Intelligent wind tunnel monitoring system for hydro-generator units based on circular track robot

Abstract

This application proposes an intelligent wind tunnel monitoring system for a hydro-generator unit based on a circular track robot. The system includes a circular track, a robot, a communication module, and a data analysis module. The circular track is installed on the inner wall of the wind tunnel and consists of multiple independent power supply sections. The robot includes a composite sensor module, a carbon brush assembly, a supercapacitor module, and a rechargeable battery. The composite sensor module monitors various parameters within the wind tunnel. The carbon brush assembly contacts the circular track, enabling each power supply section to supply power to the robot, forming a three-tiered power supply architecture combined with the supercapacitor module and rechargeable battery. The communication module transmits monitoring data to the data analysis module. The data analysis module processes the data using a multimodal data fusion algorithm to determine if any anomalies have occurred within the wind tunnel. This system utilizes a circular track robot to automatically inspect various parameters within the wind tunnel, improving the accuracy, comprehensiveness, and real-time nature of wind tunnel monitoring.

Description

Technical Field

[0001] This application relates to the field of hydropower equipment monitoring technology, and in particular to an intelligent monitoring system for wind tunnels of hydropower generator sets based on a circular track robot. Background Technology

[0002] Currently, condition monitoring of hydro-generator units is a crucial aspect of hydropower station operation. Among these components, the wind tunnel, a sealed cavity enclosing the stator and rotor of the hydro-generator unit, is a key part of hydropower equipment, thus requiring precise monitoring of the status of all equipment within the wind tunnel.

[0003] In related technologies, manual inspection is generally used, or monitoring is achieved by setting up corresponding monitoring equipment for the parameters to be monitored (such as monitoring the temperature at a specific point on the stator using a temperature-sensing resistor). However, in practical applications, the monitoring methods mentioned above suffer from problems such as low monitoring efficiency, limited monitoring range leading to incomplete monitoring results, and susceptibility to interference during the monitoring process. Summary of the Invention

[0004] This application aims to at least partially address one of the technical problems in the related art.

[0005] Therefore, the purpose of this application is to propose an intelligent monitoring system for wind tunnels of hydro-generator units based on a circular track robot. This system utilizes a circular track robot to automatically inspect various parameters in the wind tunnel, such as pipeline leakage, stator and rotor status, and environmental parameters. It has high monitoring efficiency and improves the accuracy, comprehensiveness, and real-time performance of wind tunnel monitoring.

[0006] To achieve the above objectives, this application proposes an intelligent wind tunnel monitoring system for a hydro-generator unit based on a circular track robot. The system includes: a circular track, a robot, a communication module, and a data analysis module; wherein,

[0007] The circular track is installed on the inner wall of the wind tunnel of the hydro-generator unit. The circular track is composed of multiple independent power supply sections. The robot travels along the circular track to perform periodic inspections. The robot includes a composite sensor module, a carbon brush assembly, a supercapacitor module, and a rechargeable battery. The composite sensor module is used to monitor the parameters of various devices and environmental parameters inside the wind tunnel. The carbon brush assembly arranged at the bottom of the robot contacts the circular track to supply power to the robot through each of the power supply sections. The supercapacitor module is used to supply power to the robot as it crosses the connection between two adjacent power supply sections, and the rechargeable battery is used to supply power to the robot in case of abnormal track power supply. The communication module is used to send the data collected by the composite sensor module to the data analysis module; the data analysis module is used to process the received data through a multimodal data fusion algorithm to determine whether an anomaly has occurred in the wind tunnel.

[0008] Optionally, in one embodiment of this application, the composite sensor module includes: an infrared thermal imager for monitoring temperature field distribution data of various devices inside the wind tunnel; a microphone array for monitoring acoustic data generated by the mechanical structure of various devices inside the wind tunnel; and a multispectral camera for monitoring images of various devices inside the wind tunnel and environmental images around the devices.

[0009] Optionally, in one embodiment of this application, the communication module includes: an optical fiber communication unit and a wireless communication unit; wherein, the optical fiber communication unit includes a rotor end and a stator end, the rotor end is arranged on the robot, the stator end is arranged on the circular track, the rotor end is used to send the data collected by the composite sensor module to the stator end, and the stator end establishes a communication connection with the data analysis module through an optical fiber line; specifically, the communication module is used to transmit various monitoring data in sequence according to data priority in an interference-affected communication mode.

