TED-AJ03-286 PREDICTION OF GRID-CONNECTED PHOTOVOLTAIC POWER OUTPUT UNDER NATURAL ENVIRONMENT

概要

論文の詳細を見る
Recently, the photovoltaic (PV) power generation system has attracted attention as a clean energy source. In particular, the residential roofing PV system connected to a power grid has become popular due to both increasing concerns regarding global warming and the continued decrease in PV manufacturing costs. Construction of an efficient PV system, including peripheral devices and the PV module, would be possible if the energy production could be predicted accurately from weather data before installation of the system. The PV output can be calcualted theoretically using the voltage-current characteristic equation of the silicon solar cell, if the solar irradiance and the cell temperature are given. The cell temperature affects the electrical output characteristics and efficiency of the cell. Unfortunately however, we cannot determine the cell temperature directly. The cell temperature depends on a number of physical and environment factors. Traditionally, the cell temperature has been determined for a specific installation from the solar irradiance and ambient temperature using an empirical equation. However the cell temperature is affected by the construction of the PV module and the manner in which the PV module is installed. In addition, when the PV module produces electric power, since the operating cell temperature and the PV output interact with each other, obtaining the cell temperature theoretically is difficult. The present study introduces a new method by which to computer the cell temperature and the PV output analytically from weather data and the physical properties of materials included in the PV module. When these data are given, the cell temperature is calculated based on the module heat balance while considering the electrical output. The calculation result then enables us to calculate the PV output based on the voltage-current characteristic equations. This computation is repeated until the interactive relation between the cell temperature and the PV output is satisfied. In order to provide the performance data needed for the method validation, a 5-kW photovoltaic test facility using polycrystalline-silicon solar cells was constructed at the Yatsushiro National College of Technology, and long-term performance monitoring under natural environment was conducted. The PV system mounted in an open rack was installed on the rooftop of a three-story building. The azimuth and slope angles of the PV arrays were fixed at 0° and 20°, respectively. The power was returned to the AC utility line through a power conditioning unit. The recorded data included irradiance, ambient temperature, wind speed, PV array temperature, DC voltage and current. At the present time, more than 12 months worth of data are available for analysis. The validity of the prediction model was evaluated by comparing the model to actual measurements. When irradiance and cell temperature were given, the PV output as calculated using the voltage-current characteristic equations was checked with respect to conformance to the measurements. The results show that the calculations are in good agreement with the measurements. Subsequently, when irradiance, ambient temperature, and wind speed were given, the cell temperature as calculated by considering the heat balance of the PV module was examined for validity. The solution for the prediction model involves the interactive relation between the PV output and cell temperature. The calculations are in relatively good agreement with the measurements, despite the simplification by the prediction model. Next, the daily PV energy production was calculated using this model. The results show that when a daily or a long-term PV energy production is predicted, the prediction result is in good agreement with the experimental result, even when the heat capacity of the PV module is assumed to be zero. Finally, the potential annual energy productions of cities around the world were estimated from local weather data for the 5-kW PV system of the present study. These results suggest that placement which provides large irradiation and low ambient temperature is suitable for the PV system.
一般社団法人日本機械学会の論文

TED-AJ03-286 PREDICTION OF GRID-CONNECTED PHOTOVOLTAIC POWER OUTPUT UNDER NATURAL ENVIRONMENT

スポンサーリンク

概要

著者

関連論文

スポンサーリンク