Development of a GUI for Parallel Connected Solar Arrays

Transcription

1 Development of a GUI for Parallel Connected Solar Arrays Nisha Nagarajan and Jonathan W. Kimball, Senior Member Missouri University of Science and Technology 301 W 16 th Street, Rolla, MO Abstract This work describes the development of a software package with a graphical user interface (GUI) for the evaluation of a solar array. The paper presents MATLAB-based simulations for various configurations of solar panels and compares these configurations on basis of power output and photovoltaic characteristics, via current-voltage and power-voltage curves. The simulation used field data to consider partial shading of the array. The computations also consider integrated power converters. Index Terms Array configurations, Graphical User Interface, Integrated Power Converters, Photovoltaic characteristics G I. INTRODUCTION lobal electrical energy consumption is steadily rising, driving an ever greater demand to increase the power generation capacity. This demand has driven significant growth in the variety of technology available to harness renewable energy. Rising fossil fuel prices and an awareness of global warming are drawing particular attention to the use of solar photovoltaic (PV) cells and their use as an alternative source of electric power is increasing by leaps and bounds [1]. Statistics show an increase of 60% from the year 2004 to 2009 [2]. Solar power offers unique advantages to customers. It is clean, creating no emissions or noise at point of use, and it has minimal environmental impact along the supply chain. It is modular, infinitely scalable, and often portable. It can power anything from a wristwatch to a multimegawatt power station [3]. A PV system is designed to draw maximum power from a PV panel or array. In the long term, this project will develop a new solar array technology that increases modularity and installation flexibility. Each module will contribute its individual maximum power, and weaker modules will not reduce the output of the whole array. Since this technology will be modular the solar arrays can be installed quickly and upgraded as needed. A first step in this project has been to develop a means to evaluate this new approach. To this end, software has been designed to compute the power output of a PV system in various scenarios through the power-voltage and current-voltage (IV) characteristics of the array. This software has a graphical user interface (GUI) that allows the user to input various scenarios and configurations to find the best configuration for the application depending on power output, shading, insolation (incident solar radiation), and other factors. This simulation tool is suitable for training and planning purposes. It predicts the performance of either a conventional array or an array built with the new technology. In a training environment, installers will be able to quickly evaluate the power output of an array using a panel that can be oriented at various angles to the sun or even subjected to uneven shading. Similarly, planners will be able to use the software to select sites, orient panels and compare panel and power converter technologies. The software that is currently available is not user-friendly and it does not consider the impact of power converters on the power output of an array. The GUI proposed here, however, analyzes data obtained from a sensor array s data acquisition system (DAQ) to capture information about temperature, insolation and other factors. It also provides a training and planning platform that allows PV system users to select the panels and configuration that best suit their needs. The program supports rapid evaluation of a wide range of insolation scenarios and array topologies. It includes a batch processing mode, in which the user provides several parameters that the software maps into many simulations. The GUI is runs in MATLAB with a simple user interface that offers minimal clutter and helpful tool tips. II. INITIAL DEVELOPMENT Before the development of the GUI, this work calculated various parameters using the user inputs or user parameters. These calculations took into consideration the new technology and conventional systems and evaluated the effects of various parameters on PV system characteristics. A. Basic Hardware and Approach A typical solar panel generates current and voltage that varies with temperature and insolation and from cell to cell due to manufacturing variation. Most solar arrays are composed of several panels, each rated for a relatively low voltage and connected in a series strings to achieve some desired system voltage. In a conventional array, each panel in a series string must conduct the same amount of current. If one panel generates less current due to shading or damage, the current of the whole array is limited, and its power output is thus disproportionately reduced. Each module of the proposed technology, however, comprises of a conventional panel and a dc-dc converter that boosts its voltage by a fixed ratio to the desired system voltage. The boosting converters use a switched capacitor topology to provide a fixed voltage gain. To incorporate these features, this work computed results for various configurations of the solar arrays using MATLAB to plot the power-voltage and current-voltage (IV) characteristics.

