Intelligent inspection device and control method for oil circuit of hydroelectric generator set

Abstract

The application discloses a kind of water turbine generator unit oil circuit intelligent inspection device and control method, it is related to unit operation and maintenance detection technical field.The device includes three-channel integrated flange butt joint assembly, oil filling flow regulating unit, multistage filtration unit, oil quality real-time detection unit, double-mode endoscopic dredging robot and AI intelligent data processing terminal.Three-channel assembly is connected unit oil circuit by single interface disassembly-free sealing;Oil filling, filtration and detection unit constitute flushing closed loop together;Robot enters oil circuit through launching channel, with double mode of passive and autonomous walking.AI terminal fuses pipeline vision, flow pressure and oil quality data, and links control oil filling flow rate and robot working mode switching.The application solves the problems of traditional oil circuit maintenance disassembly cumbersome, process fragmentation and small-diameter probe difficulty, realizes disassembly-free full-process integrated operation and preventive intelligent operation and maintenance.

Description

Technical Field

[0001] This invention relates to the field of generator operation and maintenance testing technology, specifically to an intelligent inspection device and control method for the oil circuit of a hydro-generator unit. Background Technology

[0002] The statements in this section are provided only as background information in connection with this disclosure and may not constitute prior art.

[0003] The hydraulic system of a hydro-generator unit undertakes core functions such as lubrication, hydraulic control, and heat dissipation. Its pipeline routing is complex, with many branches and large pipe diameters. Long-term operation can easily lead to defects such as sludge deposition, pipe wall corrosion, and cracks, making it a core maintenance target for unit operation and maintenance.

[0004] Currently, the relevant maintenance technologies in the industry mainly suffer from the following deficiencies: 1. Cumbersome pipeline connection methods and long maintenance downtime: When the existing dedicated oil filter testing device is connected to the unit's oil circuit, the original oil pipes and flanges need to be disassembled in sections, which is complicated and significantly reduces the power generation time.

[0005] 2. Fragmented processes and low efficiency of multi-equipment collaboration: Sludge flushing, impurity filtration, oil quality testing and pipeline endoscopy belong to different equipment. Each maintenance requires repeated docking, emptying and oil filling, which can easily cause oil loss and secondary contamination of the unit's precision components by impurities.

[0006] 3. Difficulty in exploring narrow and winding oil passages: Most existing general-purpose pipeline robots are fully autonomous and cannot adapt to the working conditions of small diameter, many bends and full of oil in turbine oil passages. They are difficult to pass through and have limited range, and lack coordination with the oil circulation system.

[0007] 4. Limited detection dimensions and lack of early warning capabilities: Existing solutions can only achieve single-time oil quality detection or single-point imaging, without physically integrating multi-source data such as oil physicochemical data, visual defects, pressure and flow, making it difficult to upgrade from "post-event maintenance" to "preventive operation and maintenance". Summary of the Invention

[0008] The purpose of this invention is to address the technical problems existing in the current maintenance of hydraulic systems of hydro-generator units, such as cumbersome pipeline disassembly and long downtime, fragmented processes involving multiple equipment, difficulties in exploring small-diameter and bent hydraulic systems, and the lack of preventive and early warning capabilities in single-dimensional detection. This invention provides an intelligent inspection device and control method for hydraulic systems of hydro-generator units. Based on a three-channel integrated single-interface docking structure and a continuously variable speed flushing and filtering closed loop, and linked to a dual-mode endoscopic unblocking robot with its flow-following / autonomous walking mode and AI multi-source data fusion analysis, this invention achieves integrated operation of the entire process without disassembly, intelligent pipeline exploration without blind spots, and preventive maintenance based on fault trends.

