Visual online monitoring system and method for abrasion of key component of impulse turbine

Abstract

The invention provides a visual online monitoring system and method for abrasion of key components of an impulse turbine. A monitoring device capable of moving around a track is carried on an outer ring annular main track of a waterwheel chamber track device through a moving device; the monitoring device comprises a multi-degree-of-freedom mechanical arm module, and an abrasion monitoring module is carried at the tail end of the multi-degree-of-freedom mechanical arm module. A monitoring device capable of moving around a track is carried on an inner ring annular positioning track of the waterwheel chamber track device through a moving device, and the monitoring device adopts a global visual monitoring instrument; the monitoring device and the monitoring device are connected with the intelligent control system; the system further comprises a data transmission and storage module and an energy supply module. Dynamic monitoring and visual presentation of abrasion of the water bucket and the spray needle can be achieved without shutdown, and the device has the advantages of being high in environmental adaptability, high in monitoring precision and automatic in operation.

Description

Technical Field

[0001] This invention belongs to the field of turbine operation and maintenance monitoring technology, and relates to the detection of erosion of flow components of impulse turbines, specifically a visual online monitoring system and method for erosion of key components of impulse turbines. Background Technology

[0002] Southwest my country is rich in hydropower resources, accounting for 65.5% of the country's exploitable resources. The region's high head and large drop topography are suitable for the operation of impulse turbine generator units. However, the sediment content of the rivers is significantly higher than the national average. The high-speed water flow carrying sediment causes severe abrasion and cavitation on key components such as turbine buckets, nozzles, and spray nozzles.

[0003] Erosion causes defects and cracks on the surface of turbine flow components, reducing hydraulic efficiency and increasing the frequency of component replacement and the risk of unit failure. Current traditional erosion monitoring methods mainly include manual visual inspection after shutdown, contact gauge measurement, water pressure testing, and localized grinding and weighing. These methods have significant drawbacks: they are cumbersome, requiring component disassembly and consuming considerable time; unit shutdowns result in significant power generation losses; they rely on human experience, leading to highly subjective test results; they cannot achieve dynamic tracking, only acquiring discrete data, lacking real-time and comprehensiveness, making it difficult to construct a full life-cycle erosion evolution model; and their monitoring dimensions are limited, failing to meet the needs of precise early warning and condition-based maintenance for large hydropower units.

[0004] Existing monitoring technologies, such as ultrasonic testing and infrared testing, can detect some defects, but ultrasonic testing is complex to operate and insensitive to minor surface abrasion, while infrared testing is susceptible to environmental temperature interference and has limited accuracy. Neither can meet the precise visual monitoring requirements of impulse turbines operating under high-speed rotation, strong water mist, and complex conditions. Therefore, developing a non-stop, real-time, accurate abrasion visualization monitoring system that can adapt to complex environments is of great significance for improving the safety and economy of turbine operation. Summary of the Invention

[0005] The purpose of this invention is to provide a visualized online monitoring system and method for the erosion of key components of an impulse turbine, which can realize dynamic monitoring and visualization of the erosion of the water bucket and nozzle without stopping the machine. It has the characteristics of strong environmental adaptability, high monitoring accuracy and automated operation, and effectively overcomes the defects of traditional monitoring methods.

[0006] To achieve the above-mentioned technical features, the purpose of this invention is as follows: a visual online monitoring system for the erosion of key components of an impulse turbine, including a waterwheel chamber track device, wherein a monitoring device capable of moving around the track is mounted on the outer ring main track of the waterwheel chamber track device via a motion device. The monitoring device includes a multi-degree-of-freedom robotic arm module, and an abrasion monitoring module is mounted at the end of the multi-degree-of-freedom robotic arm module. The abrasion monitoring module includes a high-definition industrial camera, a laser rangefinder, a supplementary lighting device, and an environmental adaptation component, which together constitute a multi-sensor fusion detection unit. The inner ring positioning track of the waterwheel chamber track device is equipped with a monitoring device that can move around the track via a motion device. The monitoring device adopts a global visual monitoring instrument. The monitoring and control devices are connected to the intelligent control system; It also includes a data transmission and storage module and an energy supply module.

[0007] Preferably, the waterwheel chamber track device adopts a double-ring concentric track structure, including an outer ring main track and an inner ring positioning track arranged coinciding with the center of the water turbine bucket. The radius of the outer ring main track is greater than the maximum rotational outer diameter of the water turbine bucket and a space is reserved for the extension and retraction of the multi-degree-of-freedom robotic arm module. Both the outer ring main track and the inner ring positioning track include guide rails. The guide rails adopt a three-layer structure: the upper layer is responsible for suspension and guidance, the middle layer is the middle slide, and the lower layer bears the load and friction functions. The motion device is equipped with dual servo motors and a planetary reducer, and achieves differential motion through inner / outer driving wheels and driven devices. It is combined with a disc brake mechanism and adopts a dual-motor redundancy design.

