卡尔曼小波神经网络风速预测

发布时间：2018-01-10 21:31

本文关键词：卡尔曼小波神经网络风速预测　出处：《电力系统及其自动化学报》2015年12期 　论文类型：期刊论文

【摘要】：风速变化的随机性使得风电并网成为当今制约风电发展的瓶颈问题。如能预测风速,并提高风速预测精度,能够有助于调度部门对风电场积极进行规划和调度,减轻风电并网对电力系统产生的不利影响。用卡尔曼滤波算法建立数据滤波模型,对原始风速数据进行一级处理,去除测量误差和系统误差;再用改进的BP小波神经网络建立风速预测仿真模型;利用卡尔曼滤波后的风速数据进行风速预测,预测结果与BP神经网络风速预测方法的预测结果对比。对比结果表明该算法预测精度高,说明该算法在处理非平稳随机数据方面具有较好的应用前景。
[Abstract]:The randomness of wind speed change makes wind power grid connection become the bottleneck of wind power development. If wind speed can be predicted and the precision of wind speed prediction can be improved, it can be helpful for dispatching department to plan and dispatch wind farm actively. In order to reduce the adverse effect of wind power grid connection on power system, the Kalman filter algorithm is used to establish the data filtering model, and the original wind speed data are processed in a primary way to remove the measurement error and system error. Then an improved BP wavelet neural network is used to establish the simulation model of wind speed prediction. Using Kalman filter wind speed data to predict the wind speed, the prediction results are compared with the BP neural network wind speed prediction method. The comparison results show that the algorithm has high prediction accuracy. It shows that the algorithm has a good application prospect in dealing with non-stationary random data.
【作者单位】：上海电机学院电气工程学院;
【基金】：上海市教育委员会科研创新重点资助项目(12ZZ197);上海市教育委员会重点学科资助项目(J51901) 上海市自然科学基金资助项目(12ZR1411600) 上海市区科委技术创新资助项目(2011MH065/2011MH089/2011MH097/2011MH099)
【分类号】：TM614
【正文快照】： 随着风电产业的迅速发展,以及风电并网技术的日益成熟,风电并网比例不断扩大。然而风的间歇性、随机性、波动性给风电并网带来了很多技术难题。如果能对风电场的风速进行准确的预测,就能够减少电力系统的无效运行成本,减轻风电并网时对电力系统的不利影响,提高风电在新能源发

【共引文献】