[0009] The technical solution of the present invention is as follows: A smart inspection device for the oil circuit of a hydro-generator set, comprising: Three-channel integrated flange docking assembly, oil filling flow regulation unit, multi-stage filtration unit, real-time oil quality detection unit, dual-mode endoscopic unblocking robot and AI intelligent data processing terminal; The three-channel integrated flange assembly integrates an oil filling channel, an oil return filtration channel, and a robot delivery channel. The three-channel integrated flange assembly is used to connect to the oil circuit port of the hydro-generator unit through a single-interface seal. The outlet of the oil filling flow regulating unit is connected to the oil circuit of the hydro-generator unit through the oil filling channel, which is used to provide an oil filling flow with stepless adjustable velocity. The multi-stage filtration unit and the real-time oil quality detection unit are connected in series in the external circuit of the return oil filtration channel. The multi-stage filtration unit, the real-time oil quality detection unit and the oil filling flow regulation unit together form a closed-loop circulating oil circuit. The dual-mode endoscopic dredging robot enters the oil circuit of the hydro-generator unit through the robot deployment channel. The dual-mode endoscopic dredging robot has two working modes: passive movement with the oil flow and autonomous walking drive. The AI ​​intelligent data processing terminal is electrically connected to the oil-filled flow regulation unit, the real-time oil quality detection unit, and the dual-mode endoscopic unblocking robot. The AI ​​intelligent data processing terminal is configured to receive flow and pressure data fed back by the oil-filled flow regulation unit, oil physicochemical data fed back by the real-time oil quality detection unit, and pipeline visual data transmitted back by the dual-mode endoscopic unblocking robot. Based on the pipeline visual data, the terminal identifies the internal state of the pipeline and sends a flow rate regulation command to the oil-filled flow regulation unit to control the dual-mode endoscopic unblocking robot to switch between the passive movement mode following the oil flow and the autonomous walking drive mode.

[0010] Furthermore, the three-channel integrated flange assembly includes a flange joint, a quick-sealing clamp, a two-way shut-off valve, and a sealing buffer gasket; the flange joint has an independently provided oil filling channel, an oil return filtration channel, and a robot delivery channel; a sealed hatch is configured on the outside of the robot delivery channel.

[0011] Furthermore, the bidirectional shut-off valve is installed in the oil filling channel and the oil return filter channel to isolate the oil circuit of the hydro-generator set from the internal oil circuit of the intelligent inspection device for the hydro-generator set oil circuit during the connection and disassembly of the single-interface sealed connection.

[0012] Furthermore, the oil filling flow regulation unit includes an oil storage tank, a variable frequency oil filling pump, an electric flow regulating valve, and a pressure sensor installed in the oil filling channel, connected in sequence; the pressure sensor is electrically connected to the AI ​​intelligent data processing terminal and is used to acquire the flow pressure data; The AI ​​intelligent data processing terminal is configured to: when the flow pressure data exceeds a preset safety threshold, control the variable frequency oil pump to reduce its speed and trigger automatic pressure relief.

[0013] Furthermore, the multi-stage filtration unit is provided with a coarse filter screen, a precision filter element and an impurity collection chamber in sequence along the oil flow direction; the multi-stage filtration unit is also configured with a bypass switching valve in parallel, as well as a differential pressure alarm sensor for monitoring the pressure difference across the precision filter element.

[0014] Furthermore, both the differential pressure alarm sensor and the bypass switching valve are electrically connected to the AI ​​intelligent data processing terminal; The AI ​​intelligent data processing terminal is configured to generate an alarm signal when the differential pressure alarm sensor detects that the differential pressure value reaches a preset blockage threshold, and control the bypass switching valve to open so as to switch to the bypass to ensure the continuity of the circulation loop.

[0015] Furthermore, the dual-mode endoscopic unblocking robot includes a waterproof body, as well as a high-definition wide-angle camera module, autonomous walking drive wheels, and a miniature retractable unblocking scraper mounted on the waterproof body; The autonomous walking drive wheel is in a retracted state in the passive movement mode following the oil flow, and extends to provide movement power in the autonomous walking drive mode; the miniature retractable cleaning scraper is configured to extend in the autonomous walking drive mode to cooperate with the oil flow to clean the sludge on the pipe wall.

[0016] Furthermore, the AI ​​intelligent data processing terminal is configured with linkage control logic: When no pipeline defect is identified based on the pipeline visual data, the oil filling flow adjustment unit is controlled to output a preset inspection flow rate, and the dual-mode endoscopic unblocking robot is controlled to be in the passive movement mode with the oil flow. When a defect or stubborn sludge point is identified in the pipeline based on the pipeline visual data, the oil filling flow adjustment unit is controlled to reduce the flow rate, and the dual-mode endoscopic unblocking robot is controlled to switch to the autonomous walking drive mode to reach the defect or stubborn sludge point.

