You are on page 1of 13

Available online at www.sciencedirect.

com

Solar Energy 91 (2013) 337349 www.elsevier.com/locate/solener

Monitoring, modelling and simulation of PV systems using LabVIEW


Aissa Chouder a, Santiago Silvestre b,, Bilal Taghezouit a, Engin Karatepe c
b a Photovoltaic Laboratory, Development Centre of Renewable Energies, BP 62 Route de lObservatoire, 16340 Bouzareh Algiers, Algeria ` cnica de Catalunya, C/Jordi Girona 1-3, Campus Nord UPC, 08034 Barcelona, Spain Electronic Engineering Department, Universitat Polite c Department of Electrical and Electronics Engineering, Ege University, 35100 Bornova, Izmir, Turkey

Received 24 July 2012; received in revised form 25 September 2012; accepted 29 September 2012 Available online 29 October 2012 Communicated by: Associate Editor Nicola Romeo

Abstract This paper presents a detailed characterization of the performance and dynamic behaviour of photovoltaic systems by using LabVIEW real-time interface system. The developed software tool integrates several types of instruments into a single system which is able to oer online measurements all data sources and comparison simulation results with monitored data in real-time. Comprehensive monitoring and analyzing of PV systems play a very important role. The proposed method is a low-cost solution to provide fast, secure and reliable system by making the system database-ready for performance analysis of PV systems. The proposed method is also applied to a grid connected PV system in the Centre de Developpement des Energies Renouvelables (CDER) in Algeria. The results show that there is a good agreement between the measured and simulation results values. The integration methodology of robust simulation and monitored data in real-time can be extended to study the fault diagnosis of a PV system. 2012 Elsevier Ltd. All rights reserved.
Keywords: PV systems; Monitoring; Modelling; Simulation

1. Introduction LabVIEWe (Laboratory Virtual Instrument Engineering Workbench) is a graphical programming language by National Instruments that uses icons instead of lines of text to create applications. Nowadays this programming environment has found its application in many scientic elds and technical engineering, so in this work we propose an integral LabVIEW platform of monitoring, modelling and simulation tools for photovoltaic (PV) systems. Many applications of LabVIEW for monitoring PV systems have been reported before in the literature (Koutroulis nez and Kalaitzakis, 2003; Forero et al., 2006; Mart rquez etal., 2009; Vergura and Natangelo, 2009; Ulieru Boho et al., 2010).

Corresponding author. Tel.: +34 934017491; fax: +34 934016756.

E-mail address: santi@eel.upc.edu (S. Silvestre). 0038-092X/$ - see front matter 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.solener.2012.09.016

On the other hand dierent commercial software solutions are available for PV systems simulation (Silvestre, 2012) and standard simulation software, as Matlab (Yusof et al., 2004; Pater and Agarwal, 2008; Karatepe et al., 2008; Chouder and Silvestre, 2012) or Pspice (Castan er and Silvestre, 2002; Silvestre et al., 2009) have been also extensively used for this purpose. Furthermore Vergura and Natangelo (2010) have integrated Matlab and Simulink for monitoring energy performances of PV plants. In this work we report the integration of monitoring, modelling and simulation of PV systems in the same environment able to give information of the system behaviour in real time. This solution allows the acquisition and control of all necessary data from the PV system, evaluate main model parameters of PV modules and array, calculate the performance ratio (PR) and Yields of the system, create HTML and XLS report les and visualize all these data and the dynamic system behaviour in real time. Moreover the integration of robust modelling and simulation gives

