首页> 外文会议>2011 3rd International Workshop on Intelligent Systems and Applications >Analysis of Energy Consumption Prediction Model Based on Genetic Algorithm and Wavelet Neural Network
【24h】

Analysis of Energy Consumption Prediction Model Based on Genetic Algorithm and Wavelet Neural Network

机译:基于遗传算法和小波神经网络的能耗预测模型分析

获取原文

摘要

This paper analyzed the enterprise process energy consumption systematically with a lot of statistic data starting from energy efficiency, and established the energy consumption prediction model based on genetic algorithm of wavelet neural network (GA-WNN). This paper made previous optimization training with genetic algorithm, which have feature of natural evolution regularity, to the weights and dilation-shift scale of wavelet neural network. Partly replaced gradient descent method of wavelet frame neural network where parameters optimization only with a single gradient direction, overcame the shortcoming that easily into the local minimum and cause oscillation effect of the single gradient descent method. Simulation results showed the effectiveness of the forecasting model, and it is feasible for solving the process energy consumption multi-factor quantitative analysis problem which general mathematical model is difficult to describe.
机译:本文系统地分析了企业过程能源消耗,从能效开始,基于小波神经网络(GA-Wnn)的遗传算法建立了能耗预测模型。本文以遗传算法制作了以前的优化训练,具有自然演化规律的特征,对小波神经网络的重量和扩张换向比例。部分替代小波帧神经网络的梯度下降方法,其中仅采用单个梯度方向的参数优化,克服容易进入局部最小的缺点并导致单梯度下​​降方法的振荡效果。仿真结果表明预测模型的有效性,解决过程能耗多因素定量分析问题是可行的,这是难以描述的一般数学模型。

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