[0017] Furthermore, the oil physicochemical data includes oil contamination level data; The AI ​​intelligent data processing terminal is equipped with a flushing closed-loop logic: when the oil contamination data exceeds the standard threshold, the oil filling flow regulation unit is controlled to gradually increase the flow rate to form turbulent flow that peels off the sludge from the pipe wall; when the oil contamination data returns to within the standard threshold, the oil filling flow regulation unit is controlled to end the flushing operation.

[0018] This invention also proposes a control method based on the above-mentioned intelligent oil circuit inspection device for hydro-generator sets, comprising the following steps: Step S1: Connect the oil circuit port of the hydro-generator set through the single-interface sealing connection of the three-channel integrated flange docking assembly; Step S2: Control the oil filling flow rate adjustment unit to establish an initial closed-loop circulation oil circuit, and the oil quality real-time detection unit monitors the oil contamination data in real time; when the oil contamination data does not meet the standard, increase the oil filling flow rate to form scouring turbulence, and at the same time, the multi-stage filtration unit simultaneously intercepts and filters the return oil; Step S3: Send the dual-mode endoscopic cleaning robot into the pipeline through the robot delivery channel, control the dual-mode endoscopic cleaning robot to perform inspection in the passive movement mode with oil flow, and transmit pipeline visual data back in real time. Step S4: The AI ​​intelligent data processing terminal receives the pipeline visual data. When the target defect point is identified, it issues an instruction to control the oil flow adjustment unit to reduce the flow rate and controls the dual-mode endoscopic dredging robot to switch to the autonomous walking drive mode for fixed-point exploration and cleaning.

[0019] Compared with existing technologies, the advantages of this invention are: 1. Single-interface, disassembly-free connection, significantly reducing downtime: Through the pioneering three-channel integrated flange docking assembly, the three channels of oil filling, oil return, and robot deployment are highly integrated. Only one docking is required to complete the pipeline connection for the entire maintenance process, reducing disassembly and assembly time by more than 70% and shortening the unit downtime maintenance time by 40% to 60%.

[0020] 2. Electromechanical-hydraulic linkage to overcome the bottleneck of small-diameter pipe inspection: For winding small-diameter oil circuits, an innovative "flow rate control-robot dual-mode linkage" control architecture is adopted. The AI ​​terminal dynamically schedules the flow rate of the oil filling pump and switches the robot (passive following the flow / autonomous walking) mode based on the visual recognition results. It uses fluid thrust to assist long-distance passage, which solves the problems of insufficient endurance and easy jamming of fully autonomous robots in oil-filled small-diameter pipes. The defect detection rate of the entire pipeline is reduced by more than 80%.

[0021] 3. Closed-loop control throughout the entire process to prevent secondary damage to precision components: A robust physical closed loop of "oil flushing - synchronous oil filtration - oil quality monitoring" has been constructed. The stubborn sludge stripped off by turbulent flushing is immediately intercepted by multi-stage filtration units. Combined with automatic bypass switching and pump and valve coordinated pressure relief safety logic, secondary wear caused by impurities flowing back to the unit's bearings and hydraulic valve cores is avoided.

[0022] 4. Multi-source data timestamp fusion enables preventative intelligent operation and maintenance: Unlike single visual algorithms, this device achieves clock alignment of multi-dimensional data such as pressure, oil quality, and vision at the underlying hardware level. It directly converts the severity level of defects into physical speed adjustment / cleaning commands and transforms all data into a single unit health record, upgrading the traditional "post-fault maintenance" to "preventative operation and maintenance", reducing the risk of sudden oil circuit shutdown by more than 60%. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments recorded in the embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0024] Figure 1 This is a schematic diagram of the overall device structure of the present invention; Figure 2 This is a schematic diagram of the closed-loop oil circuit circulation and signal connection of the device of the present invention. Detailed Implementation

[0025] It should be noted that relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0026] The features and performance of the present invention will be further described in detail below with reference to embodiments.

