Scholarlink

<p>The rising integration of renewable energy sources and dynamic load profiles has introduced significant uncertainty into modern power systems, challenging their reliability and operational resilience. This study presents a comprehensive framework for reliability enhancement and load shedding mitigation through degradation-aware optimal resource utilization and demand-side load shifting. To address renewable intermittency, Long Short-Term Memory (LSTM) networks are employed to forecast solar PV and wind generation by capturing spatio-temporal uncertainties, ensuring more accurate representation of renewable availability in the optimization model. The proposed methodology jointly optimizes generation scheduling, battery energy storage system (BESS) dispatch, and demand response (DR) participation while explicitly considering battery degradation cost and cycle-life limitations. A stochastic optimization model is developed to reduce the overall operational expenditure, including degradation and reliability penalties, subject to system constraints. The optimized scheduling is evaluated using GAMS (General Algebraic Modeling System), and sequential Monte Carlo simulation (MCS) is applied to estimate key reliability indices such as Expected Energy Not Supplied (EENS). The framework is implemented on the IEEE 24-Bus Reliability Test System (RTS) under varying load and renewable uncertainty scenarios. The simulation outcomes reveal that the integrated operation of BESS and DR significantly reduces involuntary load shedding and improves reliability indices compared to conventional scheduling approaches. The EENS value decreases by 64.39%, 52.62%, and 36.19% for cases 3, 2, and 1, respectively, as the number of line outages increases. The incorporation of degradation-aware modelling ensures sustainable battery operation without compromising system performance. Overall, the proposed approach provides a robust and economically viable solution for reliability-oriented operational planning in renewable-rich power systems.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Stochastic Reliability Enhancement of Renewable-Rich Power System via LSTM Forecasting and Degradation-Aware Resource Optimization

  • 期刊论文

  • 发表日期：

    2026年1月27日

查看全文
  • Smriti Singh,
  • Sagar Bhati,
  • R. K. Saket

摘要

The rising integration of renewable energy sources and dynamic load profiles has introduced significant uncertainty into modern power systems, challenging their reliability and operational resilience. This study presents a comprehensive framework for reliability enhancement and load shedding mitigation through degradation-aware optimal resource utilization and demand-side load shifting. To address renewable intermittency, Long Short-Term Memory (LSTM) networks are employed to forecast solar PV and wind generation by capturing spatio-temporal uncertainties, ensuring more accurate representation of renewable availability in the optimization model. The proposed methodology jointly optimizes generation scheduling, battery energy storage system (BESS) dispatch, and demand response (DR) participation while explicitly considering battery degradation cost and cycle-life limitations. A stochastic optimization model is developed to reduce the overall operational expenditure, including degradation and reliability penalties, subject to system constraints. The optimized scheduling is evaluated using GAMS (General Algebraic Modeling System), and sequential Monte Carlo simulation (MCS) is applied to estimate key reliability indices such as Expected Energy Not Supplied (EENS). The framework is implemented on the IEEE 24-Bus Reliability Test System (RTS) under varying load and renewable uncertainty scenarios. The simulation outcomes reveal that the integrated operation of BESS and DR significantly reduces involuntary load shedding and improves reliability indices compared to conventional scheduling approaches. The EENS value decreases by 64.39%, 52.62%, and 36.19% for cases 3, 2, and 1, respectively, as the number of line outages increases. The incorporation of degradation-aware modelling ensures sustainable battery operation without compromising system performance. Overall, the proposed approach provides a robust and economically viable solution for reliability-oriented operational planning in renewable-rich power systems.