[0008] Preferably, the end of the multi-degree-of-freedom robotic arm module is equipped with a universal interface and a rubber shock absorber, and realizes three-dimensional motion of circular movement, axial extension and multi-angle rotation, and is equipped with an inertial measurement unit; The inertial measurement unit includes three ring laser gyroscopes and three quartz accelerometers, with an operating temperature range of -40℃ to +85℃, and features shock resistance and waterproofing. The intelligent control system includes a hierarchical processor, a motion control unit, an environmental sensing unit, an image processing unit, and an abrasion analysis unit. It adopts a distributed architecture of a main processor and a secondary processor. The main processor is the global decision center, the secondary processor is responsible for image acquisition and processing and motion control, and the environmental sensing unit integrates air pressure, temperature, and humidity sensors. The data transmission and storage module adopts industrial Ethernet transmission, supports real-time data transmission, local caching and remote server storage, and realizes the visualization and historical traceability of monitoring data through the SCADA system. The energy supply module includes a wide-temperature lithium iron phosphate battery and a waterproof wireless charger, supporting automatic charging and unmanned operation.

[0009] Preferably, the emergency braking response time of the disc brake mechanism is ≤0.5s, and in the dual-motor redundancy design, the other motor can maintain 50% load operation when one motor fails. The high-definition industrial camera has a resolution of at least 24.4MP, a frame rate of at least 15fps, an IP67 protection rating, a sapphire lens with an integrated anti-fog heating film, a miniature centrifugal fan and a double-layer explosion-proof microporous flow equalization plate, a built-in near-infrared intensifier and a global shutter, and communicates with the secondary processor via a GPIB bus; the secondary processor performs image denoising, sharpening, and histogram equalization preprocessing through a high-speed data acquisition card. The supplementary lighting device is a high-brightness pulsed light source with a ring array design, a color temperature of 4200K±200K, an illumination angle of 1.6°-11°, and features continuous zoom and strobe synchronization functions. It can dynamically adjust the light intensity and illumination angle based on data from the environmental sensing unit.

[0010] Preferably, the monitoring device is equipped with a long-focus high-definition waterproof camera with an optical zoom of 6-180mm, supports 8 million pixels, and has a sensor size of ≥1 / 1.8". It can ensure noise control under low light conditions, and is equipped with a supplementary light device, a near-infrared enhanced F1.6 aperture and a large target surface sensor. It has a waterproof rating of IP68 and an aluminum alloy anti-corrosion coating on its shell. The environmental adaptation components include a 316L stainless steel protective housing, double sealing rings, a liquid cooling system, and a temperature and humidity sensor. The protective housing is filled with a 10mm thick silicone buffer layer and has a streamlined shape. The interface is sealed with multiple layers of rubber, and a drainage hole is reserved at the bottom. The temperature and humidity sensor has a measurement range of -40℃ to +80℃ and 0 to 100%RH, with a temperature accuracy of ±0.3℃ and a humidity accuracy of ±2%RH.

[0011] Preferably, the image processing unit of the intelligent control system integrates a super-resolution reconstruction algorithm and an instance segmentation algorithm, which can realize the identification of abrasion areas and crack detection; the abrasion analysis unit divides the monitoring results into three levels: slight, moderate and severe, and triggers corresponding early warning mechanisms for different levels.

[0012] Preferably, the wide-temperature lithium iron phosphate battery has an operating temperature of -35℃ to +65℃, a cycle life of ≥4000 cycles, and an IP68 protection rating; the wireless charger has a power of 300W-24V-10A, an IP68 protection rating, and triggers a return-to-home charging command when the battery's usable capacity is below 20%.

[0013] Preferably, the global visual monitoring instrument of the inner ring positioning track is used to follow the monitoring device in the same direction and record the operating status of the multi-degree-of-freedom robotic arm module. When the multi-degree-of-freedom robotic arm module fails, it can temporarily replace some of its functions, or understand the internal situation of the waterwheel chamber through manual control. The intelligent control system supports a multi-condition adaptive monitoring strategy. When multiple nozzles are working, monitoring points are set in the area of ​​non-working nozzles. When the turbine bucket is idling or stationary, full-area coverage is achieved through the circular motion of the track. The monitoring frequency is dynamically adjusted according to the sediment content and flow velocity of the water flow.

