Smartphone-Based, Field-Deployed AI Inspection System for Power Distribution Infrastructure

Sampath Balaji Masabathula; Srija Vanapalli; Gangadhar Kanchi; Sreenivasa Rao V. A. N; Vijaya Lalitha; Sasank Chilamkurthy

cs.CVOct 13, 2026

Smartphone-Based, Field-Deployed AI Inspection System for Power Distribution Infrastructure

Authors: Sampath Balaji Masabathula, Srija Vanapalli, Gangadhar Kanchi, Sreenivasa Rao V. A. N, Vijaya Lalitha, Sasank Chilamkurthy

Organizations: Eastern Power Distribution Company of Andhra Pradesh Limited · Andhra University College of Engineering, Visakhapatnam, India · Von Neumann AI

Abstract

Routine inspection of power distribution infrastructure remains challenging for utilities operating geographically dispersed networks, where conventional manual inspection can be difficult to perform consistently and at scale. This work presents a smartphone-based AI inspection system for assessing targeted distribution-infrastructure conditions from images captured by field personnel. The proposed workflow integrates image-quality and viewing-perspective validation with AI-assisted assessment of pole lean, crossarm and top-cleat tilt, voltage level, vegetation-clearance risk, and distribution-transformer (DTR)-ground clearance. The system integrates deep-learning models with monocular depth estimation, geometric analysis, and rule-based processing within a common multi-stage inspection workflow designed for practical field operation. The system was pilot deployed across five selected distribution feeders within the Eastern Power Distribution Company of Andhra Pradesh Ltd. (APEPDCL), where 1,222 defect instances were identified across the assessed categories and subsequently tracked through a rectification and verification workflow, with 84 defect instances recorded as rectified during the study period. Feeder-level interruption records were also examined over the study period; however, the available observations do not provide sufficient evidence to establish a reliability improvement attributable to the proposed system. A curated multi-module field-captured dataset, PD-Defect, is also publicly released to support reproducible research in distribution-infrastructure inspection. The field deployment demonstrates the practical integration of smartphone-based AI-assisted inspection into field inspection and maintenance activities, providing a mechanism for systematic identification, documentation, rectification tracking, and verification of targeted distribution-infrastructure defects.

Figures & tables

Fig. 1 : Overall workflow of the proposed smartphone-based AI-assisted inspection system. Smartphone-captured field images undergo input-quality assessment and are processed through the corresponding pole or DTR–ground clearance assessment pathways. Task-specific validation and assessment are then performed, and the resulting inspection outputs are recorded for rectification tracking and post-rectification verification.

Class	Split	Acc	Prec	Rec	F1
Pole Lean Assessment
Leaned_Pole	Test	89.94	88.51	90.59	89.53
	Val	93.92	93.70	97.54	95.58
Straight_Pole	Test	91.06	90.15	97.54	93.70
	Val	95.58	93.42	95.95	94.67
Crossarm and Top-Cleat Tilt Assessment

TABLE I : Image-Level Classif. Metrics on Val. & Test Sets

Class	Split	Acc	Prec	Rec	F1
Full-Image
Valid	Test	91.67	88.32	91.67	89.96
	Val	89.47	88.15	89.47	88.81
perspective-distorted	Test	89.33	92.41	89.33	90.85
	Val	89.26	90.48	89.26	89.86
Pole-Top Crop

TABLE II : Perspective-Validation Metrics on Val. & Test Sets

Class	Split	Prec	Rec	[email protected]	Angular MAE ( ∘ )
Pole (Lean)
Pole	Test	96.86	93.40	97.75	0.29
	Val	96.47	90.33	98.19	0.60
Pole-top Component
Pole	Test	99.24	98.10	99.39	0.37
	Val	96.85	94.49	98.09	0.38

TABLE III : Pose Estimation and Derived Angular Metrics

Fig. 2 : Representative outputs showing (a) detected crossarm, V-crossarm, and top-cleat components with tilt angles of $2.86^{\circ}$ , $7.26^{\circ}$ , and $10.52^{\circ}$ , respectively, estimated from axes defined by their detected keypoints. Although tilt is detected for the V-crossarm and top-cleat, their angles remain below the $15^{\circ}$ threshold and are therefore classified as non-defective; and (b) detected high-tension (HT) and low-tension (LT) structures for voltage-level classification.

Fig. 3 : Representative pole-lean assessment showing a detected pole with an apparent inclination of $17.82^{\circ}$ , estimated from the pole axis defined by its detected keypoints, resulting in a leaned classification.

Feeder	Surveyed	Defective	Pole Lean	Crossarm	Top-Cleat	DTR–Ground Clearance	Vegetation Clearance	Total
11KV BONDADA	557	157	16	12	11	0	128	167
11KV PERUMALI	807	214	62	25	12	1	125	225
11KV YERRAKONERU	853	333	43	19	21	4	272	359
11KV BADIVANIPETA	836	317	19	19	9	4	287	338
11KV A.RAVIVALASA	757	120	17	18	7	3	88	133
Total	3,810	1,141	157	93	60	12	900	1,222

TABLE IV : Defect Identification Across the Five Selected Feeders. The Defective column denotes distinct poles associated with identified defects; Crossarm and Top-Cleat denote tilt-related defect instances.

