Intelligent pipeline detection and maintenance method and system based on multi-source new energy supply

Abstract

The present invention discloses an intelligent pipeline detection and maintenance method and system based on multi-source renewable energy energy supply, comprising the following steps: collecting energy through a photovoltaic power generation module, a fluid energy recovery module and a thermoelectric conversion module respectively, and inputting the collected energy into an energy storage device; an intelligent energy management system monitors the power generation status of the photovoltaic power generation module, the fluid energy recovery module and the thermoelectric conversion module in real time through sensor data of each module and a deep learning model, and predicts the power generation trend of each module; monitoring the charging status of the photovoltaic power generation module, the fluid energy recovery module and the thermoelectric conversion module based on a first-order RC circuit model, and predicting the charge status using a Kalman filter algorithm; and sorting the charging of multiple tasks based on a data fusion algorithm according to the maintenance tasks performed by a multi-degree-of-freedom robotic arm to ensure that the multi-degree-of-freedom robotic arm always has sufficient energy support during the execution of the tasks.

Description

Technical Field

[0001] The present invention belongs to the technical field of pipeline detection and maintenance systems, and specifically relates to an intelligent pipeline detection and maintenance method and system based on multi-source renewable energy energy supply. Background Art

[0002] Existing pipeline inspection and maintenance systems primarily rely on battery power or external wired power. However, these traditional power supply methods present numerous problems. On the one hand, battery capacity is limited, and endurance is severely insufficient, resulting in short operating times and difficulty meeting the continuous operation requirements of long-distance pipelines, greatly limiting the efficiency and scope of pipeline inspection and maintenance work. On the other hand, pipelines often contain abundant renewable energy sources such as sunlight, fluid kinetic energy, and temperature differences, but existing robots fail to fully utilize these resources, resulting in significant energy waste. Furthermore, traditional wired charging methods require the robot to return to a fixed charging point, which not only increases operational complexity but also severely impacts inspection efficiency and reduces the flexibility and timeliness of pipeline maintenance work. Therefore, there is an urgent need for a pipeline inspection and maintenance system that incorporates new energy technologies to enhance endurance, improve inspection and maintenance efficiency, and fill the gaps in existing technology in this area. An existing AI-based gas pipeline leak detection method (Bian Zhiyuan. Research on the Application of Artificial Intelligence in New Energy Gas Pipeline Leak Detection [R]. Beijing: Beijing Gas User Service Co., Ltd., 2025) uses a deep learning algorithm to analyze gas flow characteristics and pipeline continuity equations to achieve high-precision leak location. Its core technology combines sensor data with machine learning models, but its energy supply still relies on traditional batteries or external power sources, with no mention of renewable energy sources. A drawback is its single energy source, which fails to utilize ambient energy sources such as sunlight and fluid kinetic energy within the pipeline. Furthermore, it lacks a dynamic energy management strategy, making it impossible to optimize the coordinated operation of multiple energy sources. Existing technologies are based on smart grid status monitoring technology based on multi-source fusion (Yan Bo, Zhang Hao, Guo Ziming, et al. Research and application of comprehensive analysis and intelligent alarm technology of power grid faults based on multi-source data fusion [J]. Power System Protection and Control, 2018, 46(8): 5-12.) which optimizes power grid monitoring by integrating multi-source data (such as equipment operation data and environmental parameters) and emphasizes data fusion and AI analysis. However, it does not design a new energy supply module for pipeline scenarios and does not solve the technical problem of energy recovery in pipelines. Summary of the Invention

[0003] The purpose of this invention is to provide an intelligent pipeline inspection and maintenance method and system powered by multiple renewable energy sources, addressing the challenges of traditional pipeline inspection and maintenance methods, such as insufficient endurance, low energy utilization efficiency, low operational efficiency, and significant safety hazards. By integrating multiple renewable energy technologies with intelligent management strategies, this invention achieves efficient, intelligent, environmentally friendly, and autonomous pipeline inspection and maintenance operations. This significantly improves the system's endurance, intelligence, operational efficiency, and adaptability, addressing the market's urgent need for efficient and intelligent pipeline inspection and maintenance.

[0004] The present invention is achieved through at least one of the following technical solutions.

[0005] The intelligent pipeline detection and maintenance method based on multi-source new energy energy supply includes the following steps:

[0006] S1, collecting energy through the photovoltaic power generation module, the fluid energy recovery module and the temperature difference thermoelectric conversion module respectively, and inputting the collected energy into the energy storage device;

[0007] S2, the intelligent energy management system uses a deep learning model to monitor the power generation status of the photovoltaic power generation module, fluid energy recovery module, and thermoelectric conversion module in real time, and predict the power generation trend of each module;

[0008] S3, monitor the charging status of the photovoltaic power generation module, fluid energy recovery module, and thermoelectric conversion module based on the RC circuit model, and predict the state of charge using the Kalman filter algorithm;

[0009] S4. According to the maintenance tasks performed by the multi-degree-of-freedom robotic arm, the charging of multiple tasks is sorted based on the data fusion algorithm to ensure that the multi-degree-of-freedom robotic arm always has sufficient energy support during the execution of detection and maintenance tasks.

[0010] Furthermore, the photovoltaic power generation module uses sensors to collect light distribution and pipeline geometry data in real time, monitors the light intensity and spatial layout in the pipeline through sensors, and uses machine learning algorithms to predict the optimal deployment position of the deployable flexible perovskite photovoltaic cell assembly to maximize light energy collection.

[0011] Furthermore, the temperature difference thermoelectric conversion module is based on a genetic algorithm, with the temperature difference in the pipeline and the characteristics of the phase change energy storage material as constraints, to optimize the structural parameters and operating temperature range of the flexible thermoelectric generator.

[0012] Furthermore, the intelligent energy management system uses the LSTM network to capture the dynamic changes in the power generation state through the memory cell state, and combines the state of charge (SOC) estimation value fused by the Kalman filter to predict the future power generation trend of each module.

[0013] Furthermore, in the RC circuit model, the open circuit voltage of the battery The terminal voltage of the battery The relationship between them is expressed as:

[0014] ;

[0015] in, Indicates the battery time The charge and discharge current at this time is positive for the charged state and negative for the discharged state; Indicates the internal resistance of the battery; Indicates the capacitance of the battery; represents the open circuit voltage of the battery, The value is related to the battery's state of charge (SOC).

