基于PSO与ELM组合算法的短期光伏发电功率预测模型
|
摘要
光伏发电功率受自然环境因素影响较大,具有很强的随机性和波动性,准确及时的光伏发电功率预测对电网的调度运行具有重要的意义.提出了一种基于PSO与ELM组合算法的短期光伏发电功率预测模型.该模型通过调整粒子群优化算法(particle swarm optimization,PSO)不同阶段的寻优重点,为极限学习机(extreme learning machine,ELM)设定出了最佳网络参数,避免了ELM随机产生输入层权值和隐含层阈值造成的网络不稳定问题.同时结合传统神经网络和ELM网络隐含层节点选取原则为组合模型,设定了最佳隐含层节点数,提高了模型预测精度.实际算例验证了组合算法模型能够有效提高短期光伏发电功率预测的预测精度.
Influenced by the natural environment factors,photovoltaic power had strong randomness and volatility,therefore an accurate and timely forecast to photovoltaic power was of significances to the power grid dispatching operation. Based on PSO and ELM combined algorithm,a short-term photovoltaic power prediction model was proposed. This model sought advantages and key points at different stages and set the optimizing algorithm( PSO). It also could avoid the problem of network instability caused by the random generation of input layer weight and hidden layer threshold by ELM. In the meantime,combining the most of traditional neural network and ELM network,the optimal hidden layer number was set for the combined model to improve the model prediction accuracy. Through the practical example,it demonstrated that the combined algorithm model could effectively improve the prediction accuracy of short-term photovoltaic power generation prediction.
Influenced by the natural environment factors,photovoltaic power had strong randomness and volatility,therefore an accurate and timely forecast to photovoltaic power was of significances to the power grid dispatching operation. Based on PSO and ELM combined algorithm,a short-term photovoltaic power prediction model was proposed. This model sought advantages and key points at different stages and set the optimizing algorithm( PSO). It also could avoid the problem of network instability caused by the random generation of input layer weight and hidden layer threshold by ELM. In the meantime,combining the most of traditional neural network and ELM network,the optimal hidden layer number was set for the combined model to improve the model prediction accuracy. Through the practical example,it demonstrated that the combined algorithm model could effectively improve the prediction accuracy of short-term photovoltaic power generation prediction.
引文
[1]郭安琪.含大规模间歇性能源的电力系统优化调度模型研究[D].济南:山东大学,2018.
[2]黎嘉明.应对大规模新能源并网的电力系统发电调度控制研究[D].武汉:华中科技大学,2017.
[3]王运.新能源发电优先调度评价技术的研究与应用[D].北京:华北电力大学,2016.
[4]于群,朴在林,胡博.基于EEMD和BP神经网络的短期光伏功率预测模型[J].电网与清洁能源,2016,32(7):132-137.
[5]张立影,孟令甲,王泽忠.基于双层BP神经网络的光伏电站输出功率预测[J].电测与仪表,2015,52(11):31-35.
[6]刘沛汉,袁铁江,梅生伟,等.基于遗传算法优化神经网络的光伏电站短期功率预测[J].水电能源科学,2016,34(1):211-214.
[7]李元诚,白恺,曲洪达,等.基于粒子群-稀疏贝叶斯混合算法的光伏功率预测方法[J].太阳能学报,2016,37(5):1153-1159.
[8]林亚芹,程轶平,安景.基于PSO混合核函数SVM对光伏最大功率点的预测研究[J].电气自动化,2014,36(5):41-43.
[9]殷豪,董朕,孟安波.基于结合混沌纵横交叉的粒子群算法优化极限学习机的短期负荷预测[J].计算机应用研究,2018,35(7):2088-2091.
[10]王文锦,戚佳金,王文婷,等.基于人工蜂群优化极限学习机的短期负荷预测[J].电测与仪表,2017,54(11):32-35.
[11]梁吉业,冯晨娇,宋鹏.大数据相关分析综述[J].计算机学报,2016,39(1):1-18.
[12]孙鑫.极限学习机结构优化及其应用研究[D].南宁:广西大学,2014.
[13] FENG G R,HUANG G B,LIN Q P,et al. Error minimized extreme learning machine with growth of hidden nodes and incremental learning[J]. IEEE transactions on neural networks,2009,20(8):1352-1357.
[14] HUANG G B,CHEN L,SIEW C K. Universal approximation using incremental constructive feedforward networks with random hidden nodes[J]. IEEE transactions on neural networks,2006,17(4):879-892.
[15] ZHU Q Y,QIN A K,SUGANTHAN P N,et al. Evolutionary extreme learning machine[J]. Pattern recognition,2005,38(10):1759-1763.
[16] HUANG G B,ZHU Q Y,SIEW C K. Extreme learning machine:theory and applications[J]. Neurocomputing,2006,70(1/2/3):489-501.