338

A. Chouder et al. / Solar Energy 91 (2013) 337349

the opportunity to compare simulation results with monitored data in real time, allowing the development of new tools for fault detection as well as new prediction models with the objective of improve the performance and reliability of PV systems, optimizing the system output to achieve higher yields. 2. PV system description The proposed method of monitoring, modelling and simulation of PV systems has been applied to a grid con veloppement nected PV system located in the Centre de De des Energies Renouvelables (CDER), Algerie. The PV system is formed by 90 PV modules (Isofoton 106Wp-12 at STC) divided in three subgenerators of 3 kWp each one. The subgenerators are formed by two parallel strings of 15 PV modules in series. Each subgenerator is connected to a single phase inverter of 2.5 kW (IG30 Fronius) that injects the generated energy into a phase of the public low voltage distribution network of the National Company (Sonelgaz) 220 V50 Hz (Hadj Arab etal., 2005). The block diagram of this PV system is shown in Fig. 1. 3. Monitoring PV systems using LabVIEW Fig. 2 shows a schematic diagram of the sensors and acquisition data of the monitoring system implemented in the PV system. Dierent sensors are included to measure irradiance, temperature, as well as current and voltages at the DC and AC sides of the system. The data acquisition is carried out using an Agilent 34970A and dedicated multiplexer module Agilent 34902A with sixteen channels. Data communication between PC and LabVIEW, where the incoming data is processed, is performed by GPIB bus.

Two pyranometers (Kipp & Zonen CM 11 type) and a reference solar cell are used to measure the irradiance. One of the pyranometers and the reference cell are installed at two dierent places of the PV plant to measure irradiance in the tilted plain. A second pyranometer measures the irradiance in the horizontal plane. Thermocouples have been used to measure the ambient temperature near the PV plant in order to predict the eective PV modules temperature. Eq. (1) is used to calculate this eective solar cell temperature, Tc. T c T a NOCT 20  C G 800 1

where Ta is the ambient temperature, G the irradiance and NOCT the Normal Operating Cell Temperature given by the PV modules manufacturer. For the measurements of currents, DC (Idc) and AC (Iac), we have used two CLSM-50 closed loop Hall eect current sensors and a dual operational amplier LM 1458N. The DC output voltage of the PV system is measured by means of a voltage divider, while the AC output voltage is measured at the secondary of the transformer used for the voltage supply of the hall sensors. All data coming from the acquisition system are processed in LabVIEW using the VI shown by Fig. 3, where the appearing coecients are used to calibrate the dierent monitored parameter. The VI allows the following tasks: Communication with the data acquisition Agilent 34970A in order to setup the dierent channels via GPIB bus (DC/AC and temperature measurements), processing the output string coming from the data acquisition, splitting the output string to the corresponding measured variable and the calibration of each channel with

Fig. 1. Schematic diagram of the CDER grid connected PV system.

A. Chouder et al. / Solar Energy 91 (2013) 337349

339

Fig. 2. Sensors and acquisition data system to monitor the PV system.

Fig. 3. VI developed to obtain the monitored data.

Table 1 Calibration factors and channels associated to each measured variable and sensitivity of the irradiance sensors used. Measured variables GI,c GI,p, GH,p Tamb VDC,meas VAC,meas IDC, IAC,meas Channel number 101 102, 103 104 105 106 107, 108 Sensors Reference cell: Rsh 0:037 X CM 11: sensitivity = 5 lV/(W m2) k type Thermocouple Resistive divider AC adapter Hall eect Calibration factor 7901.4 19,9203.2 Direct measure in Agilent 34,970 29 220/18 1.6

340

A. Chouder et al. / Solar Energy 91 (2013) 337349

the corresponding scaling factor. Table 1 shows the calibration factors and channels associated to each measured variable as well as the sensitivity of the irradiance sensors used. 4. Description of PV system modelling and simulation 4.1. Modelling the PV module The model of the PV module is based on the one diode model of the solar cell shown in Fig. 4, where G and T are irradiance and temperature respectively, Iph is the photo generated current depending on irradiance and temperature conditions, D is the diode modelling the P/N junction of the solar cell and Rsh and Rs are the shunt and series resistances respectively, modelling the power losses in the device (Overstraeten and Mertens, 1986; Castan er and Silvestre, 2002). The output current of the solar cell, I, can be written as: I I ph I d I sh 2

include various branches in parallel. If we consider Ns solar cells in series in each branch and a total number of Np branches for a PV module, the total number of solar cells forming the PV module is Ns Np. So, Eq. (6) can be conveniently scaled to obtain a similar equation for the I(V) characteristic of a PV module, taking into account the following equations (Castan er and Silvestre, 2002; Chenni et al., 2007; Karatepe et al., 2007): I m N pI V m N sI Rsm Ns Rs Np Ns Rsh Np 7 8 9 10

