计算机与现代化 ›› 2025, Vol. 0 ›› Issue (07): 77-82.doi: 10.3969/j.issn.1006-2475.2025.07.011

• 人工智能 • 上一篇    下一篇

基于改进鲸鱼算法的Bi-LSTM电力负荷预测

  


  1. (1.贵州电网兴义供电局,贵州 兴义 562499; 2.贵州电网安顺供电局,贵州 安顺 561099; 3.贵州电网电力调度中心,贵州
    贵阳 550002; 4.南瑞继保电气有限公司,江苏 南京 211102; 5.南京理工大学,江苏 南京 210094)
  • 出版日期:2025-07-22 发布日期:2025-07-22
  • 作者简介:作者简介:饶弘宇(1997—),男,贵州兴义人,助理工程师,学士,研究方向:电力调度自动化,E-mail: 634199599@qq.com; 陈馨(1999—),女,贵州安顺人,助理工程师,学士,研究方向:电网调度自动化,E-mail: cx2198732483@163.com; 陈胜(1986—),男,贵州六盘水人,高级工程师,硕士,研究方向:电网调度自动化,E-mail: 568786829@qq.com; 夏天(1989—),男,贵州安顺人,工程师,硕士,研究方向:电力系统调度自动化,E-mail: gzxiatian@126.com; 沈力(1990—),男,江苏南京人,工程师,硕士,研究方向:电力系统自动发电控制,E-mail: shenl@nrec.com; 通信作者:吕广强(1974—),男,山东济南人,副教授,博士,研究方向:电力系统优化调度,E-mail: lgqiang1008@163.com。

Power Load Forecasting of Bi-LSTM Based on Improving Whale Algorithm


  1. (1. Guizhou Power Grid Xingyi Power Supply, Xingyi 562499, China; 2. Guizhou Power Grid Anshun Power Supply, Anshun 561099, China; 3. Guizhou Power Grid Dispatch Control Center, Guiyang 550002, China; 4. NR Electric Co., Ltd, Nanjing 211102, China; 5. Nanjing University of Science and Technology, Nanjing 210094, China)
  • Online:2025-07-22 Published:2025-07-22

摘要:
摘要:新能源并网后电力系统稳定运行的关键是电力负荷预测的准确性。为此本文提出一种基于双向长短期记忆神经网络结合注意力机制的电力负荷预测模型。针对神经网络中超参数难以选取最优的问题,采用鲸鱼算法对超参数进行优化设计。为解决鲸鱼优化算法全局搜索和局部开发分配不均的问题,提出一种用分段非线性收敛因子调节的方法;同时对其后期开发能力不足的问题,提出一种自适应权重和随机差分变异组合方法,以实现超参数的优化。通过仿真验证了本文所提方法在超参数设计上的有效性,并验证了基于改进鲸鱼算法的电力负荷预测的准确性和有效性。



关键词: 关键词:电力负荷预测, 双向长短期记忆神经网络, 超参数, 鲸鱼优化算法, 注意力机制

Abstract:
Abstract: The accuracy of the power load prediction is the key to ensure the stable operation of power system after new energy grid connection. A power load forecasting model based on Bi-LSTM neural network and attention mechanism is proposed. Aiming at the problem that it is difficult to select the optimal hyperparameters in neural network, the whale algorithm is used to optimize the hyperparameters. To solve the problem of uneven distribution of global search and local development of whale optimization algorithm, a method of adjusting the piecewise nonlinear convergence factor is proposed. To solve the insufficient late-stage development ability, a method of combining adaptive weights and random difference variation is proposed for the hyperparameters optimization. Simulation verifies the effectiveness of the proposed method in hyperparameter design, and verifies the accuracy and effectiveness of power load forecasting based on improved whale algorithm.

Key words: Key words: power load forecasting, bidirectional long-term and short-term memory neural network, hyperparameters, whale optimization algorithm, attention mechanism

中图分类号: