type, probability, and timestamp, to the remote monitoring center via the data remote transmission module.

[0103] The remote monitoring platform of this invention is an intelligent integrated platform 10 with a built-in 4G / 5G DTU or industrial Ethernet module. All operating data (temperature, humidity, speed, pressure pulsation characteristic values, status indicators) are packaged and uploaded to the cloud server or remote monitoring center in 1-minute intervals. The platform provides a web interface and a mobile APP, supporting:

[0104] (1) Real-time data dashboard;

[0105] (2) Historical trend curve analysis;

[0106] (3) Fault alarm recording and query;

[0107] (4) Generation of wind turbine health status assessment report;

[0108] (5) Automatic dispatch of maintenance work orders.

[0109] The key point of this invention lies in integrating systemic requirements such as green environmental protection, high efficiency and energy saving, state perception, and intelligent diagnosis through interdisciplinary technological fusion, and implementing them into a specific, engineerable hardware product and software system. Specifically, as follows:

[0110] (1) A specific combination of aerodynamic configuration and structural features

[0111] This system protects against the aforementioned environmentally friendly mixed gas (density between SF6 and air) with a forward-curved, high-acceleration centrifugal impeller featuring a specific number of blades (e.g., 52 blades), a specific blade outlet angle (e.g., 35°), and an inlet angle (e.g., 17.27°). This impeller can meet the performance requirements (static pressure and flow rate) of the original SF6 fan system while maintaining the original installation dimensions and speed. As a substitute medium for SF6, it is not limited to 8% C4F7N / 92% CO2, but can also include other proportions of C4F7N / CO2 mixed gases, or C5F... 10 Environmentally friendly insulating gases such as O / CO2, dry air and fluorinated ketone mixtures can all be refitted and optimized for impellers using the aerodynamic design method of this invention, provided their physical properties are known.

[0112] (2) A specific fault diagnosis method based on pressure pulsation signal

[0113] The protection system employs a complete process: collecting pressure pulsation signals from within the wind turbine casing, transforming them into images using a Gram angle field (GADF), and then inputting the images into a convolutional neural network (CNN) for blade fault type identification and early warning. This is the core algorithmic process for achieving intelligent diagnosis. Besides pressure pulsation signals, vibration acceleration sensors can be installed at the non-drive end of the motor, or motor current characteristic analysis (MCSA) can be used to indirectly monitor the impeller's mechanical condition. In terms of signal processing algorithms, in addition to GADF-CNN, short-time Fourier transform (STFT) can be used to generate time-spectrum graphs, or wavelet transform combined with support vector machine (SVM) and other "signal-to-image + AI recognition" or traditional machine learning schemes can be used for fault diagnosis. However, GADF-CNN has advantages in feature preservation and recognition accuracy.

[0114] (3) A specific manufacturing process to improve the reliability of riveted impellers

[0115] One manufacturing method involves welding and overlaying at the bending and riveting joints of centrifugal fan blades to create a reinforced zone, thereby eliminating stress concentration and improving fatigue life. For strengthening stress concentration points on the impeller, if welding and overlaying are not used, local laser hardening of the blade bending area before riveting can be considered to improve surface hardness and fatigue strength. Alternatively, integral casting or additive manufacturing (3D printing) of the impeller can be used during the design phase to fundamentally eliminate the connection interface. However, these solutions may increase manufacturing costs or be limited by materials and processes.

[0116] (4) A multi-sensor collaborative layout logic and intelligent control strategy

[0117] The protection system employs specific location logic, such as placing the transformer temperature sensor near the high-voltage winding and the ambient temperature and humidity sensor at the lowest point of the outermost edge of the radiator. It also incorporates a hierarchical parallel intelligent control logic based on the fusion of multi-source information, which includes temperature control speed regulation, anti-condensation risk avoidance, and blade fault diagnosis and alarm.

[0118] (5) An integrated system integration and connection solution

[0119] The protection system integrates all sensor signal lines on the wind turbine body into a single multi-core aviation connector, achieving a unified physical integration design for fast and reliable connectivity. Furthermore, remote data transmission is not limited to 4G / 5G or industrial Ethernet; in specific scenarios, low-power wide-area network technologies such as LoRa and NB-IoT, or dedicated fiber optic networks, can also be used for data transmission.

[0120] In summary, this invention seeks to protect a complete, interconnected technical solution, rather than a single, isolated improvement. Its scope covers everything from the design and manufacturing methods of the core component (impeller), to the construction logic of the state-aware network, to the algorithmic flow of intelligent diagnosis and control, and the integrated form of engineering implementation, thus constructing a robust technical protection barrier.

