Degradación de módulos CIS tras un año de exposición exterior en un enclave soleado.

1,429 views

Published on

Dr. Gustavo Nofuentes Garrido. UJAEN. España
XVII Simposio Peruano de Energia Solar
IV Conferencia Latinoamericana de Energía Solar
Blog: solucionessolares.blogspot.com

Published in: Business, Technology
0 Comments
0 Likes
Statistics
Notes
  • Be the first to comment

  • Be the first to like this

No Downloads
Views
Total views
1,429
On SlideShare
0
From Embeds
0
Number of Embeds
2
Actions
Shares
0
Downloads
4
Comments
0
Likes
0
Embeds 0
No embeds

No notes for slide
  • The used outdoor measurement system is installed in the laboratory in the High Technical School building of the University of Jaén (Spain, latitude 38ºN, longitude 3º W, with a Continental-Mediterranean climate). The calibration in STC of all significant electrical parameters of two 40-Wp ShellTM ST40 CIGS modules,–modules A and B, from now on- was entrusted to the accredited independent laboratory (AIL) CIEMAT-IER (Madrid) at the beginning of our work. Additionally, a small 5-Wp Shell TM ST5 CIGS module was also calibrated to play the role of an irradiance sensor of the same technology as the modules to be tested. two four-wire RTD Pt100 pasted at the backskin of each one of the two calibrated PV modules to measure the cell temperature. A current-voltage curve tracer (IVCT) has been developed using a power supplier (KepcoTM BOP36-12M), a datalogger (AgilentTM 34970A), a class 0·5-60mV/4A shunt resistor, a calibrated 5-Wp Shell TM ST5 CIGS module to measure the incident irradiance and two four-wire RTD Pt100 pasted at the backskin of each one of the two calibrated PV modules to measure the cell temperature. Labview TM software running on a personal computer drives this IVCT. Consequently, the I-V curve and temperature of a cell of each PV module together with the incident irradiance are periodically scanned every two minutes.
  • Using a linear regression of maximum power –normalized to G = 1000 W·m -2 - vs. cell temperature, as depicted in figures 5 and 6, yields cell maximum power temperature coefficients for modules A and B, respectively. It is worth mentioning that these values were found to be better than those provided by the manufacturer. The figure on the RHS of the slide shows the module maximum power normalized to 1000 W·m-2 and corrected to 25 ºC using the above calculated coefficients. A very good stability over time is highly noticeable. A bold horizontal line represents the result of the measurement of module peak power provided by the AIL in its certificate of calibration. The distance between the two horizontal dashed lines equals the 95% confidence interval provided by the above certificate of calibration. In this paper, the module performance ratio ( PR M ) has been calculated as the ratio of the module delivered energy per watt peak to the collected irradiation in kWh·m -2 . It stems from these excellent figures that the CIGS modules have experienced a negligible decrease in their maximum power in STC.
  • Using a linear regression of maximum power –normalized to G = 1000 W·m -2 - vs. cell temperature, as depicted in figures 5 and 6, yields cell maximum power temperature coefficients for modules A and B, respectively. It is worth mentioning that these values were found to be better than those provided by the manufacturer. The figure on the RHS of the slide shows the module maximum power normalized to 1000 W·m-2 and corrected to 25 ºC using the above calculated coefficients. A very good stability over time is highly noticeable. A bold horizontal line represents the result of the measurement of module peak power provided by the AIL in its certificate of calibration. The distance between the two horizontal dashed lines equals the 95% confidence interval provided by the above certificate of calibration. In this paper, the module performance ratio ( PR M ) has been calculated as the ratio of the module delivered energy per watt peak to the collected irradiation in kWh·m -2 . It stems from these excellent figures that the CIGS modules have experienced a negligible decrease in their maximum power in STC.
  • Using a linear regression of maximum power –normalized to G = 1000 W·m -2 - vs. cell temperature, as depicted in figures 5 and 6, yields cell maximum power temperature coefficients for modules A and B, respectively. It is worth mentioning that these values were found to be better than those provided by the manufacturer. The figure on the RHS of the slide shows the module maximum power normalized to 1000 W·m-2 and corrected to 25 ºC using the above calculated coefficients. A very good stability over time is highly noticeable. A bold horizontal line represents the result of the measurement of module peak power provided by the AIL in its certificate of calibration. The distance between the two horizontal dashed lines equals the 95% confidence interval provided by the above certificate of calibration. In this paper, the module performance ratio ( PR M ) has been calculated as the ratio of the module delivered energy per watt peak to the collected irradiation in kWh·m -2 . It stems from these excellent figures that the CIGS modules have experienced a negligible decrease in their maximum power in STC.
  • Using a linear regression of maximum power –normalized to G = 1000 W·m -2 - vs. cell temperature, as depicted in figures 5 and 6, yields cell maximum power temperature coefficients for modules A and B, respectively. It is worth mentioning that these values were found to be better than those provided by the manufacturer. The figure on the RHS of the slide shows the module maximum power normalized to 1000 W·m-2 and corrected to 25 ºC using the above calculated coefficients. A very good stability over time is highly noticeable. A bold horizontal line represents the result of the measurement of module peak power provided by the AIL in its certificate of calibration. The distance between the two horizontal dashed lines equals the 95% confidence interval provided by the above certificate of calibration. In this paper, the module performance ratio ( PR M ) has been calculated as the ratio of the module delivered energy per watt peak to the collected irradiation in kWh·m -2 . It stems from these excellent figures that the CIGS modules have experienced a negligible decrease in their maximum power in STC.
  • Maximum power measured under real sunlight has been compared to modelled maximum power during an experimental campaign ranging from July 2006 to March 2007. This modelled maximum power has been calculated using measured values of G and TC together with the calibrated values of all significant electrical parameters as input data for some tried simple algebraic methods. In the terms defined in a following section, the root mean square error (RMSE) and the mean bias error (MBE) derived from measured vs. modelled maximum power have been analysed. Both RMSE and MBE provide the guidance to know which algebraic method(s) fit(s) best the experimental data so as to be recommended.
  • Degradación de módulos CIS tras un año de exposición exterior en un enclave soleado.