[0027] Example 1 Please refer to the following: Figure 1 , Figure 2This embodiment provides an intelligent inspection device for the oil circuit of a hydro-generator set, specifically applicable to the periodic maintenance, troubleshooting, sludge removal, pipe wall defect detection, and life trend assessment of the governor pressure oil circuit, guide bearing lubrication oil circuit, thrust bearing cooling oil circuit, and hydraulic operation control pipeline of large and medium-sized mixed-flow and axial-flow hydro-generator sets.

[0028] The intelligent oil circuit inspection device for the hydro-generator unit includes: a three-channel integrated flange docking assembly, an oil filling flow regulation unit, a multi-stage filtration unit, a real-time oil quality detection unit, a dual-mode endoscopic unblocking robot, and an AI intelligent data processing terminal.

[0029] The three-channel integrated flange assembly integrates an oil filling channel, an oil return filtration channel, and a robot delivery channel. The three-channel integrated flange assembly is used to connect to the oil circuit port of the hydro-generator unit through a single-interface seal.

[0030] Specifically, the three-channel integrated flange assembly includes a flange joint, a quick-sealing clamp, a two-way shut-off valve, and a sealing buffer gasket. The flange joint has an independently configured oil filling channel, an oil return filtration channel, and a robot delivery channel. A sealing hatch is configured on the outside of the robot delivery channel. Furthermore, to address the issue of pressurized oil leakage during the robot's cable-pulling movement, a dynamic sealing component (such as an adaptive dynamic sealing gland or a lip seal) is coaxially installed on the inside of the sealing hatch or at the entrance of the robot delivery channel. The inner diameter of the dynamic sealing component is interference-fitted with the outer diameter of the wired flexible transmission cable, ensuring that the cable maintains a pressure-bearing and sealed state inside the pipeline during passive flow or active winding sliding.

[0031] Due to the limited flange diameter of the original turbine unit (such as DN80 or DN50 standard flanges), in this embodiment, the three channels inside the flange joint adopt a "concentric sleeve type" or "parallel and high-polymer isolation sealing" structural arrangement. The three channels share the same unit flange mating surface and do not physically interfere with each other. The bidirectional shut-off valve is installed in the oil filling channel and the oil return filter channel to isolate the turbine generator set oil circuit from the internal oil circuit of the turbine generator set oil circuit intelligent inspection device during the connection and disassembly through the single-interface sealing connection, ensuring no leakage during the disassembly and assembly process, and realizing rapid connection of multiple sections of the original unit pipeline without disassembly. Compared with the existing technology, the disassembly and assembly time can be reduced by more than 70%.

[0032] The outlet of the oil filling flow regulating unit is connected to the oil circuit of the hydro-generator unit through the oil filling channel, which is used to provide an oil filling flow with stepless adjustable velocity.

[0033] Specifically, the oil filling flow regulation unit includes an oil storage tank, a variable frequency oil filling pump, an electric flow regulating valve, and a pressure sensor installed in the oil filling channel, which are connected in sequence; the pressure sensor is electrically connected to the AI ​​intelligent data processing terminal and is used to acquire the flow and pressure data.

[0034] (Safety Logic Explanation): The AI ​​intelligent data processing terminal is configured to: when the flow and pressure data exceed the preset safety threshold, control the variable frequency oil pump to reduce its speed, and simultaneously control the opening of the electric flow regulating valve or switch to the pressure relief circuit, triggering automatic pressure relief through pump-valve coordinated control to ensure the safety of the unit pipeline; during normal flushing speed regulation, the AI ​​terminal also achieves stepless and precise control of the flushing oil flow by jointly adjusting the pump's operating frequency and the flow cross-sectional area of ​​the electric flow regulating valve.

[0035] The multi-stage filtration unit and the real-time oil quality detection unit are connected in series in the external circuit of the return oil filtration channel. The multi-stage filtration unit, the real-time oil quality detection unit and the oil filling flow regulation unit together form a closed-loop circulating oil circuit of "oil filling flushing - synchronous oil filtration - oil quality monitoring" to avoid secondary contamination of the unit by the stripped impurities.

