Globally, large-scale photovoltaic (PV) systems are being installed to achieve maximum power generation efficiency. However, this often results in severe power fluctuations, impacting the stability of the power system. To enhance the power system stability, it is crucial to effectively distribute the times of maximum power output. This can be achieved by making changes to the inclination angle and azimuth of PV modules at power plants. The modeling results indicate a prediction error of 3.65% relative to the actual measurements. To assess system stability based on each installation method, new concepts called demand gradient (DG) and demand gradient average (DGA) were introduced and analyzed using the load pattern in Metropolitan Seoul. The electricity usage was categorized into four divisions: residential (dwelling), public, service, and manufacturing. The analysis revealed that optimal inclination angles for the dwelling, public, service, and manufacturing sectors were 60°, 30°, 30°, and 15°, respectively. As for installation azimuths, the optimal values were 0°, 80°, 160°, and 230°, respectively. When both the inclination angle and azimuth were varied simultaneously, their optimal installation methods included inclination angles of 60°, 60°, 35°, and 50°, and azimuths of 330°, 210°, 230°, and 190°for the dwelling, public, service, and manufacturing sectors, respectively. Among the various scenarios, the most significant improvement in grid stability was observed in the existing south-facing installation of DGA in the dwelling sector, demonstrating a remarkable 77% reduction from 0.1 to 0.057.