Rshm

where Iph is the photo generated current, Id is the diode current and Ish is the current across Rsh. These currents are given as: I ph G I ph;ref lIcc T T ref Gref 3

where G and T are respectively the irradiance and temperature conditions of work, Gref and Tref are irradiance and temperature at standard test conditions (STCs): 1000 W/m2 and 25 C, Iph,ref is the photo generated current at STC and lIcc the temperature coecient of current.     V Rs I I d I sat exp 1 4 nV t where Isat is the is the reverse saturation current of diode, n is the diode ideality factor and Vt the thermal voltage. I sh V Rs I Rsh 5

where parameters with subscript m stands for the PV module. The model that allows to obtain the I(V) characteristic of a PV module, considering Eqs. (6)(10), has been implemented in LabVIEW environment. The I(V) and P(V) characteristics as well as the coordinates of the maximum power point (MPP) of both, solar cell and PV module are available results of the developed model. The input parameters for the calculations are: Open circuit voltage of the solar cell (Voc), short circuit current of the solar cell (Isc), Rs, Rsh, Ns, Np, G, T, lIcc and the solar cell ideality factor (n). The owchart for the obtention of the I(V) and P(V) characteristics is shown in Fig. 5. 4.2. Inverter model The inverter model developed in LabVIEW environment is based in the performance model inverter presented by King et al. (2007). The AC output power of the inverter, PAC,sim is dened by the following equation:
 P AC;sim   P aco C A B P dc B C P dc B2 AB 11

Eq. (2) can be rewritten, considering Eqs. (3)(5), as       V RsI V Rs I I I ph I sat exp 1 nV t Rsh

where Paco is the maximum AC output power for inverter at reference or nominal rating conditions, Pdc is the DC power at the inverter input and parameters A, B and C are given by the following equations: A P dco 1 C 1 V DCsim V dco B P so 1 C 2 V DCsim V dco C C o 1 C 3 V DCsim V dco 12 13 14

Eq. (6) is an implicit and not linear equation than gives the I(V) characteristic of the solar cell. Commercial photovoltaic modules are composed by association of solar cells in series forming a branch. Some higher power PV modules

Fig. 4. Equivalent circuit of the solar cell.

where VDCsim is the DC voltage at the inverter input, Vdco and Pdco are respectively the DC voltage and power inputs at which the AC-power rating is achieved at the reference rating condition, Pso is the DC power required at the inverter input to start working properly and C1, C2 and C3 are empirical coecients to adjust the PAC(PDC) characteristic of the inverter. The values of main parameters involved in Eq. (11) used in this work for modelling the inverters are shown in Table 2.

A. Chouder et al. / Solar Energy 91 (2013) 337349

341

Fig. 5. Flowchart for the obtention of the I(V) and P(V) characteristics.

Table 2 Values used in the performance model inverter. Paco (W) 2692 Pdco (W) 2900 Vdco (V) 275.6 Pso (W) 25 Co (W)1 6.67 105
t X Einst 0

C1 (V)1 0.00296

C2 (V)1 0.00458

C3 (V)1 0.0255

4.3. Grid connected PV system simulation The main objective of the simulation of grid connected PV system is to obtain expected evolution of voltages and currents at the DC side of the system as well as at the AC side, at the inverter output. So, simulation results will give the expected behaviour, in a dynamic way, of the whole system taking into account real conditions of climate parameters. From simulation results the values of power (P) and energy, instantaneous (Einst) and cumulative (Ecum) energies, can be evaluated as follows: Einst P Dt 15

Ecum

16

The simulation of the whole grid connected PV system is based on the models presented above for PV modules and for the inverter and is carried out also in LabVIEW environment. The owchart of the simulation process is depicted in Fig. 6. The calculation of the power and cumulative and instantaneous energies, AC and DC, generated by the photovoltaic system in dynamic regime is evaluated using another VI.

342

A. Chouder et al. / Solar Energy 91 (2013) 337349

Fig. 6. Flowchart of the simulation process.

