Design And Implementation Of Data Collector And Abnormal Data Calibration For Photovoltaic String

Data analysis is the basis of fault detection,power station monitoring,and operation management of photovoltaic power plants.Affected by external environmental interference and the nonlinear errors of devices during the process of photovoltaic data collection,there are problems such as many abnormal interferences and low availability exist generally in the collected data,which unable to support the demand of data analysis and hinders the pace of digital power station construction.Thus it is of great significance to research the photovoltaic data acquisition devices with stable working performance and abnormal data calibration algorithms.In this paper,the photovoltaic data acquisition system and the calibration module of abnormal data are designed according to the correlation between photovoltaic meteorological data and output data.The main work is as follows:(1)A photovoltaic data collector based on ARM is designed to provide data support for calibration and analysis of data.The collector includes a main control module,a data storage module based on SD card,a data transmission module based on long range radio(LoRa)transmission technology,and an acquisition module for collecting data such as current,voltage,temperature,humidity,irradiance,speed and direction of wind and so on.(2)By analyzing the correlation between photovoltaic meteorological data and output data,a calibration model of photovoltaic abnormal data based on dual-weighted kernel partial least squares algorithm(DLW-KPLS)is proposed.The model can predict the photovoltaic output data,identify the anomalies of the collected data,and calibrate the data,which filter the interference of abnormal sampling effectively.(3)The modules of data storage,processing and display are designed on the host computer.The data collected by the collector is calibrated and then put it into the database to store.When the control center is accessed,the data is called for display,which realizes the complete functions of data collection,transmission,processing,storage and display.The experimental results show that:(1)The data collector supports the collection that data of current,voltage and which is multi-meteorological related to photovoltaic strings,which can realize wireless data transmission with stable performance.(2)The error of experimental data is improved by the DLW-KPLS photovoltaic abnormal data calibration model proposed in this paper obviously,which indicates that this method can play an important role in the calibration of photovoltaic abnormal data.