...
  • 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>Neural computing & applications >A new adaptive fuzzy inference system for electricity consumption forecasting with hike in prices
【24h】

A new adaptive fuzzy inference system for electricity consumption forecasting with hike in prices

机译：电价上调的新型自适应模糊推理系统

获取原文
获取原文并翻译 | 示例
           
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Large increase or hike in energy prices has proven to impact electricity consumption in a way which cannot be drawn from historical data, especially when price elasticity of demand is not significant. This paper proposes an integrated adaptive fuzzy inference system (FIS) to estimate and forecast long-term electricity consumption when prices experience large increase. To this end, first a novel procedure for construction and adaptation of Takagi-Sugeno fuzzy inference system (TS-FIS) is suggested. Logarithmic linear regressions are estimated with historical data and used to construct an initial first-order TS-FIS. Then, in the adaptation phase, expert knowledge is used to define new fuzzy rules which form a new secondary FIS for electricity forecasting. To show the applicability and usefulness of the proposed model, it is applied for forecasting of annual electricity consumption in Iran where removing energy subsidies has resulted in a hike in electricity prices. Gross domestic product (GDP), population and electricity price are three inputs for the initial TS-FIS. A questionnaire survey was conducted to collect the expert estimation on possible change in electricity per capita, change in electricity intensity and the ratio of GDP elasticity to population elasticity when price hikes. Based on the information collected, a fuzzy rule base is formed and used to construct the secondary FIS which is used for electricity forecasting until 2016. Furthermore, the performance of the proposed model of this paper is compared with three other models namely ANFIS, ANN and one-stage regression in terms of their mean absolute percentage error. The comparison shows a superior performance for the proposed FIS model.
机译：事实证明，能源价格的大幅上涨或上涨会以无法从历史数据中得出的方式影响用电量，尤其是在需求价格弹性不大的情况下。本文提出了一种集成的自适应模糊推理系统（FIS）来估计和预测价格大幅上涨时的长期用电量。为此，首先提出了一种构建和修改高木-杉野模糊推理系统（TS-FIS）的新颖方法。用历史数据估计对数线性回归，并将其用于构建初始一阶TS-FIS。然后，在适应阶段，专家知识将用于定义新的模糊规则，从而形成用于电力预测的新的二次FIS。为了显示该模型的适用性和实用性，将其用于预测伊朗的年度电力消耗，因为取消能源补贴已导致电价上涨。国内生产总值（GDP），人口和电价是初始TS-FIS的三个输入。进行了问卷调查，以收集有关价格上涨时人均电力可能变化，电力强度变化以及GDP弹性与人口弹性之比的专家估计。根据收集到的信息，形成一个模糊规则库，并将其用于构建用于电力预测的二次FIS，直到2016年。此外，将本文提出的模型的性能与其他三个模型ANFIS，ANN和ANFIS进行比较。就其平均绝对百分比误差而言的一阶段回归。比较表明，所提出的FIS模型具有优越的性能。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 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