[0036] Specifically, the multi-stage filtration unit is sequentially arranged with a coarse filter screen, a precision filter element, and an impurity collection chamber along the oil flow direction. The multi-stage filtration unit is also equipped with a bypass switching valve and a differential pressure alarm sensor for monitoring the pressure difference across the precision filter element. Both the differential pressure alarm sensor and the bypass switching valve are electrically connected to the AI ​​intelligent data processing terminal. The AI ​​intelligent data processing terminal is configured to generate an alarm signal and control the bypass switching valve to open when the differential pressure alarm sensor detects that the differential pressure value reaches a preset clogging threshold, thereby switching to the bypass to ensure the continuity of the circulation loop.

[0037] It should be noted that, in order to prevent large particles of sludge from directly entering the next stage and damaging the precision sensor after switching the bypass, a high-flow safety coarse filter is also provided in the bypass; at the same time, when the AI ​​intelligent data processing terminal opens the bypass, it will control the variable frequency oil pump to reduce its speed to maintain a minimum safe circulation.

[0038] The real-time oil quality detection unit can specifically consist of an online particle size sensor, a moisture sensor, a viscosity sensor, an acid value sensor, and a data acquisition module, connected in series on the clean oil side of the return oil filtration channel. The oil physicochemical data includes oil contamination data (such as water content, aging degree, etc.).

[0039] The dual-mode endoscopic cleaning robot enters the oil circuit of the hydro-generator unit through the robot deployment channel. The robot has two operating modes: passive movement with the oil flow and autonomous driving. The overall outer diameter of the robot is precisely matched to the inner diameter of the deployment channel, and its turning radius is smaller than the bending radius of the smallest diameter oil circuit of the hydro-generator unit (such as a DN50 pipe), ensuring smooth passage through complex and narrow branch pipes.

[0040] Specifically, the dual-mode endoscopic drainage robot includes a waterproof body, and mounted on the waterproof body a high-definition wide-angle camera module, an LED fill light, autonomous driving wheels, a miniature retractable drainage scraper, and a wired flexible transmission cable. To overcome the imaging blurring caused by light scattering and refraction in oil-filled closed pipes, the LED fill light is equipped with an anti-glare lens and a polarizing filter assembly, and the lens of the high-definition wide-angle camera module is coated with an oleophobic and anti-fouling coating; thereby eliminating halo interference in the oily environment and ensuring high-definition extraction of defect features. The cable synchronously transmits video and control signals, ensuring signal stability in the oil-filled closed environment.

[0041] The autonomous walking drive wheel is in a retracted state in the passive movement mode following the oil flow, and extends through an internal micro motor in the autonomous walking drive mode to provide forward, backward, and turning power; the micro retractable cleaning scraper is configured to extend in the autonomous walking drive mode (e.g., driven by a micro electromagnetic push rod) to cooperate with the turbulent oil flow to mechanically scrape away stubborn sludge from the pipe wall.

[0042] The AI ​​intelligent data processing terminal is electrically connected to the oil flow regulation unit, the real-time oil quality detection unit, and the dual-mode endoscopic unblocking robot. It receives flow and pressure data, oil quality physicochemical data, and pipeline visual data transmitted back by the dual-mode endoscopic unblocking robot. To solve the fusion judgment error caused by inconsistent sampling frequencies of different sensors, the AI ​​intelligent data processing terminal embeds a multi-source data synchronous receiving module. It scores and timestamps the flow and pressure data, oil quality physicochemical data, and pipeline visual data through a unified clock source, ensuring that multi-dimensional state data at the same physical location of the pipeline are accurately bound on the same time profile. Based on the pipeline visual data, it identifies the internal state of the pipeline and issues flow rate adjustment commands to control the working mode of the robot.

[0043] In this embodiment, it should be noted that, as those skilled in the art will understand, the AI ​​intelligent data processing terminal can be composed of an industrial touch screen host (IPC), a data acquisition module (DAQ), and a programmable logic controller (PLC). Its "pipeline defect AI identification model" can employ a pre-trained, mature deep learning neural network (e.g., ResNet or YOLO architecture) based on turbine oil circuit defect samples (such as pipe wall cracks, corrosion, weld defects, sludge accumulation, etc.). Its linkage control is not an abstract algorithm, but rather converts the identification results into PLC control electrical signals output to each actuator. Furthermore, after identifying an anomaly, the pipeline defect AI identification model automatically marks the specific location, size, and severity level of the defect in the pipeline's 3D model or image. Based on this severity level and the predicted sludge adhesion intensity, the AI ​​intelligent data processing terminal adaptively calculates and matches the corresponding turbulent scouring velocity range, achieving precise closed-loop stepless speed regulation.

