Utilities are spending millions to inspect and maintain field assets, which are lifeline to transmission and distribution. Despite that, vegetation encroachments in conductor corridors, wildfire, natural calamities like earthquake and deteriorated asset conditions are the main reasons behind faults and blackouts/brownouts. Proactive inspections of field assets, fast decision making, and repair / maintenance / replacements thereof regularly can prevent such incidents. Even regulations mandate utilities to do so, but right from the inspection to necessary repair works for the vast field asset base in a scheduled time is a herculean task for a utility. It involves spending of huge resources in terms of skilled hands and truck-rolls apart from Capex spend on assets.
While maximum repair and maintenance work for field assets are reactive and sometimes routine, some of the Utilities have adopted condition-based maintenance for the critical field assets like transformers. The core pillar of arriving at a condition-based maintenance of the field assets is the historical operational data from different assets - like loading pattern, threshold violations, and various other sensor data. But those do not give information about the condition of secondary assets that do not generate operational data; like towers, poles, cross-arms, insulators, pipelines, joints etc. Age-old foot patrol visual inspection to monitor the condition of such assets has already proven inadequate. Thus, some utilities have taken resort to costly helicopter inspections or more recently drone aided inspections which capture image/video/ LiDAR data of assets in the field. This visual data is then interpreted manually to identify anomalies and defects in the respective assets. Often due to obscured clarity, reinspection is recommended.
While capturing image data from the field and analyzing it at the backend by an array of skilled professionals has proven to be effective to some extent, utilities are facing multiple challenges in automating and scaling this in a synchronous standardized approach. Questions like “how many glass insulators were found defective in the last quarter” are still addressed manually and with human subjectivity of risk averseness that widely varies.
TCS UVision, the Machine Vision Platform for Utilities, addresses the utility need to automate, standardize, and obtain accurate anomaly detections in field asset inspections and vegetation management problems.
This platform gives an end-to-end value realization, stakeholder engagement and automated, scalable, machine first output to utilities in both asset inspection and vegetation encroachment processes. The platform (with seamless mobility and web-support) is capable of ingesting data from multiple sources (including satellite imagery), in a cloud-hosted (recommended) or on-premises environment and perform automated damage/anomaly detection of individual assets and vegetation corridors. The core AI-ML solution is augmented with a data driven prioritization process, protocol adherence, edge analytics, and field work process/annotation management in the upstream; and enterprise ecosystem integration with meaningful visualization, in-map view, analytics, and initiating workflow for repair/rectification process in the downstream.
A bouquet of prebuilt, containerized, fine-tuned, outcome-oriented AI-ML algorithms and features developed and patented by TCS Utilities Research & Innovation Labs automates the platform engine in the anomaly detection process. Built-in plug and play conflated data platform, UI/UX for end-users, the work process flow for stakeholders, and sub-second query responses come along with TCS UVision platform.
TCS is engaged in different Utilities for Distribution Asset damage identification & quantification, Transmission Tower asset damages and Vegetation Encroachment detection. We are even engaged with Gas & Water Utilities for their respective asset anomalies detection.
Apart from bringing in exponential savings (both Opex & Capex) and improving accuracy, this platform helps utilities to reach their sustainability goal in reducing carbon footprint to a considerable extent.