[17] HUANG G B,ZHU Q Y,SIEW C K. Extreme learning machine:a new learning scheme of feedforward neural networks[C]∥IEEE International Joint Conference on Neural Networks. Budapest,2004:985-990.
[18] HUANG G B,LIANG N Y,RONG H J,et al. On-line sequential extreme learning machine[C]∥IASTED International Conference on Computational Intelligence. Calgary,Alberta,2005:232-237.
[19]王小川,史峰,郁磊. MATLAB神经网络43个案例分析[M].北京:北京航空航天大学出版社,2013.
[20]蒋林利.改进的PSO算法优化神经网络模型及其应用研究[D].厦门:厦门大学,2014.
[21]温阳东,李龙剑.基于LDIW-PSO算法的BP神经网络在压力传感器中的应用[J].化工自动化及仪表,2014,41(9):1031-1034.
[22]陈前宇,陈维荣,戴朝华.电力系统无功优化多目标处理与算法改进[J].电力系统保护与控制,2014,42(5):129-135.
[23]毛恒.粒子群优化算法的改进及应用研究[D].泉州:华侨大学,2008.
[2]黎嘉明.应对大规模新能源并网的电力系统发电调度控制研究[D].武汉:华中科技大学,2017.
[3]王运.新能源发电优先调度评价技术的研究与应用[D].北京:华北电力大学,2016.
[4]于群,朴在林,胡博.基于EEMD和BP神经网络的短期光伏功率预测模型[J].电网与清洁能源,2016,32(7):132-137.
[5]张立影,孟令甲,王泽忠.基于双层BP神经网络的光伏电站输出功率预测[J].电测与仪表,2015,52(11):31-35.
[6]刘沛汉,袁铁江,梅生伟,等.基于遗传算法优化神经网络的光伏电站短期功率预测[J].水电能源科学,2016,34(1):211-214.
[7]李元诚,白恺,曲洪达,等.基于粒子群-稀疏贝叶斯混合算法的光伏功率预测方法[J].太阳能学报,2016,37(5):1153-1159.
[8]林亚芹,程轶平,安景.基于PSO混合核函数SVM对光伏最大功率点的预测研究[J].电气自动化,2014,36(5):41-43.
[9]殷豪,董朕,孟安波.基于结合混沌纵横交叉的粒子群算法优化极限学习机的短期负荷预测[J].计算机应用研究,2018,35(7):2088-2091.
[10]王文锦,戚佳金,王文婷,等.基于人工蜂群优化极限学习机的短期负荷预测[J].电测与仪表,2017,54(11):32-35.
[11]梁吉业,冯晨娇,宋鹏.大数据相关分析综述[J].计算机学报,2016,39(1):1-18.
[12]孙鑫.极限学习机结构优化及其应用研究[D].南宁:广西大学,2014.
[13] FENG G R,HUANG G B,LIN Q P,et al. Error minimized extreme learning machine with growth of hidden nodes and incremental learning[J]. IEEE transactions on neural networks,2009,20(8):1352-1357.
[14] HUANG G B,CHEN L,SIEW C K. Universal approximation using incremental constructive feedforward networks with random hidden nodes[J]. IEEE transactions on neural networks,2006,17(4):879-892.
[15] ZHU Q Y,QIN A K,SUGANTHAN P N,et al. Evolutionary extreme learning machine[J]. Pattern recognition,2005,38(10):1759-1763.
[16] HUANG G B,ZHU Q Y,SIEW C K. Extreme learning machine:theory and applications[J]. Neurocomputing,2006,70(1/2/3):489-501.
[17] HUANG G B,ZHU Q Y,SIEW C K. Extreme learning machine:a new learning scheme of feedforward neural networks[C]∥IEEE International Joint Conference on Neural Networks. Budapest,2004:985-990.
[18] HUANG G B,LIANG N Y,RONG H J,et al. On-line sequential extreme learning machine[C]∥IASTED International Conference on Computational Intelligence. Calgary,Alberta,2005:232-237.
[19]王小川,史峰,郁磊. MATLAB神经网络43个案例分析[M].北京:北京航空航天大学出版社,2013.
[20]蒋林利.改进的PSO算法优化神经网络模型及其应用研究[D].厦门:厦门大学,2014.
[21]温阳东,李龙剑.基于LDIW-PSO算法的BP神经网络在压力传感器中的应用[J].化工自动化及仪表,2014,41(9):1031-1034.
[22]陈前宇,陈维荣,戴朝华.电力系统无功优化多目标处理与算法改进[J].电力系统保护与控制,2014,42(5):129-135.
[23]毛恒.粒子群优化算法的改进及应用研究[D].泉州:华侨大学,2008.