[0044] The specific linkage logic includes: 1. Robot and Flow Rate Linkage Logic: When no pipeline defects are detected based on the pipeline visual data, the oil-filled flow rate adjustment unit is controlled to output a preset inspection flow rate, and the dual-mode endoscopic cleaning robot is controlled to be in the passive movement mode following the oil flow (suitable for rapid initial screening of straight pipe sections and long pipelines); when a defect point or stubborn sludge point (such as a bend or branch pipe) is detected based on the pipeline visual data, the oil-filled flow rate adjustment unit is controlled to reduce the flow rate, and the dual-mode endoscopic cleaning robot is controlled to switch to the autonomous walking drive mode to reach the defect point or stubborn sludge point for high-definition magnified exploration or scraping.

[0045] 2. Flushing closed-loop logic: When the oil contamination data exceeds the standard threshold, the oil filling flow regulation unit is controlled to gradually increase the flow rate to form turbulent flow to peel off the sludge from the pipe wall; when the oil contamination data returns to within the standard threshold, the oil filling flow regulation unit is automatically fed back and controlled to end the flushing operation.

[0046] Furthermore, this embodiment takes the in-depth overhaul of the speed regulating pressure oil circuit (matching DN80 standard flange) of a large and medium-sized hydro-generator unit as an example. The specific workflow (steps S1-S4) is as follows: Step S1 (Quick Connection): Connect the turbine generator set's oil circuit port via the single-interface sealed connection of the three-channel integrated flange connection assembly. Specifically: close the unit's original oil circuit valves, open the device's bidirectional shut-off valve, and complete all operational pathway connections in a single connection, without needing to disassemble the original pipelines throughout the process.

[0047] Step S2 (Circulation Detection and Flushing Removal): The oil filling flow rate regulation unit is controlled to start the variable frequency oil filling pump to establish an initial low flow rate closed-loop circulation oil circuit. The real-time oil quality detection unit simultaneously collects oil contamination data such as particle size and moisture content. When the detection result shows that the oil particle contamination exceeds the standard (indicating the presence of sludge deposition), the oil filling flow rate is gradually increased to 1.2 m / s to form flushing turbulence, stripping away the greasy deposits on the pipe wall. At the same time, the multi-stage filtration unit simultaneously filters the impurities carried by the return oil through coarse filtration and fine filtration.

[0048] Step S3 (Full-area endoscopic inspection): The dual-mode endoscopic unblocking robot is sent into the pipeline through the robot delivery channel. The oil filling unit lowers the flow rate to 0.8m / s and controls the dual-mode endoscopic unblocking robot to complete a rapid inspection of a straight pipe section (e.g., 20m long) in the passive movement mode following the oil flow. The pipeline visual data is then transmitted back to the AI ​​terminal in real time.

[0049] Step S4 (Targeted Detection and Collaborative Unblocking): The AI ​​intelligent data processing terminal receives the pipeline visual data. When it initially identifies target defect points such as pipeline bends (e.g., stubborn sludge buildup and localized corrosion spots), it issues a command to control the oil-filled flow regulating unit to reduce its speed to 0.3 m / s and simultaneously controls the dual-mode endoscopic unblocking robot to extend its drive wheels and switch to the autonomous walking drive mode. The robot then reaches the target point for high-definition magnification to confirm the size and location of the defect. At the same time, it extends a micro scraper to work with localized turbulence to complete targeted cleaning.

[0050] After cleaning is completed, continuous cyclical testing continues. At the same time, the AI ​​intelligent data processing terminal stores all the data from a single maintenance into the built-in historical database dedicated to the turbine oil circuit, establishing a single unit oil circuit health record. By comparing multi-cycle testing data, it outputs the oil sludge deposition rate and pipe wall corrosion development trend curve, predicts the failure cycle, and automatically generates an oil circuit health trend analysis report containing maintenance cycle adjustment suggestions.