[0014] Another aspect of the present invention provides a method for visual online monitoring of erosion of key components of an impulse turbine. The method is implemented using the aforementioned monitoring system and includes the following steps: Step 1, Monitoring System Deployment: A waterwheel chamber track device is laid on the inner wall of the top of the waterwheel chamber. The outer ring main track is equipped with a multi-degree-of-freedom robotic arm module, and the inner ring positioning track is equipped with a global vision monitoring instrument. The abrasion monitoring module is integrated into the end of the multi-degree-of-freedom robotic arm module. The connection and debugging with the intelligent control system and data transmission and storage module are completed, and a reference image database of the initial surface state of the key components of the water turbine is established. Step 2, monitor path planning: The intelligent control system pre-sets key monitoring points for the turbine runner diameter, number of nozzles, structural parameters, nozzle working status, water flow velocity, and operating conditions. It also plans the movement path, dwell time, and shooting angle of the degree-of-freedom robotic arm module to ensure no blind spots in detection. Step 3, Dynamic Adaptation Monitoring: The free-degree robotic arm module moves along the outer ring main track to the preset point and is positioned with the assistance of a global vision monitoring instrument; a high-definition industrial camera is triggered to capture the image the moment the water bucket leaves the jet area, and the supplementary lighting device dynamically adjusts the light intensity and angle; the environmental sensing unit monitors environmental parameters in real time, and activates the anti-fog device when the humidity is too high and the liquid cooling system when the temperature is too high; the monitoring device moves with the monitoring device, monitors the operation of the monitoring device and provides timely feedback to the terminal; Step 4, Abrasion Data Acquisition: The abrasion monitoring module collects surface images and multi-source data of the component and transmits them synchronously to the image processing unit of the intelligent control system. Step 5, Analysis of Abrasion Characteristics: The image processing unit optimizes image quality through super-resolution reconstruction algorithm, identifies abrasion areas and cracks using instance segmentation algorithm, and extracts geometric features of abrasion area, depth, and distribution density; combined with auxiliary imaging and ranging data and comparison with the benchmark image database, the abrasion analysis unit determines the abrasion level. Step 6, Data Transmission and Early Warning: Monitoring data is transmitted to the remote control terminal via industrial Ethernet and SCADA system, and the erosion area, degree and development trend are visualized; continuous monitoring is performed when there is slight erosion, reminders and early warnings are triggered and repair suggestions are pushed when there is moderate erosion, and emergency warnings are triggered and component replacement suggestions are pushed when there is severe erosion. Step 7, System Maintenance and Calibration: The monitoring module's lenses and sensors are automatically cleaned and calibrated weekly, and the monitoring path and frequency are dynamically adjusted. The baseline image database is updated monthly, and the wear level determination algorithm is optimized based on actual maintenance records.

[0015] Preferably, the turbine structural parameters in step 1 include the runner diameter and the number of nozzles, and the reference image database is an initial surface state image database of key components of the turbine; In step 3, the camera should capture the image the instant the water bucket leaves the jet area to avoid water splashes obscuring the view. The instance segmentation algorithm mentioned in step 5 is the YOLO instance segmentation algorithm. The geometric features include abrasion area, depth, and distribution density, which can accurately extract the edge contour and extension trajectory of the abrasion area and crack. The criteria for determining the degree of abrasion in step 6 are as follows: slight abrasion is localized micro-scratches or point-like abrasion, with shallow cracks less than 1 mm wide; moderate abrasion is localized continuous scratches or sheet-like pits, with shallow cracks ≥ 1 mm wide; severe abrasion is large-area deep wear, with a depth > 1 mm or an abrasion area exceeding one-third of the corresponding area or the appearance of penetrating cracks.

[0016] The present invention has the following beneficial effects: 1. This invention adopts an integrated architecture of a double-ring track and a multi-degree-of-freedom robotic arm, coupled with a multi-sensor fusion detection unit, which can realize dynamic monitoring of water tank and nozzle erosion without stopping the machine, avoiding the power generation loss caused by the shutdown maintenance of traditional methods, and significantly improving monitoring efficiency.

[0017] 2. This invention integrates technologies such as machine vision, infrared-assisted vision enhancement, and laser ranging, combined with super-resolution reconstruction and instance segmentation algorithms, to achieve accurate identification of abrasion areas and cracks. It has high monitoring accuracy and can comprehensively capture dynamic changes in abrasion, providing accurate data support for unit operation and maintenance.

[0018] 3. This invention is equipped with complete environmental adaptation components and energy supply modules, including a 316L stainless steel protective shell, liquid cooling heat dissipation system, anti-fog device, and other environmental adaptation components. The energy supply module supports wide temperature range of -35℃ to +65℃ and has a protection level of IP67 or higher. It can adapt to complex working conditions such as high humidity, high water mist, strong vibration, and large temperature difference. It is especially suitable for the operating environment of impulse turbines in high sediment-laden watersheds in Southwest China and has strong stability.

[0019] 4. The intelligent control system of this invention supports multi-condition adaptive monitoring strategies. The monitoring path and frequency can be dynamically adjusted according to the actual working conditions. At the same time, the monitoring data is visualized and remotely warned through the SCADA system, reducing the cost of manual intervention and improving the level of intelligent operation and maintenance.