2 Fig 1. Equivalent circuit of a solar cell, representing the single exponential model of a solar cell. A PV cell can be represented by an equivalent circuit, as shown in Fig. 1 [4]. The characteristics of such a cell can be obtained using standard equations [5]. Many researchers have considered the effects of different insolation levels on a single solar cell; however, the IV and power-voltage characteristics of a single module do not predict the presence of multiple steps and peaks, which are common among the photovoltaic characteristics of large PV arrays that receive non uniform insolation. Therefore, this work used the datasheet values of a solar panel to calculate its series resistance, shunt resistance and diode quality (or ideality factor) and generalized the results for a number of solar panels. The general current-voltage characteristic of a PV panel based on the single exponential model is calculated as v irs nv v ir s t s i Iph Io e 1 (1) Rsh where: I ph is the photogenerated current at Standard Test Conditions (STC), I o is the dark saturation current at STC, R s is the panel series resistance, R sh is the panel shunt resistance, and N s is the number of cells in the panels connected in series. In the above equation, V t is the junction thermal voltage given by AkTstc Vt (2) q where A is the diode quality (ideality) factor, k is the Boltzmann s constant, q is the charge of an electron, T stc is temperature at STC in degrees Kelvin [4]. To calculate the required parameters, equation (1) can be written for the three key points of the current-voltage characteristic: the short-circuit point, the maximum power point, and the open-circuit point [4]. Solution of these equations using a numerical method such as the Newton- Raphson method yields the parameters series resistance, shunt resistance and diode quality. With these parameters, performance at any temperature or insolation level may be computed with the same set of equations. challenge in using a PV source is to analyze its nonlinear output characteristics, which vary with temperature and insolation. These characteristics become more complicated when the insolation is not uniform over the entire array, as in partly cloudy conditions, resulting in multiple peaks [5]. In a conventional array, if a panel is shaded, its current decreases, limiting the output of the entire series. Although other panels in the array may be capable of greater output, they are limited by the weakest panel. The computations performed here considered partial shading. If the voltage of one string is far enough below that of the other strings voltage, the shaded string will contribute no power to the system. One shaded or damaged panel can eliminate the contribution of an entire string, and can even affect the power production of the other strings. This condition must be considered when computing the parameters and characteristics. Initially, this work assumed that all panels in the array are equally shaded or unshaded. Once the computations were complete for unshaded panels, cases of partial shading were considered. In addition to the incident radiation on a panel, temperature also plays a vital role in determining the characteristics of a panel. The effect of temperature is calculated as [6] 3 r T 1 1 Io( T) Io exp r r (3) T m' Vt V t where, I o (T) denotes the reverse saturation current for a given temperature and V t (T) is the thermal voltage. The subscript r denoted a value defined for reference operating conditions. Here, ε= 1.12 ev is the silicon energy band gap, m is the equivalent PN junction factor, which is given by m =m r /N s 2, where N s denotes the number of cells connected in series [ref] III. DEVELOPMENT OF THE GUI The GUI was been developed to be user friendly and uncluttered. It includes a number of screens for which the user can enter various required parameters. It also includes tool tips, which give the user more specific instructions for each input. The GUI focuses on the the physical and electrical arrangement of the panels in the array. The user can provide known shading characteristics, or allow the software to calculate shading scenarios automatically. The program generates several plots and computes the power delivered by the array depending on configuration. The user beings by entering the geometrical location to plot a graph of altitude versus azimuth of the sun. Latitude, entered in decimal degrees, and day of the year are the primary inputs. This plot shows the number of useable sun hours for any given day of any given year based on latitude. Figure 2 shows a screenshot of this step. B. Effect of shading and temperature The effect of shading on a solar panel is crucial. A major