    1. 1. DEGRADACIÓN DE MÓDULOS CIS TRAS UN AÑO DE EXPOSICIÓN EXTERIOR EN UN ENCLAVE SOLEADO Nofuentes G., Fuentes M., Aguilera J., De la Casa, J. Plazaola, C. (*) Grupo de Investigación IDEA. Escuela Politécnica Superior. Universidad de Jaén. Campus de Las Lagunillas, s/n. 23071-Jaén ( España ). Tel: +34 953 212 434. Fax: +34 953 211 967. E-mail: [email_address] (*) Universidad Tecnológica de Panamá IV Conferencia Latino Americana de Energía Solar (IV ISES_CLA) y XVII Simposio Peruano de Energía Solar (XVII- SPES) Cuzco (Perú), 1-5 noviembre de 2010
    2. 2. ESQUEMA DE LA PRESENTACIÓN <ul><li>INTRODUCCIÓN </li></ul><ul><li>OBJETIVO </li></ul><ul><li>METODOLOGÍA </li></ul><ul><li>CONFIGURACIÓN EXPERIMENTAL </li></ul><ul><li>RESULTADOS EMPÍRICOS </li></ul><ul><li>CONCLUSIONES </li></ul>
    3. 3. INTRODUCCIÓN <ul><li>Módulos fotovoltaicos basados en materiales de estructura calcopirita (CIS, CIGS): </li></ul><ul><ul><li>Presencia aún marginal en el mercado </li></ul></ul><ul><ul><li>Gran potencial: algunas previsiones de fabricación de hasta 1 GWp/año en la próxima década </li></ul></ul><ul><ul><li>Existencia de módulos con eficiencias de célula en torno al 13% en CEM. Varios fabricantes proporcionan garantías del 90% de la potencia pico durante 10 años </li></ul></ul><ul><ul><li>Estabilidad a sol real escasamente documentada y no comprendida satisfactoriamente. C ontroversia acerca del uso de simuladores solares para calibrar la potencia pico en CEM antes y después de la exposición a la intemperie </li></ul></ul>
    4. 4. OBJETIVO <ul><ul><li>Una auditoría técnica sobre la tecnología CIS/CIGS puede asesorar a diseñadores de sistemas, entidades financieras e inversores </li></ul></ul><ul><ul><li>El estudio de la estabilidad de la potencia pico a lo largo del tiempo en climas soleados puede conducir a un empleo confiado de estos módulos en áreas geográficas con alto nivel de irradiación </li></ul></ul><ul><ul><li>OBJETIVO </li></ul></ul><ul><ul><li>Contribución al estudio de la degradación a sol real </li></ul></ul><ul><ul><li>de módulos CIS </li></ul></ul>
    5. 5. METODOLOGÍA <ul><ul><li>Análisis de datos experimentales de funcionamiento a sol real durante 12 meses de dos módulos CIS: </li></ul></ul><ul><ul><ul><li>Evolución de potencia máxima normalizada y corregida en temperatura para CEM </li></ul></ul></ul><ul><ul><ul><li>Cálculo del factor de rendimiento ( performance ratio, en inglés) de cada módulo ( PR M ) </li></ul></ul></ul><ul><ul><li>Comparación de los valores de potencia máxima en CEM de dichos módulos suministrados por un Laboratorio Acreditado Independiente (CIEMAT-IER) antes y después de la exposición a la intemperie. </li></ul></ul>