[0020] 5. The system of this invention adopts a modular design, and the track, robotic arm and monitoring module can be independently disassembled and replaced, which has good expandability and versatility. It can be adapted to different models and capacities of impulse turbines, and has good expandability and versatility, with broad application prospects.

[0021] 6. The distributed processor architecture of this invention enables parallel task processing. The track drive adopts a dual-motor redundancy design. The distributed processor architecture enables parallel task processing, and the failure of a single module does not affect the overall operation of the system. The energy supply module supports automatic charging and low-battery return. The redundancy design of the track drive further improves stability under complex working conditions.

[0022] 7. Multi-sensor fusion and high-speed data links ensure that abrasion information is transmitted without loss and with low latency, providing support for accurate analysis.

[0023] 8. The remote monitoring and automated early warning proposed in this invention significantly reduce the cost of manual intervention and achieve full lifecycle digital management by combining adaptive strategies. Attached Figure Description

[0024] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0025] Figure 1 This is a schematic diagram of the overall structure of the monitoring device of the present invention.

[0026] Figure 2 This is a schematic diagram of the overall structure of the monitoring device of the present invention.

[0027] Figure 3 This is a bottom view of the installation positions of the waterwheel chamber track device, robotic arm, and monitoring device of the present invention.

[0028] Figure 4 This is a schematic diagram of the cooperation between the guide rail and the motion device of the present invention.

[0029] Figure 5 This is a demonstration diagram of the monitoring capabilities of this invention.

[0030] Figure 6 This is a schematic diagram of the primary and secondary processor architecture of the present invention.

[0031] Figure 7 This is a schematic diagram of the abrasion monitoring module-camera unit and secondary processor architecture of the present invention.

[0032] Figure 8This is a schematic diagram of the chassis track device-drive system architecture of the present invention.

[0033] Figure 9 This is a schematic diagram of the overall collaborative architecture of the system of the present invention.

[0034] Figure 10 This is a flowchart of the monitoring method of the present invention.

[0035] Figure 1 In the middle: 11 is a high-definition industrial camera, 12 is a supplementary lighting device, 13 is a rubber shock absorber, 14 is a motion device, 15 is a dual servo motor, and 16 is a planetary reducer. Figure 2 In the middle: 23 is a telephoto, high-definition, waterproof camera; Figure 3 In the middle: 31 is the monitoring device, 32 is the surveillance device, 33 is the outer ring main track, and 34 is the inner ring positioning track; Figure 4 In the middle: 41 represents the guide rail; Figure 5 In the middle: 51 is the waterwheel chamber, 54 is the water turbine bucket, and 55 is the water turbine nozzle; Figure 10 In Chinese: L1 is a schematic diagram of the early stage of the solution process, L2 is a schematic diagram of the mid-stage of the solution process, and L3 is a schematic diagram of the late stage of the solution process. Detailed Implementation To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0036] Example 1: See Figure 1-10A visualized online monitoring system for the erosion of key components of an impact turbine includes a turbine chamber track device. A monitoring device 31, capable of moving around the track, is mounted on the outer ring main track 33 of the turbine chamber track device via a motion device 14. The monitoring device 31 includes a multi-degree-of-freedom robotic arm module, with an erosion monitoring module mounted at its end. The erosion monitoring module includes a high-definition industrial camera 11, a laser rangefinder, a supplementary lighting device 12, and environmental adaptation components, collectively forming a multi-sensor fusion detection unit. A monitoring device 32, capable of moving around the track, is mounted on the inner ring positioning track 34 of the turbine chamber track device via the motion device 14. The monitoring device 32 employs a global vision monitoring instrument. The monitoring device 31 and monitoring device 32 are connected to an intelligent control system. The system also includes a data transmission and storage module and an energy supply module. All modules are integrated through mechanical, electrical, and communication interfaces, enabling dynamic monitoring of erosion in the water tanks and nozzles without stopping the turbine.

[0037] Furthermore, the waterwheel chamber track device adopts a double-ring concentric track structure, with the center of the outer ring main track 33 and the inner ring positioning track 34 both coinciding with the center of the water turbine bucket 54; the radius of the outer ring main track 33 is 300mm larger than the maximum rotating outer diameter of the water turbine bucket, reserving space for the extension and retraction of the multi-degree-of-freedom robotic arm module.

[0038] Furthermore, both the outer ring main track 33 and the inner ring positioning track 34 include guide rails. The cross-section of the guide rails is in a three-layer form: the upper layer is responsible for suspension guidance, the middle layer is the middle slide, and the lower layer is responsible for load-bearing and friction. The track material is 316L stainless steel with surface hardening treatment and a hardness ≥45HRC.

[0039] Furthermore, in order to enable the monitoring and control devices to operate on the corresponding outer ring main track 33 and inner ring positioning track 34, a corresponding motion device is required. The motion device 14 adopts a configuration of dual servo motors 15 and planetary reducers 16, with differential operation of the inner / outer drive wheels to achieve stepless speed regulation of 0-0.5m / min; the disc brake mechanism has an emergency braking response time of ≤0.5s and has dual motor redundancy function, so that when one motor fails, the other motor can maintain ≥50% of the rated load to continue operating.

