Scholarlink

It is crucial to forecast accurately short-term power load for ensuring the safe and stable operation of a power system. In this paper a method for short-term power load forecasting based on an algorithm of Improved Parrot Optimization to optimize a CNN-BiGRU model is proposed. Firstly, the Spearman correlation coefficient method is adopted to sift out the highly correlated features with power load as inputs, and the CNN-BiGRU model is constructed. Secondly, the shortcomings of the traditional Parrot Optimizer algorithm are improved, and the Improved Parrot Optimization is utilized to find the optimal solutions for the key hyperparameters of the overall model. Finally, comparative experiments are conducted using load data from Australia. The results indicate that the IPO-CNN-BiGRU model has exhibited higher predictive performance than other models.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Research on Short-Term Power Load Forecasting Based on IPO-Optimized CNN-BiGRU

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Qun Ge,
  • Yalei Li,
  • Zhenfeng Wang

摘要

It is crucial to forecast accurately short-term power load for ensuring the safe and stable operation of a power system. In this paper a method for short-term power load forecasting based on an algorithm of Improved Parrot Optimization to optimize a CNN-BiGRU model is proposed. Firstly, the Spearman correlation coefficient method is adopted to sift out the highly correlated features with power load as inputs, and the CNN-BiGRU model is constructed. Secondly, the shortcomings of the traditional Parrot Optimizer algorithm are improved, and the Improved Parrot Optimization is utilized to find the optimal solutions for the key hyperparameters of the overall model. Finally, comparative experiments are conducted using load data from Australia. The results indicate that the IPO-CNN-BiGRU model has exhibited higher predictive performance than other models.