Robot na May Sariling Pagsisiyasat para sa 110kV Transmission Lines: disenyo at Pagpapatupad ng Sistema na may Tatlong Braso na Nakasabit

Abstract

Upang harapin ang mga inherent na limitasyon ng manual na pagsusuri at aerial survey para sa high-voltage transmission lines, ipinakilala ng proposal na ito ang isang autonomous inspection robot na espesyal na disenyo para sa 110 kV power lines. Mayroon itong isang inovatibong three-arm suspended mechanical structure, na nagbibigay ng autonomous crawling, obstacle negotiation, online power harvesting, at multi-fault diagnosis. Layunin nito na awtomatik at intellectualizein ang line inspection, na siyang makakapag-angkop na pag-improve ng efficiency at seguridad ng grid operation at maintenance habang pinabababa ang cost.

I. Project Background and Objectives

1.1 Background: Challenges of Traditional Inspection Methods

Ang high-voltage transmission lines, na patuloy na nakalantad sa outdoor environment, ay madaling makuha ng mga kapansanan tulad ng broken strands at wear dahil sa mechanical tension, electrical flashover, at material aging, kaya nangangailangan ng regular na pagsusuri. Ang kasalukuyang mga pamamaraan ay may malaking bottlenecks:

• Manual Inspection: Labor-intensive, inefficient, high-risk, at malaki ang pagkaka-limitado ng panahon at terreno.
• Drone Aerial Surveying: Mataas na operational cost, limited endurance, subject sa airspace control at adverse weather, at mahirap para sa close-range defect detection.

1.2 Objectives: An Intelligent Inspection Alternative

Layunin ng proyektong ito na magdesinyo ng autonomous inspection robot para sa 110 kV high-voltage transmission lines na maaaring palitan ang manual labor. Ang core objectives ay kinabibilangan ng:

• Functional Autonomy: Makamit ang autonomous crawling at precise obstacle negotiation (e.g., crossing vibration dampers and clamps) sa lines.
• Intelligent Detection: Integrate visual at infrared sensors upang automatic na i-identify at i-diagnose ang typical faults tulad ng broken strands.
• Energy Self-Sufficiency: Gumamit ng non-contact inductive power harvesting technology para sa online self-replenishment, na nagbibigay-daan sa long-distance inspection.
• Maximized Efficiency: Malaking pag-improve ng inspection efficiency at data accuracy, na siyang nagbabawas ng operational costs at safety risks.

II. Core Technical Solutions

2.1 Innovative Mechanical Structure Design: High Mobility and Stability

• Overall Structure: Nag-aadopt ng isang three-arm suspended configuration na naglalaman ng mga advantage ng multi-segment separated at wheel-arm composite mechanisms, na balanse ang efficiency ng wheeled movement at stability ng inchworm-like creeping. Ang kabuuang timbang ay humigit-kumulang 29 kg.
• Key Components:
• Flexible Arms: Ang front at rear arms ay gumagamit ng double four-bar linkage mechanism, na pinapatakbo ng kabuuang 16 motors, na nagbibigay ng independent o coordinated pitch motion na may joint stiffness-flexibility smooth transition capability upang ma-adapt sa complex line conditions.
• Drive Unit: Gumagamit ng high-power Swiss Maxon DC motors na may center-separated drive wheels, na nagbibigay ng malakas na obstacle-crossing ability (capable of passing vibration dampers) at gradeability (routine 60°, up to 80° with braking).
• Braking Unit: Nag-aadopt ng spiral-crank slider self-locking mechanism upang mabigyan ng epektibong pag-iwas sa accidental slipping o falling sa slope traversal o obstacle negotiation.
• Kinematic Validation: Inverse kinematics analysis batay sa CCD iterative algorithm; ang simulations ay nagpapakita ng convergence sa loob lamang ng 7 iterations, na epektibong nagpapatunay ng kakayahan ng robot na makamit ang complex poses tulad ng crossing suspension clamps at 45° turn jumpers.