3 Fig 2. Geometrical data screen: User inputs include day of the year and latitude. Fig 4. Custom parameters: An example of a tool tip providing the user by giving more specific instructions on how to enter the data is also seen. Fig 3. Choice screen: User can choose a predefined panel or custom panel. On the next screen the user enters more specific information about the array configuration. The system plots the actual PV characteristics depending on the solar panel type, its parameters, and the array configurations. On the next screen, shown in Figure 3, the user selects a solar panel either from the library of panels or by entering datasheet parameters or custom values, from a dropdown menu. If the user chooses a standard built-in panel, the parameters are displayed for that panel, and the user can either confirm the choice or go back to select another panel. Figure 4 shows the screen that appears if the user chooses to enter parameters manually. The program also features error messages. For examples, it displays a warning box if the user enters data that does not conform to the required data type (such as a character string where numerical values are required). Also, if the user leaves a required field blank, a warning message appears, as shown in Figure 5. Once the user confirms the panel parameters, a screen appears asking whether the user would like to include an integrated power converter (IPC). This boosting converter is a dc-dc converter based on switched capacitor topology and provides a fixed voltage gain. In the simulations performed here, it was modeled as an ideal transformer with a fixed resistance. If such an IPC is connected, the panels are considered to be connected in parallel with each other. If not, the strings of panels connected in series are considered to be Fig 5. An error message generated when a required field is left blank. connected in parallel. Figure 6 shows a screenshot of the IPC. Radio buttons allow the user to include an IPC. If an IPC is included, editable text boxes allow him to indicate number of turns (or the turns ratio) and the fixed resistance in ohms. If not, these additional buttons remain inactive. Following the IPC screen a final screen (shown in Figure 7) allows the user to select configuration and insolation options. Depending on the configuration needed for the array, the user chooses the number of panels in series and parallel. If an IPC has been included, then the series input is disabled, and the user can indicate only the number of panels in parallel. The user can choose to import the insolation data in either of two ways: the user can plot insolation as a time series by using the Browse button to enter insolation as a.csv file from the data acquisition (DAQ) system. The experiments performed here used the Tern R-Engine-A DAQ system. Alternatively, the user can allow the software to generate random insolation dat, using the Max and the Min fields to specify the insolation range in W.m -2. In this case, a pop-up is generated with the random values displayed in W.m -2. Once all the data parameters have been entered, the computations begin and a pop-up box show their progress. The system plots the power-voltage (P-V), current-voltage (I- V) and power-current (P-I) curves for the selected configuration, and displays a pop up box showing the maximum power obtained, the output voltage and the output

4 TABLE I COMPARISON OF THE TWO CASES WITH RANDOM DATA (SCENARIO 1) NO IPC WITH IPC SERIES 3 N/A PARALLEL 4 12 TURNS RATIO N/A 10 RESISTANCE (Ω) N/A 10 OUTPUT VOLTAGE (V) OUTPUT CURRENT (A) MAXIMUM POWER (W) Fig 6. User inputs for integrated power converter. (a) Fig 7. Configuration screenshot. Number of panels in series and/or parallel can be entered via editable text boxes as well as insolation data patters. current. Figure 8 shows the sample plots for power-voltage, current-voltage and power-current curves. IV. SIMULATION FOR EXAMPLE SCENARIOS To verify the advantage of this new parallel connection technology over the conventional technology, this work ran a number of scenarios, with both random and field data. The following compares the results, primarily on the basis of power output and energy. A. Scenario 1 This scenario relied on random sets of data, comparing. output plots with and without an IPC. Both cases used ahe Uni-Solar 68 panel, and the insolation data for both were randomly generated in a range between 500 and 1000 W.m -2. Table I compares the parameters and output numbers for the two cases. Power output was greater when an IPC was present Figure 9 plots the results for an array without an IPC. Figure 10 shows the characteristics of an array with an IPC. It presents 12 panels in a conventional series-parallel configuration, with each series string having 3 panels in series and 4 such parallel connected strings. Following this, we use real experimental field data to compare different scenarios in a similar way. B. Scenario II Fig 8. Characteristics of the selected configuration: (a) I-V Characteristics, P-V Characteristics, and P-I characteristics. A second scenario used experimental or field data. The sensor array used Uni-solar 68 panels. It was set up at various test sites with varied insolation patterns. Each simulation used 400 readings from the test site, taken 4 seconds apart; the readings were recorded using the DAQ. Table II shows the properties of this scenario. In addition to the characteristic plots, insolation of each panel can also be plotted over time. This plot indicates each panel s short circuit current which is proportional to insolation, and thus is indicative of the scaled insolation of each panel as well. Figure 11 shows the plots for the configuration without an IPC with insolation data obtained from the field experiments. Figure 12 shows the same plots when an IPC is present. V. CONCLUSION Development of this interface was prompted by the need for a simple GUI. This new program supports rapid evaluation of a wide range of insolation scenarios and array topologies. Parallel-connected solar arrays are inherently superior to conventional arrays if there is any variation in insolation across the array, such as partial shading or variation in orientation. To support this, new power converters based on this technology, which is based on integrated power converters not only retains the advantages of conventional