    6. 6. CONFIGURACIÓN EXPERIMENTAL Módulos A y B, desde ahora Dos módulos calibrados Shell TM ST40 de 40 Wp Módulo calibrado Shell TM ST5 de 5Wp: sensor de irradiancia DTR Pt100: Sensor de temperatura de célula Fuente de alimentación (Kepco TM BOP36-12M) + Unidad de conmutación (Agilent TM 34970A) + Labview TM sobre PC Trazador de curvas Tensión-Intensidad Relés de alta potencia. Es posible secuenciar la obtención de curvas V-I de hasta 4 módulos JAÉN Lat. 38ºN; Long. 3ºO Clima continental-mediterráneo
    7. 7. RESULTADOS EMPÍRICOS (1/5) Regresión lineal: el coeficiente de temperatura para la potencia máxima (gamma) resultó ser mejor para ambos módulos que el proporcionado por el fabricante (-0.0048 ºC -1 vs -0,006 ºC -1 ))!!! Método de T.Strand et al.: módulo A (1/2) Tc (ºC) Potencia máxima normalizada a 1000 W·m -2 (W)
    8. 8. RESULTADOS EMPÍRICOS (2/5) Método de T.Strand e al.: módulo A (2/2) Potencia máxima normalizada y corregida en temperatura para CEM (W)
    9. 9. RESULTADOS EMPÍRICOS (3/5) Método de T.Strand e al.: módulo B Potencia máxima normalizada y corregida en temperatura para CEM (W)
    10. 10. RESULTADOS EMPÍRICOS (4/5) Factor de rendimiento Performance Ratio ( PR M ) Donde: E = Energía producida en el período de tiempo bajo análisis(Wh); H = Irradiación incidente durante el período de tiempo bajo nanálisis (Wh·m -2 ); G* = 1000 W·m -2 ; P* MOD,M = Potencia pico calibrada por un Laboratorio Acreditado Independiente (W) PR M anual = 0,93 para el módulo A PR M anual = 0,89 para el módulo B
    11. 11. Calibración en CEM de la potencia máxima de los módulos en el CIEMAT-IER antes y después de la campaña experimental Junio 2006 Julio 2007 Degradación del 4,3% (módulo A) y 5,3% (módulo B) RESULTADOS EMPÍRICOS (5/5) Wp
    12. 12. CONCLUSIONES <ul><li>La potencia máxima normalizada a 1000 W·m -2 y corregida a 25 ºC –método de T. Strand et al. – muestra una buena estabilidad para los dos módulos ensayados durante el período analizado de doce meses </li></ul><ul><li>Los valores del factor de rendimiento o performace ratio medidos para los módulos A y B (0,93 y 0,89, respectivamente) apoyan la conclusión anterior </li></ul><ul><li>La degradación experimentada por los módulos antes y después de la campaña experimental de doce meses es inferior al 5,5%, según datos proprcionados por un Laboratorio Acreditado Independiente </li></ul><ul><li>Estos resultados optimistas deben ser tomados con cautela debido a las siguientes razones: </li></ul><ul><ul><li>Muestra muy limitada (2 módulos) de un único fabricante </li></ul></ul><ul><ul><li>Otras consideraciones de índole tecnológica (el factor de forma puede verse afectado de forma distinta para los módulos fabricados por diferentes fabricantes de módulos CIS, p. ej.) </li></ul></ul><ul><li>Actualmente se siguen ensayando y midiendo módulos CIS en la Universidad de Jaén, con resultados positivos </li></ul>
    13. 13. Gracias por su atención Dr. G. Nofuentes Tel: +34.953.212.434 Fax: +34.953.211967 E-Mail: gnofuen@ujaen.es IV Conferencia Latino Americana de Energía Solar (IV ISES_CLA) y XVII Simposio Peruano de Energía Solar (XVII- SPES) Cuzco (Perú), 1-5 noviembre de 2010

    ×