基于大数据的发电量预测分析数据平台建设

现有电力企业发电指标统计渠道较分散，发售配电各个环节产生的数据种类繁杂且体量较大，因此针对电力数据质量差、已挖掘价值低的问题，本文采用大数据预测神经网络和数据平台可视化展示的方法，以某发电企业为研究对象，构建了发电量预测的模型及以此为核心的数据平台。发电量预测的应用服务支持PC和APP端，涵盖了某电力企业所有控股电厂及新能源项目，以滚动预测的模式迭代优化算法模型，提高发电量预测精度，有利于安排生产、燃料采购、物资计划，进一步做好污染物排放预测和管控。平台重点实现了数据多维管理和用户界面的可视化展示。数据层面，从源头做好数据采集清洗和数据库实时数据交互；用户层面，更直观形象展示大数据预测从获取数据、数据分析及预处理、数据训练到预测的四个阶段。

Construction of Data Platform for Power Generation Forecast and Analysis Based on Big Data

Currently power companies have more scattered statistics channels for power generation indicators, and the processes of power generation, power sale and power distribution have produced large variety and large amount of data. Therefore, in order to solve the problem of poor data quality and low value of mined data, this paper introduces the neural network method of big data prediction and the method of visual display, taking a certain power generation enterprise as the research object, and builds a model for power generation forecast and a data platform based on it. The application service of power generation forecast supports the PC and APP, covering all the power plants and new energy projects of the certain power generation enterprise. Iterative optimization of the algorithm model in a rolling prediction mode to improve the accuracy of power generation prediction, which is advantageous to power generation arranging, fuel procurement, material planning and further improving the prediction and control of pollutant emissions. The platform focuses on the realization of multi-dimensional data management and visual display of the user interface. At the data level, the platform performs data collection, cleaning and database real-time interaction from the source, and at the user level, the platform intuitively displays data acquisition, data analysis and preprocessing, data training and data prediction, which are the four stages of big data prediction.