[0016] Furthermore, the multi-degree-of-freedom robotic arm uses cameras and ultrasonic sensors to obtain environmental images and structural data. The environmental images and structural data are pre-processed by the edge computing module and input into the lightweight CNN network to extract the spatial characteristics of pipeline defects, form a pipeline health status assessment matrix, and feedback defect information.

[0017] Furthermore, the multi-degree-of-freedom robotic arm dynamically plans the motion trajectory based on the model predictive control algorithm (MPC), and performs repair or cleaning operations in combination with the feedback defect information;

[0018] The model-based predictive control algorithm (MPC) is:

[0019] ;

[0020] ;

[0021] in, express The position of the robotic arm joint at the moment; represents the joint velocity; Indicates the mass of the robotic arm; express The joint torque input at any moment; represents the damping coefficient.

[0022] A system for implementing the intelligent pipeline inspection and maintenance method based on multi-source renewable energy energy supply includes a photovoltaic power generation module, a fluid energy recovery module, a temperature difference thermoelectric conversion module, an intelligent energy management system, a wireless charging module, and a pipeline inspection and maintenance module;

[0023] The photovoltaic power generation module includes a deployable flexible perovskite photovoltaic cell assembly, a sensor, a micro servo motor or a shape memory alloy, wherein the sensor, the micro servo motor or the shape memory alloy are all connected to the deployable flexible perovskite photovoltaic cell assembly;

[0024] The fluid energy recovery module transmits the data collected by the flow rate sensor and pressure sensor to the intelligent control system, which dynamically adjusts the blade angle of the micro-turbine generator through the intelligent control system to optimize the energy recovery efficiency;

[0025] The temperature difference thermoelectric conversion module uses genetic algorithms to optimize the structural parameters and operating temperature range of the flexible thermoelectric generator based on the data collected by the temperature sensor;

[0026] Intelligent energy management system, used to monitor the power generation status of photovoltaic power generation modules, fluid energy recovery modules and thermoelectric conversion modules, and optimize power distribution;

[0027] The wireless charging module monitors the charging status of the photovoltaic power generation module, fluid energy recovery module, and thermoelectric conversion module using a first-order RC circuit model, and uses a Kalman filter algorithm to predict the state of charge.

[0028] The pipeline inspection and maintenance module includes a multi-degree-of-freedom robotic arm, which is connected to an energy storage device to provide stable power for the robotic arm.

[0029] Furthermore, the intelligent energy management system collects the power generation parameters of each module in real time through the irradiance sensor of the photovoltaic power generation module, the turbine speed sensor of the fluid energy recovery module, and the thermocouple array of the temperature difference thermoelectric module, and adjusts the power output weight of each module in combination with the deep learning model to ensure the maximum energy utilization efficiency of the system.

[0030] Furthermore, a replaceable cutting head, welding head or cleaning brush tool head is provided at the end of the multi-degree-of-freedom robotic arm.

[0031] Compared with the existing technology, the beneficial effects of the present invention are:

[0032] 1. The pipeline inspection and maintenance system of the present invention comprehensively utilizes new energy technologies to achieve autonomous energy supply and ensure long-term stable operation of the system in the pipeline.

[0033] 2. The present invention uses an intelligent energy management system to utilize the LSTM network to capture the dynamic change law of the power generation state through the memory cell state, and combines the state of charge (SOC) estimation value fused by the Kalman filter to predict the future power generation trend of each module.

[0034] 3. The present invention uses artificial intelligence algorithms to distribute energy through intelligent energy management and wireless charging, dynamically optimizes the working status of different new energy modules in real time, and ensures the rational use of energy.

[0035] 4. The present invention's thermoelectric conversion module utilizes temperature differences to generate electricity in environments with significant temperature differences, such as those found in hot and cold pipelines or deep-sea pipelines. Using a genetic algorithm, the module optimizes the structural parameters and operating temperature range of the flexible thermoelectric generator, using the temperature difference within the pipeline and the properties of the phase-change energy storage material as constraints.

[0036] 5. The pipeline inspection and maintenance module of this invention integrates a camera, ultrasonic sensor, and a robotic arm, providing powerful inspection and maintenance capabilities. The robotic arm features multi-degree-of-freedom motion and can be equipped with interchangeable tool heads, such as cutting heads, welding heads, and cleaning brushes, to accommodate a variety of inspection and maintenance tasks. Equipped with high-precision sensors and advanced control algorithms, it enables high-precision operation and control, ensuring efficient and accurate pipeline inspection and maintenance.

[0037] 6. The fluid energy recovery module of this invention integrates a micro-turbine generator, cleverly utilizing the flow of gas or liquid within a pipeline to generate electricity. The adjustable turbine blade angle optimizes energy recovery efficiency based on varying flow rates, ensuring efficient power generation under various fluid conditions. This fully converts fluid kinetic energy into electrical energy, replenishing the system's energy supply. The micro-turbine generator is equipped with an intelligent control system that automatically adjusts the blade angle based on the flow rate and direction of the fluid within the pipeline to optimize energy recovery efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 The figure is a flow chart of an embodiment of an intelligent pipeline detection and maintenance system based on multi-source new energy energy supply.

[0039] Figure 2 Schematic diagram of the structure of the cleaning module according to an embodiment of the present invention.

[0040] Figure 3 A partial schematic diagram of a cleaning module provided in an embodiment of the present invention.

[0041] Figure 4 Schematic diagram of the structure of the cleaning module according to an embodiment of the present invention.

[0042] Figure 5 A structural diagram of a multi-degree-of-freedom robotic arm provided in an embodiment of the present invention.

[0043] Figure 6 This is a diagram of the drive and transmission structure provided by an embodiment of the present invention.

[0044] Figure 7 A top view of a multi-degree-of-freedom robotic arm provided in an embodiment of the present invention.