[0051] After the maintenance is completed, the work enters the final stage. The oil filling flow regulation unit is controlled to circulate until the oil quality meets the standard and then stops operating. A retrieval command is issued to the dual-mode endoscopic unblocking robot. The robot is driven in reverse by its autonomous walking drive wheels and smoothly exits the pipeline with the help of the external wired flexible transmission cable winding tension. Finally, the bidirectional shut-off valve is closed to release the pipeline residual pressure, the flange assembly is removed, and the maintenance is completed.

[0052] In summary, this embodiment solves the pain point of long disassembly and assembly time in traditional methods by using a single flange three-channel integrated structure, overcomes the problem of small-diameter curved oil circuit exploration by using flow rate regulation and robot dual-mode linkage, and achieves integrated intelligent maintenance throughout the entire process through multi-source data AI closed-loop control of flushing, filtering, detection, and endoscopy, which greatly reduces the probability of sudden failures and maintenance costs.

[0053] The embodiments described above merely illustrate specific implementation methods of this application, and while the descriptions are detailed and specific, they should not be construed as limiting the scope of protection of this application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the technical solution of this application, and these modifications and improvements all fall within the scope of protection of this application.

[0054] This background section is provided to generally present the context of the invention. The work of the currently named inventors, the work to the extent described in this background section, and aspects of this section that did not constitute prior art at the time of application are neither expressly nor impliedly acknowledged as prior art to the invention.

Claims

1. An intelligent inspection device for the oil circuit of a hydro-generator set, characterized in that, include: Three-channel integrated flange docking assembly, oil filling flow regulation unit, multi-stage filtration unit, real-time oil quality detection unit, dual-mode endoscopic unblocking robot and AI intelligent data processing terminal; The three-channel integrated flange assembly integrates an oil filling channel, an oil return filtration channel, and a robot delivery channel. The three-channel integrated flange assembly is used to connect to the oil circuit port of the hydro-generator unit through a single-interface seal. The outlet of the oil filling flow regulating unit is connected to the oil circuit of the hydro-generator unit through the oil filling channel, which is used to provide an oil filling flow with stepless adjustable velocity. The multi-stage filtration unit and the real-time oil quality detection unit are connected in series in the external circuit of the return oil filtration channel. The multi-stage filtration unit, the real-time oil quality detection unit and the oil filling flow regulation unit together form a closed-loop circulating oil circuit. The dual-mode endoscopic dredging robot enters the oil circuit of the hydro-generator unit through the robot deployment channel. The dual-mode endoscopic dredging robot has two working modes: passive movement with the oil flow and autonomous walking drive. The AI ​​intelligent data processing terminal is electrically connected to the oil-filled flow regulation unit, the real-time oil quality detection unit, and the dual-mode endoscopic unblocking robot. The AI ​​intelligent data processing terminal is configured to receive flow and pressure data fed back by the oil-filled flow regulation unit, oil physicochemical data fed back by the real-time oil quality detection unit, and pipeline visual data transmitted back by the dual-mode endoscopic unblocking robot. Based on the pipeline visual data, the terminal identifies the internal state of the pipeline and sends a flow rate regulation command to the oil-filled flow regulation unit to control the dual-mode endoscopic unblocking robot to switch between the passive movement mode following the oil flow and the autonomous walking drive mode.

2. The intelligent inspection device for the oil circuit of a hydro-generator unit according to claim 1, characterized in that, The three-channel integrated flange assembly includes a flange joint, a quick-sealing clamp, a two-way shut-off valve, and a sealing buffer gasket; the flange joint has an independently configured oil filling channel, an oil return filtration channel, and a robot delivery channel; the robot delivery channel is equipped with a sealed hatch on its outer side.

3. The intelligent inspection device for the oil circuit of a hydro-generator unit according to claim 2, characterized in that, The bidirectional shut-off valve is installed in the oil filling channel and the oil return filter channel to isolate the oil circuit of the hydro-generator set from the internal oil circuit of the intelligent oil circuit inspection device of the hydro-generator set during the connection and disassembly of the single-interface sealed connection.