[0010] Optionally, in one embodiment of this application, the annular track includes: a plurality of insulating inlay components, the plurality of insulating inlay components being used to separate each of the power supply sections; each of the insulating inlay components being connected to a fixed busbar of the annular track via an elastic compensation mechanism.

[0011] Optionally, in one embodiment of this application, multiple sets of carbon brushes are respectively provided on the positive and negative circuits of the robot. Each set of carbon brushes further includes a bidirectional spring mechanism for adjusting the pressure applied to the carbon brushes. The pressure applied to the carbon brushes is determined based on the contact condition and wear condition between the carbon brushes and the annular track.

[0012] Optionally, in one embodiment of this application, the data analysis module is specifically used for: performing image recognition on the images of the various devices and the environmental images, determining the target component with newly added bright spots based on the reflection characteristics in the multi-round inspection images; analyzing the temperature field distribution data of the target component in the multi-round inspection, and determining whether the target component has leaked based on the temperature change trend of the target component.

[0013] Optionally, in one embodiment of this application, the data analysis module is specifically used to: compare the temperature field distribution data of the various devices with the historical temperature data of the corresponding devices under the same load conditions; and determine that any device has an overheating abnormality if the temperature difference between the temperature field distribution data of any device and the historical temperature data is greater than a temperature difference threshold, or if the temperature field distribution data is greater than a temperature threshold.

[0014] Optionally, in one embodiment of this application, the data analysis module is specifically used to: compare the voiceprint data generated by the mechanical structure of the various devices with the reference voiceprint spectrum of the corresponding devices under normal operating conditions; determine the frequency band of the voiceprint data, and combine the voiceprint comparison result and the weight corresponding to the frequency band to perform mechanical fault assessment on the various devices.

[0015] Optionally, in one embodiment of this application, the data analysis module is specifically used to: perform weighted calculations by combining the temperature field distribution data, the acoustic print data, and the image data to obtain a comprehensive fault score for each device; and to issue a fault warning when the comprehensive fault score exceeds a scoring threshold.

[0016] Optionally, in one embodiment of this application, the inspection operation height of the robot is kept horizontal with the upper end of the stator of the hydro-generator set, and the annular track is fixed by a wall-fixed foundation arranged around the inner wall of the wind tunnel, or by a hoisting bracket erected from the top of the wind tunnel downwards.

[0017] The technical solutions provided by the embodiments of this application offer at least the following beneficial effects: This application utilizes a circular track robot for automated inspection, avoiding safety hazards to workers caused by the harsh environment inside the wind tunnel. The robot can continuously and uniformly complete a full-circle inspection at a set frequency, significantly improving inspection efficiency and preventing missed inspections. Furthermore, this application expands the monitoring range, achieving comprehensive and multi-dimensional accurate monitoring through composite sensors, enabling a complete understanding of the operating status of various equipment within the wind tunnel. Moreover, this application employs a three-level power supply architecture consisting of a segmented power supply track, a supercapacitor, and a built-in rechargeable battery, coupled with a redundant power extraction design using bidirectional spring-pressurized carbon brushes. This allows for power supply during track segment transitions and sudden track power outages, ensuring stable power supply under normal, segment-crossing, and emergency operating conditions, guaranteeing no data loss throughout the monitoring process. Furthermore, this application utilizes multi-modal data fusion for early and accurate fault identification and tiered alarms, significantly improving the early fault detection rate, reducing false alarm rates, and preventing unit accidents caused by sudden faults. Therefore, this application improves the accuracy, comprehensiveness and real-time performance of wind tunnel monitoring, which is conducive to ensuring the safe and stable operation of hydro-generator units.

[0018] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0019] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1 This is a schematic diagram of the structure of a wind tunnel intelligent monitoring system for a hydro-generator unit based on a circular track robot, as proposed in an embodiment of this application. Figure 2 This is a schematic diagram of the structure of a specific robot proposed in an embodiment of this application. Detailed Implementation

[0020] Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain the present invention, and should not be construed as limiting the present invention.

[0021] It should be noted that the environment inside the wind tunnel of the hydro-generator unit is extremely harsh, with not only high-decibel noise but also strong electromagnetic interference. This makes manual inspection not only inefficient but also poses a serious threat to the health of the inspection personnel. The temperature-measuring resistors used in the relevant embodiments have a limited monitoring range, only able to monitor the temperature at specific points inside the stator, failing to cover critical parts such as the ends, making it difficult to comprehensively grasp the temperature status of the equipment. Furthermore, leaks in the water supply pipeline inside the wind tunnel are difficult to detect in their early stages, and sudden leaks can easily trigger stator short-circuit shutdown accidents, severely affecting the stable operation of the equipment.