[0045] In the figure, 1-claw, 2-connecting rod, 3-claw arm, 4-upper arm, 5-lower arm, 6-connecting rod, 7-hexagon socket head screw, 8-rear small arm, 9-bolt, 10-third stud, 11-second stud, 12-first support, 13-motor plate, 14-DC brushless reduction motor, 15-first servo, 16-base plate, 17-lower base, 18-aluminum metal base plate, 19-first stud, 20-second support, 21-middle base, 22-upper base, 23-small arm, 24-support seat, 25-connecting rod gear, 26-connecting rod shaft, 27-second servo, 28-steering disc, 29-deep groove ball bearing, 30-hexagonal thin nut, 31-electronic control board, 32-dual-axis servo, 33-second pipeline robot splicing plate, 34-third pipeline robot splicing plate, 35-fourth pipeline robot splicing plate, 36-first pipeline robot splicing plate, 37-first pipeline robot side plate, 38-second pipeline robot side plate, 39-third pipeline robot side plate, 40-pipeline robot cover plate, 41-wheel, 42-first pipeline robot connecting plate, 43-second pipeline robot connecting plate, 44-spring, 45-connecting block, 46-working plate, 47-drive shaft, 48-pipeline robot cleaning parts, 49-third pipeline robot connecting plate, 50-fourth pipeline robot connecting plate, 51-fifth pipeline robot connecting plate, 52-sixth pipeline robot connecting plate, 53-first screw, 54-second screw, 55-fourth pipeline robot side plate. DETAILED DESCRIPTION

[0046] In order to make the purpose, features, and advantages of the present invention more obvious and easy to understand, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described below are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0047] To better understand the present invention, this embodiment first introduces an intelligent pipeline inspection and maintenance system based on multi-source renewable energy power supply, including: a photovoltaic power generation module, a fluid energy recovery module, a temperature difference thermoelectric conversion module, an intelligent energy management system, a wireless charging module, and a pipeline inspection and maintenance module.

[0048] The photovoltaic power generation module includes an expandable flexible perovskite photovoltaic cell assembly, a sensor assembly, a micro servo motor or a shape memory alloy; the expandable flexible perovskite photovoltaic cell assembly is connected to the micro servo motor or the shape memory alloy, and is driven by the micro servo motor or the shape memory alloy to achieve dynamic adjustment of the expansion angle and position of the expandable flexible perovskite photovoltaic cell assembly.

[0049] As an embodiment, the sensor assembly includes a light intensity sensor (such as a photodiode array) and a spatial layout sensor (such as a lidar or a depth camera). The light intensity sensor and the spatial layout sensor are installed inside the pipeline to collect light distribution and pipeline geometry data in real time. The collected light distribution and pipeline geometry data are transmitted to the edge computing module. The edge computing module (Shi W, Cao J, Zhang Q, et al. Edge Computing: Visionand Challenges[J]. IEEE Internet of Things Journal, 2016, 3(5): 637-646.) analyzes the sensor data, combines the machine learning algorithm to predict the optimal deployment position, and generates control instructions for controlling the micro servo motor or shape memory alloy.

[0050] As an example, the deployable flexible perovskite photovoltaic cell assembly is composed of a transparent conductive oxide, a perovskite light absorption layer, electron transport materials such as titanium dioxide (TiO2) and tin oxide (SnO2), and metal electrodes. Sensors, micro-servo motors, or shape memory alloys are used to control the deployable flexible perovskite photovoltaic cell assembly to its optimal position, storing electrical energy in an energy storage device. Transparent conductive oxides are made from flexible materials such as polyimide (PI), polyethylene terephthalate (PET), indium tin oxide (ITO), or fluorine-doped tin oxide (FTO).

[0051] The unfoldable flexible perovskite photovoltaic cell module is a flexible perovskite solar cell (Jeon et al., "Perovskite Solar Cells with High Efficiency and Flexibility", Nature Energy, 2020). Its flexible substrate (such as PET or PI film) is combined with a perovskite light-absorbing layer, and tri(p-tolyl)phosphine (TTP) interface passivation treatment is used to reduce surface defects and improve carrier mobility, thereby achieving an energy conversion efficiency of 23.81%.

[0052] The fluid energy recovery module includes flow sensors and pressure sensors for real-time monitoring of fluid velocity and direction. These sensors, installed within the pipeline, transmit the collected data to an intelligent control system, which dynamically adjusts the micro-turbine generator's blade angle to optimize energy recovery efficiency. This intelligent control system uses a particle swarm optimization (PSO) algorithm to optimize blade angle and rotational speed in real time, ensuring efficient power generation under varying fluid operating conditions.

[0053] The micro-turbine generator is equipped with an intelligent control system that can automatically adjust the blade angle according to the flow rate and direction of the fluid in the pipeline to optimize energy recovery efficiency. The micro-turbine generator housing uses a polyimide (PI) film substrate. After the intelligent control system is initialized, the flow rate sensor and pressure sensor begin to monitor parameters such as the flow rate, flow direction, and pressure of the fluid in the pipeline. Based on this monitoring data, the intelligent control system optimizes the blade angle and rotation speed of the micro-turbine generator to maximize energy recovery efficiency. During operation, the micro-turbine generator uses the kinetic energy of the fluid flow in the pipeline to drive the rotor and stator inside the generator through the rotation of the turbine, converting mechanical energy into electrical energy to power the system.

[0054] The temperature difference thermoelectric conversion module includes a temperature sensor for real-time monitoring of temperature difference and a flexible thermoelectric generator. The data collected by the temperature sensor installed in the pipeline is optimized using a genetic algorithm to optimize the structural parameters and operating temperature range of the flexible thermoelectric generator.

[0055] In one embodiment, the flexible thermoelectric generator (FTEG) features a serpentine folded array structured on a polyimide (PI) substrate and a bismuth telluride (Bi2Te3) thermoelectric film. The serpentine array is bonded to the inside of the pipeline robot's housing via thermally conductive silicone. A phase-change energy storage material (paraffin wax / expanded graphite PCM composite) is incorporated into the flexible thermoelectric generator (FTEG) to maintain temperature stability and energy supply, ensuring stable operation. This flexible thermoelectric generator (FTEG) utilizes a flexible thermoelectric generator (FTEG) structure and optimizes its design to improve thermoelectric conversion efficiency. The device can generate an open-circuit voltage of 173 millivolts and a power output of 4.5 milliwatts at a temperature gradient of 50K. This embodiment incorporates a phase-change energy storage material to effectively improve the efficiency of temperature differences, stably converting temperature differences into electrical energy, providing additional energy for the system and enhancing its energy self-sufficiency. The flexible thermoelectric generator (FTEG) utilizes a flexible thermoelectric generator (FTEG) structure and optimizes its design to improve thermoelectric conversion efficiency.

[0056] In this embodiment, the intelligent energy management system monitors the power generation status of the photovoltaic power generation module, fluid energy recovery module, and thermoelectric conversion module in real time by fusing multi-source sensor data and a deep learning model, and optimizes power distribution. At the same time, the photovoltaic output power of the deployable flexible perovskite photovoltaic cell assembly, the turbine speed of the micro-turbine generator, the temperature difference voltage inside and outside the pipeline, the environmental images obtained by the camera on the inner wall of the pipeline, and the structural data of the ring obtained by the ultrasonic sensor are pre-processed by the edge computing module and input into a lightweight CNN network (Goodfellow I, Bengio Y, Courville A. Deep Learning[M]. MIT Press, 2016.) to extract the spatial characteristics of pipeline defects and form a pipeline health status assessment matrix.