[0040] Furthermore, the monitoring device is installed on the outer ring main track 33, which can rotate 360° continuously along the track; the robotic arm body has a three-joint structure, realizing radial extension and retraction of 0-800mm and pitch rotation, with a repeatability of ±0.2mm.

[0041] Furthermore, the robotic arm end is equipped with a universal interface and a rubber shock absorber 13, which allows for quick replacement of the abrasion monitoring module; it integrates an IMU, paired with a three-axis ring laser gyroscope and a three-axis quartz accelerometer, with an operating temperature of -40 to +85℃, to suppress imaging jitter caused by unit vibration and water flow impact.

[0042] Furthermore, the monitoring device is equipped with a high-definition industrial waterproof camera 11 with a resolution of 24.4MP, a frame rate of 15fps, and a global shutter; the lens is made of sapphire material and integrates an anti-fog heating film, and is equipped with a miniature centrifugal fan and a double-layer explosion-proof microporous flow equalization plate to ensure a light transmittance of ≥95% in water mist environments; it has a built-in near-infrared enhanced CMOS sensor that can communicate with the secondary processor.

[0043] Furthermore, the monitoring device is equipped with a supplementary lighting device 12, which uses a ring array pulsed LED with a color temperature of 4200K±200K and an adjustable illumination angle of 1.6°-11°. It is synchronized with the camera shutter strobe to eliminate reflections and shadows.

[0044] Furthermore, the monitoring device is equipped with a long-focus, high-definition, waterproof camera with an optical zoom of 6-180mm, supporting 8 megapixels, and a sensor size ≥1 / 1.8". It can ensure noise control in low light conditions, and is equipped with an F1.6 aperture and a large-area sensor. It has an IP68 waterproof rating and an aluminum alloy anti-corrosion coating.

[0045] Furthermore, the intelligent control system adopts a master-slave processor distributed architecture. The master processor is responsible for global path planning, data fusion, and remote interaction; the first sub-processor is responsible for the motion control of the robotic arm; the second sub-processor is responsible for controlling the operation of the motion device 16; the third sub-processor performs image acquisition and preprocessing, denoising, sharpening, and histogram equalization of the images, and storing and transmitting them to the terminal; the fourth sub-processor integrates EDSR super-resolution reconstruction and YOLO instance segmentation algorithms to achieve 0.5mm-level abrasion area recognition and 1mm-level crack detection, and divides the results into three levels—slight, moderate, and severe—through the abrasion analysis unit, triggering corresponding early warning mechanisms.

[0046] Furthermore, the environmental adaptation components include a 316L stainless steel protective shell, double sealing rings, a liquid cooling system, and a temperature and humidity sensor. The protective shell is filled with a 10mm thick silicone buffer layer and has a streamlined shape to reduce water flow impact resistance. The interface uses multi-layer rubber sealing, and a drainage hole is reserved at the bottom. The temperature and humidity sensor has a measurement range of -40℃ to +80℃ and 0 to 100%RH, with a temperature accuracy of ±0.3℃ and a humidity accuracy of ±2%RH. It monitors environmental parameters in real time and triggers adaptation measures.

[0047] Furthermore, the image processing unit of the intelligent control system integrates super-resolution reconstruction algorithm and instance segmentation algorithm, which can realize the identification of erosion areas at the 0.5mm level and the detection of cracks of 1mm and above; the erosion analysis unit divides the monitoring results into three levels: slight, moderate and severe. Slight erosion is manifested as local micro-scratches or point-like erosion, and shallow cracks with a width greater than 1mm; moderate erosion is manifested as local continuous scratches or sheet-like pits; severe erosion is manifested as large-area deep wear with a depth greater than 1mm, erosion area exceeding one-third of the corresponding area or penetrating cracks. Different levels trigger corresponding early warning mechanisms.

[0048] Furthermore, the system is equipped with an energy supply module, including a wide-temperature lithium iron phosphate battery and a waterproof wireless charger; operating temperature -35℃ to +65℃, cycle life ≥4000 cycles, protection level IP68; the wireless charger has a power of 300W-24V-10A, protection level IP67, and is installed in a safe position on the side wall of the water turbine room, supporting automatic charging and unmanned operation, and triggering a return-to-base charging command when the battery's available capacity is below 20%.

