光伏电站远程监控及发电预测系统的研究

发布时间：2018-04-19 12:37

本文选题：光伏电站 + 远程监控　；参考：《辽宁工程技术大学》2014年硕士论文

【摘要】：随着科学技术的迅速发展和能源危机的不断加重,人们越来越重视寻找新能源。光伏由于有种种优点被人们普遍利用。随着光伏电站的大规模兴起,配套的监控设备也在逐步完善。人们已经不仅仅满足于现场观测各种实时数据,而是希望更科学智能地管理电站。电站作为一种新能源,不同于传统的常规能源,由于其发电功率的间歇性波动,成为并网的一大瓶颈。如何在保证安全的情况下,最大限度的消除功率波动给电网带来的影响已经成为研究的热点。如何将功率预测与电站的监控管理有机的联系在一起是本文研究的重点。本文首先介绍了光伏电站监控系统的研究现状,论述了功率预测作为光伏电站监控有机组成的重要性。结合现场实际情况对系统总体框架和功能进行了详细设计。其次,考虑到光伏电站布线不方便,选用了无线通信模块WiFi和GPRS进行数据的传输,详细叙述了通信过程。通过协议TCP、UDP的比较,提出了一种通信协议,来解决通信过程中存在的数据丢包和实时性问题,再次,利用灰色关联法寻找历史数据中与待测日相似度较高日期,通过BP神经网络训练后,得出待预测日的发电功率。通过与传统的算法与物理算法比较,得出此算法的适用范围。最后,利用C++builder6.0构建了电站本地监控,完成了科学管理电站的要求。
[Abstract]:With the rapid development of science and technology and the aggravation of energy crisis, people pay more and more attention to finding new energy.Photovoltaic is widely used because of its various advantages.With the large-scale rise of photovoltaic power plants, supporting monitoring equipment is also gradually improving.People are not only satisfied with the field observation of real-time data, but also want to manage the power station more scientifically and intelligently.As a new energy, power station is different from conventional energy, because of its intermittent fluctuation of power generation, it becomes a bottleneck of grid connection.How to eliminate the influence of power fluctuation on power grid to the maximum extent has become a hot topic in the case of ensuring safety.How to combine power prediction with monitoring and management of power station is the focus of this paper.This paper first introduces the research status of photovoltaic power station monitoring system, and discusses the importance of power prediction as an organic component of photovoltaic power plant monitoring system.The overall frame and function of the system are designed in detail according to the actual situation in the field.Secondly, considering the inconvenient wiring of photovoltaic power station, the wireless communication module WiFi and GPRS are selected for data transmission, and the communication process is described in detail.Based on the comparison of TCP / UDP protocol, a communication protocol is proposed to solve the data packet loss and real-time problems in the process of communication. Thirdly, the grey correlation method is used to find the date of high similarity between the historical data and the day to be tested.After BP neural network training, the generation power of the day to be predicted is obtained.By comparing with the traditional algorithm and the physical algorithm, the application range of the algorithm is obtained.Finally, the local monitoring of the power station is constructed by using C builder6.0, and the requirement of scientific management of the power station is fulfilled.
【学位授予单位】：辽宁工程技术大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TM615

【参考文献】