[0022] To improve the aforementioned monitoring situation, some related embodiments have proposed monitoring schemes using multiple types of fixed sensors and visual inspection equipment. However, the solutions in these embodiments also have significant limitations: fixed sensors are limited by their installation location, resulting in a very limited coverage area and making it difficult to achieve comprehensive monitoring of the equipment; wireless transmission methods suffer from extremely poor signal stability in environments with strong electromagnetic interference, easily leading to data loss or transmission interruptions, and failing to guarantee reliable transmission of monitoring data; visual inspection technology, in harsh environments with low light and strong interference, suffers from severely affected image quality, resulting in a significantly increased false alarm rate and making it difficult to accurately identify the actual status of the equipment.

[0023] Therefore, this application proposes an intelligent monitoring system for wind tunnels of hydro-generator units based on a circular track robot. This system utilizes a circular track robot to automatically inspect various parameters in the wind tunnel, such as pipeline leakage, stator and rotor status, and environmental parameters.

[0024] The following description, with reference to the accompanying drawings, illustrates an intelligent wind tunnel monitoring system for a hydroelectric generator set based on a circular track robot, as proposed in an embodiment of the present invention.

[0025] Figure 1 This is a schematic diagram of the structure of an intelligent wind tunnel monitoring system for a hydro-generator unit based on a circular track robot, as proposed in an embodiment of this application. Figure 1 As shown, the monitoring system includes: a circular track 10, a robot 20, a communication module 30, and a data analysis module 40.

[0026] The circular track 10 is installed on the inner wall of the wind tunnel of the hydro-generator unit. The circular track 10 is composed of multiple independent power supply sections. The robot 20 travels along the circular track 10 to perform periodic inspections.

[0027] like Figure 2 As shown, the robot 20 includes a composite sensor module 21, a carbon brush assembly 22, a supercapacitor module 23, and a rechargeable battery 24. The composite sensor module 21 is used to monitor the parameters of various devices and environmental parameters in the wind tunnel. The carbon brush assembly 22, which is arranged at the bottom of the robot 20, is in contact with the circular track to supply power to the robot 20 through each power supply segment.

[0028] The supercapacitor module 23 is used to supply power to the robot 20 when the robot crosses the connection between two adjacent power supply segments, and the rechargeable battery 24 is used to supply power to the robot 20 in case of abnormal track power supply.

[0029] The communication module 30 is used to send the data collected by the composite sensor 21 module to the data analysis module 40; the data analysis module 40 is used to process the received data through a multimodal data fusion algorithm to determine whether an anomaly has occurred in the wind tunnel.

[0030] Specifically, the circular track robot system used in this application includes a high-precision circular guide rail installed on the inner wall of the wind tunnel, and a robot that moves along the guide rail to perform inspections. The robot 20 rotates at a constant speed under the drive of a servo motor, and the robot 20 should have explosion-proof function.

[0031] The rotating track robot of this application integrates a composite sensor module containing various types of detection equipment, enabling it to monitor various data to be monitored within the wind tunnel. Furthermore, the rotating track robot also integrates a composite power supply system, which can continuously power the robot in normal operation mode and in special modes such as power supply anomalies, ensuring the stability of monitoring.

[0032] Furthermore, the communication module 30, which employs an anti-interference design, sends the data monitored by the robot 20 to the data analysis module 40. The data analysis module 40 can perform remote data analysis to determine whether there are any abnormal conditions inside the wind tunnel based on the monitoring data. For example, the data analysis module 40 could be a data processing device in a remote control center or a cloud server.

[0033] Based on the above embodiments, the various modules in the intelligent monitoring system of this application will be described in detail below.

[0034] In one embodiment of this application, the composite sensor module includes: an infrared thermal imager for monitoring temperature field distribution data of various devices inside the wind tunnel; a microphone array for monitoring acoustic data generated by the mechanical structure of various devices inside the wind tunnel; and a multispectral camera for monitoring images of various devices inside the wind tunnel and environmental images around the devices.

[0035] For example, infrared thermal imagers can monitor the temperature field distribution of components such as the stator end and connectors of a hydroelectric generator set. High-sensitivity microphone arrays monitor the noise signatures emitted by various mechanical devices within the wind tunnel for noise spectrum analysis and to detect anomalies in the mechanical structure. Multispectral cameras can capture image data of various devices within the wind tunnel, as well as environmental data of the surrounding environment (such as the ground beneath the devices), allowing image recognition algorithms to identify the reflective characteristics of water stains and oil stains in the images.