[0057] The wireless charging module monitors the charging status of the photovoltaic power generation module, the fluid energy recovery module, and the temperature difference thermoelectric conversion module by the energy storage device based on a first-order RC circuit model, and uses a Kalman filter algorithm to predict the charge status.

[0058] The pipeline inspection and maintenance module includes a multi-degree-of-freedom robotic arm connected to an energy storage device that provides stable power. The end of the multi-degree-of-freedom robotic arm can be equipped with interchangeable tool heads, such as cutting heads, welding heads, and cleaning brushes, to accommodate a variety of inspection and maintenance tasks. Before the system is operational, the module performs initialization checks on the high-definition camera, ultrasonic sensor, robotic arm, and other equipment to ensure proper function. Based on the pre-set inspection and maintenance tasks, a tool head replacement mechanism at the end of the robotic arm automatically replaces the corresponding tool head.

[0059] As an example, Figures 5 to 7 As shown, the multi-degree-of-freedom robotic arm of this embodiment includes a base structure, a drive and transmission structure, and an arm structure. The base structure is connected to the arm structure via bolts to ensure the stability and reliability of the arm structure. The base structure is equipped with a cleaning module for cleaning sediment and dirt from the inner wall of the pipe.

[0060] The base structure is the basic support structure of the robot arm and includes a lower base 17, a middle base 21, an upper base 22, a support base 24, and a base plate 16. Each base is fixed by bolts to form a multi-layer frame, and the multi-layer frame is fixed to the base plate 16. The support base 24 is the core installation platform of the drive and transmission system.

[0061] like Figure 5 As shown, the lower base 17 is fixed to the bottom plate 16 by a first stud 19, providing overall stability as the base of the robot arm.

[0062] The middle base 21 is mounted on the lower base 17 via the second stud 11 and carries the movement function of the robotic arm.

[0063] The upper base 22 is mounted on the middle base 21 via the third stud 10 as an intermediate support layer. A deep groove ball bearing 29 is provided between the middle base 21 and the upper base 22.

[0064] The support base 24 is fixed to the top of the upper base 22 by bolts 9 and is used to fix the drive and transmission structure.

[0065] The motor plate 13 is fixed to the underside of the base plate 16. A M15×8 brushless DC reduction motor 14 is mounted on the motor plate 13. This M15×8 brushless DC reduction motor 14 is connected to a first servo 15. The first servo 15 is fixed to the side of the lower base 17. Its output shaft is connected to a circular metal steering wheel 28. Rotating the steering wheel by the first servo 15 controls the direction of the robotic arm. An aluminum base plate 18 is positioned above the first servo 15 on the lower surface of the middle base 21.

[0066] The driving and transmission structure includes a connecting rod gear 25, a connecting rod shaft 26, and a connecting rod rack. Figure 6 As shown, one end of the connecting rod shaft 26 is embedded in the center hole of the connecting rod gear 25, and the other end is connected to the connecting rod rack via a keyway, achieving the conversion of rotational motion to linear motion. A second servo 27 is mounted on the support base 24 to drive the connecting rod gear 25. The second servo 27 is secured to the support base 24 via a thin hexagonal nut 30. The connecting rod rack is bolted to the bottom of the connecting rod 2 and meshes with the connecting rod gear 25. The second servo 27 drives the connecting rod gear 25, which in turn drives the connecting rod rack in linear motion, thereby driving the connecting rod 2 to extend and retract.

[0067] The arm structure comprises an upper arm 4, a lower arm 5, and auxiliary supports. One end of the upper arm 4 is hinged to one end of the lower arm 5. The other end of the upper arm 4 is connected to the claw arm 3 via a connecting rod 2. The connecting rod 2 is driven by the linear motion of the connecting rod rack, enabling the arm structure to extend and retract. The other end of the lower arm 5 is connected to a connecting rod 6, which is mounted on a support base 24 via a hexagon socket head cap screw 7. The claw arm 3 is terminated with a replaceable claw 1. A deep groove ball bearing is installed at the hinged joint between the claw arm 3 and the connecting rod 2, providing low-friction support and enabling the rotation of tool heads (such as cutting and welding heads, which can be installed externally) while ensuring flexible rotation.

[0068] The auxiliary support includes the rear arm 8 and the arm 23. The rear arm 8 is mounted on the base plate 16 via the first support member 12, and the arm 23 is mounted on the base plate 16 via the second support member 20, further stabilizing the structure. The rear arm 8 and the arm 23 are located on both sides of the support base 24 to enhance the structural rigidity.

[0069] like Figures 2 to 4As shown, the cleaning module includes a first motor, a main frame, a pipeline robot cleaning component 48, and a wheel assembly. The main frame includes the second pipeline robot splicing plate 33, the third pipeline robot splicing plate 34, the fourth pipeline robot splicing plate 35, and the first pipeline robot splicing plate 36. These two panels form a box-like load-bearing structure. The main frame is also equipped with multiple robot side panels (first pipeline robot side panel 37, second pipeline robot side panel 38, and third pipeline robot side panel 39). The main frame is secured to the inner wall of the pipeline using multiple M10×60 hexagonal flange nuts, providing a rigid support. A first pipeline robot cover plate 40 is secured to the first pipeline robot splicing plate 36. The first pipeline robot cover plate 40 and the fourth pipeline robot side panel 55 form a closed protective layer. The first motor is secured to the second pipeline robot splicing plate 33, located at the front end of the main frame.

[0070] The pipeline robot cleaning component 48 is connected to the output shaft of the 60kg dual-axis servo 32 through a coupling, driving the cleaning component to rotate. The dual-axis servo 32 is fixed to the second pipeline robot splicing plate 33 by a first screw 53 and a second screw 54.

[0071] The power of the first motor is transmitted through a coupling to the transmission assembly (which includes the connecting rod gear 25, the drive shaft 47 of the pipeline robot's cleaning unit 48, and other components). The coupling is secured to the second pipeline robot's splicing plate 33 using a 62-GB / T 16674-1996 hexagonal flange nut M10×60 to ensure stable power transmission. The connecting rod gear 25 and the drive shaft 47 are connected by a keyway, transmitting the rotational power to the drive shaft of the pipeline robot's cleaning unit 48.

