Research On Noise Of Dataacquisition And Data Compression In Power Recording System

Power monitoring device can help to achieve close-coordination of theentire power grid, and it is critical to improve the reliability of power supply.Real-time and accurate data of power system is an important basis to ensure thestable operation of power monitoring device, however, if power monitoringdevices keep on working for long time, the massive data will inevitably affectthe performance of the power grid. In the situation of the global development ofsmart grid and the gradual expansion of power structure, efficient datacompression scheme is an effective solution for improving communicationcapability and storage costs. In order to increase data compression ratio, itdepends on not only optimizing compression algorithm, but also improving thecorrelation of data by restraining sampling noise.This thesis is based on an actual power recording system, firstly introducesthe hardware structure and the noise situation of the entire system. Then,analyzes the amplitude noise, phase noise and clock jitter from data acquisitionsystem by theoretical calculation and building Matlab model. Respectively,testing the relationship between the compression performance and noise fromsignal conditioning circuit, phase locked loop, clock jitter of ADC by the methodof single noise input, and then determine the main noise source according to datacompression ratio and propose some suggestion and method of controlling effectfrom signal conditioning circuit noise and PLL phase noise. In regard to thesteady-state sampling data of power system, come up with an improved datacompression scheme based on two-dimensional lifting wavelet transform, whichcan get better performance on SNR and compression ratio than the originalmethod, and through online testing the feasibility of the compression scheme canbe verified.The research combines the sampling noise and the result data compression,and takes the compression ratio as an index for evaluating data acquisitionsystem. It is able to provide theoretical basis for improving quality of samplingand compression in power recording system.