4. The intelligent inspection device for the oil circuit of a hydro-generator unit according to claim 1, characterized in that, The oil filling flow regulation unit includes an oil storage tank, a variable frequency oil filling pump, an electric flow regulation valve, and a pressure sensor installed in the oil filling channel, which are connected in sequence; the pressure sensor is electrically connected to the AI ​​intelligent data processing terminal and is used to acquire the flow pressure data; The AI ​​intelligent data processing terminal is configured to: when the flow pressure data exceeds a preset safety threshold, control the variable frequency oil pump to reduce its speed and trigger automatic pressure relief.

5. The intelligent inspection device for the oil circuit of a hydro-generator unit according to claim 1, characterized in that, The multi-stage filtration unit is provided with a coarse filter screen, a precision filter element and an impurity collection chamber in sequence along the oil flow direction; the multi-stage filtration unit is also provided with a bypass switching valve in parallel, as well as a differential pressure alarm sensor for monitoring the pressure difference across the precision filter element.

6. The intelligent inspection device for the oil circuit of a hydro-generator unit according to claim 5, characterized in that, The differential pressure alarm sensor and the bypass switching valve are both electrically connected to the AI ​​intelligent data processing terminal. The AI ​​intelligent data processing terminal is configured to generate an alarm signal when the differential pressure alarm sensor detects that the differential pressure value reaches a preset blockage threshold, and control the bypass switching valve to open so as to switch to the bypass to ensure the continuity of the circulation loop.

7. The intelligent inspection device for the oil circuit of a hydro-generator unit according to claim 1, characterized in that, The dual-mode endoscopic unblocking robot includes a waterproof body, as well as a high-definition wide-angle camera module, autonomous walking drive wheels, and a miniature retractable unblocking scraper mounted on the waterproof body; The autonomous walking drive wheel is in a retracted state in the passive movement mode following the oil flow, and extends to provide movement power in the autonomous walking drive mode. The miniature retractable cleaning scraper is configured to extend in the autonomous walking drive mode to cooperate with the oil flow to clean the sludge on the pipe wall.

8. The intelligent inspection device for the oil circuit of a hydro-generator unit according to claim 7, characterized in that, The AI ​​intelligent data processing terminal is equipped with linkage control logic: When no pipeline defect is identified based on the pipeline visual data, the oil filling flow adjustment unit is controlled to output a preset inspection flow rate, and the dual-mode endoscopic unblocking robot is controlled to be in the passive movement mode with the oil flow. When a defect or stubborn sludge point is identified in the pipeline based on the pipeline visual data, the oil filling flow adjustment unit is controlled to reduce the flow rate, and the dual-mode endoscopic unblocking robot is controlled to switch to the autonomous walking drive mode to reach the defect or stubborn sludge point.

9. The intelligent inspection device for the oil circuit of a hydro-generator unit according to claim 1, characterized in that, The oil physicochemical data includes oil contamination level data; The AI ​​intelligent data processing terminal is equipped with a flushing closed-loop logic: when the oil contamination data exceeds the standard threshold, the oil filling flow regulation unit is controlled to gradually increase the flow rate to form turbulent flow that peels off the sludge from the pipe wall; when the oil contamination data returns to within the standard threshold, the oil filling flow regulation unit is controlled to end the flushing operation.

10. A control method for the intelligent oil circuit inspection device of a hydro-generator unit according to any one of claims 1-9, characterized in that, Includes the following steps: Step S1: Connect the oil circuit port of the hydro-generator set through the single-interface sealing connection of the three-channel integrated flange docking assembly; Step S2: Control the oil filling flow rate adjustment unit to establish an initial closed-loop circulation oil circuit, and the oil quality real-time detection unit monitors the oil contamination data in real time; when the oil contamination data does not meet the standard, increase the oil filling flow rate to form scouring turbulence, and at the same time, the multi-stage filtration unit simultaneously intercepts and filters the return oil; Step S3: Send the dual-mode endoscopic cleaning robot into the pipeline through the robot delivery channel, control the dual-mode endoscopic cleaning robot to perform inspection in the passive movement mode with oil flow, and transmit pipeline visual data back in real time. Step S4: The AI ​​intelligent data processing terminal receives the pipeline visual data. When the target defect point is identified, it issues an instruction to control the oil flow adjustment unit to reduce the flow rate and controls the dual-mode endoscopic dredging robot to switch to the autonomous walking drive mode for fixed-point exploration and cleaning.

Images