分布式光伏发电功率预测与监测平台

发布时间：2018-06-22 16:48

本文选题：分布式光伏发电系统 + LabVIEW　；参考：《山东建筑大学》2017年硕士论文

【摘要】：在全球科技与经济发展的推动下能源消费需求也随之急剧增长,并且伴随着极端天气和雾霾的频繁出现,传统能源的弊端越来越突出,人们对低碳的追求和新能源开发利用的渴望也愈来愈强烈。在物质利益与能源节约的权衡与取舍中,太阳能凭借免输送,零污染,无噪声以及取之无尽、用之无绝的优势,使其在新兴能源领域中的地位尤为凸显。灵活高效的分布式光伏发电系统是太阳能应用领域中普及面最广的一个重要分支,但是当其并入电网运行后,光伏系统出力易受光能波动性、间歇性、随机性等影响使其成为一个不可控源,对电网的高效、安全、稳定运行带来严峻挑战。通过将提前预测出的光伏发电功率作为基准,来制定发电计划、及时判断系统运行情况并作出相应解决措施,是最终保障电力系统可靠、持久、稳定运行的有效举措。本文以提高光伏发电输出功率预测模型的精度和运算速率为目的,在潍坊某电力科技公司的分布式光伏发电系统及其数据采集平台的基础上,利用LabVIEW编写数据监测平台来实时显示并保存光伏数据与气象数据。对采集到的实验数据进行分析与处理后,用改进的IHCMAC算法搭建光伏发电输出功率预测模型。主要研究工作如下:1.探索分析光伏发电系统组成结构及特性。利用Matlab软件搭建太阳能光伏电池数学模型,并得到其输出特性。借助本地微型气象站采集的气象数据和光伏系统的功率数据,从理论上分析光伏系统输出功率与各要素的相关性关系,排除不必要因素对预测结果的干扰。2.利用LabVIEW软件搭建分布式光伏发电系统监测平台。开发具有数据采集、显示、预测与保存功能的监测平台来实时观测光伏系统运行状态和气象条件的变化。为便于后续预测工作直接调用,将数据保存为Excel格式。该监测平台预测功能模块输入参数的获取也是通过调用已保存为Excel格式的历史实测数据实现的。3.构建实时监测数据的分析与处理模型。借助历史数据对光伏系统输入输出因素进行关联性分析,获取预测模型的主要影响因素,并使用均值填充法、GESD算法、Z-score算法对实验数据进行预处理,使得实验数据更加完备、准确,为预测工作提供合理、可靠的数据基础。4.建立分布式光伏发电功率预测模型。针对短期光伏预测中存在的计算量大预测精确度低等难题,在对现场数据进行采集、处理与定量分析的基础上,基于改进的IHCMAC算法,提出了短期智能预测算法并对其加以改进,利用采集的气象参数、光伏发电数据,构建了光伏发电功率预测模型,并通过性能评价验证了该算法的有效性。
[Abstract]:Driven by the development of global science and technology and economic development, the demand for energy consumption has also increased sharply, and with the frequent emergence of extreme weather and smog, the disadvantages of traditional energy have become more and more prominent. The pursuit of low-carbon and the desire for the development and utilization of new energy are becoming more and more intense. In the trade-off between material benefits and energy conservation, solar energy is especially prominent in the emerging energy field because of its advantages of no transportation, zero pollution, no noise and endless use. The flexible and efficient distributed photovoltaic power generation system is one of the most popular and important branches in the solar energy application field. However, when it is incorporated into the power grid, the photovoltaic system is vulnerable to the fluctuation and intermittence of light energy. The influence of randomness makes it an uncontrollable source, which brings severe challenges to the efficient, safe and stable operation of power grid. It is an effective measure to ensure the reliable, lasting and stable operation of the power system in the end by taking the photovoltaic power generation power predicted in advance as the benchmark to formulate the power generation plan, judge the operation of the system in time and make the corresponding measures to solve the problem. In order to improve the precision and calculation rate of photovoltaic output power prediction model, this paper is based on the distributed photovoltaic power generation system and its data acquisition platform of a power company in Weifang. LabVIEW is used to compile data monitoring platform to display and save photovoltaic data and meteorological data in real time. After analyzing and processing the collected experimental data, an improved IHCMAC algorithm is used to build a photovoltaic output power prediction model. The main research work is as follows: 1. To explore and analyze the structure and characteristics of photovoltaic power generation system. The mathematical model of solar photovoltaic cell is built by Matlab software, and its output characteristics are obtained. With the help of meteorological data collected by local micrometeorological station and the power data of photovoltaic system, the correlation between output power of photovoltaic system and various elements is analyzed theoretically, and the interference of unnecessary factors to forecast results is eliminated. LabVIEW software is used to build a distributed photovoltaic system monitoring platform. A monitoring platform with the functions of data acquisition, display, prediction and storage is developed to observe the changes of the operating state and meteorological conditions of photovoltaic system in real time. The data is saved in Excel format to facilitate the direct call of the subsequent prediction work. The acquisition of input parameters of the function module of the monitoring platform is also realized by calling the historical measured data saved as Excel format. The analysis and processing model of real-time monitoring data is constructed. With the help of historical data, the relationship between input and output factors of photovoltaic system is analyzed, and the main influencing factors of the prediction model are obtained, and the experimental data are preprocessed by using the mean filling GESD algorithm and Z-score algorithm, which makes the experimental data more complete. Accurate, provide reasonable and reliable data base for forecasting work. 4. A distributed photovoltaic power prediction model is established. In view of the difficult problems in short-term photovoltaic prediction, such as large amount of calculation and low precision of prediction, on the basis of collecting, processing and quantitative analysis of field data, based on the improved IHCMAC algorithm, a short-term intelligent prediction algorithm is proposed and improved. Based on the collected meteorological parameters and photovoltaic power generation data, a photovoltaic power prediction model is constructed, and the effectiveness of the algorithm is verified by performance evaluation.
【学位授予单位】：山东建筑大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TM615

【相似文献】