基于E-Seq2Seq技术的数据驱动型机组组合智能决策方法

Data-driven Intelligent Decision-making Method for Unit Commitment Based on E-Seq2Seq Technology

在能源技术变革日新月异、人工智能技术与电力系统深度融合的背景下,研究具有高适应性、高精度的机组组合智能决策方法具有重要意义。该文结合门限循环神经网络(gated recurrent unit,GRU)提出一种基于E-Seq2Seq(expand sequence to sequence,E-Seq2Seq)技术的数据驱动型机组组合智能决策方法。首先研究并梳理机组组合模型输入输出序列的类型与结构,形成机组组合弹性多序列映射型样本;然后研究提出一种适用于弹性多序列映射型样本的E-Seq2Seq技术;在此基础上,以GRU为神经元构建机组组合深度学习模型,并最终提出一种基于E-Seq2Seq技术的数据驱动型机组组合智能决策方法。基于IEEE118节点系统、Python环境的算例验证该文方法的正确性和有效性。

It is now a background of the energy technologies change rapidly and the integration of artificial intelligence technology and power system. It is significant to study the decision-making method of unit commitment with high adaptability and high precision. In this study, a data-driven intelligent decision-making method for unit commitment based on Expand Sequence to Sequence(E-Seq2 Seq) technology in combination with gated recurrent unit(GRU) neurons was proposed. Firstly, the type and structure of input and output sequence of unit commitment model were studied and analyzed, and the flexible multi-sequence mapping sample of unit commitment was formed. Then an E-Seq2 Seq technique for flexible multi-sequence mapping samples was proposed. On this basis, GRU was used as a neuron to construct a deep learning model of unit commitment. Finally, a data-driven intelligent decision-making method for unit commitment based on E-Seq2 Seq technology was proposed. Examples based on IEEE118-bus system and Python environment verifies the correctness and effectiveness of the proposed method.