Finally, the expected values of the system yields: Reference yield (Yr), array yield (Ya) and nal yield (Yf), as well as the performance ratio (PR), can be evaluated from the simulation results using the following expressions (Ha berlin and Beutler, 1995; Commission of the European Communities, 1997): R Dt G Yr 0 17 Gref R Dt Edc Ya 0 18 Po R Dt Eac Yf 0 19 Po PR Yr Yf 20

expected ve main parameters (Iph, Isat, n, Rs, Rsh) (Chouder and Silvestre, 2012) has been integrated in LabVIEW environment as a separated module. The developed procedure nds out the ve parameters at STC (1000 W/m2 and 25 C) and then calculates them for any other real operating condition. The eect of variation of the parameter values of a single PV module aects the entire PV system performance (DAlessandro et al., 2011).Our method includes parameter extraction techniques for the whole system that allow a good estimation of the system output. If some parameter of a PV module changes and important dierences between simulation results and monitored data are observed, is necessary to run again the parameter extraction algorithm and check again if simulation results show a good agreement with monitored data. Main results obtained, applying the parameters extraction procedure, are summarized in Table 3. The visualization panel performing this task is shown in Fig. 8. The validation of the procedure is carried out by comparing real measurement of IV characteristics measured at outdoor conditions and the simulated one generated by introducing the expected ve model parameters. The result of this comparison is consolidated by plotting the error curve between simulated and measured IV characteristics and quantifying the main error indicator as seen in Fig. 9 and reported in Table 4. These values are below previous reported values for errors obtained in simulations of PV modules (Mahmoud et al., 2012; Villalva et al., 2009). The resolution of Eq. (6) shows an increasing error between simulation and measurement near the open circuit voltage, as can be seen in Fig. 9. It is essentially due to numerical stability, because the number of points generated by the simulation is less than the number of points given by the IV tracer PVPM2516, so the calculation of the error is also aected at high voltage. 5.2. Analysis of PV system simulation Results The variable climates used in the simulations carried out to analyze and validate the simulation procedure of the PV system, presented before in Section 4.2, are temperatures and irradiances obtained from the monitoring system shown in Fig. 10. The samples have been measured with a time step of 1 min. The simulation of the PV system behaviour, following the ow chart shown in Fig. 6, was done in real time. Main results obtained for DC voltage and current at the output of the PV array are shown in Fig. 11. As can be seen, a good agreement is obtained between simulation results and monitored data. Main dierences observed between 15.00 h and 18.00 h for the current are due to inverter disconnection forced by grid disturbances at the end of the day. The results obtained for the generated power at the DC and AC sides of the system are shown in Fig. 12, where the eect of the inverter disconnection is present again. Table 5 summarizes the errors observed between monitored values and simulation results for power and energy

where Gref is the irradiance at STC, G is the measured irradiance and Po is the nominal PV system power. Fig. 7 shows the evolution of Yr along a day and its value for this day obtained from LabVIEW simulations. 5. Results and discussion 5.1. PV module model validation In order to validate the PV module model used to predict the whole PV system performance, a procedure based on outdoor measurement and analytic derivation of the

A. Chouder et al. / Solar Energy 91 (2013) 337349

343

Fig. 7. Reference yield. Table 3 Extracted parameters for Isofoton 106-12 PV module at 762 W/m2 of irradiance and 26.2 C of temperature. Parameter Value Iph (A) 5.1 Io (A) 1.9 10
8

n 1.14

Rs (X) 0.33

Rsh (X) 144

Isc (A) 5.07

Voc (V) 20.4

Im (A) 4.53

Vm (V) 16.31

Fig. 8. Visualization panel of parameters extraction of PV module.

generated by the PV system, giving a good approach to the accuracy of the simulation procedure. Root mean square errors (RMSEs) below 4% are obtained for power and energy delivered by the PV array, these values are below previous RMSE values reported in the literature in simulations of PV systems (Chouder and

Silvestre, 2009) and in the same order of magnitude than MSE errors obtained using neural networks algorithms (Yu and Chang, 2011). The whole system has been monitored along the months of June, August and September 2012 using the presented procedure. Simulations of the dynamic system behaviour

344

A. Chouder et al. / Solar Energy 91 (2013) 337349

Fig. 9. Plot of the error between measured and simulated IV characteristic (G = 762 W/m2, T = 26.2 C).

Table 4 Main error values between measured and simulated parameters. Relative error Value EI sc % 0.97 EV oc % 1.07 EI m % 1.63 EV m % 0 RMSEA 0.13 RMSE% 2.66

Fig. 10. Monitored irradiance and temperature proles.

have been carried out in the same period of time. Fig. 13 shows the values obtained for the reference yield, Yr, for the second and third weeks of June and Fig. 14 shows the same information for dierent days corresponding to these 3 months.

