Due to the growing energy consumption and importance of the environment protection, demand for research for developing alternative technologies and renewable resources to substitute polluting fossil fuels increases continuously. One of the most important renewable sources is solar energy. Although a small section of solar energy reaches the earth's surface, this small amount is the main source of renewable energy in the earth. Nowadays, the most important applications of solar energy include: solar thermal energy for home uses, industrial and power plants, and direct conversion of sunlight into electricity by photovoltaic panels. The electric power production of photovoltaic panels depends on the direction and angle of incidence of solar radiation on the panels and climatic conditions of the location where the panels are installed. In certain applications, it may not be possible to install a large number of photovoltaic panels in a certain desirable direction. Instead, panels are installed in different directions and slopes due to space constraints. In this case, it is desirable to install the panels as close as possible to the optimum direction. In the present project, the optimum direction of leaves of a tree-shaped photovoltaic systems is specified. Two methods were used to achieve this objective. In the first method the optimal angles for each hour of a day are calculated using a genetic algorithm then the number of panels which would be positioned in each direction is indicated. In the second method the optimum angles and the number of panels for each angle have been calculated simultaneously by using a genetic algorithm. Results show that the best performance of such a tree occurs when all panels of the tree-shaped system are sloped as close as possible to the local latitude, though at different azimuth angles. The effect of seasonal, bi-monthly, and daily adjustments of each panel slope on the photovoltaic-tree performance has been investigated and the results are reported. When the azimuth angel for each panel on the tree is fixed, seasonally adjustment of the panel slope increases the radiation on the panel by only 2.38% while adjusting the slope of the panel bi-monthly leads to 2.88% increase in solar radiation on the panel. Also, optimization of seasonal slope for different azimuth angles, and optimization of hourly azimuth angle for different slopes have been performed. In addition to these, seasonal optimum slope for hourly adjusted azimuth angle has been calculated. Effect of considering constant clearness index and variable clearness index throughout each day is also reported. Considering a constant clearness index throughout each day leads to overestimation of performance improvement of the tree-shaped system. Results show that with seasonal optimum slope, hourly adjustment of azimuth angle leads to the greatest increase in radiation on the panel. This amounts to 24.11% for Varzaneh and 19.12% for Bojnoord. These values are based on the assumption constant clearness index throughout each day. Without assumption this amounts to 18.81% for Varzaneh and 15.35% for Bojnaord. Keywords : Photovoltaic panel; solar energy; optimization; slope and azimuth angles