[0049] Furthermore, in the aforementioned dual planetary orbit structure, the outer ring main track is equipped with an abrasion monitoring device, and the inner ring positioning track is equipped with a monitoring camera, which acts as a supervisor. When the outer ring device is working, the camera moves in the same direction as it and records the operating status of the robotic arm. When the robotic arm malfunctions, the inner ring device can temporarily replace some of the functions of the monitoring device, or it can be manually controlled to understand the internal condition of the waterwheel chamber. Furthermore, the intelligent control system supports a multi-condition adaptive monitoring strategy. When multiple nozzles are working, monitoring points are set in the area of ​​non-working nozzles to avoid water flow impact and water mist interference. When the water bucket is idling or stationary, full area coverage is achieved through the circular motion of the track. The monitoring frequency is dynamically adjusted according to the sediment content and flow velocity of the water flow. The monitoring frequency is automatically increased when the sediment content is high during the flood season and appropriately reduced during the dry season.

[0050] Example 2: Step 1, Monitoring System Deployment: Double-ring tracks are laid on the inner wall of the turbine's waterwheel chamber. The outer ring main track is used to carry the multi-degree-of-freedom robotic arm module, while the inner ring positioning track is used to install global vision monitoring instruments. High-definition industrial cameras, laser rangefinders, and supplementary lighting devices are integrated into the end of the robotic arm. The connection and debugging with the intelligent control system and data transmission and storage module are completed, and a reference image database of key components of the turbine is established.

[0051] Step 2, monitor path planning: The intelligent control system, based on the turbine's structural parameters, such as the runner diameter and the number of nozzles, and operating conditions, such as the nozzle working status and water flow velocity, pre-sets key monitoring points such as the water-facing surface of the water bucket, the water-dividing blade, the root, the water outlet edge, and the exposed parts of the nozzles, and plans the robotic arm's movement path, dwell time, and shooting angle to ensure no monitoring blind spots.

[0052] Step 3, Dynamic Adaptation Monitoring: The robotic arm moves along the outer ring main track to the preset monitoring point and is positioned with the assistance of a global vision monitoring instrument; a high-definition industrial camera is triggered to capture images the moment the water bucket leaves the jet area to avoid water splashes obscuring the view; the supplementary lighting device dynamically adjusts the light intensity and angle according to the ambient light intensity; temperature and humidity sensors monitor environmental parameters in real time; when the humidity is too high, the anti-fog device is activated; when the temperature is too high, the liquid cooling system is activated; the monitoring device moves with the monitoring device, monitors the operation of the monitoring device, and provides timely feedback to the terminal.

[0053] Step 4, Abrasion Data Acquisition: The image acquisition system uses a high-definition industrial camera and its auxiliary equipment to clearly capture images of the component surface. It collects surface condition data for key parts of the impact turbine bucket and nozzle. All data are synchronously transmitted to the image processing unit of the intelligent control system via the GPIB bus.

[0054] Step 5, Analysis of Abrasion Characteristics: The image processing unit first optimizes the image quality by using a super-resolution reconstruction algorithm to remove noise and blur interference; then it uses the YOLO instance segmentation algorithm to identify the abrasion area and cracks, and extracts geometric features such as abrasion area, depth, and distribution density; combined with auxiliary imaging and ranging data and comparison with the benchmark image database, the abrasion analysis unit determines the abrasion level.

[0055] Step 6, Data Transmission and Early Warning: Monitoring data is transmitted to the remote control terminal via industrial Ethernet and SCADA system, and the erosion area, degree and development trend are presented in a visual manner. When the erosion level reaches the moderate level, an alert is triggered, and when it reaches the severe level, an emergency alert is triggered, and targeted maintenance suggestions are pushed, such as continuous monitoring of slight erosion, repair of moderate erosion, and replacement of parts for severe erosion.

[0056] Step 7, System Maintenance and Calibration: The monitoring module's lenses and sensors are automatically cleaned and calibrated weekly, and the monitoring path and frequency are dynamically adjusted based on operational data. The benchmark database is updated monthly, and the wear level determination algorithm is optimized based on actual maintenance records to ensure long-term stable monitoring accuracy.

[0057] Although the preferred embodiments of the present invention have been described above in conjunction with the accompanying drawings, the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many specific modifications under the guidance of the present invention without departing from the spirit of the invention and the scope of protection of the claims, and these modifications all fall within the scope of protection of the present invention.

Claims

1. A visualized online monitoring system for erosion of key components of an impulse turbine, characterized in that, The device includes a waterwheel chamber track device, on which a monitoring device (31) capable of moving around the track is mounted via a motion device (14) on the outer ring main track (33) of the waterwheel chamber track device. The monitoring device (31) includes a multi-degree-of-freedom robotic arm module, and the end of the multi-degree-of-freedom robotic arm module is equipped with an abrasion monitoring module. The abrasion monitoring module includes a high-definition industrial camera (11), a laser rangefinder, a supplementary lighting device (12), and an environmental adaptation component, which together constitute a multi-sensor fusion detection unit. The inner ring positioning track (34) of the waterwheel chamber track device is equipped with a monitoring device (32) that can move around the track via a motion device (14). The monitoring device (32) adopts a global visual monitoring instrument. The monitoring device (31) and the surveillance device (32) are connected to the intelligent control system; It also includes a data transmission and storage module and an energy supply module.