The comparison between monitoring (meas) and simulation results (sim) for the array and nal yields are given in Figs. 15 and 16 for the same weeks of June. This information is enlarged in Figs. 17 and 18 for several days for these 3 months.

A. Chouder et al. / Solar Energy 91 (2013) 337349

345

Fig. 11. Evolution of voltage and current at the output of the PV array, measured (meas) and simulation results (sim).

Fig. 12. Evolution of the power generated by the PV system, measured (meas) and simulation results (sim).

Table 5 Indicators of the accuracy of the simulation procedure. PDC Mean error RMSE RMSE% 61.13 W 6826 W 2.73 PAC 80.97 W 95.17 W 3.81 EDC 0.17 Wh 0.19 Wh 2.73 EAC 0.22 Wh 0.26 Wh 3.81

6. Conclusion This works presents an integral LabVIEW platform of monitoring, modelling and simulation of grid connected PV systems. In the same platform, we propose the modelling of the PV module identied with outdoor measurements of IV curves in order to extract the main PV module parameters. The PV module modelling and extraction parameters procedure has been successfully validated experimentally. For the dynamic behaviour modelling of the PV system, an accurate model of the Inverter is included. The inverter model allows to predict AC output power as a function of DC input voltage and DC input power. The simulation methodology of the PV system in real dynamic conditions of work has also been been validated successfully in a grid connected PV system located in Argelia. The developed platform allows the acquisition and control of all necessary data from the PV system, the simulation in real time of the whole PV system working in

On June 19 we removed intentionally one string from the whole PV array to show the eectiveness of the model to detect malfunctions. So in this day the simulation results gave the result of a healthy system, while the monitored data gives lower output power. As can be seen in Figs. 15 and 16, the comparison of monitored data and simulation results gives the opportunity to detect a malfunction in the system. Finally the values obtained for the performance ratio are given in Figs. 19 and 20. As can be seen a good agreement is found between simulation results and monitored data.

346

A. Chouder et al. / Solar Energy 91 (2013) 337349

Fig. 13. Reference yield June 2012.

Fig. 14. Reference yield along summer 2012.

Fig. 15. Array yield June 2012.

dynamic behaviour, calculates the performance ratio, PR and Yields of the system, create HTML and XLS report les and visualize all these data and the dynamic system

behaviour in real time. This toolbox results in robust modelling, advanced simulation incorporating predictions of system output with respect to solar resource, local weather

A. Chouder et al. / Solar Energy 91 (2013) 337349

347

Fig. 16. Final yield June 2012.

Fig. 17. Array yield along summer 2012.

Fig. 18. Final yield along summer 2012.

348

A. Chouder et al. / Solar Energy 91 (2013) 337349

Fig. 19. Performance ratio June 2012.

Fig. 20. Performance ratio along summer 2012. Alessandro, V., Guerriero, P., Daliento, S., Gargiulo, M., 2011. A D straightforward method to extract the shunt resistance of photovoltaic cells from currentvoltage characteristics of mounted arrays. SolidState Electronics 63, 130136. ndez, J., Gordillo, G., 2006. Development of a Forero, N., Herna monitoring system for a PV solar plant. Energy Conversion and Management 47, 23292336. Ha berlin, H., Beutler, Ch., 1995. Normalized representation of energy and power for analysis of performance and on-line error detection in PVsystems. In: Proceedings of the 13th EU PV Conference on Photovoltaic Solar Energy Conversion, October 1995, Nice, France, pp. 934937. Hadj Arab, A., Cherfa, F., Chouder, A., Chenlo, F., 2005. Grid connected photovoltaic system at CDER-Algeria. In: Proceedings of the 20th European Solar Energy Conf. and Exhibition, June 2005, Barcelona, Spain. Karatepe, E., Boztepe, M., C olak, M., 2007. Development of a suitable model for characterizing photovoltaic arrays with shaded solar cells. Solar Energy 81, 977992. Karatepe, E., Hiyama, T., Boztepe, M., C olak, M., 2008. Voltage based power compensation system for photovoltaic generation system under partially shaded insolation conditions. Energy Conversion and Management 49 (8), 23072316.

