摘要
目前风电场都是使用单一的功率爬坡预测模型,其泛化能力较差,预测精度低。通过分析支持向量机和极限学习机两种单一功率的爬坡预测模型,研究了这两种模型的权值选取方法,提出了一种组合功率爬坡预测模型。该模型使用基于粒子群算法的改进算法对上述两种单一模型的权值进行权重优化,形成了一种新的高精度的预测系统。然后对该系统进行建模仿真。仿真结果验证了该预测模型的有效性,其预测精度较之单一的预测模型有了很大的提高。
At present,wind farms use a single power ramp prediction model,which has poor generalization ability and low prediction accuracy.By analyzing the two single-power ramp-up prediction models of support vector machine and extreme learning machine,the weight selection methods of these two models are studied,and a combined power hill-climbing prediction model is proposed.The model uses the improved algorithm based on particle swarm optimization algorithm to optimize the weights of the above two models,and forms a new high-precision prediction system.Then the system is modeled and simulated.The simulation results verify the effectiveness of the prediction model,and its prediction accuracy is greatly improved compared with the single prediction model.
作者
李永馨
王鸿
王致杰
王海群
LI Yongxin;WANG Hong;WANG Zhijie;WANG Haiqun(Shanghai Dianji University,Shanghai 201306,China)
出处
《电力系统保护与控制》
EI
CSCD
北大核心
2019年第18期115-120,共6页
Power System Protection and Control
基金
电力传输与功率变换控制教育部重点实验室开放课题资助(2016AB14)
国家自然科学基金项目资助(51477099)~~
关键词
粒子群算法
风场功率爬坡
组合预测
建模仿真
particle swarm optimization
wind power uphill
combination forecast
modeling and simulation
作者简介
李永馨(1995—),女,硕士研究生,研究方向为分布式发电并网技术。E-mail:751090538@qq.com。