[0072] The main frame is also equipped with multiple welded plates: the first welded plate secures the first motor; the second welded plate supports the sidewalls of the first, second, and third pipeline robot side panels 37, 38, and 39; the third welded plate forms the bottom load-bearing structure; and the fourth welded plate provides sealing. Each welded plate is joined together using screws and nuts.

[0073] like Figure 3 As shown, the fourth pipeline robot side plate 55 is also welded and spliced ​​with the high-strength load-bearing frame to simultaneously fix the transmission components of the first motor and the pipeline robot cleaning part 48; the high-strength load-bearing frame includes the first pipeline robot connecting plate 42, the second pipeline robot connecting plate 43, the third pipeline robot connecting plate 49, the fourth pipeline robot connecting plate 50, the fifth pipeline robot connecting plate 51, and the sixth pipeline robot connecting plate 52.

[0074] The first pipeline robot connecting plate 42 and the second pipeline robot connecting plate 43 are spliced ​​together through hexagonal flange nuts to form side wall supports. The third pipeline robot connecting plate 49, the fourth pipeline robot connecting plate 50, the sixth pipeline robot connecting plate 52 and the fifth pipeline robot connecting plate 51 are combined to form the bottom load-bearing layer, which is cushioned by springs.

[0075] A work plate 46 is mounted in the middle of the main frame. A pipe robot cleaning unit 48 is secured to this plate via a connecting block 45. The drive shaft of the pipe robot cleaning unit 48 meshes with the connecting rod gear 25. The wheel assembly of the pipe robot cleaning unit 48 includes symmetrically spaced wheels 41, which are mounted to one side of the main frame via shafts and bearings. Springs 44 are installed between the symmetrically spaced wheels to preload the wheels.

[0076] The base plate 16 of the base structure is connected to the main frame through the first pipeline robot splicing plate 36 to achieve integrated detection and cleaning functions. The DC brushless reduction motor 14, the first steering gear 15 and the first motor are coordinated and powered by an intelligent energy management system.

[0077] The pipeline cleaning principle of this embodiment: The cleaning module completes the task of cleaning the inner wall of the pipeline through the following steps:

[0078] 1).Power transmission: The motor drives the wheel group to rotate, pushing the cleaning module to move in the pipeline.

[0079] 2) Cleaning unit operation: The pipeline robot cleaning unit 48 cleans the sediment and dirt on the inner wall of the pipeline through rotation or vibration.

[0080] 3) Spring buffer: The spring 44 absorbs the impact of the uneven inner wall of the pipe, ensuring that the cleaning module maintains good contact with the inner wall of the pipe, thereby improving the cleaning effect.

[0081] The M15×8 brushless DC motor 14 and first servo 15 are powered by an energy storage device. This energy is derived from photovoltaic power generation, fluid energy recovery, and thermoelectric power generation modules, supplemented by a wireless charging module. Energy collected by the photovoltaic, fluid energy, and thermoelectric power generation modules is stored in the energy storage device, which prioritizes powering the brushless DC motor 14, first servo 15, and sensors.

[0082] The electric control board 31 of the intelligent energy management system optimizes power distribution through the edge computing module (Shi W, et al. IEEE IoT Journal, 2016).

[0083] like Figure 1 As shown, this embodiment provides an intelligent pipeline detection and maintenance method based on multi-source new energy energy supply, including the following steps:

[0084] S1. Energy collection and conversion, including the following steps:

[0085] S11. The photovoltaic power generation module collects light distribution and geometric data within the pipeline, monitors the light intensity and spatial layout within the pipeline through sensors, and uses a machine learning algorithm to predict the deployment position of the deployable flexible perovskite photovoltaic cell assembly to maximize light energy collection. The collected light energy is input into the energy storage device, specifically including the following steps:

[0086] S111. Collecting Light Data: Install light intensity sensors (such as photodiode arrays) and spatial layout sensors (such as lidar or depth cameras) inside the pipeline to collect real-time light distribution and pipeline geometry data. Light intensity sensors (such as photodiode arrays) measure light intensity by converting optical signals into electrical signals. The output current of the photodiode array is proportional to the incident light intensity and can be expressed as:

[0087] ;

[0088] in, is the output current, is the sensor sensitivity, is the incident light power. The output current of the photodiode array is input to the edge computing module for real-time collection of light intensity data.

[0089] S112. Spatial Data Collection: LiDAR acquires distance information by emitting laser pulses and measuring the flight time of the reflected light. The distance calculation formula is:

[0090] ;

[0091] in, is the straight-line distance between the laser radar and the reflection point on the inner wall of the pipe, is the speed of light, For flight time.

[0092] S113. Predicting the optimal deployment position of a deployable flexible perovskite photovoltaic cell module: Using edge computing modules to analyze LiDAR and depth camera data, combined with machine learning algorithms (such as reinforcement learning), we predict the optimal deployment position of a deployable flexible perovskite photovoltaic cell module and generate control instructions. The reinforcement learning algorithm uses a Q-learning algorithm, and the update rule is:

[0093] ;

[0094] in, Status Take action of value, is the learning rate, For reward, is the discount factor, For the next state Take action Maximum value.

[0095] S114. Deployment and adjustment of a flexible, deployable perovskite photovoltaic cell assembly to maximize energy collection, with the collected energy input into an energy storage device: The flexible perovskite photovoltaic cell assembly is driven by a micro-servo motor or shape memory alloy, enabling dynamic adjustment of the deployment angle and position. The micro-servo motor receives control commands to drive the rotational joints of the photovoltaic cell assembly's biomimetic folding structure. The duty cycle of the PWM signal in the control command is linearly related to the motor's rotation angle, which can be expressed as:

[0096] ;

[0097] in, is the rotation angle of the micro servo motor, is the proportionality constant, is the on-time of the PWM signal, is the period of the PWM signal.

[0098] S12, the fluid energy recovery module automatically adjusts the blade angle of the micro-turbine generator with adjustable blade angle according to the flow rate and flow direction of the fluid in the pipeline to optimize the energy conversion efficiency, and inputs the collected energy into the energy storage device, specifically comprising the following steps:

[0099] S121. Monitoring fluid parameters: Monitoring the flow rate of the fluid in the pipeline through flow rate sensors and pressure sensors , flow direction and pressure Parameters such as flow rate. Used to calculate fluid kinetic energy ( is the fluid density, is the turbine cross-sectional area), which is used as the input of the PSO algorithm optimization objective function; the flow direction is used to determine whether the fluid flow direction matches the turbine rotation direction and dynamically adjust the blade deflection angle; the pressure Used to detect abnormal fluid conditions in pipelines (such as blockage or leakage) and trigger protection mechanisms.

