光伏发电输出功率的预测方法研究

发布时间：2018-06-05 10:39

本文选题：光伏发电输出功率 + 环境因素　；参考：《沈阳农业大学》2017年硕士论文

【摘要】：太阳能作为人类最为理想的清洁能源,与其他能源相比其优势自然不言而喻,而光伏发电也成为人类利用这种能源的一种重要方式。影响光伏发电效果最为明显的因素就是环境因素,由于环境的随机性和不确定性,也导致了光伏输出功率的不稳定,这无疑会对电网的调度和潮流的调控的产生影响。所以如果能够预先估计下一时段光伏发电的输出功率,对光伏发电的输出功率进行相应的预测,必将减少电网运行的风险同时也更加便于人们合理使用电能,使能源利用的效率得到了提高,间接的减少了使用能源过程中对环境带来的污染。本课题以MATLAB软件为建模平台,首先对大量数据进行分析和处理,通过数据证实了环境温度、相对湿度和辐照强度这三个环境因素与光伏发电输出功率之间的相互关系,通过他们相互之间的性质和联系,最终选用了多元线性回归的方法并建立模型,通过大量数据的分析和七次异常值的筛选,得出回归方程,然后对回归方程进行回归效果的显著性等检验,最终将通过检验的回归模型应用于晴朗天气状况下的光伏功率预测,效果稳定符合要求。但是在非晴朗天气状况下,基于多元线性回归的预测模型的预测效果并不理想,所以针对非晴朗天气状况下选用BP神经网络的方法重新搭建模型,在输入层对其进行改进和优化,在三个环境因素的基础上加入了前三天的历史功率共同作为模型的输入进行训练,得出的结果符合预期要求,能够实现在非晴朗天气状况下光伏输出功率的预测。
[Abstract]:Solar energy as the most ideal clean energy, compared with other sources of energy, its advantages are self-evident, and photovoltaic power generation has become an important way to use this energy. The most obvious factor affecting the effect of photovoltaic power generation is the environmental factor. Because of the randomness and uncertainty of the environment, it also leads to the instability of photovoltaic output power, which will undoubtedly have an impact on the dispatching of power grid and the regulation of power flow. So if we can estimate the output power of photovoltaic power generation in advance and forecast the output power of photovoltaic power generation, it will reduce the risk of grid operation and make it more convenient for people to use electricity reasonably. The efficiency of energy use has been improved, indirectly reducing the environmental pollution caused by the use of energy. In this paper, MATLAB software is used as the modeling platform. Firstly, a large number of data are analyzed and processed. Through the data, the relationship between three environmental factors, environmental temperature, relative humidity and radiation intensity, and the output power of photovoltaic power generation is proved. Through their mutual properties and relations, the multivariate linear regression method is used and the model is established. Through the analysis of a large number of data and the screening of the seven abnormal values, the regression equation is obtained. Finally, the regression model is applied to the photovoltaic power prediction in sunny weather, and the effect is stable enough to meet the requirements. But under the condition of non-sunny weather, the prediction effect of the prediction model based on multivariate linear regression is not ideal, so the BP neural network method is used to rebuild the model under the condition of non-sunny weather. The input layer is improved and optimized, and the historical power of the first three days is added to the three environmental factors for training as the input of the model, and the results are in line with the expected requirements. The photovoltaic output power can be predicted under the condition of non-sunny weather.
【学位授予单位】：沈阳农业大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TM615

【相似文献】