[0036] This application adopts a three-level power supply architecture for the robot 20, consisting of a segmented track, a supercapacitor, and a rechargeable battery, which can solve the problem of continuous power supply when crossing segmented tracks and during maintenance power outages.

[0037] In one embodiment of this application, the circular track includes: a plurality of insulating inserts for separating various power supply sections; each insulating insert is connected to a fixed busbar of the circular track via an elastic compensation mechanism.

[0038] Specifically, in this embodiment, the segmented power supply track has insulating material embedded inside the circular track 10. Multiple insulating inlay components divide the circular track 10 into 4-6 independent power supply segments, each with an arc length of 60 to 90 degrees. The robot 20's power acquisition module is a carbon brush assembly installed at its bottom. The carbon brush assembly contacts the copper busbars on the power supply segments, thereby supplying power to the robot 20 from the power supply segments.

[0039] It should be noted that if a complete metal track is used, the circular track 10 will be a huge circular conductor, and the current will flow throughout the entire loop, leading to problems such as unstable voltage and a large fault range. Once a problem occurs, the entire track needs to be replaced, which is very difficult to construct.

[0040] Therefore, in this embodiment, the insulating device is embedded at the junction of adjacent power supply sections. Utilizing the non-conductive physical property of the insulated component, the entire circular metal conductive track is physically separated and electrically insulated, making each power supply section an independent conductive unit. This ensures that each section of the track can normally supply power to the robot while preventing short circuits between sections. When the robot 20 travels on the circular track 10, each power supply section serves as the physical structure through which the robot 20 directly contacts and draws power, providing a continuous DC power supply to the robot 20 and simultaneously charging the robot 20's built-in lithium battery pack.

[0041] Each insulated component is connected to the fixed busbar of the circular track via a flexible compensation mechanism. The fixed busbar is the external fixed power supply trunk line that provides power to the segmented power supply track. It is the external power supply head of the entire monitoring system. The copper busbar of each section of the conductive track is connected to the external fixed busbar through the flexible compensation mechanism to deliver stable DC power to each independent power supply section of the circular track.

[0042] Because the wind tunnel contains the core cavity of the hydroelectric generator unit, it continuously generates a large amount of heat during operation, causing significant fluctuations in ambient temperature. Simultaneously, the circular track itself is made of metal, and temperature changes trigger thermal expansion and contraction, potentially leading to poor contact between the track and the power supply lines. Therefore, this embodiment equips each insulating inlay component (indirectly, each power supply segment) with an elastic compensation mechanism. The elastic deformation of this mechanism adaptively matches the thermal expansion and contraction displacement of the corresponding power supply segment, preventing contact problems caused by thermal expansion and contraction.

[0043] Furthermore, the annular track 10 in this embodiment also includes a power supply slip ring to achieve rotational power supply. The power supply slip ring is a conductive connection component designed for robots that perform rotational movements. Its core function is to avoid the wire tangling problem caused by traditional power cables when the robot rotates. This application achieves continuous power supply during robot rotation through a rotational contact method, which can prevent severe wire tangling caused by the wires circling and inspecting with the robot.

[0044] In one embodiment of this application, multiple sets of carbon brushes are respectively provided on the positive and negative circuits of the robot. Each set of carbon brushes further includes a bidirectional spring mechanism, which is used to adjust the pressure applied to the carbon brush. The pressure applied to the carbon brush is determined based on the contact condition and wear condition between the carbon brush and the annular track.

[0045] Specifically, this application installs a bidirectional spring-pressurized carbon brush assembly at the bottom of the robot 20, with multiple bidirectional spring-pressurized carbon brush assemblies set for both the positive and negative poles to achieve a redundant design, and the pressure of the carbon brush assembly is adjustable.

[0046] By adjusting the pressure of the carbon brush assembly, power interruption of the annular track 10 can be prevented, and the wear rate of the equipment can be controlled. Since the annular track 10 cannot be perfectly circular in practical applications, and the robot 20 will experience slight vibrations during operation, the elasticity of the bidirectional spring mechanism ensures that the carbon brushes remain in close contact with the track surface, compensating in real time for gap changes caused by mechanical tolerances, vibration, or thermal expansion and contraction, ensuring a consistently unobstructed current path from the annular track 10 to the robot 20. Furthermore, by adjusting the spring pressure, an optimal balance point can be found. This avoids both insufficient pressure leading to high contact resistance, sparking, or even power outages, and excessive pressure increasing frictional resistance and accelerating wear on the carbon brushes and track.