[0100] S122. Blade Angle Adjustment: The micro-turbine generator is equipped with an intelligent control system that adjusts the blade angle of the micro-turbine generator based on the intelligent optimization algorithm of particle swarm optimization (PSO) to construct the objective function in is the blade angle, is the rotation speed, is the energy conversion efficiency, is the pressure loss, is the weight coefficient. Iterative optimization through PSO algorithm and , by simulating the social behavior of particle swarms, we can find the optimal solution in the search space. The speed update formula is:

[0101] ;

[0102] ;

[0103] in, Indicates the The particle in The first iteration Dimensional speed; represents the inertia weight; and represents the learning factor; and Represents a random number; Indicates the The best historical position of the particle Dimensional component; The first one represents the global best position Dimensional component; Indicates the The particle in The first iteration Dimensional location.

[0104] S13. The thermoelectric conversion module utilizes the temperature difference between the inside and outside of the pipeline and uses a genetic algorithm to iteratively calculate the optimal operating temperature range of the flexible thermoelectric generator. It stabilizes energy supply through phase change energy storage materials and inputs the collected energy into the energy storage device. Specifically, the module includes the following steps:

[0105] S131. Temperature monitoring: Real-time monitoring of the temperature inside the pipeline through multiple temperature sensors (unit: K) and the temperature outside the pipe , calculate the temperature difference Temperature difference As the input parameter of the genetic algorithm, it is used to construct the fitness function ,in Output power for thermoelectric generator; Phase change energy storage material (PCM) according to Dynamically adjust the phase change temperature range to maintain the stable operating temperature of the thermoelectric generator.

[0106] S132. Thermoelectric performance optimization: Using the optimization method based on genetic algorithm (GA), the temperature difference The structural parameters (such as the number of thermocouples, leg length) and operating temperature range of the flexible thermoelectric generator are optimized based on the characteristics of the phase change energy storage material. The crossover and mutation operations can be expressed as:

[0107] ;

[0108] ;

[0109] in, and Indicates the and parent individuals; represents the cross coefficient; represents the offspring individual; Indicates variable asynchronous length; represents a random noise vector, is the number of iterations. The optimization goal is to maximize the thermoelectric conversion efficiency :

[0110] ;

[0111] in is the thermoelectric figure of merit, which is determined by the properties of the phase change energy storage material. and They are the hot end and cold end temperatures, which are collected in real time by temperature sensors.

[0112] S2. Intelligent energy management is achieved through an intelligent energy management system, including the following steps:

[0113] S21. Calculate energy efficiency:

[0114] The intelligent energy management system monitors the power generation status of photovoltaic power generation modules, fluid energy recovery modules, and thermoelectric conversion modules in real time by integrating multi-source sensor data with a deep learning model (Goodfellow I, Bengio Y, Courville A. Deep Learning[M]. MIT Press, 2016.), and optimizes power distribution. The energy utilization efficiency is calculated as follows:

[0115] ;

[0116] in, Indicates energy utilization efficiency; Indicates the useful power in the entire system, that is, the power that is effectively utilized; Indicates the total power of the entire system. By calculating the ratio of useful power to total power, the energy utilization efficiency is obtained. ,when When the power generation efficiency is high and stable, the energy storage device has low charge and discharge losses, and the intelligent energy management system optimizes power distribution through AI algorithms to make the useful power ratio significant), the system can meet the real-time operation needs and even accumulate redundant power in the energy storage device. At this time, the energy is considered to be sufficient. On the contrary, if The useful power is too low (for example, due to insufficient sunlight, the efficiency of photovoltaic power generation decreases, the fluid flow rate is too low, or the temperature difference is too small, resulting in insufficient energy recovery), so that the useful power cannot cover the minimum energy consumption requirements of core functions such as robot arm operation, sensor operation and data transmission. Even if there is some power in the energy storage device, the system may still be judged as insufficient energy due to the power gap.

[0117] S22. Monitoring and predicting power generation status, including the following steps:

[0118] The photovoltaic module's irradiance sensor, the fluid energy recovery module's turbine speed sensor, and the thermocouple array of the thermoelectric module collect power generation parameters from each module in real time. An intelligent energy management system (combined with a deep learning model) adjusts the power output weights of each module to ensure maximum energy efficiency. The intelligent energy management system uses an LSTM network (Hochreiter S, Schmidhuber J. Long Short-Term Memory[J]. Neural Computation, 1997, 9(8): 1735-1780.) to capture the dynamic changes in power generation status through memory cell states. Combined with the state of charge (SOC) estimate fused by the Kalman filter, it predicts the power generation trend of each module in the next 30 minutes.

[0119] S3, wireless charging and state of charge (SOC) prediction, includes the following steps:

[0120] S31. State of Charge Prediction

[0121] The state of charge (SOC) of the energy storage device battery is obtained based on a first-order RC circuit model. By monitoring the battery's terminal voltage and charge and discharge current, combined with the battery's internal resistance and capacitance, the battery's open circuit voltage can be predicted, and the battery's state of charge (SOC) can be obtained. This allows for accurate monitoring of the charging state. Specifically, the following are included:

[0122] Battery status monitoring: By monitoring the terminal voltage and charge and discharge current of the energy storage device battery, combined with the battery's internal resistance and capacitance, the battery's open circuit voltage is predicted.

[0123] State of charge SOC estimation: Based on the first-order RC circuit model, the open circuit voltage of the battery The terminal voltage of the battery The relationship between them is expressed as:

[0124] ;

[0125] in, Indicates that the energy storage device is at time The charge and discharge current is positive when charging and negative when discharging. The charge and discharge current at this time is positive for the charged state and negative for the discharged state; Indicates the internal resistance of the energy storage device; Represents the equivalent capacitance of the energy storage device; Indicates the open circuit voltage of the energy storage device, which is related to the state of charge (SOC) of the energy storage device and can be obtained through experimental calibration. During the charging process, the terminal voltage of the energy storage device is monitored. and charge and discharge current , combined with the internal resistance of the energy storage device and capacitors , the open circuit voltage of the energy storage device can be predicted , and then obtain the battery's state of charge SOC, realize accurate monitoring of the charging state, and ensure that the battery is always in good working condition.

