DEVELOPMENT OF ONLINE MONITORING SYSTEM FOR A GRID CONNECTED SOLAR PHOTOVOLTAIC PLANT

Abstract

This Doctoral Thesis entitled “Development of online monitoring system for a grid connected solar photovoltaic power plant” is aimed to develop the methods/techniques/concepts to improve the performance of1the photovoltaic (PV) plants. Though the amount of energy1that a PV module generate1depends mainly on the incident irradiance, there are different factors/faults/losses (such as, ambient temperature, shading effects, dust or snow accumulation on the surface of the solar panels, maximum power point tracking error and electrical disconnection) which may negatively affect the energy yield. These factors would directly affect the PV module operating parameters and hence its performance. Therefore, it is essential for a PV plant to have an effective and dedicated monitoring system, in order to detect and locate the loss/fault at module level, and to perform the subsequent diagnosis. This thesis starts with the literature overview of various PV monitoring systems, which includes, detailed overview1of all the major PV monitoring1and evaluation1techniques in terms of their1relative performances, sensors used1along with their working principles,1controller(s) used in data acquisition systems,1data transmission methods, and data storage1and analysis. Moreover, a detailed review of popular fault detection techniques with their relative performances, addressing all major types of faults in PV systems is covered. An overview1is made of the PV1modules operational parameters1and their dependence1on applying1atmospheric1conditions. These1are: (i) Short1circuit1current: 𝐼𝑠𝑐 (ii) Open1circuit1voltage: 𝑉𝑜𝑐 (iii) Current1at1maximum1power1point: 𝐼𝑚𝑝 (iv) Voltage1at1maximum1power1point: 𝑉𝑚𝑝 (v) Power1at1maximum1power1point: 𝑃𝑚𝑝 The1translation procedures1of referred parameters1from Standard Test Conditions1(STC) to Real Operating1Conditions (ROC) are1defined. The1proposed model is1validated using1consolidated data1obtained in the1Solar Energy Laboratory, Alternate Hydro Energy Center (AHEC), Indian Institute of Technology Roorkee (IITR) through a PV analyzer of HT instruments. The1procedure1applied1has1been1to1generate1the ii I-V curve values1from proposed model, according1to reported ambient1conditions and then1the results1are compared1in order to1quantify1the resulting1Relative Error1(𝑅𝐸). A total1of 10 sets of1measurements have been1made under different1conditions.The 𝐼 − 𝑉 curves1at STC have been obtained1and further translated to ROC.1The Relative Error (RE) of main1operational parameters1such as short circuit current (𝐼𝑠𝑐),1open circuit voltage (𝑉𝑜𝑐 ), power at maximum1power point (𝑃𝑚𝑝 ) and1corresponding voltage (𝑉𝑚𝑝 ) and1current (𝐼𝑚𝑝 ) as well1has been1calculated. The1same procedure has been1applied to the Smart1Monitoring and Communications Module1(SMCM) obtained experimental results generated1in the Solar Energy Laboratory installed JJP60F230 PV modules,1once they have been previously validated.1PV modules has been equipped1with one SMCM,1which has been in charge1of monitoring its operational1parameters (𝑉, 𝐼 𝑎𝑛𝑑 𝑇𝑚 ), and further1transmit them to the1Central Control System (CCS), by means1of the Power Line Communication (PLC)1technology using already1existing DC power lines.1The master-SMCM1version is in1charge of data routing to1the CCS for further1processing. Additionally,1it processes the parameters1of the calibrated cell1which gives the effective1incident irradiance1and the ambient1temperature1sensor. A new low cost monitoring system is developed in this thesis, which uses SMCM configuration,1and the PLC based1physical communications1layer. The SMCM is a1smart device based1on TI MSP1430 high performance and low1cost micro controller. Considering the SMCM as1an electronic1measurement device, its associated1uncertainty is1quantified. Moreover, an online monitoring system with new fault diagnosis technique is proposed for the solar PV systems operating under grid-tied and off-grid modes. This technique is experimentally validated using the PV system built at Solar Energy Laboratory. A user friendly web application is developed for easy access of monitored data via Internet. Experimental evaluations are carried out to demonstrate the effectiveness of proposed technique in detecting the various fault occurrences in both grid-tied and off-grid PV system. Furthermore, another low cost online monitoring and fault diagnosis technique with optimized voltage sensor locations, and corresponding web application are proposed for gird-tied PV systems. Economic analysis is carried out to study the cost effectiveness of this improved fault iii detection technique considering different values of interest rate and energy tariffs. Experimental results are obtained to demonstrate the effectiveness of proposed fault detection technique. Finally, a health1monitoring method1for PV systems1based on k-nearest neighbors (kNN) is proposed. To implement and1validate the proposed method in computer programs,1a new approach for1modeling PV systems is proposed1that only requires1information from1manufacturer’s1datasheet reported1under normal-operating1cell temperature (NOCT)1conditions and1standard-operating test conditions (STCs). The proposed1model precisely represents characteristics1of PV systems at different temperatures, as the temperature1dependency of parameters1such as ideality factor,1series resistance, and thermal voltage is1considered in the proposed1model