AI-Enabled Digital Twins and Power System Asset Management for Long-Term Grid Investment Planning

<title>Abstract</title> <p>This paper presents a research-oriented synthesis of the study on OptimTwin, an integrated framework that combines artificial-intelligence-driven digital twins and power system asset management to support long-term grid investment planning. The reviewed work addresses a critical challenge facing modern power systems: the need to manage aging infrastructure, increasing renewable penetration, operational uncertainty, and investment constraints through methods that are predictive, data-driven, and economically defensible. Drawing on related literature in smart-grid digital twins, predictive maintenance, IoT-enabled monitoring, neural-network-based decision support, and power asset management, this synthesis positions OptimTwin within the broader transition from reactive maintenance toward continuously synchronized cyber-physical planning. The study uses IRENA FlexTool, a deep feed-forward neural network surrogate, and an extended IEEE/NREL 118-bus benchmark system to compare base, investment, and OptimTwin planning scenarios. Reported findings indicate that lifecycle-cost-aware planning can support high variable renewable energy penetration while improving investment indicators. This document restates the source study in original language, organizes key quantitative information into tables and figures, and evaluates the framework from the perspective of research engineering, with emphasis on reliability, lifecycle economics, methodological transparency, and utility-scale deployment readiness.</p>