Roboticong May Pagsisiyasiya para sa 110kV Transmission Lines: disenyo at pagpapatupad ng isang Tatlong-Braso na Nakasabit na Sistema

Abstract

Upang tugunan ang inherent na limitasyon ng manual na inspeksyon at aerial survey para sa high-voltage transmission lines, ipinakilala ng proposal na ito ang isang autonomous inspection robot na partikular na disenyo para sa 110 kV power lines. Mayroon itong isang inobatibong three-arm suspended mechanical structure, na naglalaman ng autonomous crawling, obstacle negotiation, online power harvesting, at multi-fault diagnosis. Layunin nito na automatizehin at intellectualizehin ang line inspection, na siyang nagpapataas ng epektividad at kaligtasan ng grid operation and maintenance habang pinabababa ang mga gastos.

I. Project Background and Objectives

1.1 Background: Challenges of Traditional Inspection Methods

Ang high-voltage transmission lines, na patuloy na nakalantad sa outdoor environments, ay madaling magkaroon ng mga defect tulad ng broken strands at wear dahil sa mechanical tension, electrical flashover, at material aging, kaya kailangan ng regular na inspeksyon. Ang kasalukuyang mga pamamaraan ay may malaking bottlenecks:

• Manual Inspection: Labor-intensive, inefficient, high-risk, at malaki ang paghahadlang ng weather at terrain.
• Drone Aerial Surveying: Mataas ang 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 lumikha ng isang autonomous inspection robot para sa 110 kV high-voltage transmission lines na kayang palitan ang manual labor. Ang mga pangunahing layunin 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: Gamitin ang non-contact inductive power harvesting technology para sa online self-replenishment, na nagbibigay-daan sa long-distance inspection.
• Maximized Efficiency: Malaking pagtaas 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: Ginagamit ang isang three-arm suspended configuration na pagsasama ng mga advantages ng multi-segment separated at wheel-arm composite mechanisms, balancing the efficiency of wheeled movement with the stability of 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 with joint stiffness-flexibility smooth transition capability upang tumugon 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: Gumagamit ng spiral-crank slider self-locking mechanism upang mabigyan ng epektibong pagpigil sa accidental slipping or falling during slope traversal or obstacle negotiation.
• Kinematic Validation: Inverse kinematics analysis batay sa CCD iterative algorithm; ang simulations ay nagpapakita ng convergence in only 7 iterations, na epektibong nagpapatotoo sa 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: Ginagamit ang isang three-layer distributed control structure (upper ground management layer, middle robot planning layer, lower execution layer), na pinagsasama 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 pinagsasama ng real-time sensor feedback para sa fully autonomous crawling at obstacle negotiation.
• Remote Control Mode: Sa napakalupit na environments, ang ground operators ay maaaring gawin ang joint-level fine manipulation o magbigay 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: Ginagamit ang split-core current transformer upang inductively harvest energy mula sa magnetic field sa paligid ng high-voltage conductor. Ang CT core ay gawa sa 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 covers a 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 safety, efficiency, at long service life.

2.4 Machine Vision Obstacle Recognition: Accurate Navigation

• Recognition Targets: Maaaring ma-identify ang 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 of the transmission line via histogram equalization and threshold segmentation.
• Feature Extraction: Extracts obstacle contours using morphological operations, analyzing left/right edge slopes as classification features.
• Recognition: Applies a fuzzy pattern recognition algorithm based on the maximum membership principle for fast and 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, ginagamit ang infrared sensor upang detect ang thermal radiation signals.
• Intelligent Diagnosis Model:
• Signal Processing: Uses the db4 wavelet base for 6-layer decomposition to filter out noise and focus on frequency bands containing fault features.
• Feature Extraction: Introduces wavelet energy entropy to characterize signal complexity, combined with peak-to-peak values of detail components, forming a 4-dimensional feature vector.
• Diagnosis Decision: Uses a 3-layer BP neural network for diagnosis. Experimental verification shows 100% accuracy on test samples and a 98% online detection success rate.

III. Solution Advantages Summary

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

IV. Current Limitations and Future Prospects

4.1 Current Limitations

• Still requires minimal manual assistance in extremely complex line environments.
• Potential for further optimization of mechanism size and obstacle-negotiation stroke for a more compact design.
• Power system starting current remains relatively high, limiting application on very low-load lines.
• Current fault detection types are mainly focused on broken strands; the range of detectable faults can be expanded.

4.2 Future Outlook

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