[0047] Furthermore, in this embodiment, two sets of carbon brushes are independently installed on the positive and negative power supply circuits of the robot. The positive carbon brushes are in contact with the positive copper busbar of the track, responsible for obtaining positive electricity from the track to provide power input to the robot. The negative carbon brushes are in contact with the negative copper busbar of the track, responsible for sending the robot's return current back to the track, completing the conductive circuit. In this embodiment, the two sets of positive and two sets of negative carbon brushes operate independently, and the two sets of brushes of the same polarity are connected in parallel. This dual redundancy ensures that power outages caused by a single set failure are avoided.

[0048] This application also integrates a supercapacitor module and a rechargeable battery into robot 20, making robot 20 an independent mobile terminal with built-in power outage buffer. If there is a brief problem with the track power supply, robot 20 can continue to operate by relying on its own energy storage components, improving the robot's lifespan and the system's reliability. Among them, the rechargeable lithium battery pack provides emergency power when the track power is lost, while the supercapacitor module provides power when crossing sections.

[0049] The following is a detailed description of the scenario in which the three-level power supply architecture of this application powers the robot 20: As a first example, the normal power supply mode. When robot 20 is running normally, the carbon brush assembly is in continuous contact with the copper busbars of the track, providing DC power while simultaneously charging the built-in rechargeable battery pack and supercapacitor of robot 20.

[0050] As a second example, the cross-segment power supply switching mode. During the inspection process of robot 20, visual analysis is performed on the images captured by the multispectral camera to determine the current position of robot 20, or the existing UWB system on site is used to determine the position information of robot 20. When it is determined that robot 20 has moved to the track segment (i.e., the connection between two adjacent power supply segments), the supercapacitor module switches to the power supply system of robot 200ms in advance based on the detected position signal, maintaining the power supply to the various sensors and communication modules 30 on robot 20. After the cross-segment is completed, it automatically switches back to the track power supply mode, with no data loss throughout the process.

[0051] As a third example, the emergency mode. If the track power supply fails completely (e.g., due to a power outage caused by maintenance of the circular track 10), the system switches to lithium battery power. The high-temperature resistant lithium battery pack serves as an emergency power source, maintaining basic monitoring functions for several minutes when the track power supply is interrupted. The robot 20 automatically enters a low-power mode, maintaining only infrared thermal imaging and leak detection functions, with the sampling frequency reduced by 50%.

[0052] In one embodiment of this application, the communication module includes: an optical fiber communication unit and a wireless communication unit; wherein, the optical fiber communication unit includes a rotor end and a stator end, the rotor end is arranged on the robot, the stator end is arranged on a circular track, the rotor end is used to send the data collected by the composite sensor module to the stator end, and the stator end establishes a communication connection with the data analysis module through an optical fiber line; the communication module is specifically used to transmit various monitoring data in sequence according to data priority in an interference-affected communication mode.

[0053] Specifically, the communication module 30 in this embodiment adopts an anti-interference design that combines optical fiber communication with wireless communication to avoid electromagnetic interference and ensure real-time transmission of key data.

[0054] The wireless communication module can be integrated inside the robot 20. This module can employ various wireless communication methods, such as a WiFi module and a LoRa module. The WiFi and LoRa modules form a parallel dual-wireless transmission path, achieving redundant design for wireless communication. Leveraging the relatively high transmission rate of the WiFi protocol, it transmits larger volumes of monitoring data. Meanwhile, utilizing the stronger anti-interference capabilities, longer transmission distance, and lower power consumption of the LoRa protocol, it transmits key alarm data and core monitoring parameters.

[0055] Furthermore, the fiber optic slip ring used in this embodiment consists of a rotor and a stator. The rotor end (i.e., the rotating end) is mounted on the robot and rotates with it. Robot data (such as high-definition video and thermal images) is converted into optical signals via an onboard network port and then sent to the rotor end through a photoelectric conversion module. The rotor end is fixed to the robot's shell or chassis, forming a single unit with the robot. When the robot moves on the circular track, the rotor end follows and rotates in the same speed and direction. The stator end (the fixed end) is mounted on a fixed bracket on the inner or outer side of the circular track and connected to a fixed optical fiber leading to the data analysis module 40.

[0056] The communication module 30 in this embodiment adopts an anti-interference communication scheme, which realizes redundant transmission based on fiber optic slip ring and wireless module. When it is switched to low data communication mode due to environmental interference, the data to be transmitted is transmitted in batches according to preset priority. For example, leakage signal takes priority over temperature data.

