BALANCING THE SURGES: HOW AI SOLVES THE RENEWABLE INTERMITTENCY PROBLEM FOR UZBEKISTAN'S GRID

This article examines the role of artificial intelligence (AI) in addressing the intermittency challenge of renewable energy sources (RES) within Uzbekistan’s national power system. As Uzbekistan advances toward its strategic target of achieving 25 GW of installed renewable energy capacity by 2030, the stochastic generation characteristics of solar photovoltaic and wind power systems pose significant risks to grid stability, frequency regulation, and dispatch efficiency. The study analyses AI-based forecasting systems, real-time load balancing algorithms, smart grid technologies, battery energy storage systems (BESS), and digital energy management infrastructure. A comparative assessment is conducted using international experiences from Germany, China, the UAE, South Korea, and the European Union. The findings demonstrate that Long Short-Term Memory (LSTM) neural networks, Deep Q-Learning algorithms, and AI-driven digital twin technologies can reduce renewable energy forecasting errors to below 5% under Uzbekistan’s climatic and operational conditions. The article further proposes practical recommendations, institutional reform measures, and policy directions aimed at accelerating the transition toward an AI-optimised and digitally managed energy system.本文探讨了人工智能（AI）在解决乌兹别克斯坦国家电力系统内可再生能源（Renewable Energy Sources，RES）间歇性难题中的作用。随着乌兹别克斯坦朝着2030年实现25吉瓦可再生能源装机容量的战略目标迈进，太阳能光伏与风电系统的随机发电特性将对电网稳定性、调频效率及调度效能构成显著风险。本研究分析了基于人工智能的预测系统、实时负荷平衡算法、智能电网技术、电池储能系统（Battery Energy Storage Systems，BESS）以及数字能源管理基础设施。本研究参考德国、中国、阿联酋、韩国及欧盟的国际经验开展对比评估。研究结果表明，在乌兹别克斯坦的气候与运行条件下，长短期记忆（Long Short-Term Memory，LSTM）神经网络、深度Q学习（Deep Q-Learning）算法以及人工智能驱动的数字孪生技术可将可再生能源预测误差降至5%以下。本文进一步提出了旨在加快向人工智能优化、数字化管理能源系统转型的务实建议、体制改革举措与政策导向。