2. The online visual monitoring system for erosion of key components of an impulse turbine according to claim 1, characterized in that, The waterwheel chamber track device adopts a double-ring concentric track structure, including an outer ring main track (33) and an inner ring positioning track (34) arranged to coincide with the center of the water turbine bucket (54). The radius of the outer ring main track (33) is greater than the maximum rotational outer diameter of the water turbine bucket (54) and a space for extension and retraction of the multi-degree-of-freedom robotic arm module is reserved. Both the outer ring main track (33) and the inner ring positioning track (34) include guide rails (41). The guide rails (41) adopt a three-layer structure of upper, middle and lower layers. The upper layer is responsible for suspension guidance, the middle layer is the middle slide, and the lower layer bears the load and friction functions. The motion device (14) is equipped with dual servo motors (15) and planetary reducers (16), and achieves differential motion through inner / outer driving wheels and driven devices. It is combined with a disc brake mechanism and adopts a dual-motor redundant design.

3. The online visual monitoring system for erosion of key components of an impulse turbine according to claim 2, characterized in that, The end of the multi-degree-of-freedom robotic arm module is equipped with a universal interface and a rubber shock absorber (13), and realizes three-dimensional motion of circular movement, axial extension and multi-angle rotation, and is equipped with an inertial measurement unit; The inertial measurement unit includes three ring laser gyroscopes and three quartz accelerometers, with an operating temperature range of -40℃ to +85℃, and features shock resistance and waterproofing. The intelligent control system includes a hierarchical processor, a motion control unit, an environmental sensing unit, an image processing unit, and an abrasion analysis unit. It adopts a distributed architecture of a main processor and a secondary processor. The main processor is the global decision center, the secondary processor is responsible for image acquisition and processing and motion control, and the environmental sensing unit integrates air pressure, temperature, and humidity sensors. The data transmission and storage module adopts industrial Ethernet transmission, supports real-time data transmission, local caching and remote server storage, and realizes the visualization and historical traceability of monitoring data through the SCADA system. The energy supply module includes a wide-temperature lithium iron phosphate battery and a waterproof wireless charger, supporting automatic charging and unmanned operation.

4. The online visual monitoring system for erosion of key components of an impulse turbine according to claim 3, characterized in that, The emergency braking response time of the disc brake mechanism is ≤0.5s. In the dual-motor redundancy design, the other motor can maintain 50% load operation when one motor fails. The high-definition industrial camera (11) has a resolution of at least 24.4MP, a frame rate of at least 15fps, an IP67 protection rating, a lens made of sapphire material and an integrated anti-fog heating film, a micro centrifugal fan and a double-layer explosion-proof microporous flow equalization plate, a built-in near-infrared intensifier and a global shutter, and communicates with the secondary processor via the GPIB bus. The secondary processor performs image denoising, sharpening, and histogram equalization preprocessing via a high-speed data acquisition card. The supplementary lighting device is a high-brightness pulsed light source with a ring array design, a color temperature of 4200K±200K, an illumination angle of 1.6°-11°, and features continuous zoom and strobe synchronization functions. It can dynamically adjust the light intensity and illumination angle based on data from the environmental sensing unit.

5. The online visual monitoring system for erosion of key components of an impulse turbine according to claim 3, characterized in that, The monitoring device (32) is equipped with a long-focus high-definition waterproof camera (23), with optical zoom of 6-180mm, supporting 8 million pixels, sensor size ≥1 / 1.8", which can ensure noise control under low light conditions. It is equipped with a supplementary light device, near-infrared enhanced F1.6 aperture and large target surface sensor, waterproof rating IP68, and the shell is made of aluminum alloy anti-corrosion coating. The environmental adaptation components include a 316L stainless steel protective housing, double sealing rings, a liquid cooling system, and a temperature and humidity sensor. The protective housing is filled with a 10mm thick silicone buffer layer and has a streamlined shape. The interface is sealed with multiple layers of rubber, and a drainage hole is reserved at the bottom. The temperature and humidity sensor has a measurement range of -40℃ to +80℃ and 0 to 100%RH, with a temperature accuracy of ±0.3℃ and a humidity accuracy of ±2%RH.

6. The online visual monitoring system for erosion of key components of an impulse turbine according to claim 3, characterized in that, The image processing unit of the intelligent control system integrates super-resolution reconstruction algorithm and instance segmentation algorithm, which can realize the identification of abrasion areas and crack detection; the abrasion analysis unit divides the monitoring results into three levels: slight, moderate and severe, and triggers corresponding early warning mechanisms for different levels.

