• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>International Conference on Power Systems >Investigation of Influence of Derived Weather Variables using two different Approaches for Modeling Day-Ahead Hourly Electric Load Power Demand Forecasting Framework with Standard Long Short-Term Memory Networks
【24h】

Investigation of Influence of Derived Weather Variables using two different Approaches for Modeling Day-Ahead Hourly Electric Load Power Demand Forecasting Framework with Standard Long Short-Term Memory Networks

机译：使用两种不同方法来调查衍生天气变量的影响，采用标准长短期内存网络建模日期每小时电负载电力需求预测框架

获取原文
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Electric Load Power Demand Forecasting (ELPDF) is a key activity in power system operation and control networks. This helps to make consumer friendly and efficient distribution network. With the integration of renewable energy sources and adaptation of various demand side response strategies, a most accurate short-term prediction of electrical load power consumption is mandatory. The proposed day-ahead ELPDF models are created on a new parallel multiple inputs and output framework using two different approaches with chosen features as per empirical knowledge obtained through continuous research. The models are developed in standard LSTM (Long Short-Term Memory) network using deep learning techniques. The statistical errors used for evaluating the performance of ELPDF are presented herewith. Irrespective of variations in weather and consumer behavior, the models proposed here yields an average forecast error of 2.8% to 3.6% for day-ahead ELPDF. Also, these models are easier to implement and individually produces the very near forecast at an accuracy of 96.4%.
机译：电负载电源需求预测（ELPDF）是电力系统操作和控制网络的关键活动。这有助于使消费者友好和有效的分销网络。随着可再生能源的整合和各种需求侧反应策略的适应，强制性的电负载功耗的最准确的短期预测是强制性的。在通过连续研究获得的经验知识的每种经验知识，使用两种不同方法在新的并行多输入和输出框架上创建了所提出的一天的ELPDF模型。该模型采用深层学习技术在标准LSTM（长短期存储器）网络中开发。这里介绍了用于评估ELPDF性能的统计误差。无论天气和消费者行为的变化如何，这里提出的模型都会产生2.8％的平均预测误差为2.8％至3.6％，以满足前一天的ELPDF。此外，这些模型更容易实现和单独生产非常接近的预测，精度为96.4％。

著录项

相似文献

  • 外文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号