摘要
太阳能发电技术的发展有利于缓解全球能源危机及环境污染等问题,而太阳辐照的精准预测是太阳能发电功率预测的基础和前提。该文提出一种基于人工神经网络耦合自回归滑动平均模型的太阳总辐照资源预测方法,并建立了相应的实验平台。该实验平台通过分布式传感器网络采集测试区域内离散的太阳总辐照资源数据,在MATLAB软件中构建模型并加以训练,利用耦合预测方法预测太阳总辐照资源数据,并通过实验仿真验证预测方法的有效性。该实验平台的构建融合了“可编程控制器”“电路基础”“MATLAB仿真”等多种课程,可以有效激发学生的创新思维与学习潜能,培养学生的科研能力和实践能力,从而推动新能源科学与工程专业的创新实验教学与实践。
The development of solar power technology is helpful for mitigating the issues of global energy crisis and environment pollution,and the precise prediction of solar irradiation is the premise and fundament of the prediction of solar power output.This paper proposes a prediction method of solar irradiation based on artificial neural network coupled with autoregressive moving average model,and develops corresponding experimental platform.This platform collects concrete solar irradiation data of test area with distributed censor network,develops model in MATLAB,utilizes coupled prediction method to forecast the solar irradiation data,and validates the effectiveness by experimental simulation.The development of this platform integrates many courses such as PLC,circuit foundation and MATLAB simulation,which can effectively enlighten students'innovative thinking and learning potential,develop students'ability of research and practice,thus motivate the innovative experiment education and exploration of new energy science and engineering.
作者
郭苏
何意
王琛
GUO Su;HE Yi;WANG Chen
出处
《科教文汇》
2020年第27期84-86,共3页
Journal of Science and Education
基金
江苏省自然科学基金(编号:BK20181308),中央高校业务费(编号:B200202174)。
关键词
太阳辐照预测
自回归滑动平均模型
神经网络
创新实验教学
solar irradiation prediction
autoregressive moving average model
neural network
innovative experiment education