Data Processing And Database Design Of Power Equipment Monitoring

With the development of the ubiquitous power Internet of Things,power equipment has also begun to develop towards intelligence,informatization,and automation.The importance of power equipment in smart grid security protection has become increasingly prominent.Traditional power equipment has a low degree of informatization and digitization,acquisition and transmission equipment are greatly affected by the environment,resulting in data deviation and a wide variety of data outliers.The existence of equipment abnormal data brings great difficulties to power system state estimation and equipment safety protection.Based on the characteristics of specific power equipment monitoring data,this paper studies data preprocessing and analysis technology to realize the cleaning of abnormal data.We restore the real state of power equipment itself to provide decision-making basis for power equipment safety protection.The main work is as follows:For the abnormal time series data of power equipment,we design data acquisition and abnormal detection processing flow,including data feature normalization,model training evaluation;we use long short-term memory recurrent neural network to design anomaly detection algorithms for long-term,short-term,and even weak time-dependent power equipment time series data.We built the power equipment monitoring experimental platform and complete the data processing software design of the platform which verify the anomaly detection algorithm.Based on the simplified model of distribution network and equipment correlation,the current load data model of power equipment is defined.We analyze the change of correlation through basic shape transformation and point-arc transformation to detect the data anomalies which is verified by simulation.Aiming at the multi-type data structures such as analog,digital,switch existing in power equipment,we dedign multi-modal data storage structure and database.The paper develop the software design of multi-modal database.We design specific test cases to verify database functions and test performance of the system during large-scale concurrent operations.