[0121] The preferred embodiments of the present invention have been described in detail above. It should be understood that those skilled in the art can make numerous modifications and variations based on the concept of the present invention without creative effort. Therefore, all technical solutions that can be obtained by those skilled in the art based on the concept of the present invention through logical analysis, reasoning, or limited experimentation on the basis of existing technology should be within the scope of protection defined by the claims.

Claims

1. A smart fan for power gas-insulated transformers, characterized in that, It includes a circulating centrifugal fan body, a sensor system, a data acquisition and processing module, an intelligent control module, and a remote monitoring platform; the circulating centrifugal fan body is connected to the gas pipeline of the gas-insulated transformer through an inlet flange and an outlet flange, and is directly driven by an electric motor through the motor main shaft connected to the flange of the circulating fan impeller; The sensor system is arranged on the body of the circulating centrifugal fan, and its arrangement follows the principles of "closeness to the heat source, accurate reflection, and ease of maintenance"; The data acquisition and processing module acquires data through the sensor system; The intelligent control module integrates a frequency converter control interface, digital output, and communication interface based on the data collected by the data acquisition and processing module. The remote monitoring platform has a built-in 4G / 5G DTU or industrial Ethernet module, and all operating data is packaged and uploaded to the cloud server or remote monitoring center in 1-minute cycles.

2. The smart fan for a gas-insulated power transformer as described in claim 1, characterized in that, The circulating centrifugal fan body is configured for 8%. An environmentally friendly mixed gas that is also compatible with FS6 gas.

3. A smart fan for a gas-insulated power transformer as described in claim 2, characterized in that, The circulating centrifugal fan body includes an impeller, the impeller flow channel adopts a strong acceleration flow channel, the impeller includes a front cover plate, a rear cover plate, blades and a hub, the blades are disposed between the front cover plate and the rear cover plate, and the hub is disposed on the rear cover plate; After the blade is riveted to the front cover plate and the rear cover plate, argon arc welding is used to repair and overlay the joint to form a smooth transition reinforcement zone.

4. A smart fan for a gas-insulated power transformer as described in claim 3, characterized in that, The impeller's rear cover plate also includes balance holes.

5. A smart fan for a gas-insulated power transformer as described in claim 3, characterized in that, The static pressure, volumetric flow rate, impeller diameter, and rotational speed of the impeller in the air medium remain the same as those of the original impeller.

6. A smart fan for a gas-insulated power transformer as described in claim 5, characterized in that, The impeller is a forward-curved centrifugal impeller, comprising 52 blades, with a blade outlet angle of 35° and an inlet angle of 17.27°.

7. A smart fan for a gas-insulated power transformer as described in claim 1, characterized in that, The sensor system includes a fan internal temperature sensor, a fan internal pressure sensor, a transformer internal temperature sensor, and an environmental sensor. The internal temperature sensor of the fan is embedded in the volute, the internal pressure sensor of the fan is installed on the tongue of the volute, the internal temperature sensor of the transformer is arranged on the insulation support of the hottest section of the high voltage winding, and the environmental sensor is an integrated temperature and humidity transmitter installed on the inner side of the collector pipe at the outermost and lowest point of the transformer radiator.

8. A smart fan for a gas-insulated power transformer as described in claim 7, characterized in that, All sensor cables are led out through a dedicated conduit and converged into a waterproof and shockproof multi-core aviation connector, which connects to the corresponding socket in the data acquisition box of the data acquisition and processing module.

9. A smart fan for a gas-insulated power transformer as described in claim 1, characterized in that, The data acquisition and processing module continuously reads the transformer's internal temperature and load current signals. It incorporates a built-in PID algorithm to calculate the required cooling airflow and outputs a 4-20mA signal to the frequency converter, continuously adjusting the fan speed to stabilize the transformer's hot spot temperature within the set range. Read the ambient temperature and humidity, and calculate the current dew point temperature. T d Estimate the surface temperature of the radiator T s ;like If a condensation risk is detected, the intelligent control module will prioritize implementing anti-condensation strategies.

10. A smart fan for a gas-insulated power transformer as described in claim 1, characterized in that, The intelligent control module includes a fault diagnosis module, which is used to determine the probability of the current state belonging to various types of faults. If the probability of any fault exceeds a set threshold, the impeller fault alarm module will be activated immediately, issuing an audible and visual alarm, and sending a warning message to the remote monitoring center through the data remote transmission module, including the fault type, probability and timestamp.

Images