Scholarlink

<p>The current paradigm shift of sustainability in the power system is largely motivated by the rapid injection/assimilation of renewable energy sources (RES) into the modern smart power grids intensifying the need to provide innovative means of managing variability, greater grid endurance, and the energy efficiency. The traditional models that have been used in the power systems to operate such as Auto Regressive Integrated Moving Average (ARIMA) do not support the non-linear relationships of the inaccurate predictions and fault detection to a restricted degree, as found through the adjustment of the grid and the performance issues. To redeem such cases, this paper proposes a smart grid model based on intelligent machine learning and is composed of Bi-Directional Long Short-Term Memory (Bi-LSTM) networks to predict load and generation more accurately and context-sensitive, collaboratively plan energy with Gradient Boosting Machines (GBM), and predict faults in real-time. The whole system model will enable better decision-making at both the inter-temporal and operational scales resulting in the best integration of the RES and intelligent grid operation. The experimental results indicate that there is an improvement in four key measures of performance: energy efficiency increased by 23%; forecasting error decreased by 17% (MAPE); fault detection accuracy increased by 28%; and the Grid Reliability Index (GRI) had been increased by 15%; overall system efficiency recorded at approximately 93%. The new system simulation and validation are completed with references to MATLAB/Simulink and Python based grid labs environment under Grid LAB-D environment which proves its reliability, adaptation and applicability in reality in deploying a sustainable smart grid.</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

Smart Grid Innovations: Integrating Renewable Energy Sources for Sustainable Power Systems

  • 期刊论文

  • 发表日期：

    2026年2月25日

查看全文
  • Pushpendra Pandey,
  • Mithun G. Aush,
  • K. Chanthirasekaran,
  • Paul Rodrigues,
  • M. Elangovan,
  • D. Anand

摘要

The current paradigm shift of sustainability in the power system is largely motivated by the rapid injection/assimilation of renewable energy sources (RES) into the modern smart power grids intensifying the need to provide innovative means of managing variability, greater grid endurance, and the energy efficiency. The traditional models that have been used in the power systems to operate such as Auto Regressive Integrated Moving Average (ARIMA) do not support the non-linear relationships of the inaccurate predictions and fault detection to a restricted degree, as found through the adjustment of the grid and the performance issues. To redeem such cases, this paper proposes a smart grid model based on intelligent machine learning and is composed of Bi-Directional Long Short-Term Memory (Bi-LSTM) networks to predict load and generation more accurately and context-sensitive, collaboratively plan energy with Gradient Boosting Machines (GBM), and predict faults in real-time. The whole system model will enable better decision-making at both the inter-temporal and operational scales resulting in the best integration of the RES and intelligent grid operation. The experimental results indicate that there is an improvement in four key measures of performance: energy efficiency increased by 23%; forecasting error decreased by 17% (MAPE); fault detection accuracy increased by 28%; and the Grid Reliability Index (GRI) had been increased by 15%; overall system efficiency recorded at approximately 93%. The new system simulation and validation are completed with references to MATLAB/Simulink and Python based grid labs environment under Grid LAB-D environment which proves its reliability, adaptation and applicability in reality in deploying a sustainable smart grid.