Study Of Remote Intelligent Monitor Approach For Photovoltaic Power Station

Since solar power generation has the advantages of high reliability,low maintenance costs,zero fuel consumption,less noise pollution and high security,the photovoltaic industry has aroused more and more attention in the world.Excellent monitoring effect is of importance for improving power generating efficiency,protecting power plant safety and reducing power losses.However in the existing monitoring system,PCS is usually the smallest monitor unit,which resulting in making some anomaly be neglected.Like individual solar panels’ off line,the traditional monitoring system could not identify this problem timely,resulting in power loss;what’s more,environmental data collected were often unable to be used effectively.This paper aimed to research photovoltaic plant remote intelligent monitoring method.By optimizing the monitoring system architecture,designing sun tracking control modules and soot formation judgment module,we achieved the goal of strengthening monitoring capacity,improving power generation efficiency and reducing power losses.Firstly,this paper developed a three-tier monitoring framework for monitoring the photovoltaic power plant.At the bottom tier,we selected solar arrays series as the smallest monitor unit to collect operating data.And there also was a small meteorological station to collect environmental data.Then the thesis designed a multi-interfaces controller to transmit data and implement some control functions in the middle tier.At the top-tier,PC and mobile terminals were chosen to be the equipment in monitoring center,with which clients could timely get the field data,analyze statistics and send controlling order to controller at any time.Secondly,the sun tracking control module was designed.The paper built a weather classification method by fusing multi-meteorological data.Different tracking control strategies were proposed to tackle various weather conditions so as to reduce the consumption of the photovoltaic tracking system.The classification models of fine-cloudy weather were trained based on the support vector machines with the multi-meteorological data in 2012 of Baoding.Then we analyzed the models with cross contrast of different seasons,different times,and different feature,for getting the most accurate one for judging weather.Finally,the thesis designed a soot formation judgment module.By comparing real generation power values with theoretical values,which was calculated based on electricity generation model,judge the extent of soot formation.We also processed the extent of soot formation in three grades:mild,moderate or severe.In addition,in view of the required function of the PV power station monitoring system,the software and hardware of monitoring scheme were researched.The hardware design includes hybrid power supply circuit,series communication interface circuit and mass storage interface circuit.The upper workstation software realized such functions as real-time data display,electricity generation data statistics,image analysis,data storage,and fault alarm,etc.