2.2 Hierarchical Intelligent Control System: Seamless Autonomy and Remote Control

• System Architecture: Nag-aadopt ng isang three-layer distributed control structure (upper ground management layer, middle robot planning layer, lower execution layer), na sinasadya ng PC/104 industrial computer at ATmega128AU microcontroller para sa real-time decision-making at execution.
• Hybrid Control Strategy:
• Autonomous Mode: Offline path planning batay sa pre-set knowledge base, na pinagsama sa real-time sensor feedback para sa fully autonomous crawling at obstacle negotiation.
• Remote Control Mode: Sa napakalalim na environment, ang ground operators ay maaaring gawin ang joint-level fine manipulation o mag-issue ng macro-commands via remote intervention, na suportado ng HD video (25-30 Hz) na inililipad mula sa robot.
• Performance Metrics: Single inspection distance ≥ 2 km, average speed ≥ 0.9 m/h, image transmission distance ≥ 2 km.

2.3 Online Inductive Power Harvesting & Intelligent Power Management: Unlimited Endurance

• Power Harvesting Principle: Gumagamit ng split-core current transformer upang inductively harvest energy mula sa magnetic field sa paligid ng high-voltage conductor. Ang CT core ay gawa ng high-permeability iron-based nanocrystalline alloy; ang optimized design ay nagbibigay ng low starting current na 32 A.
• Power System: Nagbibigay ng stable rectified voltage; ang output power ay sumasaklaw sa line current range na 32 A hanggang 10 kA. Nakakabit ng 24 V/12 A·h intelligent Li-ion battery pack na gumagamit ng three-stage charging algorithm, na may over-temperature protection para sa seguridad, efficiency, at long service life.

2.4 Machine Vision Obstacle Recognition: Accurate Navigation

• Recognition Targets: Accurately identifies key obstacles tulad ng suspension clamps, straight-line jumper clamps, at turn jumper clamps.
• Algorithm Flow:
• Positioning: Coarse positioning via sub-block grayscale analysis, precise identification ng transmission line via histogram equalization at threshold segmentation.
• Feature Extraction: Extracts obstacle contours gamit ang morphological operations, analyzing left/right edge slopes bilang classification features.
• Recognition: Applies a fuzzy pattern recognition algorithm based on the maximum membership principle para sa mabilis at accurate obstacle type identification.
• Performance: Single image processing time ≈ 108 ms; reliably identifies typical obstacles, providing real-time input for obstacle-negotiation decisions.

2.5 Broken Strand Intelligent Diagnosis: Accurate Fault Warning

• Detection Principle: Batay sa phenomenon ng localized resistance increase at temperature rise dahil sa broken strands, gumagamit ng infrared sensor upang detektohin ang thermal radiation signals.
• Intelligent Diagnosis Model:
• Signal Processing: Gumagamit ng db4 wavelet base para sa 6-layer decomposition upang filter out noise at focus sa frequency bands na may fault features.
• Feature Extraction: Introduces wavelet energy entropy upang characterize signal complexity, combined sa peak-to-peak values ng detail components, forming a 4-dimensional feature vector.
• Diagnosis Decision: Uses a 3-layer BP neural network para sa diagnosis. Experimental verification shows 100% accuracy sa test samples at 98% online detection success rate.

III. Solution Advantages Summary

• High Adaptability: Three-arm flexible structure nagbibigay ng excellent obstacle negotiation at terrain adaptability.
• High Autonomy: Hybrid control system nagbibigay ng long-distance autonomous inspection na may remote intervention capability.
• Long Endurance: Innovative online power harvesting fundamentally solves endurance limitations.
• Accurate Detection: Integration ng machine vision at infrared thermography na may intelligent algorithms ensures high fault-recognition accuracy.
• Safe and Economical: Replaces high-risk manual work, reducing safety hazards at long-term operational costs.

IV. Current Limitations and Future Prospects

4.1 Current Limitations

• Maaari pa ring kailanganin ng minimal na manual assistance sa napakalalim na line environments.
• Potential para sa further optimization ng mechanism size at obstacle-negotiation stroke para sa mas compact na disenyo.
• Ang power system starting current ay maaari pang mataas, limiting application sa very low-load lines.
• Ang current fault detection types ay mainly focused sa broken strands; ang range ng detectable faults ay maaaring i-expand.

4.2 Future Outlook

• Mechanism lightweighting at balance optimization upang improve ang obstacle-negotiation efficiency at stability.
• Integration ng multi-sensor navigation upang enhance ang positioning at environmental perception accuracy.
• Optimization ng power harvesting circuit upang further reduce ang starting current at expand ang application range.
• Expansion ng fault diagnosis library upang include ang defects tulad ng damaged insulators at contamination.
• Enhancement ng reliability ng robot, improving industrial-grade protection (e.g., dustproof, waterproof, at EMC capabilities).