[0057] In one embodiment of this application, the robot's inspection operation height is kept horizontal with the upper end of the stator of the hydro-generator set, and the circular track is fixed by a wall-fixed foundation arranged around the inner wall of the wind tunnel, or by a hoisting bracket erected from the top of the wind tunnel downwards.

[0058] Specifically, this embodiment requires that the robot 20 maintain a horizontal position at the final inspection height with the upper end of the stator of the unit, thereby ensuring that the sensors can accurately align with the core monitoring areas such as the stator end and pipelines. Two installation and deployment methods can be used to achieve this requirement.

[0059] The first method involves constructing a mounting base on the annular sidewall of the wind tunnel, level with the upper end of the stator, along the circumferential direction. A high-precision annular guide rail is then fixed to this wall base. The robot 20 is mounted on the guide rail and performs a circumferential rotational inspection along the wall, with the working height precisely aligned with the upper end of the stator.

[0060] The second option is to build a hoisting bracket from the top plate of the wind tunnel cavity downwards if there is insufficient installation space on the wind tunnel wall (e.g., dense pipes on the side of the wall or obstruction by other equipment). The circular guide rail is then suspended below the top plate through the bracket, and the horizontal height of the guide rail remains consistent with the upper end of the stator. The robot 20 is hoisted on the guide rail to perform circumferential rotation inspection. The working height remains unchanged, only the installation carrier is changed from the wall to the top.

[0061] The data analysis module 40 of this application incorporates various intelligent data analysis algorithms and anomaly diagnosis models, enabling it to analyze the data monitored by the composite sensor module 21 and determine in real time whether any anomalies exist. For example, using a dynamic leakage detection algorithm, it compares image features at the same location within a continuous rotation cycle of the robot 20 to determine changes in ground reflectivity and metal surface humidity, and then combines this with temperature change trends to identify early signs of leakage. Another example is the use of a multimodal fault early warning model, which integrates abnormal noise detected by voiceprint recognition, component temperature determined by thermal imaging analysis, and leakage phenomena identified by visual inspection; these three channels of data are then weighted to calculate the fault probability.

[0062] In one embodiment of this application, the data analysis module is specifically used for: performing image recognition on images of various devices and environmental images; determining the target component with newly added bright spots based on the reflection characteristics in the multi-round inspection images; analyzing the temperature field distribution data of the target component in the multi-round inspection; and determining whether the target component has leaked based on the temperature change trend of the target component.

[0063] For example, during leak detection, the robot 20 completes a patrol every 30 minutes, acquiring images of pipe connections using a multispectral camera. By comparing the images with the previous 10 historical images at the pixel level, it detects new bright spots at the flange joint (which can be determined based on the reflective characteristics of water stains). Simultaneously, it checks whether the temperature at this location has decreased (due to the evaporative heat absorption effect). After confirming a leak through these two checks, an alarm is triggered.

[0064] In one embodiment of this application, the data analysis module is specifically used to: compare the temperature field distribution data of various devices with the historical temperature data of the corresponding devices under the same load conditions; and determine that any device has an overheating abnormality if the temperature difference between the temperature field distribution data of any device and the historical temperature data is greater than the temperature difference threshold, or if the temperature field distribution data is greater than the temperature threshold.

[0065] For example, during stator overheating monitoring, an infrared thermal imager scans the stator end to generate a three-dimensional temperature field cloud map. The current three-dimensional temperature field cloud map is then compared with historical data under the same load conditions. If the temperature difference between the current and historical temperatures in any region is greater than 5℃, or if the temperature at a certain point in the current temperature field is greater than 80℃, the region is determined to be experiencing localized overheating and is considered an abnormal region.

[0066] In one embodiment of this application, the data analysis module is specifically used to: compare the acoustic fingerprint data generated by the mechanical structure of various devices with the reference acoustic fingerprint spectrum of the corresponding devices under normal operating conditions; determine the frequency band of the acoustic fingerprint data; and combine the acoustic fingerprint comparison results and the weights corresponding to the frequency bands to perform mechanical fault assessment on various devices.

[0067] Specifically, in this embodiment, when monitoring mechanical structure anomalies, the characteristics of mechanical faults are converted into mathematically calculable fault probabilities to construct an accurate early warning model. Since there is a correspondence between mechanical fault mechanisms and acoustic signature frequencies, different mechanical faults produce different vibration and noise frequencies. Therefore, this embodiment pre-sets corresponding weights for different acoustic signature frequency bands.

