摘要
电-气综合能源系统中多能负荷之间的耦合程度不断增加,提升了能源系统调度和运行的难度。为此,对零碳排放下电-气综合能源系统多能负荷预测方法进行了研究。分析零碳排放下电-气综合能源系统的运行架构。以气象因素为影响因子,运用灰色关联度分析法获得多能负荷与各因子的相关性。将相关性分析结果与系统历史多能负荷数据共同作为输入数据,构建基础长短期记忆(LSTM)预测模型。结合樽海鞘群算法(SSA)优化模型关键参数,获得优化LSTM预测模型,实现系统多能负荷预测。试验结果表明:冷负荷与电负荷的关联度为0.88;热负荷与电负荷的关联度为0.681;实际预测平均绝对百分误差低于0.45。该方法预测效果理想,为系统最优调度与运行规划奠定了基础。
The increasing degree of coupling between multi-energy loads in the integrated electricity-gas energy system increases the difficulty of energy system scheduling and operation.For this reason,the multi-energy load forecasting method for integrated electricity-gas energy system with zero carbon emission is investigated.The operational architecture of the integrated electricitygas energy system with zero carbon emission is analyzed.The correlation between the multi-energy load and each factor is obtained by using gray correlation analysis with meteorological factors as the influencing factors.The results of the correlation analysis and the historical multi-energy load data of the system are jointly used as input data to construct the basic long short-term memory(LSTM)prediction model.The key parameters of the model are optimized by combining with salp swarm algorithm(SSA)to obtain the optimized LSTM prediction model and realize the system multi-energy load prediction.The experimental results show that the correlation between cold load and electric load is O.88;the correlation between heat load and electric load is 0.681;the average absolute percentage error of the actual prediction is lower than 0.45.This method has good predictron effect,and lays the foundation for the optimal scheduling and operation planning of the system.
作者
舒舟
欧莉玲
何丰
田诗语
SHU Zhou;OU Liling;HE Feng;TIAN Shiyu(Shenzhen Power Supply Bureau,Shenzhen 518000,China;Shenzhen Electric Power Development and Design Institute,Shenzhen 518000,China)
出处
《自动化仪表》
CAS
2024年第2期116-121,共6页
Process Automation Instrumentation
关键词
电-气综合能源系统
零碳排放
相关性分析
多能负荷预测
长短期记忆预测模型
灰色关联度
樽海鞘群算法
气象因素
Integrated electricity-gas energy system
Zero carbon emission
Correlation analysis
Multi-energy load forecasting
Long short-term memory(LSTM)forecasting model
Gray correlation
Salp swarm algorithm(SSA)
Meteorological factors
作者简介
舒舟(1989-),男,硕士,工程师,主要研究方向为电气工程,E-mail:shuzhou189@163.com。