and system behaviour. The output results obtained from this toolbox could allow the inclusion in the same platform an algorithm for automatic supervision and fault detection of the PV system. References
Castan er, L., Silvestre, S., 2002. Modelling Photovoltaic Systems using Pspice. John Wiley & Sons Ltd., Chichester. Chenni, R., Makhlouf, M., Kerbache, T., Bouzid, A., 2007. A detailed modeling method for photovoltaic cells. Energy 32, 17241730. Chouder, A., Silvestre, S., 2009. Analysis model of mismatch power losses in PV systems. Journal of Solar Energy Engineering 131 (2), 024504 024509. Chouder, A., Silvestre, S., 2012. Modeling and simulation of a grid connected PV system based on the evaluation of main PV module parameters. Simulation Modelling Practice and Theory 20 (1), 4658. Commission of the European Communities, 1997. Guidelines for the Assessment of Photovoltaic Plants, Document B, Analysis and Presentation of Monitoring Data. Issue 4.3. Joint Research Centre.

A. Chouder et al. / Solar Energy 91 (2013) 337349 King, D.L., Gonzalez, S., Galbraith, G.M., Boyson, W.E., 2007. Performance Model for Grid-Connected Photovoltaic Inverters. Sandia Report: SAND2007-5036. Koutroulis, E., Kalaitzakis, K., 2003. Development of an integrated dataacquisition system for renewable energy sources systems monitoring. Renewable Energy 28 (1), 139152. Mahmoud, Y., Xiao, W., Zeineldin, H.H., 2012. A simple approach to modeling and simulation of photovoltaic modules. IEEE Transactions on Sustainable Energy 3 (1), 185186. nez Boho rquez, M.A., Enrique Go mez, J.M., Andu rquez, jar Ma Mart J.M., 2009. A new and inexpensive temperature-measuring system: Application to photovoltaic solar facilities. Solar Energy 83, 883890. Overstraeten, R.J., Mertens, R.P., 1986. Characterisation and Testing of Solar Cells and Modules. Physics, Technology and Use of Photovoltaic. Hilger. Bristol. Pater, H., Agarwal, V., 2008. MATLAB-based modeling to study the eects of partial shading on PV array characteristics. IEEE Transactions on Energy Conversion 23 (1), 302310. Silvestre, S., 2012. Review of system design and sizing tools. In: Mcevoy, A., Markvart, T., Castan er, L. (Eds.), Practical Hadook of Photovoltaics. Fundamentals and Applications. Elsevier, Oxford, pp. 684694. Silvestre, S., Boronat, A., Chouder, A., 2009. Study of bypass diodes conguration on PV modules. Applied Energy 86 (9), 16321640.

349

Ulieru, V.D., Cepisca, C., Ivanovici, T.D., Pohoata, A., Husu, A., Pascale, L., 2010. Measurement and analysis in PV systems. In: Proceedings of the International Conference on Circuits, ICC 2010, WSEAS, July 2010, Korfu Island, Greece, pp. 137142. Vergura, S., Natangelo, E., 2009. Labview interface for data analysis of PV. In: Proceedings of the International Conference on Clean Electrical Power, June 2009, Capri, Italy, pp. 236241. Vergura, S., Natangelo, E., 2010. LabviewMatlab integration for analyzing energy data of PV plants 2010. In: Proceedings of the International Conference on Renewable Energies and Power Quality (ICREPQ10), March 2010, Granada, Spain, pp. 635640. Villalva, M.G., Rafael, Gazoli.J., Ruppert, Filho.E., 2009. Comprehensive approach to modeling and simulation of photovoltaic arrays. IEEE Transactions on Power Electronics 24 (5), 11981208. Yu, T.S., Chang, H.T., 2011. The forecast of the electrical energy generated by photovoltaic systems using neural network method. In: Proceedings of Conference on Electric Information and Control Engineering (ICEICE), April 2011, Wuhan, China, pp. 27582761. Yusof Y., Sayuti, S.H., Abdul Latif, M., Mohd. Zamri Che Wanik, 2004. Modeling and simulation of maximum power point tracker for photovoltaic system. In: Proceedings of the Power and Energy Conference, November 2004, Kuala Lumpur, Malaysia, pp. 8893.

You might also like