[0068] The mid-frequency band (2kHz - 4kHz) has a weight of 60%. This band is typically the region where the fault characteristic frequencies of bearing components (inner ring, outer ring, and rolling elements) are located. Once the energy in this band changes abruptly, it is highly likely that the bearing has suffered early damage, which is one of the most common and serious faults in hydroelectric generators. Therefore, this band is given the highest weight.

[0069] The mid-to-high frequency band (4kHz - 8kHz) has a weight of 30%. This frequency band may be related to gear meshing failures (such as pitting, scuffing), friction, or impact responses caused by slight loosening.

[0070] The weight of the remaining frequency bands is set to 10%. For example, the high frequency band (>8kHz) is often related to background noise caused by poor lubrication, slight friction, etc., and the fault directionality is relatively weak, so the weight is set to the lowest.

[0071] In this embodiment, the robot 20 compares the voiceprint generated by a certain device currently collected with the reference voiceprint spectrum stored in the voiceprint library during normal operation. If the currently collected voiceprint does not match the reference voiceprint, the frequency band of the currently collected voiceprint is further determined, and the probability of mechanical failure of the device is calculated according to the corresponding weight.

[0072] In one embodiment of this application, the data analysis module is specifically used to: perform weighted calculations by combining temperature field distribution data, acoustic fingerprint data, and image data to obtain a comprehensive fault score for each device; and issue a fault warning when the comprehensive fault score exceeds a scoring threshold.

[0073] For example, this embodiment performs comprehensive diagnosis of hydropower unit anomalies. First, mechanical anomaly monitoring (such as bearing wear, loose bolts, and insulation degradation) is performed on the hydropower station units. This is done by comparing the data with a baseline acoustic signature in a database to monitor energy fluctuations in relevant frequency bands in real time. Then, combined with data from an infrared thermal imager, analysis shows that the temperature gradient at key connection points exceeds 2°C / cm, indicating a temperature anomaly. Visual vibration detection results from multispectral cameras are also acquired. The three-channel data are fused and output as a comprehensive fault score using an established fault probability model (which pre-assigns weights to various data types, such as 40% for acoustic signature data, 30% for temperature data, and 30% for visual data). A level-three warning is triggered when the comprehensive score exceeds a threshold (e.g., 0.75).

[0074] As an example, in the process of early leakage warning through multimodal fusion, the leakage fault score of the current equipment is determined by combining three channels of data: polarized light vision (reflecting water stain reflectivity), infrared thermometry (reflecting the evaporation heat absorption effect), and acoustic analysis (reflecting droplet impact noise). When the leakage fault score is greater than the corresponding threshold, an early leakage warning is issued to the staff.

[0075] In summary, the intelligent monitoring system for a hydro-generator wind tunnel based on a circular track robot, as described in this application, utilizes an automated inspection method with a circular track robot. This avoids safety hazards to operators caused by the harsh environment inside the wind tunnel. The robot can continuously and uniformly complete a full circle inspection at a set frequency, significantly improving inspection efficiency and preventing missed inspections. Furthermore, the system expands the monitoring range, achieving comprehensive and multi-dimensional accurate monitoring through composite sensors, enabling a complete understanding of the operating status of various equipment within the wind tunnel. The system employs a three-level power supply architecture consisting of a segmented power supply track, a supercapacitor, and a built-in rechargeable battery, coupled with a redundant power extraction design using bidirectional spring-pressurized carbon brushes. This allows for power supply even during track segment transitions and sudden power outages, ensuring stable power supply under normal, segment transition, and emergency operating conditions, guaranteeing no data loss throughout the monitoring process. Moreover, the system utilizes multi-modal data fusion for early and accurate fault identification and tiered alarms, significantly improving the early fault detection rate, reducing false alarm rates, and preventing unit accidents caused by sudden faults. Therefore, this system improves the accuracy, comprehensiveness, and real-time performance of wind tunnel monitoring, which is conducive to ensuring the safe and stable operation of the hydro-generator unit.

[0076] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0077] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "multiple" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0078] In this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," "linking," and "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components, unless otherwise explicitly limited. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0079] In this invention, unless otherwise explicitly specified and limited, "above" or "below" the second feature can mean that the first feature is in direct contact with the second feature, or that the first feature is in indirect contact with the second feature through an intermediate medium. Furthermore, "above," "over," and "on top" of the second feature can mean that the first feature is directly above or diagonally above the second feature, or simply that the first feature is at a higher horizontal level than the second feature. "Below," "below," and "under" the second feature can mean that the first feature is directly below or diagonally below the second feature, or simply that the first feature is at a lower horizontal level than the second feature.

