BP神经网络和云算法的电力营销数据处理方法

Power Marketing Data Processing Method Based on BP Neural Network and Cloud Algorithm

由于电力营销在运行过程中产生大量形态各异的数据,随之产生繁多的数据库,在有效的时间内快速寻求目标数据成为一道难题。针对该问题,该研究在数据管理过程中融入了BP神经网络和云算法,构建了大数据处理平台SP-DPP。SP-DPP软件平台具有卓越的吞吐量与加速比,能够承载多种数据类型,具有较大的数据存储容量,并且能够在较短的时间内处理批量数据,使得处理数据的能力大大提高。该研究通过BP网络算法模型计算、训练电力营销管理系统数据样本,然后将模型接收的电力数据按照误差逆传播算法训练多层前馈网络,大大提高电力营销管理系统数据处理的精确度,云算法技术实现了数据在线即时计算。实验结果表明,该方法数据处理精度较高,误差较小,提高了大数据处理能力。

Due to the power marketing in the process of operation to produce a large number of different forms of data,resulting in a variety of databases,finding target data quickly in an effective time has become a challenge.In response to this problem,we incorporate BP neural network and cloud algorithm in the data management process and build a big data processing platform SP-DPP.SP-DPP software platform has excellent throughput and acceleration ratio,can carry a variety of data types,has a large data storage capacity,and can process batch data in a relatively short time,making the ability to process data greatly improved.This study calculates and trains the data samples of the power marketing management system through the BP network algorithm model,and then trains the power data received by the model according to the error back propagation algorithm to train the multi-layer feedforward network,which greatly improves the accuracy of data processing in the power marketing management system.Cloud computing technology enables real-time online data calculation.The test shows that the proposed method has higher data processing precision and less error,which improves the processing ability of big data.