基于二层分解技术的短期负荷预测研究

钢铁用户的增多会使地区含有大量的冲击负荷,传统的预测方法难以捕捉该地区的负荷变化规律,预测精度不足。为提高含大量负荷地区的负荷预测的精度和泛化性,提出一种基于可变模式分解与奇异谱分析相结合的二层分解技术(VMD-SSA)和改进鲸鱼算法(IWOA)优化极限学习机(ELM)的短期负荷预测模型。通过实例证明,相比于其它模型,所提混合模型能充分掌握负荷的变化规律,有效提高了含大量负荷地区的负荷预测的精度和泛化能力。

Short-term load forecasting using extreme learning machine based on two-layer decomposition technique

The increase of iron and steel users will lead to a large amount of load in an area. The traditional forecasting method is difficult to capture the law of load variation in this area, and the prediction accuracy is insufficient. In order to improve the accuracy and generalization of power system load forecasting in an area with a large amount of impact load, a new method for short-term load forecasting using extreme learning machine (ELM) based on a two-layer decomposition technique (VMD-SSA) and improved whale optimization algorithm (IWOA) is proposed. Examples show that compared with other models, the proposed hybrid model can grasp the change regulation of load, which is superior to other models in forecasting ability of accuracy and generalization in areas with large loads.