[0126] S32. Wireless charging

[0127] The wireless charging module uses multi-band magnetic resonance technology to receive energy from external wireless charging base stations and monitors the charging status based on the state of charge (SOC) prediction method. The system can use magnetic resonance wireless charging to replenish energy at key nodes in the pipeline, achieving a flexible and convenient charging method, further improving the system's endurance and operating efficiency. The wireless charging module uses multi-band magnetic resonance technology, which can efficiently receive energy within different frequency ranges and is compatible with various types of wireless charging base stations. Its charge status can be monitored through the state of charge (SOC) prediction method to ensure the safety and effectiveness of the charging process. Specifically including:

[0128] Energy reception: The wireless charging module receives energy from an external wireless charging base station.

[0129] Charging monitoring: The Kalman filter algorithm is used to weight and fuse different sensor data to improve the accuracy of SOC estimation. The prediction equation and update equation of the Kalman filter algorithm are:

[0130]

[0131] ;

[0132]

[0133] ;

[0134] ;

[0135]

[0136] ;

[0137] in, and Respectively expressed in Prior and posterior estimates of moments; and denote the prior estimation error covariance and the posterior estimation error covariance respectively; represents the state transition matrix; represents the control input matrix; represents the control input; represents the process noise covariance; represents the measurement residual; Represents the residual between the measured and predicted values; Indicates the measured value; represents the measurement matrix; represents the measurement noise covariance; represents the Kalman gain; Indicates Moment based on The predicted value of the current state based on all information at the moment (including historical state and control input). The ^ above it is an estimated value rather than a true value. For example, in the state of charge (SOC) prediction of a pipeline inspection system, It is the current SOC prediction value calculated based on the SOC, charge and discharge current and other parameters at the previous moment, without considering new measurement data. express The covariance matrix of the state estimation error after the measurement update at the moment reflects the credibility of the estimated value at that moment. The smaller its value, the closer the estimate is to the true state. For example, in the system, if the error of the SOC estimation at the previous moment is significantly reduced after the sensor data correction, then Will be smaller. Quantifies the uncertainty of the residual between the measured and predicted values ​​and is used to calculate the Kalman gain is a key parameter used to balance the weight of prediction and measurement. For example, if the voltage measurement noise of the temperature difference power generation module in the pipeline is large ( High), then will increase, leading to Reduce and weaken the impact of measurement data on state updates. Identity matrix It is used in the covariance update equation in the Kalman filter to achieve iterative correction of the covariance by multiplying the unit matrix with the Kalman gain and the measurement matrix.

[0138] S4. Based on the maintenance tasks performed by the multi-DOF manipulator, the charging of multiple tasks is sorted based on the data fusion algorithm (see: Chen Zhengyu et al. A review of data fusion research in wireless sensor networks [J]. Computer Application Research, 2011, 28(5): 1601-1604). The charging power is allocated according to the expected energy consumption and time of the tasks to ensure that the multi-DOF manipulator always has sufficient energy support during the task execution. The maintenance task performed by the multi-DOF manipulator includes the following steps:

[0139] S41. Defect Detection

[0140] The multi-degree-of-freedom robotic arm uses cameras and ultrasonic sensors to acquire environmental images and structural data. After preprocessing by the edge computing module, the data is input into a lightweight CNN network to extract the spatial features of pipeline defects and form a pipeline health status assessment matrix. The specific steps include:

[0141] S411. Image data processing: A high-definition camera captures high-resolution images of the inner wall of the pipeline. Using visual recognition technology (see Chinese patent CN109242830A, a machine vision detection method based on deep learning), it detects defects such as cracks, deformations, sediments, and obstacles. The detected defects are then sent to a multi-degree-of-freedom robotic arm.

[0142] S412. Ultrasonic data processing: Ultrasonic sensors are embedded in the robotic arm or the front end of the system. They transmit and receive ultrasonic signals to detect structural anomalies such as changes in pipe wall thickness, internal corrosion, or hidden cracks.

[0143] S42. Maintenance task execution

[0144] The motion trajectory of the multi-degree-of-freedom robotic arm is dynamically planned based on the model predictive control algorithm (MPC), and combined with the feedback defect information, repair or cleaning operations are performed.

[0145] The model-based predictive control algorithm (MPC) is:

[0146] ;

[0147] ;

[0148] in, express The position of the robotic arm joint at the moment; represents the joint velocity; Indicates the mass of the robotic arm; represents the input joint torque; Represents the damping coefficient; Based on the model predictive control algorithm (MPC), the motion trajectory of the robot arm can be predicted in real time, and the output of the joint torque can be optimized according to the prediction results to achieve high-precision operation and control. During the movement of the robot arm, the position of the robot arm joint is monitored. and joint velocity , combined with the mass of the robotic arm , input joint torque and damping coefficient , which can precisely control the movement of the robotic arm, ensuring its stable operation and precise maintenance in complex pipeline environments to achieve high-precision trajectory tracking.

[0149] This embodiment integrates the aforementioned modules (photovoltaic power generation module, fluid energy recovery module, thermoelectric conversion module, intelligent energy management system, wireless charging module, and pipeline inspection and maintenance module) through a layered physical layout design (external energy collection, internal energy conversion and storage, and front-end execution unit) and the coordinated integration of power and information channels, forming a fully closed-loop chain from energy capture, storage, intelligent allocation, to task execution. Energy collection primarily uses photovoltaics and turbine generators to provide basic energy, while the thermoelectric module supplements ambient energy. The energy storage device stores energy and smoothes energy supply fluctuations to ensure continuous system operation. The robotic arm performs core operational tasks, while wireless charging provides emergency energy replenishment. The intelligent management system dynamically optimizes energy allocation and task priorities using a deep learning model, ultimately achieving efficient, autonomous, and long-lasting operation in complex pipeline environments. First, the intelligent energy management system dynamically allocates power resources for modules such as photovoltaics and fluid kinetic energy. When necessary, wireless charging modules are used to replenish energy at pipeline nodes. This system then combines the SOC value of the energy storage device predicted by the Kalman filter to generate an optimal charging strategy to ensure the continuity of maintenance tasks.

[0150] As a preferred embodiment, a flexible perovskite photovoltaic cell module can adopt flexible photovoltaic cells and combine them with a high-reflectivity coating on the inner wall of the pipe to significantly improve the efficiency of light energy utilization. By using tri(p-tolyl)phosphine (TTP) for interface passivation treatment, its flexible substrate (such as PET or PI film) is combined with the perovskite light-absorbing layer. The tri(p-tolyl)phosphine (TTP) interface passivation treatment reduces surface defects and improves carrier mobility, thereby improving energy conversion efficiency.