7. The online visual monitoring system for erosion of key components of an impulse turbine according to claim 3, characterized in that, The wide-temperature lithium iron phosphate battery operates at a temperature of -35℃ to +65℃, has a cycle life of ≥4000 cycles, and an IP68 protection rating; the wireless charger has a power of 300W-24V-10A, an IP68 protection rating, and triggers a return-to-home charging command when the battery's usable capacity is below 20%.

8. The online visual monitoring system for erosion of key components of an impulse turbine according to claim 3, characterized in that, The global visual monitoring instrument of the inner ring positioning track (34) is used to follow the monitoring device (31) to move in the same direction and record the operating status of the multi-degree-of-freedom robotic arm module. When the multi-degree-of-freedom robotic arm module fails, it can temporarily replace some of its functions or understand the internal situation of the waterwheel chamber through human control. The intelligent control system supports a multi-condition adaptive monitoring strategy. When multiple nozzles are working, monitoring points are set in the area of ​​non-working nozzles. When the turbine bucket is idling or stationary, full-area coverage is achieved through the circular motion of the track. The monitoring frequency is dynamically adjusted according to the sediment content and flow velocity of the water flow.

9. A method for visualized online monitoring of erosion of key components of an impulse turbine, characterized in that, The method is implemented using the monitoring system described in any one of claims 1-8, and includes the following steps: Step 1, Monitoring System Deployment: A water turbine waterwheel chamber track device is laid on the inner wall of the top of the water turbine waterwheel chamber (51). The outer ring main track (33) is equipped with a multi-degree-of-freedom robotic arm module, and the inner ring positioning track (34) is equipped with a global visual monitoring instrument. The abrasion monitoring module is integrated into the end of the multi-degree-of-freedom robotic arm module, and the connection and debugging with the intelligent control system and data transmission storage module are completed to establish a reference image database of the initial surface state of the key components of the water turbine. Step 2, monitor path planning: The intelligent control system pre-sets key monitoring points for the turbine runner diameter, number of nozzles, structural parameters, nozzle working status, water flow velocity, and operating conditions. It also plans the movement path, dwell time, and shooting angle of the degree-of-freedom robotic arm module to ensure no blind spots in detection. Step 3, Dynamic Adaptation Monitoring: The free-degree robotic arm module moves along the outer ring main track (33) to the preset point and is positioned with the assistance of the global vision monitoring instrument; the high-definition industrial camera is triggered to capture the image the instant the water bucket leaves the jet area, and the supplementary lighting device dynamically adjusts the light intensity and angle; the environmental sensing unit monitors the environmental parameters in real time, and starts the anti-fog device when the humidity is too high and the liquid cooling heat dissipation system when the temperature is too high. The monitoring device moves with the monitoring device, supervises the operation of the monitoring device and provides timely feedback to the terminal. Step 4, Abrasion Data Acquisition: The abrasion monitoring module collects surface images and multi-source data of the component and transmits them synchronously to the image processing unit of the intelligent control system. Step 5, Analysis of Abrasion Characteristics: The image processing unit optimizes image quality through super-resolution reconstruction algorithm, identifies abrasion areas and cracks using instance segmentation algorithm, and extracts geometric features of abrasion area, depth, and distribution density; combined with auxiliary imaging and ranging data and comparison with the benchmark image database, the abrasion analysis unit determines the abrasion level. Step 6, Data Transmission and Early Warning: Monitoring data is transmitted to the remote control terminal via industrial Ethernet and SCADA system, and the erosion area, degree and development trend are visualized; continuous monitoring is performed when there is slight erosion, reminders and early warnings are triggered and repair suggestions are pushed when there is moderate erosion, and emergency warnings are triggered and component replacement suggestions are pushed when there is severe erosion. Step 7, System Maintenance and Calibration: The monitoring module's lenses and sensors are automatically cleaned and calibrated weekly, and the monitoring path and frequency are dynamically adjusted. The baseline image database is updated monthly, and the wear level determination algorithm is optimized based on actual maintenance records.

10. The method for visual online monitoring of erosion of key components of an impulse turbine according to claim 9, characterized in that, The turbine structural parameters mentioned in step 1 include the runner diameter and the number of nozzles. The reference image database is an initial surface state image database of key components of the turbine. In step 3, the camera should capture the image the instant the water bucket leaves the jet area to avoid water splashes obscuring the view. The instance segmentation algorithm mentioned in step 5 is the YOLO instance segmentation algorithm. The geometric features include abrasion area, depth, and distribution density, which can accurately extract the edge contour and extension trajectory of the abrasion area and crack. The criteria for determining the degree of abrasion in step 6 are as follows: slight abrasion is localized micro-scratches or point-like abrasion, with shallow cracks less than 1 mm wide; moderate abrasion is localized continuous scratches or sheet-like pits, with shallow cracks ≥ 1 mm wide; severe abrasion is large-area deep wear, with a depth > 1 mm or an abrasion area exceeding one-third of the corresponding area or the appearance of penetrating cracks.

Images