[0080] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A wind tunnel intelligent monitoring system for a hydro-generator unit based on a circular track robot, characterized in that, include: The system comprises a circular track, a robot, a communication module, and a data analysis module; among which, The circular track is installed on the inner wall of the wind tunnel of the hydro-generator unit. The circular track is composed of multiple independent power supply sections. The robot travels along the circular track to perform periodic inspections. The robot includes a composite sensor module, a carbon brush assembly, a supercapacitor module, and a rechargeable battery. The composite sensor module is used to monitor the parameters of various devices and environmental parameters inside the wind tunnel. The carbon brush assembly arranged at the bottom of the robot contacts the circular track to supply power to the robot through each of the power supply sections. The supercapacitor module is used to supply power to the robot as it crosses the connection between two adjacent power supply sections, and the rechargeable battery is used to supply power to the robot in case of abnormal track power supply. The communication module is used to send the data collected by the composite sensor module to the data analysis module; the data analysis module is used to process the received data through a multimodal data fusion algorithm to determine whether an anomaly has occurred in the wind tunnel.

2. The system according to claim 1, characterized in that, The composite sensor module includes: Infrared thermal imagers are used to monitor the temperature field distribution data of various equipment inside the wind tunnel; Microphone arrays are used to monitor acoustic data generated by the mechanical structure of various devices inside the wind tunnel. Multispectral cameras are used to monitor images of various equipment inside the wind tunnel and the surrounding environment.

3. The system according to claim 1, characterized in that, The communication module includes: an optical fiber communication unit and a wireless communication unit; wherein, The optical fiber communication unit includes a rotor end and a stator end. The rotor end is arranged on the robot, and the stator end is arranged on the circular track. The rotor end is used to send the data collected by the composite sensor module to the stator end, and the stator end establishes a communication connection with the data analysis module through an optical fiber line. The communication module is specifically used to transmit various monitoring data in sequence according to data priority in the case of interference-affected communication mode.

4. The system according to claim 1, characterized in that, The circular track includes: Multiple insulating inserts are used to separate each of the power supply segments; Each of the insulating inlay components is connected to the fixed busbar of the annular track via an elastic compensation mechanism.

5. The system according to claim 1, characterized in that, Multiple sets of carbon brushes are respectively installed on the positive and negative circuits of the robot, and each set of carbon brushes further includes: A bidirectional spring mechanism is provided for adjusting the pressure applied to the carbon brush, wherein the pressure applied to the carbon brush is determined based on the contact condition and wear condition between the carbon brush and the annular track.

6. The system according to claim 2, characterized in that, The data analysis module is specifically used for: Image recognition is performed on the images of the various devices and the environmental images, and the target component with newly added bright spots is determined based on the reflection characteristics in the multi-round inspection images. The temperature field distribution data of the target component during multiple rounds of inspection are analyzed, and the leakage of the target component is determined based on the temperature change trend of the target component.

7. The system according to claim 2, characterized in that, The data analysis module is specifically used for: The temperature field distribution data of the various devices are compared with the historical temperature data of the corresponding devices under the same load conditions. If the temperature difference between the temperature field distribution data and the historical temperature data of any device is greater than the temperature difference threshold, or if the temperature field distribution data is greater than the temperature threshold, it is determined that any device has an overheating abnormality.

8. The system according to claim 2, characterized in that, The data analysis module is specifically used for: The acoustic signature data generated by the mechanical structure of the various devices are compared with the reference acoustic signature spectrum of the corresponding devices under normal operating conditions. The frequency band of the voiceprint data is determined, and mechanical fault assessment of the various devices is performed by combining the voiceprint comparison results and the weights corresponding to the frequency bands.

9. The system according to claim 2, characterized in that, The data analysis module is specifically used for: By combining the temperature field distribution data, the acoustic print data, and the image data, a weighted calculation is performed to obtain a comprehensive fault score for each device. A fault warning is issued when the overall fault score exceeds the score threshold.

10. The system according to claim 1, characterized in that, The robot's inspection operation height is kept horizontal with the upper end of the stator of the hydro-generator unit. The circular track is fixed by a wall-fixed foundation arranged around the inner wall of the wind tunnel, or by a hoisting bracket erected from the top of the wind tunnel downwards.

Images