[0151] As another embodiment, the system of the present invention can also be equipped with a fiber optic receiving device (Chen et al., "Optical Fiber-Based Daylighting System for Indoor Applications", SolarEnergy, 2021) to guide external light sources into the pipeline, further enhancing the photovoltaic power generation capacity and providing continuous and stable power support for the system.

[0152] The preferred embodiments of the present invention disclosed above are intended only to help illustrate the present invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the present invention to the specific embodiments described. Obviously, numerous modifications and variations are possible based on the contents of this specification. These embodiments are selected and described in detail in this specification to better explain the principles and practical applications of the present invention, so that those skilled in the art can better understand and utilize the present invention.

Claims

1. An intelligent pipeline inspection and maintenance method based on multi-source new energy energy supply is characterized by: The following steps are involved: S1. Collecting energy through a photovoltaic power generation module, a fluid energy recovery module, and a temperature-difference thermoelectric conversion module, respectively, and inputting the collected energy into an energy storage device; the photovoltaic power generation module collects light distribution and geometric data in the pipeline, monitors the light intensity and spatial layout in the pipeline through sensors, and uses a machine learning algorithm to predict the deployment position of a deployable flexible perovskite photovoltaic cell assembly to maximize light energy collection, and inputting the collected light energy into the energy storage device, specifically comprising the following steps: S111. Light Data Collection: A photodiode array and lidar are installed inside the pipeline to collect real-time light distribution and pipeline geometry data. The photodiode array measures light intensity by converting optical signals into electrical signals. The output current of the photodiode array is proportional to the incident light intensity and is expressed as: ； in, is the output current, is the sensor sensitivity, is the incident light power; the output current of the photodiode array is input to the edge computing module for real-time collection of light intensity data; S112. Spatial Data Collection: LiDAR acquires distance information by emitting laser pulses and measuring the flight time of the reflected light. The distance calculation formula is: ； in, is the straight-line distance between the laser radar and the reflection point on the inner wall of the pipe, is the speed of light, is the flight time; S113. Predicting the optimal deployment position of a deployable flexible perovskite photovoltaic cell module: Using edge computing modules to analyze LiDAR and combine machine learning algorithms to predict the optimal deployment position of a deployable flexible perovskite photovoltaic cell module and generate control instructions. S114. Deployment and adjustment of a deployable flexible perovskite photovoltaic cell assembly to maximize energy collection, which is then fed into an energy storage device. The flexible perovskite photovoltaic cell assembly is driven by a micro-servo motor or shape memory alloy to achieve dynamic adjustment of the deployment angle and position. S2, the intelligent energy management system uses a deep learning model to monitor the power generation status of the photovoltaic power generation module, fluid energy recovery module, and thermoelectric conversion module in real time, and predict the power generation trend of each module; S3, monitor the charging status of the photovoltaic power generation module, fluid energy recovery module, and thermoelectric conversion module based on the RC circuit model, and predict the state of charge using the Kalman filter algorithm; S4. According to the maintenance tasks performed by the multi-degree-of-freedom robotic arm, charging of multiple tasks is sorted based on data fusion, so that the multi-degree-of-freedom robotic arm always has sufficient energy support during the execution of detection and maintenance tasks.

2. The intelligent pipeline detection and maintenance method based on multi-source new energy energy supply according to claim 1 is characterized in that: The temperature difference thermoelectric conversion module is based on a genetic algorithm and takes the temperature difference in the pipeline and the characteristics of the phase change energy storage material as constraints to optimize the structural parameters and operating temperature range of the flexible thermoelectric generator.

3. The intelligent pipeline detection and maintenance method based on multi-source new energy energy supply according to claim 1 is characterized in that: The intelligent energy management system uses the LSTM network to capture the dynamic changes in the power generation state through the memory cell state, and combines the state of charge (SOC) estimation value fused by the Kalman filter to predict the future power generation trend of each module.

4. The intelligent pipeline detection and maintenance method based on multi-source new energy energy supply according to claim 1 is characterized in that: The multi-degree-of-freedom robotic arm uses cameras and ultrasonic sensors to obtain environmental images and structural data. After pre-processing by the edge computing module, the environmental images and structural data are input into the lightweight CNN network to extract the spatial characteristics of pipeline defects, form a pipeline health status assessment matrix, and feedback defect information.

5. An intelligent pipeline detection and maintenance system that implements the intelligent pipeline detection and maintenance method based on multi-source new energy energy supply as described in claim 1, characterized in that: Including photovoltaic power generation module, fluid energy recovery module, temperature difference thermoelectric conversion module, intelligent energy management system, wireless charging module, pipeline detection and maintenance module; The photovoltaic power generation module includes a deployable flexible perovskite photovoltaic cell assembly, and also includes a sensor, a micro servo motor or a shape memory alloy, wherein the sensor, the micro servo motor or the shape memory alloy are all connected to the deployable flexible perovskite photovoltaic cell assembly; The fluid energy recovery module transmits the data collected by the flow rate sensor and pressure sensor to the intelligent control system, which dynamically adjusts the blade angle of the micro-turbine generator through the intelligent control system to optimize the energy recovery efficiency; The temperature difference thermoelectric conversion module uses genetic algorithms to optimize the structural parameters and operating temperature range of the flexible thermoelectric generator based on the data collected by the temperature sensor; The intelligent energy management system is used to monitor the power generation status of the photovoltaic power generation module, fluid energy recovery module and thermoelectric conversion module, and optimize the distribution of electricity; The wireless charging module monitors the charging status of the photovoltaic power generation module, fluid energy recovery module, and thermoelectric conversion module using the RC circuit model, and uses the Kalman filter algorithm to predict the state of charge. The pipeline inspection and maintenance module includes a multi-degree-of-freedom robotic arm, which is connected to an energy storage device to provide stable power for the robotic arm.

6. The intelligent pipeline detection and maintenance system according to claim 5, characterized in that: The intelligent energy management system collects the power generation parameters of each module in real time through the irradiance sensor of the photovoltaic power generation module, the turbine speed sensor of the fluid energy recovery module, and the thermocouple array of the temperature difference thermoelectric module. It adjusts the power output weight of each module in combination with the deep learning model to maximize the energy utilization efficiency of the system.

7. The intelligent pipeline detection and maintenance system according to claim 5 or 6, characterized in that: The end of the multi-degree-of-freedom robotic arm is provided with a replaceable cutting head, welding head or cleaning brush tool head.

Images