为了快速寻找到全局最优解,提高加热炉能耗的预测精度,提出基于改进差分进化算法的步进式加热炉能耗预测方法.首先,采用小波包原理对步进式加热炉的历史运行能耗数据展开去噪处理,提高能耗数据精度;其次,采用时间卷积网络建立步进式加热炉能耗预测模型;最后,采用差分进化算法对预测模型进行求解,采用禁忌搜索算法对适应度值排名前10％的优良个体进行进一步的搜索,使差分进化算法的全局最优解寻求能力得到改进提升.实验结果表明,所提出预测方法去噪效果较高,预测结果与实际能耗结果基本一致,并且最长预测耗时不超过2 min.
Energy Consumption Prediction of Walking Beam Furnace Based on Improved Differential Evolution Algorithm
In order to quickly find the global optimal solution and improve the prediction accuracy of the energy consumption of the reheating furnace,a method for predicting the energy consumption of the walking beam reheating furnace based on the improved differential evolution al-gorithm is proposed.Firstly,the wavelet packet principle is used to de-noise the historical operation energy consumption data of the walking beam heating furnace to improve the accuracy of energy consumption data;Secondly,the time convolution network is used to establish the en-ergy consumption prediction model of the walking beam furnace;Finally,the differential evolution algorithm is used to solve the prediction model,and the tabu search algorithm is used to further search the top 10％of the excellent individuals with fitness values,so that the global optimization search ability of the differential evolution algorithm is improved.The experimental results show that the proposed prediction method has high denoising effect,the prediction results are basically consistent with the actual energy consumption results,and the longest prediction time is less than 2 minutes.
万方数据
维普