5 (a) (a) Fig 9. Array characteristics for scenario 1 when an IPC is present: (a) I-V characteristics, P-V characteristics, and P-I characteristics Figure 10. Characteristics for scenario 1 when an IPC is not present: (a) I-V characteristics, P-V characteristics, and P-I characteristics topologies but is integrated to reduce the logistics burden. The software serves as a training tool to permit comparison of various configurations and scenarios. Future work will include completion of a user s guide to the software, with detailed instructions. Also, a number of scenarios derived from field data will be analyzed. VI. ACKNOWLEDGMENTS This research was sponsored by the Leonard Wood Institute in cooperation with the US Army Research Laboratory and was accomplished under Cooperative Agreement Number W911NF The views and conclusions contained in this document are those of the authors, and do not represent TABLE II COMPARISON OF THE TWO CASES WITH FIELD DATA (SCENARIO 2) NO IPC SERIES 2 N/A PARALLEL 4 8 TURNS RATIO N/A 10 RESISTANCE (Ω) N/A 10 OUTPUT VOLTAGE (V) OUTPUT CURRENT (A) MAXIMUM POWER (W) WITH IPC

6 (a) (a) Fig 11. Array characteristics based on experimental data (scenario 2) when IPC is not present: (a) I-V characteristics, P-V characteristics, and P-I characteristics the official policies, either expressed or implied, of the Leonard Wood Institute, the Army Research Laboratory or the US Government. The US Government is authorized to reproduce and distribute reprints for government purposes, notwithstanding any copyright notation hereon. VII. REFERENCES [1] R. Ramaprabha and B. L. Mathur, "Modeling and Simulation of Solar PV Array under partial shaded conditions," to be presented at IEEE International Conference on Sustainable Energy Technologies, Fig 12. Array characteristics based on experimental data (scenario 2) when IPC is present: (a) I-V characteristics, P-V characteristics, and P-I characteristics [2] T. R. E. P. N. f. t. s. Century, "Global status report 2010." [3] B. Jigjid, "Photovoltaics: An energy option for Sustainable Development," to be presented at 3rd World Conference on Photovoltaic Energy Conversion, [4] D. Sera, R. Teodorescu, and P. Rodriguez, "PV Panel model based on datasheet values," vol. IEEE International Symposium on Industrial Electronics, pp [5] H. Patel and V. Agarwal, "MATLAB-based modeling to study the effects of partial shading on PV array characteristics," IEEE Trans. Energy Convers, vol. 23, pp [6] M. A. Vitorina, L. V. Hartmann, A. M. N. Lima, and M. B. R. Correa, "Using the model of the solar cell for determining the maximum power point of photovoltaic systems," to be presented at European Conference on Power Electronics and Applications, 2007.