Power Data Quality Optimization and Evaluation Based on BPNN

Abstract

With the continuous improvement of the information technology and communications of Smart Grid, the electric power big data environment has been formed. The data shows diversity and multi-source characteristics. How to ensure the quality of power data in the computer organization under the condition of heterogeneity is the premise of making relevant decisions. This paper firstly gives the definition of Data Space of power enterprises, analyzes the factors affecting the quality of data in the computer environment, and gives the relevant architecture of processing power data in the data space. Secondly, based on business flow and Petri net in the computer environment, this paper constructs the data flow and quality control model of the front and back platforms. The former represents the data flow in the power business and abstracts it to form Petri net computer information flow, so that the data can achieve the effect of cleaning while flowing in the business process. Finally, an evaluation index system is built and back-propagation neural network (BPNN) is used to determine the weight, a case study is given to verify the effectiveness of the proposed method.