Feeder	Apr.	May	Jun.	Jul.	Aug.	Sep.
11KV A.RAVIVALASA	59	79	91	65	42	86
11KV BADIVANIPETA	54	40	52	29	30	63
11KV BONDADA	32	68	36	53	17	42
11KV PERUMALI	34	62	79	50	51	78
11KV YERRAKONERU	48	78	81	76	40	65
Total	227	327	339	273	180	334

TABLE V : Feeder-Wise Monthly Interruption Records Across the Study Period

Industrial Inspection

Industrial inspection uses sensing, computer vision, and robotics to detect, measure, and explain defects and verify product or infrastructure quality.

See all 120 papers

Jun 11, 2026

Multi-Modal Agents for Power Distribution Defect Detection: An Evaluation of Foundation Models

The power distribution network is critical to reliable electricity delivery, yet traditional inspection methods face limitations in semantic understanding, generalization, and closed-loop automation. To address these challenges, this paper proposes a Multi-Modal Agent framework specifically for power distribution defect detection. Central to this study is the systematic evaluation of multimodal foundation models as unified cognitive engines. We rigorously assess their integrated performance across three critical capabilities: (1) Perception, where the model must accurately identify equipment and generate expert-level descriptions of defects; (2) Reasoning, where the model interprets visual findings to diagnose causes, assess severity, and plan maintenance strategies based on domain knowledge; and (3) Tool Usage, where the model acts as an autonomous operator to execute actions -- such as querying knowledge bases or generating work orders -- to achieve closed-loop maintenance. To support this evaluation, a domain-specific evaluation dataset and a comprehensive benchmark are developed. Experimental results demonstrate the strengths and limitations of current foundation models in these three dimensions, providing empirical evidence for deploying autonomous agents in high-stakes industrial environments.

Quan Quan

Industrial Anomaly Detection Multimodal Reasoning AI Agent Benchmarks Multimodal Agents Multimodal Model Evaluation Power Systems

Ask PDF

Feb 27, 2026

Autonomous Inspection of Power Line Insulators with UAV on an Unmapped Transmission Tower

This paper introduces an online inspection algorithm that enables an autonomous UAV to fly around a transmission tower and obtain detailed inspection images without a prior map of the tower. Our algorithm relies on camera-LiDAR sensor fusion for online detection and localization of insulators. In particular, the algorithm is based on insulator detection using a convolutional neural network, projection of LiDAR points onto the image, and filtering them using the bounding boxes. The detection pipeline is coupled with several proposed insulator localization methods based on DBSCAN, RANSAC, and PCA algorithms. The performance of the proposed online inspection algorithm and camera-LiDAR sensor fusion pipeline is demonstrated through simulation and real-world flights. In simulation, we showed that our single-flight inspection strategy can save up to 24 % of total inspection time, compared to the two-flight strategy of scanning the tower and afterwards visiting the inspection waypoints in the optimal way. In a real-world experiment, the best performing proposed method achieves a mean horizontal and vertical localization error for the insulator of 0.16 +- 0.08 m and 0.16 +- 0.11 m, respectively. Compared to the most relevant approach, the proposed method achieves more than an order of magnitude lower variance in horizontal insulator localization error.

Václav Riss, Vít Krátký, Robert Pěnička +1

Robotic Inspection Industrial Inspection UAV Navigation Camera-LiDAR Fusion

Ask PDF

Mar 24, 2026

From Prediction to Diagnosis: Reasoning-Aware AI for Photovoltaic Defect Inspection

Reliable photovoltaic defect identification is essential for maintaining energy yield, ensuring warranty compliance, and enabling scalable inspection of rapidly expanding solar fleets. Although recent advances in computer vision have improved automated defect detection, most existing systems operate as opaque classifiers that provide limited diagnostic insight for high-stakes energy infrastructure. Here we introduce REVL-PV, a vision-language framework that embeds domain-specific diagnostic reasoning into multimodal learning across electroluminescence, thermal, and visible-light imagery. By requiring the model to link visual evidence to plausible defect mechanisms before classification, the framework produces structured diagnostic reports aligned with professional photovoltaic inspection practice. Evaluated on 1,927 real-world modules spanning eight defect categories, REVL-PV achieves 93% classification accuracy while producing interpretable diagnostic rationales and maintaining strong robustness under realistic image corruptions. A blind concordance study with a certified solar inspection expert shows strong semantic alignment between model explanations and expert assessments across defect identification, root-cause attribution, and visual descriptions. These results demonstrate that reasoning-aware multimodal learning establishes a general paradigm for trustworthy AI-assisted inspection of photovoltaic energy infrastructure.

Dev Mistry, Feng Qiu, Bo Chen +4

Department of Electrical and Computer Engineering, Illinois Institute of Technology, Chicago, IL 60616, USA · Argonne National Laboratory, Lemont, IL 60439, USA · Smart Grid Emerging Technology Department, Commonwealth Edison, Chicago, IL 60618, USA +2

Fault Detection Industrial Inspection Multimodal Reasoning Multimodal Model Interpretability

Ask PDF

Abstract

Industrial Inspection

Explore similar work

Multi-Modal Agents for Power Distribution Defect Detection: An Evaluation of Foundation Models

Autonomous Inspection of Power Line Insulators with UAV on an Unmapped Transmission Tower

From Prediction to Diagnosis: Reasoning-Aware AI for Photovoltaic Defect Inspection